FIRST

AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE

ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


MONITORING OF LAND DRAINAGE IN EAST NILE
DELTA USING REMOTE SENSING

H. M. Habib
1
, E. A. Zaghloul
2
, A.A. Hassan
3
1. National Authority of Remote Sensing and Space Sciences, NARSS, P.O. Box 1564 Alf
Maskan, Cairo, EGYPT
2. Prof. Director, Water and Engineering Applications Division, NARSS, P.O. Box 1564 Alf
Maskan, Cairo, EGYPT
3. Prof. of Environmental Hydrology, Faculty of Eng., Ain Shams Univ., 1, Saryat st., Abbasia,
Cairo, EGYPT
ABSTRACT
The drainage system of Egypt is divided to four main drainage areas, the eastern
part is considered as the one with the highest volumes flowing to the Mediterranean
Sea. Fluctuations of the water levels and flows through the year as a result of
different agricultural water uses according to the cropping pattern in the area has its
own impact on the main drains and the receiving body of lake Manzala before
discharging to the Mediterranean Sea. The case is always under field investigation
and studies through national agencies to monitor its quality and quantity as a part of
a main strategic plan of reuse of drainage water in the agricultural activities.
Remote sensing data from Landsat TM and other data sources were tested to show
the ability for acquiring valuable information regarding detection of drainage water
variation in the region.
Key words
Remote sensing; Landsat Thematic Mapper; land drainage; satellite-derived hydrological data;
eastern delta; Egypt
INTRODUCTION
Land drainage and water abstraction for agriculture often fundamentally modify the natural
hydrological functioning of adjacent or downstream wetlands from an environmental point of
view, while land drainage flows constitute a vital water sources to be reused, when meet the
limits, for agricultural activities where a lack of fresh water exists.
Although remote sensing has great potential for addressing some of the deficiencies of limited
hydrological data, applied hydrology has not readily embraced remote sensing as a useful source
of data. Engman (1996) explains that this may be because existing techniques and data have only
been sufficient for limited applications. In fact, most of the advances in remote sensing for
459
hydrology have been in areas where monitoring methods were unsatisfactory or limited, or areas
where data were scarce or non-existent. Engman (1996) also argues that, to meet the challenges
of the needs of modern hydrology and to contribute to its future progress, there is a need for (a)
more, better, and different spatial and temporal data that cannot be provided nor maintained by
traditional hydrological instruments for various reasons including costs and feasibility, and (b)
invaluable, long-term data that can be used for model development validation.
The results reported herein are based on using a system for hydrology parameters monitoring
using land observation. The aim is to develop and acquire hydrological data for drainage
catchments and using a remote sensing software for indirectly estimating drain wet width data
and consequently discharges using satellite imagery. Satellite-derived data have the potential for
the establishment of recent (two decade) historical and contemporary hydrological conditions and
for providing calibration data for hydrological models (Shepherd et al., 1999).
MONITORING OF LAND DRAINAGE
Most of the Egyptian drains carry water of relatively good quality that can be reused for
irrigation purposes. The major drains in the delta area discharge their water either directly to the
Mediterranean Sea, to Manzala Lake that is connected to the sea, or to be reused with the flow in
El-Salam canal. Therefore, the accurate determination of their discharges at different locations is
important to achieve a better management of the system.
Unfortunately, at few locations, especially those near the outfall of each drain, it was not
possible to derive such reliable discharge estimates. At such locations the stages were varying
due to the tidal and/or wind effects. In some locations continuous velocity recorders were
installed, and reliable estimates of the discharges were obtained by measuring the velocity at one
point together with the corresponding stage. However, velocity recorders are rather expensive,
their operation is frequently interrupted by the growth of shells, weeds, and/or any bulky floating
object.
Some techniques can be used to provide simple indicators of the discharges for such drains,
and other sites where direct measures are difficult to be carried out, using some simple
procedures from satellite images for management purposes.
REMOTE SENSING FOR WATER
Remote sensing is broadly defined as collecting and interpreting information about an object
or a target without being in physical contact with it (Sabins, 1986). The development and
deployment of manned and unmanned earth satellites since the sixties provided images of the
earth. Remote sensing with its wide area spatial coverage and synoptic view provides an
opportunity to assess the environmental changes of large areas and thus provides information that
is not readily available by other means.
There are numerous existing sensors that are suitable or providing invaluable data for
validation of hydrological models and for monitoring hydrological conditions. These sensors
vary mainly in (a) the frequency of the observations, which can vary from one over a few days to
one every 16 days (e.g. Landsat TM) or more, depending on the satellite-sensor and the orbit, and
(b) spatial resolution, which can range from 1 m (e.g. IKONOS) to 30 m (e.g. Landsat TM) or
more, depending on the sensor.
Space-borne radar altimetry has also been used to directly determine stage variations in large
lakes (Birkett, 1995) and large rivers (Birkett, 1998), such as those in the Amazon basin.
More recently, Alsdorf et al. (2000) improved the resolution of these remotely sensed
measurements using interferometric processing of Multi-temporal synthetic aperture radar (SAR)
data from the SIR-C (spaceborne imaging radar-C) mission to provide variations in floodplain
460
water level response to changing river discharge within an error of the order of 0.1 m. Radar
altimetry shows great promise for providing enhanced spatial monitoring of river water level
changes, by using new satellite SAR missions such as the European Space Agency's ENVISAT
RA 2 (radar altimeter) instrument.
Techniques using radar imagery offer a major advantage over visible and near- infrared
sensors for applications in hydrology due to their all-weather and day/night capabilities.
However, these approaches still suffer from three main drawbacks when compared with those
based on visible and, near-infrared sensors, such as Landsat TM. Firstly, the interpretation of
SAR imagery is much less straightforward than for the visible/infrared range. In addition, the
presence of wind-induced waves or emergent vegetation can roughen the surface of open water
bodies, making them difficult to discriminate from other non-flooded land surface types when
using single frequency and polarization SAR data. Finally, whereas the archives of visible and
near-infrared sensors are almost twenty years long, most radar archives only date back as far as
the late 1990s (Cudlip et al. 1990).
Gupta & Banerji (1985) used Landsat MSS-derived data (reservoir areas) to monitor water
volumes in water bodies in terrains of known topography. Two shortcomings of Landsat MSS
imagery compared to Landsat TM are the coarser spatial resolution (80 m pixels compared to a
Landsat TM pixel of 30 m) and the smaller number of spectral bands. Landsat MSS data have
only four spectral bands, whereas TM imagery has seven, Table 1, covering the visible, near-
infrared, mid-infrared and thermal infrared of the electromagnetic spectrum. It therefore provides
extra information in the mid-infrared and thermal infrared bands. This multi-spectral nature of
sensors such as Landsat TM provides an additional advantage over radar imagery. Of the seven
spectral bands provided by Landsat TM, three (bands 4, 5 and 7, which have, respectively,
wavelengths equal to 0.76-0.9 m, 1.55-1.75 m and 2.08- 2.35 m) are particularly sensitive to
the presence of water.
However, with regards to remote sensing of water using Landsat TM data, most of the
published studies are related to mapping the extent and frequency of river inundation. For
example, Pope et al. (1992) used Landsat TM together with airborne synthetic aperture radar data
to successfully identify and map the intermittent flood extent of meandering systems of imperfect
surface drainage in an area north of Nairobi, Kenya.
SATELLITE-DERIVED EFFECTIVE WET WIDTH
SHYLOC software is a tool that is being used to analyze satellite images to derive information
and create solutions for a specific problem by using definite criteria to integrate remotely sensed
data with other spatial and tabular data for analysis and estimation of streams surface widths as
indication of flows.
The main step is to calculate the DN values, which is gathered during satellite sweeps, the
onboard satellite sensors measure the intensity of the electromagnetic energy reflected or emitted
from the earth's surface in definite range of electromagnetic waves called bands, and then it is
stored on broad the satellite in form of numerical values (digital numbers or DNs), which are
known as raw data. These DNs are stored on magnetic tapes and then transferred to computers
for manipulation by specific software packages designed to achieve this object. Image processing
and enhancement are basically ways to change and alter the original raw data to bring out visual
details and multiple features of interest that were poorly expressed in the original data.
Shepherd et al. (1999) are amongst the first to exploit the use of the partial pixel approach,
which assumes that ditch-carrying image pixels consist of only two components; water and
homogeneous land cover, (which can be grassland, bare soil or cultivated land), to measure
461
surface water area using satellite imagery. Al-Khudltairy et al., (2001a), studied the partial pixel
approach, which forms the basis of the SHYLOC software by using the ditch index to estimate
total water surface area and thereafter dividing satellite-derived estimates of water surface areas
by satellite-derived ditch lengths to deduce effective wet ditch widths at various spatial scales
ranging from sections of drains to an entire drain system.
The SHYLOC software estimates the proportion of water and non-water features creating the
mixed reflectance from drain-carrying pixels, and converts them into appropriate surface areas. It
is worthwhile noting that the satellite-derived wet drain width is an effective value due to the
spatial resolution (30 m) of the Landsat TM images used in this study. In other words, estimates
are not being made of actual water surface area, drain length nor wet drain width from Landsat
imagery. Instead, the satellite-derived data are used to develop satellite-derived relationships
between such data and available data.
METHODOLOGY
Study data
Data required to complete this study vary from multi-temporal digital satellite data to hard
copy maps and collected available flow data. Satellite data required to fulfill the needs to
complete this study should be of high spatial resolution to allow identification of the various
features of the area, and of high spectral sensitivity to enable discrimination of the major classes
represented in the area of study.
Five multi-spectral Landsat TM images, acquired for the area between J une 2000 and
December 2000, were used to evaluate the relationships between satellite-derived effective "wet"
drain widths and available data of flows. Only two Landsat TM images were used in this study
because they were cloudily free during the period that makes them more available for analysis
processes, the first one is for J uly 2000 and the second one is for November 2000.
Data processing
Image processing and enhancement are basically ways to change and alter the original satellite
raw data to bring out visual details. There is a diversity of image processing and enhancement
techniques that can be applied in environmental studies. It is common that an image go through a
series of such techniques in order to obtain a final product that has the necessary details intended
for a specific application.
The image processing was carried out by using ERDAS-Imagine, which is a raster geographic
information and image processing software ERDAS is a popular system because it offers a set of
powerful user friendly programs which can be based on personal computers, or can be integrated
into an intelligent workstation, or utilize the power of a mini or mainframe computer. The
ERDAS system is capable of performing most of the commonly known advanced techniques in
image processing, such as image classification, filtering, image enhancement and image
rectification. While digitizing the drain course was done using ArcView (a GIS software) through
geocoding of geographic data into a computer as vector data with every point has which X and Y
coordinate of the drain selected segment.
All the images underwent geometric corrections using the ERDAS-Imagine software package.
The images were geometrically corrected.
Enhancement is the modification of an image to alter its impact on the viewer. Generally,
enhancement distorts the original digital values. There are many types of remote sensing
enhancements that are used for different applications like spectral enhancement, linear contrast
enhancement, non-linear stretch enhancements and filtering. Nonlinear contrast stretch
(histogram equalization) is the one used in this study.
462
Application of SHYLOC to Bahr Hadous Drain
Bahr Hadous drain Figure 1 is the second largest drain, in the eastern delta drainage
catchments, according to the carried flow. It receives a great importance due to its contribution,
as a surface water source, for El Salam canal that delivers water from the Nile delta, crossing
under Suez Canal, to the new reclaimed area in Sinai. The drain has seasonal water flow variation
according to the cropping pattern in the served area. The drain shows an average flow of 22.5
Million m
3
for J uly 2000, while it reached up to 63 Million m
3
for November 2000.
The SHYLOC software uses the digitized data of the drain and the Landsat image for both
J uly and November 2000 to produce an effective wet width of the drain during those two dates.
RESULTS AND CONCLUSION
By executing the software the data of J uly 200 produced an effective wet width of 39.5 m,
while those of November 2000 showed an effective width of 46.87. The temporal changes in
satellite-derived effective wet drain widths are in agreement with flow data over time. The wet-
derived width is an estimation of the surface water width on that time and is not the actual width,
a statistical relation to be established for some measurements with the actual width to have a
permanent relation between satellite derived one and filed data.
These results proved that the partial pixel method could be used for providing information
about some hydrological parameters such as surface water widths that could be correlated with
field measurement data.
The method was used in area where historical data was available, however it could also be
used for areas where routinely monitoring are not exist or areas with difficulties for management
activities.
Table 1 Characteristics of TM sensor

Bands Wavelength Resolution
(m) m

Band 1 0.45-0.52 30
Band 2 0.52-0.60 30
Band 3 0.63-0.69 30
Band 4 0.76-0.90 30
Band 5 1.55-1.75 30
Band 6 10.40-12.50 120
Band 7 2.08-2.35 30

REFERNCES
AI-Khudhairy, D. H. A., Thompson, J . R., Gavin, H. & Hamm, N. A. (1999) Hydrological
modelling of a drained grazing marsh under agricultura1land use and the simulation of
restoration management scenarios. Hydrol. Sci. J. 44(6),943-971.
Al-Khudhairy, D. H. A., Hoffmann, V. & Leemhuis, C. (2001a) SHYLOC User Manual, Version
2.001, BUR 19745 EN. European Commission, Italy.
Al-Khudhairy, D. H. A., Leemhuis, C., Hoffmann, V., Calaon R., Shepherd, I. M., Thompson, J.
R., Gavin, H., Gasca- Tucker, D., Refstrup Sorenson, H., Refsgaard, A., Bilas, G., Zalidis,
G. & Papadimos, D. (200Ib) Innovative technologies for scientific wetland management,
onservation and restoration. In: Remote Sensing and Hydrology 2000 (ed. by M. Owe, K.
Brubaker, J . Ritchie & A. Rango) (Ptoc. Santa Fe Symp., April 2000), 491--494.
Bahr Hadous Drain
N
Figure 1 Landsat TM November 2000
463
Alsdorf, D. E., Melack, J . M., Dunne, T., Mertes, L. A. K., Hess, L. L. & Smith, L. C. (2000)
Interferometric radar measurements of water level changes on the Amazon flood plain.
Nature 404 (6), 174-177.
Birkett, C. M. (1995) The contribution of TOPEX/POSEIDON to the global monitoring of
climatically sensitive lakes. J. Geophys. Res. 100(25), 179-204.
Birkett, C. M. (1998) The contribution of the TOPEX (NP A) radar altimeter to the global
monitoring of large rivers and wetlands. Wat. Resour. Res. 34 (5), 1223-1240.
Cudlip, W., Ridley, J .K., and Rapley, C. G., (1990) The use of satellite radar altimetry for
monitoring wetlands. In Proceedings 16
th
Annual Conference of the Remote Sensing
Society. Remote Sensing and Global Change. 19-21 September 1990, Swansea, Nottingham,
UK, pp .207-216
Engman E. T. (1996) Remote sensing applications to hydrology: future impact. Hydrol. Sci. J.
41(4), 637-M8.
Engman, E. T. & Gurney, R. J . (1991) Remote Sensing in Hydrology. Chapman and Hall,
London, UK.
Gavin, H. (2000) The hydrology of the Elmley Marshes, North Kent, UK. Pill Thesis, University
of London, UK.
Gupta, R. P. & Banerji, S. (1985) Monitoring of reservoir volume using Landsat data. J. Hydrol.
77, 159-170.
Helwa, F., (1995) Environmental Impact of Drainage Water Reuse Projects on Northern lakes of
Egypt, Proc. Regional Conference & International Symposium on Environmental
hydrology, 10-12 October, 1995, Cairo, Egypt, 398-402.
Hollis, G. E. & Thompson, J . R. (1998) Hydrological data for wetland management. C1WEMI2,
9-17.
J ensen, J .R.: Ramsey, E.W.; Mackey, H.E.; Christensen, E.L. and Shartiz, R.R. (1987) Inland
wetland change detection using aircraft MSS. J ournal of Photogrametric Engineering and
Remote Sensing, Vol. 53, No. %, pp. 521-529.
Mather, P. M. (1999) Computer Processing of Remotely-Sensed 1mages. J ohn Wiley & Sons,
Chichester, West Sussex, UK.
Pope, K. 0., Sheffner, E. J ., Linthicum, K. J ., Bailey, C. L., Logan, T. M., Kasichke, E. S.,
Baimey, K., Njogu, A. R. & Roberts, C. R. (1992) Identification of Central Kenyan rift
valley fever versus vector habitats with Landsat TM and evaluation of their flooding status
with airborne imaging radar. Remote. Sens. Environ. 40, 185-196. Research Systems (1999)
ENVI User's Guide, Version 3.2. RS Boulder, Colorado, USA.
Sabins, F. F. (1986) Remote sensing principals and interpretation. Second edition, W:H:
Freeman and company, New York.
Shepherd, I. M., Al-Khudhairy, D., Kaiser, C., Thompson, J . R., Zalidis, G., Hadjiakiannakis, S.
& Refsgaard, A. (1999) The use of "SHYLOC" for wetland restoration and management.
In: Restoring Wetland Functions. (ed. by N. Kontos, V. Takavakoglou & S.
Chatzigiannakis) (Proc. Tech. Meeting, Thessaloniki, Greece, 15-17 March 1998), 40-53.
Tech. Bull. Greek Biotope Wetland Center, Thermi, Greece. (A MedWet publication).
Whitelaw, A., howes, S., Fletcher, P., and Rast, M., (1994), Remote sensing applications in
hydrological modeling. SPIE (International Society for Optical Engineering) Proceedings,
2314, 618-627.
Zaghloul, E. A., ( 2002) Management System of Water Resources in Northeastern Nile Delta,
Internal report , NARSS, Cairo, Egypt
464

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005










Development of River Nile Water Quality Information
System

Noha Donia
1
and Dr. Hanan Farag
2

1. ABSTRACT
As the pollution problems are getting worse and the pressure on resources
increases, transnational issues will multiply. So, the quality of the Nile
discharges to the Mediterranean Sea through the main two branches
(Damietta and Rossetta) should be managed through management
information system. To investigate and control the pollution problem,
diverse amount of information is required to arrange in form enables to store
and retrieve the data. Therefore, River Nile water quality information
system is designed and implemented as a tool for identification of the most
common and important environmental problems of Nile river and
determination of its risk factors. An integrated system for water quality data
has been developed. The system includes the data about the water quality of
the river, the pollution sources, the historical measurements, the water
quality standards. The output of the system includes the calculation of the
water quality index and the comparison with standards. Some statistical
analysis has been conducted within the system. The aggregated information
obtained from the system will be useful in preparing periodic reports to
decision making. Graphical reports are also constructed for several purposes
education, training, evaluation, and planning of environmental research.

2. INTRODUCTION

No doubt that now is the era of information as the world has become a small village.
Nearly most of the things can be done electronically but the water quality data still relies
465
on pens, papers and post and often several staff to book a single outpatient appointment,
the manual methods have been incomplete, costly and slow. The use of computer based
automation to support, supplement or replace all manual methods which reduce the cost
and improve data management.

Creating a data model that accurately describes the structure of a set of data is a precursor
to any correctly designed database (Dun and Bradstreet, 1995). Data modeling is the
formal process of analyzing and reducing descriptions of information into separate data
components, establishing the nature and direction of relationships between those
components, and thereby building a structure for the data that automatically enforces the
rules needed to maintain data integrity. A logical data model recognizes change as part of
data management and provides a generic structure that permits later extensions without
affecting the validity of data already in the database. A data model also is used to create
and maintain documentation of all elements in the database, and thus provides a common
language and reference for all users. In addition, a data model is used as a template for
implementation of a physical database design and provides a guarantee that data entered
into the database meet a predetermined level of detail and accuracy.

Water quality data is currently stored as data set or an ASCII delimited text file
containing ambient water quality monitoring data associated with samples collected
during a particular month or entire year. This type of data management system is
inefficient both in terms of its structure and its accessibility to users. Each agencys data
are stored as individual monthly data sets with a predefined horizontal structure that
precludes the entry of additional parameters and makes quality assurance procedures
difficult to execute. Furthermore, there is a significant amount of redundant data. Each
record in every data set contains information that pertains solely to the water quality
station, not to the samples and resultant parameter values. Because of these weaknesses,
it is essential to develop a relational database management system (RDBMS) for water
quality data.

The water quality information System was developed to store water quality information.
The developed system is a computer based system that is capable of storing retrieving,
storing, processing, linking, distributing, and reproducing immediately on demand all
needed water quality information. The purpose of this paper is to describe the Rosetta
database by describing the data model and its physical implementation in MS Access.

3. REVIEW OF ENVIRONMENTAL INFORMATION SYSTEMS
Through the inventory of environmental information systems and applications using the
state of the art information technology, Oracle Developer 6i and Oracle 8i , two systems
were developed in this field. One of the systems was developed by Egyptian
Environmental Affairs Agency (EEAA). This system was divided to two modules one of
them to receive industrial data related to facilities and industrial activities either for the
facility level or the detailed process level regarding quantity and quality. The second
module was developed to receive data and information for environmental current
situation of urban areas. The system is designed to manage and analyze industrial data
(EEIS, 2001).
466
Another system includes information on the observation wells within the Nubian
Sandstone Aquifer System. The groundwater database such as water level, draw downs,
stratigraphy, hydraulic parameters water quality parameters. Developed Nubian Aquifer
Regional Information System (NARIS) as an integrated regional information system
among the countries sharing the aquifer thus ensuring the sharing, exchange and flow of
information. The database was developed using the developed information technology,
i.e. Oracle developer 6i and Oracle8i. (CEDARE, 2001).
An information system was developed (Farag,2004). The system is constructed database
for all the surface stream type. Oracle software is the software used for developing the
database and graphical user interface (GUI). River Nile is one of the streams taking into
account, but the system is generalized which it is make the analysis of Nile water quality
status is not specified. Table (1) is shown an inventory was done for environmental and
water quality databases are existing in Ministry of Water Resources and Irrigation and
National Water Research Center's institutes.


Table 1 Information Systems and Databases of MWRI and NWRC Institutes
Sector Name Database and Information Systems
Types
Remarks
Planning
Sector, MWRI
GIS database, Nile Forecasting
database and pollution sources
database
For pollution sources data base
not linked with the water
streams.
The Irrigation
Sector (IS),
MWRI
Concerning mainly data on water
levels in irrigation canals
Did not develop any GIS
system. Most of data exists in
simple spreadsheet format
The Nile Water
Affairs Sector

Data concerning mainly with
discharged records collected by
stations on the main Nile and its
tributaries outside Egypt.
The data are stored as hard
copy and spread sheets without
a well-designed irrigation
database and information
system.
The
Management
Information
Center
Developing a DSSs by analyzing vital
information and providing this
information to the decision maker.

NWRC Developing the Egyptian water map.
This will be capable of displaying
water quantity and quality.

467
The Nile
Research
Institute
Developed database system using
FoxPro as DBMS software.
This database concerns mainly
with water quantity data. The
water quality data is stored in
spreadsheets. Developing
database concerning with the
water quality and quality is
under construction.
The Drainage
Research
Institute
The database for the water quality of
the drains in delta is constructed using
Access DBMS software.

The Research
Institute for
Groundwater
(RIGW)
Water quality database of the
monitored wells is constructed using
SQL as DBMS and Visual Basic as the
development language.

Strategic
Research Unit
(SRU)
The Egypt system database includes
six main modules Agriculture,
Economics, Environment,
Groundwater, Demography and
Irrigation. The database is constructed
using Access software and linking to
GIS software through Advanced
Visual System (AVS)
Egypt system is considered as
information system.
(Source: National Water Quality Monitoring and Availability, 2001)
4. STUDY AREA DESCRIPTION
The study region is located in Rosetta Branch one of the two Nile Branches. The Rosetta
Branch extends about 240 km long the Delta Barrage at kilometer 954 to the north
forming the West Delta Branch, the average width is about 200 m. Rosetta Branch
constitutes an important waterway for the River Nile, see figure (1). In the Rosetta
Branch there are two sources of pollution which potentially affect and deteriorate its
quality of water (Farag, 1997).

The first source is the agricultural wastewater that comes from five agricultural drains
located along the branch (Rahawy drain, Sabal drain, Tala drain, South Tahrir drain and
Zaweit El-Bahr drain). The quantities and characteristics of wastewater from agricultural
lands are highly variable. The most important pollutants found in runoff from agricultural
areas are sediments, plant nutrients, crop residues, inorganic salts and minerals, chemical
fertilizers and pesticides. In addition to that, these agricultural drains also receive
domestic wastes from 55 towns and villages distributed along the branch. It is estimated
that only 5% of the villages population have access to sanitary waste disposal facilities
and it has been observed that these drains receive domestic wastewater from villages.

468
The second source is located at Kafr El-Zayat city industrial zone. Kafr El-Zayat city is
located on a sharp curved bend of Rosetta Branch between Km 120.1 and Km 123.8
measured from Delta Barrage. The Salt and Soda factory and the Maliya company are
discharging their wastewater directly and continuously to this part of the river. Industrial
wastewater are highly variable in both quality and quantity, depending mainly on the
product produced. These may include toxic metals, chemicals, organic materials, and
biological contaminants.

The high organic and solid contents of the above mentioned wastes significantly affected
the water quality in Rosetta Branch as well as the groundwater aquifers in Giza. No water
is released in this Branch except during the High Aswan Dam closure period, which is
considered an important factor in increasing the pollution of this branch. Therefore, an
adequate pollution control of the effluents in order to improve the water quality of the
branch (donia, 2002).

The scope of work for developing the information system consists of two modules. First
module handle the input and output of Water Quality data which have been collected
from different sources. The objective of developing a water quality database for the
Rosetta River Investigation was to compile information on specific parameters that define
the nature of the stream and river environment: water chemistry data, and the physical
parameters such as temperature, DO, and pH. Streamflow data are included as an integral
part to interpretation of constituents reported in units of concentration. Data related to
biotic measures were not compiled; however, the database could be expanded to include
those parameters in the future. The module was built using Relational Database
technology which enable different types of data and different operations to handle it.
Second module was an interface module to other systems that is working in Water
Quality data such as GIS modules and this will enable producing Water Quality maps.

















469






























Figure 1 Schematic Diagram of Rosetta Branch and the sampling sites

5. THE WATER QUALITY DATABASE STRUCTURE

Development of the water quality database began with a series of discussions about water
quality terms and the logical groupings and relationships between information elements.
The major data objects were identified and ways were developed to represent them in an
information system. When a general water quality concept could be realized in more than
one data structure, the alternatives were presented, the implications of each were
discussed, and the contextually accurate structure determined and incorporated into the
model. Exceptions to data generalizations and required/optional data elements were
identified, and the emerging model was evaluated and revised. This process continued in
an iterative fashion until all the information elements implied in the water quality
database requirements were set in the data structure, and the factual statements
470
represented by the relationships between all data objects were approved as accurate by
the workgroup. Because water quality specialists participated in the database design
phase, they ensured that it contained those specific, categorical, and classification
parameters needed to store, describe, validate, list, summarize, and aggregate their water-
quality information, (Steven Tessler, 2002)
Building a database and communicating its features are enhanced by establishing certain
rules of construction and standards for uniformity. Design principles provide
standardized rules for creating and assembling the database and facilitate communication
about how different elements of the database function together. The water quality
database model was built using (MS Access) as shown in figures (2, 3) The physical
implementation of the database was created in MS Access by providing table and field
identities, discrete properties (field type, size, default values), relationships, and all key
field constraints that ensure data integrity. The following sections describe the method
used for the visualization of database objects, goals for table construction and
establishing formal relationships, the rationale for extensive use of domain tables, and
naming standards applied to tables and fields ,(USEPA, 1998).
5.1 Conceptual Design
Monitoring and testing results do not stand alone: the location, time, methodology, and
other information also must be documented. The purpose of a database is to store
information in a useful way. The developed database structure provides many avenues for
complete and detailed data documentation, also all historical data are included in the
database. These historical data may be used to evaluate trends or to supplement analysis
when present data are not sufficient for evaluation.

5.2 The Relational Database
.
A relational database is a collection of formally described tables that can be edited or
expanded in many different ways without having to reorganize the database tables. A new
table can be added to the database without modifying all existing tables. Data are entered
into tables based on subject and related by a key that makes the records within any given
table unique. The columns of a table are called fields; the rows are called records.
471


Figure 2. The Developed Water Quality Information System Main Screen



Figure 3. The Sampling Site Entry Screen







472
Information about each station (sample site) is recorded in the table TBLStation. Each
record (row) contains information about one station. Fields include station name, a
unique identification number, location description, latitude, longitude, etc. The table
TBLSample contains information about samples collected and has a record for each
sample; the fields include a unique sample number, date, time, method, and unique
station number. These unique numbers or keys provide the link from one table to the
next. Information about the station is linked to each sample taken at that station without
repeating the station information for each sample. In the same way, each sample is related
to the results table TBLResults by a sample number that is uniquely assigned when the
sample and results records are added. Five-digit parameter codes are used to identify
individual constituents analyzed in the sample. The parameter table
TBLParameter_Codes then may be combined with the results table to view the full name
for the parameter using the parameter code.

The process of removing redundant data from a relational database by separating
information into smaller tables is called normalization. A normalized database generally
improves performance, lowers storage requirements, and makes it easier to change the
application to add new features. A data model is a conceptual representation of data
structures required by a database. Data structures include data objects, associations
between data objects, and rules that govern operations on the objects. The data model
focuses on required data and how it should be organized rather than on what operations
will be performed. A data model is independent of hardware or software constraints.
Rather than try to represent the data as a database would see it, the data model focuses on
representing the data as the user sees it in the real world. It serves as a bridge between the
concepts that make up real-world events and processes, and the physical representation of
those concepts in a database, (Eastman, Toledano, and Hutchinson, 1994).

5.3 Data Model Description

The developed information system data model describes water quality monitoring and
data as a complex but related process. Figure 3 shows the conceptual representation of
the data model implemented in the developed database. Monitoring stations are located
along rivers, streams, and lakes. Selected stations are sampled as part of a specific
monitoring project. Individual samples are collected and shipped to a laboratory for
analysis of specified parameters. The results of the analysis are the numerical values of
each parameter analyzed. Results also include the values of field-measured parameters,
such as temperature and stream flow. Each arrow in the diagram designates a separate
table in the developed database. Individual tables are related through unique identifiers.
As described above, a sample is identified by a sample number, and attributes include
information about the monitoring station and monitoring project in addition to sample
descriptors such as sampling date, sampling depth, medium, etc. The sample number is
included in a table of laboratory and field data results linking the values to a particular
sample.
.


473

Figure 3. Schematic Representation of The Developed Database Model

Rivers as spatial features are part of a geographical coverage, and the link to stations is
established by spatial location. Laboratories are not included at this stage because the
information is often unknown and not readily available from original data sources. The
table TBLIDLocations is part of the database, but it is not included in any of these
categories. It describes the source from which samples were taken (river, drains and
factories). The information also is used for the database maintenance and batch data
import. For discussion purposes, actual table names in the developed database are
italicized and actual field names are within quotation marks.

A station is described in the table TblStation_Information. Station locations may be
displayed in a Geographical Information System (GIS) environment using latitude and
longitude, which were determined for each station from the original data source or from
the station description and 1:100,000 scale topographic maps. In addition, the station
location in the stream network is established by river section name. Other attributes
include various station codes: Station_ID represents a unique identifier within the
database. Other fields describe the stations attributes. For example, Station_Type
identifies by a code whether the station is located on a river, lake, wetland, canal, etc. The
description of the code used in Station_Type is given in a lookup table,
TBLSTation_Type, which provides the stations Primary_type and Secondary_type.
The lookup table also indicates whether the station is located on a natural or an artificial
water body. A sample is described in the table TBLSample by the station where it was
taken, sampling date and time, sampling depth and a monitoring project under which it
was collected. Parameter codes are defined in the table TBLParameter_Codes, which
includes a verbal description, both full and abbreviated, and reporting units. Additional
related tables associate parameters with a parameter group. the main parameter group
(basic inorganic, nutrients, metals, organics, etc.); and the third number indicates the
constituent subgroup for example, nitrogen in the nutrients group, or pesticides in the
organics group). The entry forms for these data are illustrated in figures (4 and 5).
474

Figure 4. The Sampling Parameters Data Entry Screen

Figure 5. The Standard of The Water Quality Parameters Screen





475
A project is described in the table TBLProjects_Programs, which includes a project name
or title for which monitoring was performed, a code for the monitoring organization,
project study area, project purpose, beginning and ending dates, and contact information.
The organization is described by its full and abbreviated names, and category. The
address, contact person and phone number, and the organization

5. 4 Statistical Module

Statistical analysis is one of the analysis tool applied in the Nile water quality information
system. The system interface enables accessing statistical software applications as
statistica 6 software. Possible outliers in data were identified using statistical methods.
Data are first evaluated separately for consistency within an individual sampling site.
Statistical evaluation of individual datasets used the following techniques within the
statistics software:
1. Basic statistics (mean, median, and standard deviation)
2. Probabilistic distribution plot, quantile plot, test for normal or log-normal distributions
3. Time-series plots
4. Scatter plots (change of parameter with flow etc.)
5. Statistical tests for suspected outliers

Multivariate analysis technique which is defined as the analysis of multiple variables in a
single relationship or a set of relationships can be applied on the stored data. Another
definition of the multivariate measurements is the use of two or more variables as
indicators of a single composite measure. The multivariate technique was used for
determining the indicator parameters based on the pollution source type.

From the historical data that is stored in the information system database, the user selects
the period of work the type of pollution source and presses the statistical analysis button
the statistica 6. software environment, multivariate analysis (Factor analysis) is ready
through running the programmed multivariate script. Changing the number of the variate
group and the loading value is allowed based on the result of the model. The drainage
waste which is coming from the agricultural drains discharged to the Nile is selected for
applying the model. Two and three principal component is examined for 100 records of
25 agricultural out fall drains and 16 measured variables. The results of the three
principal components are more reliable. The first variate consists of 7 variables which are
more significance. The second one consists of tree and the last variate consists of 2
parameters. Figure (6) shows one of the analysis output , the three principal factors and
the significance of the parameters in each principal component.




476


Figure 6. Principal Component Analysis OutPut (Statistics Module)
6. IMPLEMENTATION AND NAVIGATION

Two queries have been designed and included with the developed database. These
queries are recommended for users with some experience with relational databases and
Microsoft Access and may be used as examples for construction of additional queries.
Advanced users are encouraged to build custom queries.

6.1 Comparison With Standards at Specific Location (Query 1)

The user can get information about the locations exceeding the standard (national and
international) for all groups of water quality data (river, drain discharge and industrial
effluent). The comparison was implemented through a module built within the system.
The output of the query is illustrated in figure (7).

6.2 Historical Data of a Parameter at Specific Location (Query 2)

The user can get information about the trend of pollution in a certain location using the
historical data at these locations. His module can be implemented for all groups of water
quality data (river, drain discharge and industrial effluent). The output of the query is
illustrated in figure (8).





477

Figure7. The Comparison With the Standard (Query 1)

Figure 8. The Site Water Quality Historical Data (Query 2)



478
7. CONCLUSION
The Nile water quality information system database was successfully developed and has
the following features:

The Database is constructed from a conveyance-based data model rather than a
site based data model, thus promoting and encouraging a water network approach
to water-quality data storage and investigation.
Nile information system handles both single-user and aggregate-user water-use
data in a single data model.
The Nile database system is implemented as a stand-alone (and portable) database
in MicrosoftAccess (MS Access) and therefore accessible to a large number of
potential users. The design flexability is enabled to customize for other relational
database system.
Statistical module is used for developing a reliable water quality indicators based
on pollution source.
The database is fully open to customization and extension and can be made
available via the World Wide Web to anyone with access to Internet browser
software.

8. REFERENCES

CEDARE, 2001, Center for Environment and Development for the Arab Region and
Europe (CEDARE). Technical report Nubian Aquifer Regional Information
System), (NARIS), version1,2001.
Donia, N. S., 2002, Water Quality Control of Rosetta Branch, Unpublished Ph.DThesis,
Institute of Environmental Studies and Researches, Ain Shams University.
Dun and Bradstreet, 1995, Dun and Bradstreet business information database: Murray
Mill, N.J .
Eastman, R., J. Toledano, and C. Hutchinson, 1994. The Malawi National
Environmental Information System. Submitted to USAID, Lilongwe.
EEIS 2001, Egyptian Environmental Affairs Agency (EEAA), Egyptian
Environmental Information System project (EEIS), Industrial Pollution Information
System Briefing Document, Version 1.1, J anuary 2001.
Farag, H. A., 1997. "Design of Water Quality Information System", Unpublished M.Sc.
Thesis, Faculty of Engineering, Ain Shams University.
Farag, H. A., 2004. "A Methodology Applied to Water Quality Management for Surface
Water" Unpublished PhD. Thesis, Faculty of Engineering, Ain Shams University.
Steven Tessler, 2002 Data Model and Relational Database Design for the New England
Water-Use Data System (NEWUDS), Geological Survey Open-File Report 01-359, 70 p.,
U.S. Department of the Interior U.S. Geological Survey, Northborough, Massachusetts,.
USEPA, 1998, Region III, Chesapeake Bay Program Office, Water Quality Database
Design and Data Dictionary RJ O Enterprises, Inc.
479

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


WATERSHED LOADING MODELS AND EXAMPLES
OF RECENT APPLICATIONS

Tanik, A., Gurel, M., Baloch, M.

ITU, Istanbul Technical University, Faculty of Civil Engineering,
Department of Environmental Engineering, 34469, Maslak, Istanbul /Turkey


ABSTRACT

Watershed modelling has become a vital tool in water quality research and management
practices with the rapid advancement in computer and information technologies. Early
watershed models were designed for the estimation of water quantities in engineering
applications such as flood forecasting, urban storm water management and many other water
resources planning activities. Since the early 1990s, an increasing emphasis on the
development of computer interfaces and application of Geographic Information System (GIS)
techniques signified the common applications of various models. Majority of these models are
simulation models for non-point source pollution that can be divided into urban run-off and
rural run-off models. In this paper, the importance of watershed modelling in the field of
integrated watershed management aiming sustainability is emphasized, continued with brief
information on historical progress of modelling approaches. The classification and types of
models is then described, typical urban and rural watershed loading models are referred.
Although watershed models also cover water quality models, the study focuses on only the
loading models. Some loading model applications from various countries selected among
those cited in literature from year 2001 onwards constitute the final section of the paper. The
objective of the study is to put forth the importance of using such models and state their most
recent application areas on selected examples together with the outcomes.

Keywords: non-point source (NPS) pollution, loading models, urban run-off models, rural
run-off models, watershed, watershed models.


INTRODUCTION

Watersheds are regions or areas with natural hydrological boundaries draining to a water
body. They include the surface and groundwater, soils, vegetation, and animals in the
drainage basin, as well as humans and their anthropogenic impacts (Reimold, 1998).The
integration, coordination, and management of human activities in a sustainable manner within
480
the boundaries of a watershed with the basic aim of protecting both the land and water
resources depend on a comprehensive understanding of the components of watersheds and
their interactions. The watershed approach includes the whole urban-rural landscape. In many
areas of the world, the changes in urban and rural areas threaten the entire economic and
social system, and the differences in land-use activities significantly impact the quality of the
land and water resources. The relative stability and function of a watershed is determined by
the rate of water inflow and outflow, pollutants, and activity patterns of the living organisms.
To understand the behaviour of pollutants arising from both point and non-point sources and
water quality issues addressed in watershed management and planning activities necessitates
the use of watershed models. As non-point sources of pollutants are temporally and spatially
uncertain and difficult to analyze, such models particularly aid to better identify and define
them. Therefore, watershed modelling has become a vital tool in water quality research and
management practices with the rapid advancement in computer and information technologies.
Watershed models enable to quantify the impacts of current, possible and planned actions on
pollution loadings and water quality. They are essential tools to address the functions and
conflicts in a watershed. Therefore, they provide an understanding of the processes involved in
especially the non-point source pollution. They can generate various alternatives according to
the specifications of the watershed which may also give rise for the planning activities, and the
impact of human-induced activities on the environment may be easily assessed via
models. Besides, the database used in the model will also be able to answer and clarify many
problems encountered in the watershed by evaluating the available data and compiling them
in an appropriate format.
Early watershed models were designed for the estimation of water quantities in
engineering applications such as flood forecasting, urban storm water management and many
other water resources planning activities such as reservoir design and water supply (Chen,
2001). Nowadays, there appears a variety of models applied for different processes to better
identify the pollution sources and therefore, it becomes important to employ them for enacting
necessary protective measures for the conservation of land and water resources of a
watershed.
In this paper, a brief historical background of watershed modelling will be mentioned
followed by information on classification and types of watershed models. Although watershed
models include both the loading and water quality models, only loading models covering
urban and rural run-off models are mentioned in the paper. The aim of this paper is to outline
some of the leading watershed models and to emphasize on how rapidly this field develops in
parallel to the development of computer and computational technology. Some selected recent
model applications will also be referred.


THE HISTORY OF WATERSHED MODELING

The historical background of watershed modelling can be divided into three stages (Chen,
2001);

1. During the mid and last 1960s, hydraulic computations and conceptual water balance
algorithms on a digital platform had been implemented. The classical and long-lasting
models like SWM, HEC-1 and SWMM laid down the theoretical and technical basis for
constructing conceptual hydrologic models which have become important tools for
watershed management and non-point source pollution control and planning.
2. As modelling techniques became more sophisticated and rapid advancement of personal
computers, numerous watershed modelling system were developed through the 1980s.
481
HSPF, CREAMS, GLEAMS, AGNPS and ANSWERS in USA and SHE and
TOPMODEL in Europe can be given as examples.
3. Since the early 1990s, the third stage has been signified by the increasing emphasis on the
development of computer interfaces and application of GIS techniques. Nowadays, there exist
an extensive literature on watershed modelling and a number of excellent computer programs
available both commercially and in the public domain for a wide range of watershed
management purposes. The field is in a constant state of development, with improvements
continually being made to existing models and new models frequently being introduced.



CLASSIFICATION OF WATERSHED MODELS

There are many categories under which watershed models are classified. Some of them are
listed below (ESCAP-UN, 1997):


The watershed model may be categorized as;
- Simulation Models which are set up to simulate the behaviour of watersheds,
catchments, river basins and various kinds of water reservoir management systems. They
are used to predict the performance and behaviour of the resource systems when various
inputs are in operation.
- Optimization Models which seek to find the best way of manipulating, designing or
operating a water source system, they are essentially decision-making models.
The watershed model may be categorized as;
- Event model which represents an event like a flood that may last from a few minutes to
several days.
- Continuous or sequential model that is operated for a long period like 100 years. These
models are used for predicting long-term catchment yield or long-term behaviour of
water quality parameters.
The watershed model may be;
- Conceptual model based on a set of equations which models the physical, chemical and
biological processes in watershed systems.
- Empirical model formulated from the relationships between system input and system
output. The processes within the system are not taken into consideration.
Model may be categorized as;
- Deterministic model which has a fixed relationship between the inputs and outputs, so
that the re-running of the model with the same input will produce the same outputs.
- Stochastic model that contains some random elements that re-running of the model will
produce variable outputs.
Model may be categorized as;
- Dynamic model that is time-dependent where the inputs and parameters can change with
time.
- Steady-state model where the inputs and coefficients are constant in time.
Model may be categorized as;
- Generic model that can be applicable for a wide range of watersheds.
- Site-specific model that is improved for a specific watershed.
482
TYPES OF MODELS
The models for integrated watershed management can be divided into receiving water
quality models and loading models (Novotny, 2003).
Receiving water models simulate the movement and spread of materials through water
bodies. An overview of some public domain models for conventional pollutants will not be
covered in this paper. It will be limited to introducing loading models.
Loading models simulate and estimate pollution generation at the source and its
movement from the source to the receiving water body. Such models are applied both to
urban and rural (agricultural) areas to determine non-point source (NPS) pollution. With the
achievements in the past decades, it is now much easier to allocate contribution of waste loads
from point sources; however, non-point sources are still a challenge to assess because of the
sophisticated process and mechanism they undergo. Therefore, NPS modelling, as an essential
component of watershed modelling, is a vital tool utilized for estimation of hydrologic
rainfall/ run-off transformation process with associated erosion, pollution build- up and wash-
off and other quality components.
There are numerous watershed models in use; especially developed within the last decade,
making it impossible to outline them all. Therefore, in this section some of the oldest but the
leading NPS models both urban and rural will be briefly introduced. In the last section on
examples of recent applications, studies conducted between years 2001-2005 are referred. It
will be seen that these recent models are not necessarily the leading models explained in this
section.


Urban Models

Most widely known urban models are mentioned below (USEPA, 1991; Novotny, 2003).
DR3-QUAL: It is a version of United States Geological Survey (USGS) Distributed Routing
Rainfall Run-off Model which includes quality simulations. Run-off generation and routing
use the kinematic wave method, and it has a parameter for estimation assistance. Quality
simulation is made by build-up and wash-off functions.
HSPF: the Hydrological Simulation Program-Fortran developed by United States
Environmental Protection Agency (USEPA) is a simulation model for prediction of run-off
from upland watersheds. This model is a development of Standford Watershed Model (SWM)
introduced in 1960s. It is a comprehensive, continuous, distributed model which includes a
variety of hydrological, fluvial, alluvial, chemical and biological processes. It incorporates the
watershed scale Agricultural Runoff Model (ARM) and urban NPS into a basin-scale analysis
framework that includes fate and transport in one dimensional stream channels. It is a large
model and requires considerable effort when applied to a watershed. The model is a part of
the USEPAs BASINS modelling system.
MOUSE: Modelling of Urban Sewers is developed by the Danish Hydraulic Institute in
cooperation with laboratories and private software firms. Model includes modules for
generation of run-off from rainfall, sewer routing and a simple routine that uses the constant
concentration approach.
STORM: Storage, Treatment, Overflow, Run-off Model, developed by the US Army Corps
of Engineers Hydrologic Engineering Centre in 1974, is the first use of continuous simulation in
urban hydrology, but the centre does not provide user support or further development and
maintenance of the program. This model is first applied to the San Francisco Master Plan for
pollution abatement. It utilizes simple run-off coefficient and unit hydrograph methods for
generation of hourly run-off depths from hourly rainfall inputs. The build-up and wash-off
formulations are used for simulation of six pre-specified pollutants.
483
SWMM: Storm Water Management Models is developed by USEPA as a single-event model
developed in 1971. It has been applied to urban hydrologic quantity-quality problems
worldwide. It is maintained and updated continuously. Later versions of SWMM can simulate
backwater, surcharging, pressure flow and looped connections and it also has various options for
quality simulation. SWMM is segmented into Run-off, Transport, Extrans,
Storage/Treatment and Statistical blocks for rainfall run-off, routing and statistical segments.
WALLINGFORD: It is developed by Hydraulic Research Ltd. in Great Britain. It consists of a
cluster of models, which includes run-off generation from rainfall (WASSP), simple and fully
dynamic sewer routing (WALLRUS and SPIDA) and a quality routine (MOSQITO).
WASS-QUAL: It is developed in 1987 by Hydraulic Research Ltd in Great Britain. The
model is a result of a joint research effort of the Water Research Centre and other British
institutions. It can be done either in a continuous or single event mode. It should not be
mistaken with the WASP model of the USEPA.


Rural (Agricultural) Models

Most widely known rural models are mentioned below (USEPA, 1991; Novotny, 2003).
Some of the previously mentioned urban models like HSPF can also be used for modelling of
agricultural watersheds, primarily erosion and movement of particulate pollutants.

AGNPS: Agricultural Non-point Source Pollution Model is developed by the US
Department of Agriculture (USDA), Agricultural Research Service (ARS) to estimate run-off
quality on nutrients and sediments and to compare the effects of various pollution control
practices. The AGNPS model includes the simulation of sediments and nutrients from
agricultural watersheds for a single-storm event or for continuous simulation. The model does
not simulate pesticides. In this model, watersheds are divided into square working areas.
These areas are examined as sub-watersheds individually. The results of the model can be
compared with other watersheds to evaluate the sources of water quality problems and to
investigate possible solutions. It also includes point sources like feedlots, wastewater
treatment plant discharges, stream bank and erosion. In AGNPS, pollutants are routed from
the top of the watershed to the outlet in a series of steps. The pollutant transport portion is
divided into two, which one is the soluble pollutants and the other part performed using
relationships between chemical concentration, sediment yield and run-off volume.
ANSWERS: Areal Non-point Source Watershed Environment Response Simulation is
developed by the Agricultural Department of Purdue University. It is an event-based,
distributed parameter model, which predicts the hydrologic and erosion response of
agricultural watersheds. It is designed to calculate peak flow rates and total surface run-off for
single events. Its application is made by dividing the watershed into a grid of square elements.
The area of this element is between 1 and 4 hectares. The output of one element becomes an
input of the adjacent element. In each element, the model simulates the processes of
interception, infiltration, surface storage, surface flow, subsurface drainage, sediment
drainage, sediment detachment, transport and deposition. Nutrients are simulated by using
correlation between chemical concentration, sediment yield and run-off volume.
ARM: Agricultural Run-off Management Model is a version of the HSPF model which can be
run independently or included in HSPF. It simulates run-off, sediment, pesticides, and
nutrient loadings from surface and sub-surface sources. It requires extensive calibration.
CREAMS: Chemicals, Run-off and Erosion from Agricultural Management Systems is
developed by the US Department of Agriculture (USDA), Agricultural Research Service
(ARS) for the analysis of agricultural best management practices for pollution control. This
model is a field scale model which consists of three separate sub-models of hydrology,
erosion/sedimentation and chemistry connected with fields. Run-off volume, peak flow,
484
evapotranspiration, soil water content and percolation are based on day. Daily sediment and
erosion yield are estimated at the edge of the field. Nutrients and pesticides are simulated,
storm load, average concentration of sediment associated and dissolved chemicals are
determined in the run-off, sediment and percolation through the root zone. By this model, user
defined activities can also be simulated. Aerial spraying, soil incorporation of pesticides,
animal waste management and agricultural best management practices can be evaluated as
user defined activities. The model has the capability of simulating up to 20 quality
components at a time.
GLEAMS: Groundwater Loading Effects of Agricultural Management Systems is
developed and maintained by the US Department of Agriculture (USDA), Agricultural
Research Service (ARS). The three main components in the model are hydrology,
erosion/sediment yield and infiltration. Precipitation is partitioned between surface run-off
and water balance on a daily basis. The soil is divided into layers, with a minimum three and a
maximum of twelve layers with various thickness used for water and pesticide routing. The
input data requirements for CREAMS-GLEAMS simulations are extensive and quite detailed.
PRZM: Pesticide Root Zone Model is developed by USEPA Environmental Research
Laboratory in Athens, Georgia. It is a one-dimensional, dynamic, compartmental model that
can be used to simulate chemical movement in unsaturated zone and below the plant root
zone. The model is divided into hydrology, which is composed of run-off and erosion and
chemical transport. Water movement within the unsaturated zone is simulated by field
capacity, wilting point and saturation water content. Irrigation applications are also used in
the model. Pesticide applications are considered in the chemical transport simulation.
Dissolved, adsorbed and vapour-phase concentrations in the soil are estimated by considering the
processes of pesticide uptake by plants, surface run-off, erosion, decay, volatilization, wash-
off, advection, dispersion and retardation.
PRS: Pesticide Run-off Simulator developed by Computer Sciences Corporation for
USEPA Office of Pesticide and Toxic Substances is based on SWRRB Model. Pesticide
application can be removed by atmospheric loss, wash-off by rainfall and leaching into the
soil. Pesticide yield is divided into a soluble fraction and an adsorbed phase based on an
enrichment ratio. The objective of this model is to simulate pesticide run-off and adsorption
onto the soil in a small agricultural watershed.
SWRRB: Simulator for Water Resources in Rural Basins was developed by US
Department of Agriculture (USDA), Agricultural Research Service (ARS) for evaluation of
basin-scale water quality. It is developed by modifying the CREAMS daily rainfall hydrology
model for application to large, complex, rural basins for simulation of weather, hydrology,
crop growth, sedimentation, nutrients and pesticide movement. The pollutant transport portion is
subdivided into two; one part handling soluble pollutants and the other part handling
sediment attached pollutants. Prediction of nutrient yields from rural basins is adopted from
CREAMS. The amount of pesticide reaching the ground is based on a pesticide application
efficiency factor. Pesticide wash-off is based on threshold rainfall amount and calculated by
empirical equations. Pesticide decay from the plants and soil are predicted by using
exponential functions based on the decay constant in the soil and half-life of pesticide on
residue.
TEHM: the Terrestrial Ecology and Hydrology Model describes soil-plant water fluxes,
interception, infiltration and storm and groundwater flow. Hydrologic part of the model is
from the Wisconsin Hydrologic Transport Model (WHTM). It is a modification of Standford
Watershed Model (SWM) and simulates soluble chemical movement, litter and vegetation
interception of the chemical, and erosion of the sorbed chemicals, chemical degradation in
soil and litter, and sorption in top layers of the soil.
485
UTM-TOX: Unified Transport Model for Toxic Materials developed by the Oak Ridge
National Laboratory for the USEPA, Office of Pesticides and Toxic Substances. UTM-TOX is
a multimedia model that combines hydrologic, atmospheric and sediment transport. It
provides a detailed simulation of soilplant processes. The Atmospheric Transport Model
(ATM) portion of UTM-TOX is a Gaussian plume model that calculates dispersion of
pollutants emitted from point, area or line sources.
WEPP: Water Erosion Prediction Project is a new generation water erosion model
developed by the ARSs National Erosion Research Laboratory. It is a continuous simulation
model, although it can be run on a single-storm basis. By continuous simulation of the model,
the important processes of erosion prediction are realized as a function of time. The output of
the continuous simulation is the time integrated estimates of erosion.

RECENT APPLICATIONS OF VARIOUS NON-POINT SOURCE (NPS)
POLLUTION MODELS

Some selected case studies from different countries on loading model applications will be
briefly referred in this section.

GLEAMS model was used to predict the soil and nutrient losses from an agricultural field in
Eastern Thailand. In this study, results show that GLEAMS is capable of producing
reasonable prediction of soil loss, run-off nitrogen and phosphorus losses and their
concentration in the crop root zone within the agreed variability of measured data (Deb et al.,
2001).
A water quality component was developed for WATFLOOD (a flood forecast
hydrological model) to deal with sediment and nutrient transport. The model uses a
distributed group response unit approach for water quantity and quality modeling. Runoff,
sediment yield and soluble nutrient concentrations are calculated separately for each land
cover class, weighted by area and then routed downstream. The model was integrated with the
GIS technology and applied to Duffin Creek Watershed that drains into Lake Ontario, 10 km
east of Metropolitan Toronto, Canada. The model is calibrated for the hydrologic response
and validated for the water quality component. In almost all the cases it was found that the
predicted sediment and nutrient yields agreed reasonably well with the measured data (Leon et
al., 2002).
An integrated model approach which is a combination of general hydrogeological
residence time model WEKU and nitrogen balance model to quantify the nitrogen loads
entering the surface waters via the groundwater path was applied to the German part of the
Elbe River having a catchment size of 150.000 km
2
. (Wendland et al., 2002) That model is
validated by the results of the MONERIS (Modelling Nutrient Emissions in the River
Systems) (Behrendt et al., 1999). At the end of the study, it is understood that the groundwater
residence time and the amount of denitrified nitrogen can vary strongly with the site in which
the nitrogen enters the saturated zone.
For the estimation of storm water quality and quantity in urban areas, the ILLUDAS
(Illinois Urban Drainage Area Simulator) was modified and added to the water quality model.
As a result, ILLUDAS-WQ was developed for the estimation of combined sewer overflow
pollution resulting from storm water run-off. The model is based upon the pollutant run-off
unit load and wash-off concept. This model was applied to two separate watersheds in the
Chongju and Daejon metropolitan areas in Korea. The application results show that the runoff
correction coefficient and wash-off coefficient depend on the storm event size and watershed
area. It is a very useful model for estimation of pollutograph from urban watershed during
rainfall (Lee et al., 2002).
486
The GIS-oriented model MODIFFUS (Model to estimate diffuse nutrient inputs into
surface waters) was developed to estimate the diffuse nutrient inputs into 38 Swiss
hydrological river catchments (Prasuhn and Mohni, 2002). The model firstly generated the
water fluxes like surface run-off, leaching, deposition etc., secondly the nutrient inputs via
various pathways were attained according to land use, crop rotation, nutrient content in soil
and etc, and lastly they are shown as maps with the help of the GIS.
Simulation experiments for analyzing nitrogen (N) leaching from arable land were
performed using the Soil and Water Integrated Model (SWIM) for a set of representative
conditions in the Saale basin (23 687 km
2
) in Central Europe. The relative importance of
natural and anthropogenic factors affecting nitrogen leaching for the Saale River basin was as
follows: (1) soil, (2) climate, (3) fertilization rate and (4) crop rotation. The simulation
experiments provide a basis for a fuzzy-rule based metamodel approach, which aims at rapid
water quality assessment of large regions (Krysanova and Haberland, 2002).
HSPF (Hydrological Simulation ProgramFORTRAN) and the SMR (Soil Moisture
Routing) models were applied to a 102 km
2
watershed in the upper part of the Irondequoit
Creek basin in central New York State, USA over a seven-year simulation period (J ohnson et
al., 2003). Despite their differences in structure and representation of hydrologic processes,
the two models simulated stream flow with almost equal accuracy. HSPF simulated winter
stream flow slightly better than SMR as a result of its complex snowmelt routine, whereas
SMR simulated summer flows better than HSPF as a result of its runoff and base flow
processes.
A case study for Posan Reservoir which is an off-stream reservoir located in North
Taiwan was conducted through AGNPS to simulate NPS pollution and BMP (Best
Management Practices). The four major objectives considered are cost, phosphorus load,
sediment load and equity. The models developed in that study are intended to support to
decision-making analysis and estimates for funding allocation for both strategies (Kao and
Chen, 2003).
The process-oriented model HERMES was used to simulate the different combinations of soil
type, groundwater level, weather conditions and crop rotation to evaluate different options
of land use, agricultural and water management to reduce nitrogen emissions and enhance
water and nutrient retardation in the Elbe river basin, Germany (Kersebaum et al.,
2003).
An ArcGIS tool, named ArcCN-Runoff, was developed to facilitate watershed-modelling
work (Zhan and Huang, 2004). Unlike raster mode, ArcCN-Runoff is designed for any shape of
polygon in order to keep irregular boundaries unaltered. Application of dissolving
techniques reduces processing time significantly. The tool can be used to design and manage
hydraulic structures and projects, to estimate future discharges, and to predict watershed
response associated with changes in topography, soil, land use, and land cover (e.g.
urbanization). It was used for estimation of runoff in two USA counties, Lyon and Osage,
Kansas as case studies.
The GIS-based Decision Support System (DSS)Drainage Runoff Input of Pesticides in
surface water, DRIPSwas developed on behalf of the German EPA (UBA) for exposure
assessment of agriculturally used pesticides in surface waters. The tool estimates the quantity of
pesticide input from non-point sources via surface runoff, tile drainage and spray drift. A
graphical user interface (GUI) was created to provide users of the DSS with easy access to the
model algorithms. Results are available as grid cell maps for the territory of Germany;
featuring monthly catchment specific pesticides in surface waters (PECsw) values (Ropke et
al., 2004).
TopManage a modelling tool was developed to demonstrate the hydrological drivers of
diffuse pollution. It adopts a new visualization methodology that allows field characterization
487
to be undertaken in a systematic manner by combining high resolution mapping and terrain
analysis (Heathwaite et al., 2004). TopManage is used, along with the field runoff
characterization, to help identify CSAs (Critical Source Areas). It was applied to three
examples of UK farming practices: (1) a sloping arable field dominated by overland flow
from bare soil, (2) a drained arable field, and (3) a field dominated by subsurface flow (with
some drains) where runoff in the local ditch was simulated.
The Library of Hydro-Ecological Modules (LHEM) was designed to create flexible
landscape model structures that can be easily modified and extended to suit the requirements of
a variety of goals and case studies. The LHEM includes modules that simulate hydrologic
processes, nutrient cycling, vegetation growth, decomposition, and other processes, both
locally and spatially. Using The Library of Hydro-Ecological Modules (LHEM) and Spatial
Modeling Environment (SME) the Patuxent Landscape Model (PLM) was built to simulate
fundamental ecological processes in the watershed scale driven by temporal (nutrient
loadings, climatic conditions) and spatial (land use patterns) forcings (Voinov et al., 2004).
Model results for the Patuxent river watershed, Maryland USA show good agreement with
data for several components of the model at several scales.
Aggregated Plot Models were developed to emphasize links between economic policy
changes and environmental outcomes at a landscape scale. Stylized farm models were used to
predict changes in household land allocation arising from agricultural policy changes, with
explicit incorporation of biophysical feedback from erosion outcomes to agricultural
productivity and subsequent crop choices made by optimizing farmers. Outcomes were
combined to predict aggregate economic and environmental impacts. The method was applied to
data from the Manupali watershed having an area of approximately 60,000 ha, in the
Philippine province of Bukidnon (Shively and Coxhead, 2004).
A framework was built within the AgriBMPWater project to compare the impact of BMPs in
terms of hydrological effectiveness, costs for the farmers and society, and their
acceptability by farmers on eight European watersheds. The hydrology of the watersheds was
described and modelled using different watershed models. The hydrologic models
implemented include SWAT with some modifications, on French and on the Italian
watersheds, BMP1top on one French watershed, HAPSU in Finland, EUROSEM,
STOTRASIM in Austria, GLEAMS in Italy, POWER in France, Italy, and Austria, EIQ in
Norway (Turpin et al., 2005).
A DEM-based Overland Flow Model was developed for computation of surface run-off
from isolated storm events. The proposed model was calibrated and verified, using rainfall
and runoff data collected on the Banha catchment in India (J ain and Singh, 2005). The model
operates on a cell basis and takes cell physical information on topography, land use and soil
from a GIS. The catchment DEM is utilized in the model to generate computational flow
direction and flow routing sequencing for each of the discredited cell of the catchment.
Effects of three soil and water conservation practices were assessed using a crop growth
simulation model (WOFOST), a Nutrient Monitoring Model (NUTMON) and a hydrological
erosion model (LISEM), which were applied at field, farm and regional scale, respectively in the
highlands of Tigray, Northern Ethiopia (Hengsdijk et al., 2005).
An integration of an agricultural economic model, RAUMIS (The Regional Agricultural
and Environmental Information System) with hydrological models GROWA98 and WEKU
was developed and used to quantify and assess impacts of alternative nitrogen reduction
measures (Gomann et al., 2005). Two German river basins namely The Ems and a sub-
catchment of the Rhine were selected as study areas in order to cover a wide range of different
landscape units with different hydrological hydro-geological and socio-economic
characteristics.
488
CONCLUSIONS

The paper attempts to make an overall review of watershed models in order to provide key
information on watershed models as a basis for achieving better planning of land-use and
water resources development. The concept of sustainable management and planning is
gaining widespread acceptance as a vision of how humans function in an ecosystem. It is clear
that the current practices need to be enriched with new strategies to establish sustainable
utilization of watershed in a controlled manner. Fate and transport of especially pollutants
arising from non-point sources and protection of water bodies from pollution need to be
clearly defined and understood by public so as to take protective measures against further
deterioration of the environment. At this point, the need of using watershed models seems to be
an obligation in the field of integrated watershed management. Watershed models and GIS are
the two important tools in the assessment of natural resources. These tools also aid to
identify the priority problems encountered in the watershed, and to implement and evaluate
sustainable management plans and actions. The iterative nature of both techniques encourages
partners to set goals and targets, and make progress based on available information which at
the same time enables continuous analysis and queries in such areas.
As seen from the content of the paper, watershed modelling gained interest especially
within the past few decades in parallel to rapid development in computer technologies. After
giving brief information on the classification and types of models, and outlining the
universally accepted basic watershed loading models on urban and rural areas addressing
especially non-point source pollution problems, application of various new models are
mentioned to provide useful information on the most recent case studies from different
countries of the world to draw the attention of those related in the field of integrated
watershed management and to emphasize on the significance of the topic.



REFERENCES

Behrendt, H., Huber, P., Kornmilch, M., Opitz, D., Schmoll, O., Scholz, G., and Uebe, R.
(1999). Nutrient Emissions into River basins of Germany, Institute of Freshwater Ecology and
Inland Fisheries, Berlin.

Chen, Y.D. (2001). Watershed Modelling for Non-point Source Water Quality Simulation:
History, Recent Development and New Trends, 5
th
International Conference on Diffuse/Non-
point Pollution and Watershed Management, Milwaukee, Proceedings in CD Rom, IWA,
J une 2001, Session 16.1.

Deb, S.K., Lida, T., and Loof, R. (2001). Application of GLEAMS Model to Predict Soil and
Nutrient Losses from the Agricultural Field in Thailand, 5
th
International Conference on
Diffuse/Non-point Pollution and Watershed Management, Milwaukee, Proceedings in CD
Rom, IWA, J une 2001, Session 16.5.

ESCAP-UN (1997). Guidelines and Manual on Land-Use Planning and Practices in
Watershed Management and Disaster Reduction, (ST/ESCAP/1781), Economic and Social
Commission for Asia and the Pacific, United Nations, 133 pages.
489
Gmann, H., Kreins, P., Kunkel, R., and Wendland, F. (2005). Model Based Impact Analysis of
Policy Options Aiming at Reducing Diffuse Pollution by Agriculture-A Case Study for the River
Ems and a Sub-Catchment of The Rhine, Environmental Modeling & Software, 20,
261-271.

Heathwaite, A.L., Quinn, P.F., and Hewett, C.J .M. (2004). Modelling and Managing Critical
Source Areas of Diffuse Pollution from Agricultural Land using Flow Connectivity
Simulation, Journal of Hydrology, xx, 116 (article in press).

Hengsdijk, H., Meijerink, G.W., and Mosugu, M.E. (2005). Modelling the Effect of Three
Soil and Water Conservation Practices in Tigray, Ethiopia, Agriculture, Ecosystems and
Environment, 105, 2940.

Jain, M.K., and Singh, V.P. (2005). DEM-based Modelling of Surface Runoff Using
Diffusion Wave Equation, Journal of Hydrology, 302, 107126.

J ohnson, M.S., Coon, W.F., Mehta, V.K., Steenhuis, T.S., Brooks, E.S., and Boll, J. (2003).
Application of two Hydrologic Models with Different Runoff Mechanisms to a Hill slope
dominated Watershed in the Northeastern US: a Comparison of HSPF and SMR, Journal of
Hydrology, 284, 5776.

Kao, J.J., and Chen, W.J. (2003). A Multiobjective Model for Non-Point Source Pollution
Control for an Off-Stream Reservoir Catchment, Water Science and Technology, 48(10), 177-
183.

Kersebaum, K.C., Steidl, J ., Bauer, O., and Piorr, H.P. (2003). Modelling Scenarios to Assess
the Effects of Different Agricultural Management and Land Use Options to Reduce Diffuse
Nitrogen Pollution into the River Elbe, Physics and Chemistry of the Earth, 28, 537545.

Krysanova, V., and Haberlandt, U. (2002). Assessment of Nitrogen Leaching from Arable
Land in Large River Basins Part I. Simulation Experiments Using a Process-Based Model,
Ecological Modeling, 150, 255275.

Lee, J .H., Yu, M.J ., Bang, K.W., Choe, J .S., and Gwan, E.M. (2002). Development of the
ILLUDIAS-WQ Model for Storm water Overflows in Urban Areas, Proceedings of 6
th
International Conference on Diffuse Pollution, Amsterdam, IWA, 30 September- 4 October
2002, 89 96.

Leon, L.F., Soulis, E.D., Kouwen, N., and Farquhar, G.J . (2002). Modelling Diffuse Pollution
with a Distributed Approach, Water Science and Technology, 45(9), 149-156.

Novotny, V. (2003). Water Quality-Diffuse Pollution and Watershed Management, 2
nd

Edition, J ohn Wiley & Sons Inc.

Prasuhn, V., and Mohni, R. (2002). GIS-Based Estimate of Phosphorus and Nitrogen Inputs
into Surface Water of the Switzerland with the model MODIFFUS, Proceedings of 6
th
International Conference on Diffuse Pollution, Amsterdam, IWA, 30 September 4 October
2002, 564 565.
490
Reimold, R.J . (1998). Watershed Management Practices, Policies and Coordination,
McGraw Hill.

Rpke, B., Bach, M., and Frede, H-G. (2004). DRIPSA DSS for Estimating the Input
Quantity of Pesticides for German River Basins, Environmental Modeling & Software, 19
(2004) 10211028.

Shively, G., and Coxhead, I. (2004). Conducting Economic Policy Analysis at a Landscape
Scale: Examples from a Philippine Watershed, Agriculture, Ecosystems and Environment,
104, 159170.

Turpin, N., Bontems, P., Rotillion, G., Barlund, I., Kaljonen, M., Tattari, S., Feichtinger, F.,
Strauss, P., Haverkamp, R., Garnier, M., Porto, A.L., Benigni, G., Leone, A., Ripa, M.N.,
Eklo, O.M., Romstad, E., Bioteau, T., Birgand, F., Bordenave, P., Laplana, R., Lescot, J .M.,
Piet, L., and Zahni, F. (2005). AgriBMPWater: Systems Approach to Environmentally
Acceptable Farming, Environmental Modeling & Software, 20, 187-196.

USEPA, (1991). Modelling of Non-point Source Water Quality in Urban and Non-urban
Areas, (EPA600 3-91 039), Office Research and Development, United States Environmental
Protection Agency, Washington, DC.

Voinov, A., Fitz, C., Boumans, R., and Costanza, R. (2004). Modular Ecosystem Modeling,
Environmental Modeling & Software, 19, 285304.

Wendland, F., Bach, M., Behrendt, H., and Kunkel, R. (2002). Integrated Modelling of
Groundwater-Borne Nitrate Intakes into the River Elbe Basin (German Part), 9th
International Conference on Watershed and River Basin Management, Edinburgh, 11-13
September 2002, IWA, Proceedings in CD Rom.

Zhan, X., and Huang, M. L. (2004). ArcCN-Runoff: An ArcGIS Tool for Generating Curve
Number and Runoff Maps, Environmental Modeling & Software, 19, 875879.
491

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005









STRATEGIC DECISION SUPPORT SYSTEM
FOR
RIVER NILE WATER QUALITY ASSESSMENT



Inas El-Gafy
(1)

1) Researcher - Strategic Research Unit- National Water Research Center,
Egypt, Email: inaa_2r@yahoo.com

Keywords: Environmental management, Decision Support System, water
quality, River Nile, water use

ABSTRACT
A Decision Support System for River Nile Water Quality Assessment
under the name DSS-RNWQA was developed through the current study. In
General DSS-RNWQA helps in the environmental management in Egypt, where
it can be used in assessing of the pollution level of the main source of fresh
water in Egypt. The developed DSS provides information and helps the decision
makers in evaluating the suitability of using River Nile for intended uses. These
uses are human drinking, irrigation of cotton; rice; wheat; corn; orange; potato;
tomato; cucumber; been; onion; barseem; and sugarcane crops, live-stock
drinking, fish life, and recreation use. The developed decision support system
has an analysis tool that can be applied to assess the improvement in the water
quality of the River Nile for certain period of time. Water quality index (WQI)
approach was applied for the evaluation process through the developed DSS.

Furthermore, the developed DSS has an information system that was
designed to store and retrieve information and data about water quality of the
Nile River for both sample and pollution point sources locations. DSS-RNWQA
was implemented through database, GIS and interactive spreadsheets analysis.
The DSS integrates models, spatial and non-spatial data and analysis tools under
user-friendly GIS-based interfaces, which confronts the decision maker with
possible measures as well as multiple management objectives.

492

INTRODUCTION

Nowadays, one of the major worldwide concerns is the deterioration of the environment. The
pollution of the environment may occur as a result of major accidents or continuous discharge
of pollutants that accumulate and cause contamination to air, water, and soil. Therefore for
the sustainability of healthy environment, there should be a strategic environmental
management plan that includes continues evaluation for the environment component, water,
air, and soil.

In Egypt the River Nile is the main source for fresh water. River Nile is sometime
exposed to pollutants due to human activities. Generally, the main sources of pollutants that
dumped into the water course are: domestic wastewater, industrial wastewater and agricultural
drains. Accordingly, water quality status on the River Nile may reach to a level that it is not
suitable for its intended usage. Therefore, it is essessntial to evaluate the contamination level
of the water in the River Nile before reaching the level of unwanted pollution. The regular
procedure for analyzing the water quality data to determine its status is usually carried out by
comparing the concentration of the on-site measured parameters to their relative standard.
This procedure does not describe the overall water quality in a way that could be easily
carried-out. Furthermore, it takes long time, effort and a specialist has to conduct this process
Therefore, there was a need for a Decision support System to carry out this process. Decision
Support System is an integrated approach for helping people make better decisions, DSS is
typically a computer program that is used by individuals or groups to facilitate the decision
making process (Fulcher et al, 1999).

The purpose of the current study is to introduce a stand alone DSS that allows the
decision maker to spatially evaluate the water quality status along the River Nile. The
methodology applied herein is to have an intensive literature review about the models and
DSSs previously developed for evaluating water quality in River Nile. In the current study it
has been decided to modify an existing module for evaluating River Nile water quality that
was developed by El-Gafy in 2004 as a part of a decision support system named DSS-
EEQRA. The modification takes place to overcome the drawbacks that are found in that
module and enhance the modified version to open wider range for decision support. The
existing module is capable of evaluating the suitability of water quality for different water
uses such as human and livestock drinking use and for irrigation of cotton and rice crops (El
Gafy, 2004).

Through the applied modification to the previous module, the new system allows the
decision maker to evaluate the suitability of using the water for irrigating ten more crops that
are wheat, corn , orange, potato, tomato, cucumber, been, onion, barseem, and sugarcane.
Water quality index (WQI) approach is applied for the evaluation process through the
developed DSS. WQI is an overall indicator of water quality obtained by aggregating several
water quality measurements into one number (Canter, 1996). Moreover, the enhancement
version extents for more analysis tools that can be applied to assess the improvement in the
water quality of the River Nile for a certain period of time for drinking water and live stock
use section. Moreover, Information system is added to the modified DSS. This part of the
developed DSS was designed to store and retrieve information and data about water quality of
the Nile River for both sample and pollution point sources locations. The user of this part will
be able to acquire information and hence endorse the following issues: i) Concentration of a
certain parameter at certain year or certain location, ii) water quality data at certain location
iii) water quality data of: industrial, agricultural, sewage drains, water treatment plants, and
493

power stations. The enhancement procedure is implemented through database, GIS and
interactive spreadsheets analysis. The developed DSS integrates models, spatial and non-
spatial data and analysis tools under a user-friendly GIS-based interface, which confronts the
decision maker with possible measures as well as multiple management objectives.

FEATURES OF THE DEVELOPED DSS

The developed decision support system contains two main parts. The first is an
Information System that was designed to store and retrieve information and data about water
quality of the Nile River for both sample and pollution point sources locations. The user of
this part will be able to acquire information and hence endorse the following issues that are
shown in Table (1): a) Concentration of a certain parameter at certain year or location, year
for certain location, b) Water quality data at certain location c) Water quality data of:
industrial, agricultural, sewage drains, water treatment plants, and power stations.

The second part is the evaluation System that was developed to evaluate the water
quality according to the different water uses. This part is designed in such a way that helps in
answering the most probable questions that may come to the decision maker mind. A list of
the proposed what-if questions that the decision maker may get answers about them through
the current decision support system is summarized in Table (2). As shown in Table (2) the
user may have a set of evaluation maps, reports and charts that give hem a wide view about
the suitability of water quality of the River Nile for human drinking, irrigation of cotton; rice;
wheat; corn; orange; potato; tomato; cucumber; been; onion; barseem; and sugarcane crops,
Live-stock drinking, Fish leaving, Fish and Human health, and recreation use.

Table (1) Features of DSS-RNWQA Information System
Item Information
Parameter Concentration of certain parameter at certain
Year for:
all location
certain location
Location Water quality data at certain location for:
all years
certain year
Plants Water quality data of:
Industrial drains
Agricultural drains
Sewage drains
Water treatment plants
power stations





494

Table (2) Features of DSS-RNWQA Evaluation System
Water Use Level of
evaluation
What...If? Output
Irrigation Macro What is going to happen if
the water body is used for
irrigation?
Map with indication of the water suitability degree
for cultivating 10 main crops
Irrigation Micro What is going to happen if
the water body is used for
irrigation?
Water suitability degree for cultivating 10 main crops
and detailed information in case of unsuitable water
quality showing the pollutant parameters and its
effects
Livestock Macro What is the suitability of
using the water body for
livestock drink?
Map showing the level of the water suitability for
live-stock drinking
Livestock Micro What is the suitability of
using the water body for
livestock drink at certain
location?
Detailed information for the degree of the suitability
of using the water either if it is positive or negative
and its impact on livestock
Drinking Macro What is the suitability of
using the water body for
drinking?
Map showing the level of the suitability of using the
water for human drinking
Drinking Micro What is the suitability of
using the water body for
drinking at certain
location?
Detailed information about the degree of suitability
of using water for drinking and more deep
information about the parameters those cause water
condition either if it is good or bad.
Water
Contact
Macro What is the suitability of
contacting the water
body?
Map showing the level of the suitability of contacting
the water body
Water
Contact
Micro What is the suitability of
contacting the water body
at certain location?
Detailed information about the degree of the
suitability of contacting the water body showing the
pollutant parameters and its effects
Fish Macro What is the suitability of
the surface water course
for fish living?
Map showing the level of the suitability of the
surface water body for fish living.
Fish Micro What is the suitability of
the surface water body
for fish living at certain
location?
Detailed information about the degree of the
suitability of using the water body for fish with high
lightning on the parameters that cause pollution for
fish
Fish and
Human
Health
Macro What is the suitability
degree of the surface
water body to produce
safe fish production for
human?
Map showing the level of the suitability of the
surface water body to produce safe fish production
for human.
Fish and
Human
Health
Micro What is the suitability
degree of the surface
water body to produce
safe fish production for
human?
Detailed information about water quality parameter
that cause problems to produce safe fish production
for human

DEVELOPMENT OF DSS-RNWQA

Design frame work
Figure (1) presents the basic design of DSS-RNWQA. It starts by collect water quality
data about the area under investigation. The user has the opportunity to select between two
options. The first is interning the information system through which the user can have
information about water quality of the Nile River for both sample and pollution point sources
locations or going through evaluation process through which the decision maker will be able
to assess the suitability of using the water for different water uses.
495




Figure (1) Basic Design of DSS-RNWQA

The evaluation process starts manipulating a set of developed water quality evaluation
indices that are: Drinking Water Quality Index, Cotton; Rice; Wheat; Corn; Orange; Potato;
Tomato; Cucumber; been; Onion; Barseem; and Sugarcane crops Irrigation Water Quality
Index, Live-stock Water Quality Index, Fish Water Quality Index, Fish and Human Water
Quality Index- Carcinogens, Fish and Human Water quality Index-Non Carcinogens, and
Contact Water Quality Index. The process ends up with quantitative and qualitative
evaluation that presents the suitability of using the water body for different uses.

Formulation
The applied indices in the evaluation system were developed after studying and
modifying the construction procedures of the previously developed water quality indices in
many countries such as United States (Canter, 1996), United Kingdom (House, 1989), and
Egypt (El-Sherbini et al, 1992; UNEP, 1995; El-Gafy, 2001, and El-Gafy 2004). The
developed indices are: Drinking Water Quality Index, Cotton Irrigation Water Quality Index,
Rice; Cotton; Wheat; barssem; Irrigation water Quality Index, Live-stock Water Quality
Index, Fish Water Quality Index, Fish and Human Water Quality Index- Carcinogens, Fish and
Human Water quality Index-Non Carcinogens, and Contact Water Quality Index. In what
follows a description for the process of developing the previous indices will be presented.

(a): Selection of Indicators for Each Water Use
Interviews with experts in different authorities in Egypt were conducted to select
water quality indicators and to determine their allowable limits. These authorities were:
National Water Research Center; Drainage Research Institute; Central Laboratory for
Environmental Monitoring; Advisory Panel for Irrigation and Drainage Projects, Ain Shams
University; Faculty of Agricultural and Faculty of Science, El-Mansora University,
Agricultural Research Center; and Desert Research Center.

496


(b): Development of Value Functions
Value functions (sub-index) to relate the concentration of each selected indicator with
its effect in the water quality were developed. Figure (2) presents an example of the developed
value functions. These functions were developed by using the literature review, interviews
with experts, or by making assumptions. Through DSS-RNWQA more that 200 value
functions are calculated. The value function can be mathematically formulated as:
( )
i i
C f I = (1)
Where:
i
I =Sub-index for Parameter i
i
C =Concentration of parameter i
0
2
4
6
8
10
0 0.001 0.002 0.003 0.004 0.005
Concentration (mg/l)
C
a
d
m
i
u
m

S
u
b
-
i
n
d
e
x


Figure (2) Value Function for Cadmium - Drinking Use

(c): Aggregation of the Sub-indices
The following equation was used to aggregate the sub-indices. This equation was
tested and recommended for calculating the water quality index (El-Gafy, 2001)
( )
i U
I Min WQI
n
= (2)
Where:

n
U
WQI =Water quality index for each water use
i
I =Sub-index for Parameter i

Implementation
Microsoft Access, EXCEL spreadsheet, and Arc view 3.1 are applied to implement the
current DSS. A set of maps, tables, quires, objects, macros and end-user interfaces were
designed and implemented. Figure (3) represents the relational schema for a part of the
implemented database.

Very poor
Poor
Good
Excellent
Moderate
497



Figure (3) The Relational Schema for The Designed Data base

DESCRIPTION OF DSS-RNWQA

In the following section, description for the developed user-interfaces and their
functions will be introduced. Moreover the outputs reports and maps from the developed
DSS will be presented. Figure (4) shows the main screen of the developed DSS. The main
screen is a window that briefly describes the system and allows the user to start navigating
through the DSS by pressing the main menu button that is shown in Figure (4). When the user
goes through the main menu screen, shown in Figure (5), he will have the ability to select the
area he would like to work on it. The user can select either to go to the data entry screen, the
information system or the evaluation system.

Data entry screen
The data entry screen was designed to input the required data for running the DSS.
These data include the concentration of different water quality parameters at the different
sampling and pollution point source locations along the River Nile. Figure (6) shows data
entry screen for sampling location. Similar screen for data entry of pollution point source
locations was developed.



498



Figure (4) Main Screen of DSS-RNWQA





Figure (5) Main Menu Screen of DSS-RNWQA

499



Figure (6) Sampling Location Data Entry Screen
Information system
Through this system the user has the opportunity to search for certain data about
sampling or pollution point source locations all over the River Nile. Where, the user will be
able to acquire information and hence endorse about concentration of a certain parameter at
certain year for all location. Figure (7) presents an output report that shows the concentration
of a certain parameter through period of time all over the River Nile. Further more, the user
may have data about a certain parameters at certain location for certain period of time. Figure
(8) shows the screen that was designed to help the user to investigate for certain parameter at
certain year and certain location in an undemanding method. Moreover, the user are able to
know water quality data of industrial, agricultural, sewage drains, water treatment plants, and
power stations. As an example, Figure (9) shows the output report about the water quality of
an industrial pollution point source location.

Figure (7) Macro Parameter Search Report

500



Figure (8) Micro Parameter Search Screen



Figure (9) Water Quality of Industrial Drain

Evaluation system
As shown in Figure (10) that presents the main screen in the evaluation system; this
process is divided into two major levels: the macro-level and the micro-level to be used by both
technical and non-technical decision-makers. The macro level is designed to present the
evaluation on the level needed by the highest level of decision makers. Through the micro
level the user will be able to know the water quality parameters that cause pollution for certain
water use at certain location, the negative effects and the standard levels of these parameters,
and the qualitative and quantitative evaluation of each parameter. For some water uses such as
drinking water a recommended treatment method is available. In what follows a detailed
description for each level of evaluation will be introduced.

501



Figure (10) Evaluation System Main Screen

Macro Evaluation Level
Figure (11) and Figure (12) present examples for the macro evaluation map and the
macro evaluation chart that may be obtained for the River Nile for certain water use. From
these map and chart the user will be able to know very poor, poor, moderate, good and
excellent water quality location for certain water use.

Table (3) presents the macro evaluation report that shows the polluting parameters for
drinking use. From this report the Total Coliform is the main parameters that cause pollution
for drinking water use in all locations.

Table (4) presents the macro evaluation report that shows the polluting parameters for
irrigation use at certain location. From this report Sodium Absorption Ratio (SAR) is the main
polluting factor for the irrigation at that location. Similar maps, charts and reports are
developed for the others water uses.

Micro Evaluation Level
Figure (13) shows the micro evaluation report of location No 21 for drinking use. As
shown in the figure Total coliform is the main polluting parameter for drinking water use in
location 21. The concentration of the Fecal Coliform at location No 21 is 2000 No/100ml
while its recommended standard is 1 No/100ml. Therefore, the water quality index at that
location is considered very poor and has a qualitative score 0 that is far from the starting score
of the very poor level. The very poor level starts from score 2 when the concentration of the
parameter is equal to the standard and extended to numerous negative number according to
how far is the concentration from the standard level. Moreover, from this report the user has
the opportunity to know the effect of the water quality parameters if it exceeds its allowable
standard. Herein, Total coliform may cause gastroenteritis, typhoid fever, dysentery cholera
and salmonella if the polluted water used without treatment.

Figure (14) presents the micro evaluation report of location No 21 for irrigation use.
As shown in this figure the parameter causes the pollution for the irrigation use in this
502

location is Sodium Absorption Ratio. In addition, the report shows the parameters
concentration and its qualitative water quality class.



Figure (11) Macro Evaluation Map




Figure (12) Macro Evaluation Chart
503


Table (3) Macro Evaluation Report of Polluting Parameters for Drinking Use
Distance
from
AHD
(km)
Date Concentration
No/100ml
DWQI
Sub-
index
Parameter Name Water
Quality Class
5 9/1/2000 750 0 Total Coliform very poor
21 9/1/2000 2000 0 Total Coliform very poor
54 9/1/2000 1300 0 Total Coliform very poor
83 9/1/2000 1000 0 Total Coliform very poor
110 9/1/2000 1400 0 Total Coliform very poor
148 9/1/2000 340 0 Total Coliform very poor

Table (4) Macro Evaluation Report of Polluting Parameters for Irrigation Use
Distance
from
AHD
(km)
Date Concentration
(Ratio)
IWQI

sub-index
Parameter Name Water
Quality
Class
1123 9/1/2000 9.54 5.97 SAR

Moderate
1136 9/1/2000 9.47 5.99 SAR Moderate
1150 9/1/2000 9.54 5.97 SAR Moderate
1166 9/1/2000 9.51 5.98 SAR Moderate




Figure (13) Drinking Micro Evaluation Report for Certain Location



504



Figure (14) Irrigation Water Micro Evaluation Report for Certain Location

CONCLUSION

Within the current study a Strategic Decision Support System for River Nile Water
Quality Assessment under the name DSS-RNWQA has been developed. DSS-RNWQA help
in the environmental management in Egypt where it can be used in assessing of the pollution
level of the main source of fresh water in Egypt.

The developed decision support system has two main parts that are the information
system and the evaluation system. The information system was developed to store and
retrieve information and data about water quality of the Nile River for both sample and
pollution point sources locations. The user of this part will be able to acquire information and
hence endorse the following issues: Concentration of a certain parameter at certain year or
location, year for certain location, b) Water quality data at certain location c) Water quality
data of: industrial, agricultural, sewage drains, water treatment plants, and power stations.

The evaluation system manipulates a set of developed water quality evaluation
indices that are: Drinking Water Quality Index, Cotton; Rice; Wheat; Corn; Orange; Potato;
Tomato; Cucumber; been; Onion; Barseem; and Sugarcane crops Irrigation Water Quality
Index, Live-stock Water Quality Index, Fish Water Quality Index, Fish and Human Water
Quality Index- Carcinogens, Fish and Human Water quality Index-Non Carcinogens, and
Contact Water Quality Index. The process ends up with quantitative and qualitative
evaluation that presents the suitability of using the water body for different uses.

The developed DSS has an analysis tools that can be applied to assess the
improvement in the water quality of the River Nile for a certain period of time for human and
live stock drinking use section The Developed decision support System was implemented
through database, GIS and interactive spreadsheets analysis.

DSS-RNWQA integrates models, spatial and non-spatial data and analysis tools under
a user-friendly GIS-based interface, which confronts the decision maker with possible
measures as well as multiple management objectives.

505


REFERENCES

Canter L. W., 1996, Environmental Impact Assessment, Second Edition, McGraw-
Hill Inc., New York, United States of America
El-Gafy I., 2001, Formulation of Surface Water Quality Index for Egypt, M. Sc.Thesis,
Institute of Environmental Studies and Research, Ain Shams University, Cairo,
Egypt.
El-Gafy I, 2004, "Decision support System for Evaluating The Environmental Quality in
Rural Areas", Ph. D.Thesis, Institute of Environmental Studies and Research, Ain
Shams University, Cairo, Egypt.
El-Sherbini A., El-Moattassem M., and Sloterdijk H. (1992), Water Quality Condition
of Rossetta Branch, Proceeding of the International Conference Nile 2000, Cairo,
Egypt.
Fulcher C., Prato T., Barnett Y., 1999, Economic and Environmental Impacts
Assessment Using WAMASS, Proceeding of the 32
nd
Hawaii International
Conference on System Science
House M. A., (1989), A Water Quality Index for River Management, J . Inst. Water and
Environment Management 3, 336-344.
UNEP, (1995), Development of Water Quality Indices for Sustainable Development: a
case study , Proceeding of the Expert Group Meeting on the Implication of
Agenda 21 for Integrated Water Management in the ESCWA Region, Amman, 2-5
October

506

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005






FUZZY FISH HABITAT MODELING

Heribert NACKEN
Prof. Dr.-Ing, Section of Engineering Hydrology,
RWTH Aachen University, Aachen 52074, Germany
Tel +49 241 8025273, Fax. +49 2418022701
nacken@lfi.rwth-aachen.de

Hani Nabhan SEWILAM
PhD, Senior Staff Scientist, Section of Engineering Hydrology,
RWTH Aachen University, Aachen 52074, Germany
Tel. +49 241 8025643, Fax. +49 241 8022702
sewilam@lfi.rwth-aachen.de

Indexing terms: Fuzzy Logic, Fish habitat, rule-based modeling

ABSTRACT
Ecohydrology is a scientific area that deals with both hydrological and
ecological processes and their interactions. Modeling these processes is
characterized as a complex task due to associated uncertainties and
nonlinearities. Many of the ecological processes are ill-defined or qualitatively
described, sharp knowledge is rather exception. Heterogeneity and variability
are main characteristics of hydrological parameters. Unfortunately, traditional
mathematical models often fail to cope with the mentioned difficulties.
According to the European Union (EU) Water Framework Directive (WFD), the
actions to be taken for ecosystem improvement should enable reaching good
ecological status for all waters in Europe before 2015. This adds the
development time and cost of our models as additional difficulties to the
ecohydrological modeling process, since the implementation of the EU WFD
requires affording many reliable ecohydrological models in a limited time
space.

507




Fuzzy logic offers potential enhancements to ecohydrological modeling
as well as many other scientific areas. It facilitates describing complicated,
uncertain and non-linear processes in a strict mathematical framework as well as
transforming available experts words and knowledge into rule-based models
with less efforts and low cost. The basic principals of fuzzy ecohydrological
models are illustrated in this paper. The advantages of using fuzzy-based
models as tools for implementing the EU WFD are discussed. The paper
introduces also a demonstration example for a fuzzy fish-habitat model. The
model shows that fuzzy logic is a reliable technique for representing many
ecohydrological processes and encourages scientists to carry out further similar
research.
EU WATER FRAMEWORK DIRECTIVE (EU WFD)
The EU WFD has the objective of reaching a Good Ecological Status by 2015. The
Directive only gives general definitions and work instructions, but leaves the details for the
member states. Therefore, researchers must actively participate to adopt technologies for
suggesting remediation action scenarios, planning and implementing them. This is necessary
to maximize protection for the current ecological quality of surface water and provide the
strongest possible incentives for the restoration of degraded waters.

Hendry et al. [1] concluded during a workshop held to discuss the implementation of
the EU WFD that computer modelling is a useful tool, but do they work? A fortune could be
spent but be no better informed at the end of it i.e. if models help understand the problem
(which is not always the case) then fine but they are not a solution in themselves and should
not be an objective unless they have a specific role to play in problem solving, understanding
and/or management. It was also mentioned that models are relatively expensive with a need
to maintain long data sets.

On the other hand, without using computer models, it will be almost impossible to
suggest remediation action scenarios for improving the ecological status, test them and decide
the ones to be implemented. For this task, many hydrological and ecological processes as
well as their interactions should be modeled. Hydrological aspects that affect ecological
status such as nutrient transport to the surface and groundwater must be quantified based on
modeling the dynamic processes. This will be essential to decide which land-use scenario is
the best to minimize diffuse sources of pollution. The complexity of such an interaction
makes modelers oversimplify their models. For instance, most diffuse sources of pollution are
based on balance concepts and not on simulating the real world dynamic processes. This is
because of the complexity and uncertainty associated with such tasks. Tasks such as the
impact of chemicals concentration on fish habitat or predicting the effects of changes in
hydrological cycle or water quality on the probability of plant species occurrence should be
quantified based on physical simulation.

Integrated models should be developed and then implemented to analyze formulated
scenarios. The ecological potentials of the analyzed scenarios should be determined. An
overall assessment should be carried out to decide the scenario to be implemented. As
508


described in the following section, the main difficulty of following this procedure is the
complexity of ecohydrological modeling.


COMPLEXITIES OF ECOHYDROLOGICAL MODELING
Ecohydrology is characterized as one of the most complex scientific areas, since it
integrates ecology into hydrological approaches, and hydrology into ecological studies [2].
Hydrology as defined by Lee [3] is the science that tries to explain the water balance
dynamics for any defined space (from point to global) and temporal scale (from seconds to
years) and their relationships with the physical and chemical transport of matter through the
hydrological cycle and with ecology. The main hydrological processes that influence the
ecosystem are evapotranspiration, precipitation, infiltration, capillary rise, groundwater flow
and surface runoff. An ecological system or ecosystem as defined by Siry [4] is the basic unit
of a synthetic and systematic study of an organism or a species and its surroundings. An eco-
system is divided into; habitat, mineral, biotic community and organic.

Mathematical modeling of ecohydrological processes faces a great difficulty, because
of the high uncertainties of the involved parameters. In addition, the data and knowledge that
describe the ecological relationships are mostly imprecise, uncertain or ambiguous.
Kampichler et al. [5] stated that an overemphasis on the precision of mathematical models
does not necessary lead to a better representation of reality. In most of the cases, ecologists
do not communicate in the form of systems of differential equations or analytical models, but
they use natural languages and qualitative reasoning for the description of ecological
relationships. Unfortunately, classical mathematical models are not well suited for dealing
with ill-defined, uncertain situations or qualitatively described processes. Therefore,
implementing artificial intelligent techniques is a suitable approach for modeling
ecohydrological processes.
ARTIFICIAL INTELLIEGENCE TECHNIQUES
In contrast of traditional mathematical modeling techniques, artificial intelligence
provides a means for processing knowledge that is represented in natural language and
enables capturing knowledge from available data. In addition artificial intelligence techniques
have the capability to handle large amount of dynamic, non-linear, noisy or uncertain data.
As defined by Dzeroski et al. [6] Artificial intelligence is the study of ways in which
computers can perform tasks that demand intelligent behavior. Artificial intelligence
techniques include artificial neural networks, fuzzy logic and genetic algorithms.

Such techniques can be the desired solution that enables researchers to describe and
assess ecohydrological processes and status in less time and with lower cost and in turn to
reach the good ecological status by 2015. Fuzzy rule-based models are recently employed in
ecohydrological modeling to perform simulation tasks that could be called qualitative
simulation. Such models enable describing uncertain and ill-defined situations in strict
mathematical frameworks. The coming sections discuss the main problems associated with
modeling of fish habitat as well as a demonstration example of implementing a fuzzy rule-
based approach to assess the suitability of the water ways for fish habitat.
FISH HABITAT AND THE WFD
Fish habitat is defined as those parts of the environment that fish depend on, directly
or indirectly, in order to carry out their life processes [7]. There are basic requirements that
should be available so that fish can successfully carry out their life processes;
509


- fish must have food to be able to reproduce and need cover to protect themselves
from predators.
- the biological, chemical and physical features of water streams must be suitable
for the reproduction process.

Water quality can affect fish directly through behavioral and physiological changes or
indirectly by affecting food supply or habitat. Extreme water quality changes may result in
physiological trauma (e.g. organ damage) or death. Water quality parameters that can affect
fish include water temperature, dissolved oxygen, pH, turbidity, ammonia, salinity, dissolved
metal concentrations and other toxic substances such as chlorinated organics, oils, pesticides,
etc [7].

It is clear that the suitability of a water stream for fish habitat is an indication for the
water quality. For the EU WFD, fish habitat can be used as an indicator for the biological
quality of water courses. Therefore, fish habitat modeling is a twofold objective; First, it
enables assessment the suitability of water bodies for fish habitat. Second, it indicates the
quality of the observed water stream and whether a remediation action is necessary or not.

A FUZZY-BASED HABITAT MODEL FOR SALMON
In this part, the first phase of a research project that aims to develop a fuzzy-based
habitat model for salmon fish is proposed [9]. The main objective of this project is to develop
a model for salmon that could be used to assess the German water bodies with respect to their
suitability for salmon habitat. The small blocks on the left side of Figure (1) represent the
input parameters of the model (refer also to Table 1). The suitability index of a water body
for fish habitat is considered the output of the model (the small block on the right side of
Figure 1). The large eight blocks in the middle of the screen are the rule-blocks, which
perform the stepwise aggregation of the values of input- and intermediate-parameters.

Table 1. The inputs and intermediate parameters of the fish model.




















Input
Parameters
Units Intermediate
Parameters
dissolved
oxygen
[mg/l]
Water velocity [m/s]
Temperature C
Physical
pH-Wert
Ammoniac [mg/l]
Cupper [mg/l]
Zink [mg/l]
Water Quality a
AFS [mg/l]
Pesticide [mg/l]
Tenside [mg/l]
Water Quality b
Patency Linguistic
Riffle-Pool Linguistic
Morphology a
Substrata Linguistic
Permeability Linguistic
Morphology b
510















Each of the input and intermediate indicators are represented as linguistic variables
with three terms. The output variable Habitat Suitability Index has five terms and uses
Center-of-Maximum (CoM) defuzzification method. Each term is described either by
triangular or trapezoidal membership function.

The membership functions were initially spaced equally and then shifted and
modified according to the opinion of fish experts and researchers. Fig (2) shows an example
of the input variable water temperature that is divided into three different membership
functions. The membership function of the set low has a degree of one from zero through 5
C. This membership value declines to zero when the value is 12 C. The same is followed to
construct both membership functions optimal and high. The approach is repeated for all
variables.

Figure 2. Membership functions for the parameter water temperature

Selecting linguistic terms that are meaningful for users are considered while
constructing the membership functions. For instance, the terms of the pesticide are
considered as follows: tolerable, damaging, deathly. Because of the unavailability of crisp
data, four out of fourteen input variables are adopted to receive qualitative input data (e.g.
511


Pool_Type). The other ten variables are crisp variables. The inputs of the crisp variables will
be transformed to linguistic terms (fuzzified) and sent to the rule-base to be manipulated with
the linguistic terms of the other four variables.

The rule-base includes fuzzy rules which are used to stepwise aggregate and assess
the input terms in order to reach the overall goal (habitat suitability index). For instance, a
rule that is used to assess and aggregate the water quality parameters (ammoniac, cooper and
zink) in one intermediate parameter (Water Quality a) can be written as follows:

IF ammoniac IS tolerable AND cooper IS tolerable AND zink IS deathly
THEN
Water_quality_a IS bad

The output term of this rule will be assessed together with the terms of the other water
quality inputs to reach an intermediate water quality term as follows:
IF water_quality_a IS bad AND water_quality_b IS bad AND pH_value IS
deathly
THEN
overall_water_quality IS bad

The output terms of the intermediate variables will be again aggregated and assessed
to give an overall assessment about the suitability of the observed waterway as follows:

IF overall_water_quality IS bad AND morphology IS unsuitable
AND water_area IS unsuitable
THEN
Habitat_Suitability IS very_low

Each rule has got a weight (from 0 to 1) that reflects it significance. Initially, all
possible rules are generated and then assessed to get the suitable weight. This process is also
carried out by fish experts and researchers. The center of maximum method is then used to
defuzzify the linguistic terms of the fish habitat suitability index. The result of the model is a
crisp suitability value of the assessed water stream that is between 0 and 1.

The first phase of the model development has shown the applicability of fuzzy logic
to develop ecohydrological models that overcome the problem of associated uncertainty of
the parameters. The model enables mixing qualitative and quantitative data in a strict fuzzy
mathematical framework. It was possible to integrate the qualitative knowledge of involved
researchers and experts during constructing both membership functions and rule-bases.
During the second phase of the research project the model will be validated and tested for
different water streams across Germany.

CONCLUSIONS
This paper scoped on the need for simple modeling techniques that enable carrying
out the required tasks of the EU WFD, mainly modeling of ecohydrological processes and
habitat prognoses. Fuzzy models normally produce compatible results comparing to the
traditional numerical methods. In addition, they are easy to interpret and require less
programming efforts and costs. Such models can be adapted to accept mixed data (fuzzy and
crisp) and carry out simulations using less number of parameters. Therefore, fuzzy rule-based
512


models are recommended to be intensively used for suggesting, analyzing and selecting the
action plans for the EU WFD.

The main difficulty of constructing fuzzy models is constructing the membership
functions and the rule-base. Combining fuzzy techniques with algorithms that enable
automatic knowledge acquisition from available data improves the process significantly. For
instance, artificial neural networks have the ability to learn new associations, new
functional dependencies and new patterns from data. Such hybrid techniques are
recommended for further ecohydrological modeling tasks.

REFERENCES

[1] Hendry et al. Implementing the EU Water Framework Directive in the Northwest: the
research agenda. Proceedings of a Research Workshop April 8th 2003, Allen Hall,
Fallowfield, Manchester.
[2] J anauer, G.A., 2000. Ecohydrology: fusing concepts and scales. Ecological Engineering
16, 9-16.
[3] Lee, J ., 1996. IAHS Newsletter 42.
[4] Siry, J ., 2001. Ideas, Actions and Values. Internet Publication:
http//fox.Rollins.edu/~jsiry/ecology.html
[5] Kampichler, Ch., Barthel j. and Wieland, R. 2000. Species density of foilage-dwelling
spiders in field margins: a simple, fuzzy-based model. In Ecological Modelling 129, 87-
99.
[6] Dzeroski, S., Grbovic, J ., Walley, W.J . and Kompare, B., 1997. Using machine learning
techniques in the construction of models: data analysis with rule induction: Ecological
Modeling 95, 95-111.
[7] Online Publications of the government of Alberta 2003 (Transportation department). Fish
Habitat Manual. http://www.trans.gov.ab.ca.
[8] Baggash, M., Sewilam, H. and Nacken, H. 2004. Fuzzy Model for Salmon Habitat. A
Diploma thesis, Aachen University of Technology.

513

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005




DSS FOR INTEDISCIPLINARY WATER
MANAGEMENT

Hani Nabhan SEWILAM
PhD, Senior Staff Scientist, Section of Engineering Hydrology,
RWTH Aachen University, Aachen 52074, Germany
Tel. +49 241 8025643, Fax. +49 241 8022702
sewilam@lfi.rwth-aachen.de

Heribert NACKEN
Prof. Dr.-Ing, Section of Engineering Hydrology,
RWTH Aachen University, Aachen 52074, Germany
Tel +49 241 8025273, Fax. +49 2418022701
nacken@lfi.rwth-aachen.de

Indexing terms: Environmental indicators, Decision Support System, participatory
management, water management.
ABSTRACT
Water resources management involves complicated social, organizational,
legal and economical issues in addition to the undoubtedly important technical
matters and environmental aspects. Management decisions have potential to be
controversial because the involved groups (ecologists, economists, hydrologists
and sociologists) hold distinct interests and unique objectives. Therefore, the
development of decision support systems (DSS) faces great difficulties not only
because of the multiple objectives and multiple participant decision making
situations but also due to the complexity of participating an team
interdisciplinary in the modelling processes. For instance, while ecologists use
natural languages and qualitative reasoning for the description of ecological
relationships, hydrologists communicate in the form of systems of differential
equations or analytical models.

The proposed concept suggests identifying specific performance
indicators to quantify the multiple objectives and assess the management
scenarios under investigation. The indicators are classified into ecological,
514

economical, social and technical indicators. To overcome the problem of
multiple participants, the Analytical Hierarchy Process (AHP) is implemented
to assign weights to the involved decision makers with respect to their
experience and background. The management scenarios will be assessed and
ranked according to the predefined performance indicators with respect to the
opinion of decision makers taking into consideration their assigned weights.
Fuzzy logic (FL) is used to facilitate transformation of qualitative knowledge
into mathematical assessment models. FL is also implemented to model and
aggregate the performance indicators. Finally, the paper introduces a case study
for implementing this concept for the management of irrigation schemes.

CONFLICT IN WATER MANAGEMENT
Many of the water systems all over the world are performing far below their technical
and financial potentials. Deterioration of water systems due to management obstacles
necessitates improving the management strategies and concepts to cope with the challenge of
providing the steadily increasing population with enough food during the new century. For
instance, management of large irrigation schemes suffers from a lack of participation between
water officials and farmers. The traditional top down management approach has failed to aid
farmers [1], because farmers often reject to implement bureaucratically decided seasonal
plans that rarely meet their objectives. There is also significant conflict associated with
integrated management of river basins because of the involvement of interdisciplinary parties
that have conflictive objectives. Often maximizing an environmental objective means
minimizing another economical objective and vice versa. For example, protecting water
quality by reducing the applied amount of pesticide leads to reducing the crop production per
unit area. In this case the environmentalists struggle to implement such strategy, because
farmers are against it. Many other examples are reported in managing water systems not only
between water users and officials but also between water official themselves.

The main challenge that will be dealt with in this paper is to propose a procedure that
enables to overcome the mentioned issues and supports the involved parties in performing the
participatory planning. In this paper, a framework for water managers and water users to
participate in setting seasonal plans for the managing of water systems will be introduced.
The suggested framework enables integrating decision makers who represent all parties
involved in water management (environmentalists, economists, engineers, sociologist, and
farmer-representatives) in the planning process.

DECISION MAKING FRAMEWORK
In general, the suggested framework that gives each of the involved management
parties the right to participate in setting management plans, assessing them and selecting the
one to be implemented. An analytical approach is suggested and implemented [2] to identify
the locations of conflict. The analysis identified the required aspects to be considered while
formulating decision alternatives and how performance indicators can be developed. The
indicators are essential to assess the performance of the management with respect to
predefined objectives. It is also recommended to test the feasibility of the suggested scenario
with respect to management restrictions before spending time for its analysis.

515


As shown in Fig (1), the decision makers together form a committee that suggests
different planning alternatives (scenarios). Each planning alternative X which satisfies the
restrictions R represents a feasible planning scenario.
..... ..... , ,
3 2 1 i
x x x x X =
(1)
where X is a decision alternative, and x is the i-th decision variable (e.g. x
1
=water fee
$/m
3
, x
2
=type of fertilizers, x
3
=allocated water m
3
/ha, etc.).
R X where:
{ } 0 ) ( , 0 ) ( , 0 ) ( , 0 ) ( = X r X r X r X r X R
leg nat tech hum
with: (2)
hum
r
=human restrictions such as preferable working time,
tech
r
=technical restrictions such as canal max. capacity,
nat
r
=natural restrictions such as soil type,
leg
r
=legal restrictions such as allowed fertilizers.
Based upon the detailed analysis of the feasible alternative X
j
, the corresponding
performance indicators will be estimated.
) (X Analysis Y = (3)
where Y is the set of indicators resulted from the simulation to assess the scenario
under observation.
{ }
44 12 11
.... ,... , Y Y Y Y Y
ki
= (4)
Formulate next Alternative
j=j+1
Satisfy
Restrictions?
Alternative
X
j
yes
no
Simulation (X
j
)
Performance
Indicators (Y
ki
)
j
Assessment (Y
ki
)
j
Last
Altarnative?
Ranking Score V
j
yes
no
Rank Alternatives
Decision
Formulation
of
Assessment
Opinions
D
a
t
a

B
a
s
e
Planning Committee
D
a
t
a

B
a
s
e

Figure (1) A Decision Making Framework

516


DECISION SUPPORT SYSTEMS
There are many definitions of what precisely a Decision Support System (DSS) is. A
definition that could be acceptable has been given by Adelman [3] who defined a DSS as an
interactive computer program that utilizes analytical methods, such as decision analysis,
optimization algorithms, and so on, for developing models to help decision makers formulate
alternatives, analyze their impacts, interpret and select appropriate options for
implementation. In order to assist water decision makers to carry out the proposed decision
making framework a DSS that includes the following subsystems should be realized:
a. Model Management Subsystem: This component will be essential to
analyze the suggested decision scenarios. It might include financial, statistical,
management science, or other quantitative models that provide the DSS analytical
capabilities, and an appropriate software management.
b. Knowledge Management Subsystem: This subsystem includes the
opinion of the decision makers that is essential to assess the proposed decision
scenarios.
c. Data management Subsystem: includes the database, which contains
the necessary data for the other subsystems which is managed using a database
management system (DBMS).
d. User interface: this is the dialogue subsystem that enables the users to
communicate with the DSS. A considerable attention should be played while
developing this component since it is the criteria that attracts the user whether to
use the system or not.

PARTICIPATORY ASSESSMENT OF MANAGEMENT SCENARIOS
This part deals only with a knowledge management subsystem as one main
component for an irrigation management DSS. The knowledge management subsystem can
be considered as an assessment model of the suggested decision scenarios. The particular
problem that should be covered by this model is threefold:
1) considering the multiple interdisciplinary criteria (technical, social,
environmental and economical).
2) considering the conflictive assessment opinions of the involved
decision makers while assessing their suggested scenarios.
3) mathematical handling of the qualitative expressed opinions.

The performance indicators considered to assess the overall performance of the
suggested management scenarios are listed in Table (1). To cope with the interdisciplinary
criteria, this work suggested a three-level hierarchical assessment approach. As stated in
Table (2), the set of indicators {Y
ki
} to be estimated through simulation for each management
scenario are placed at the bottom level of a hierarchical structure. The indicators {Y
ki
} should
be rated and aggregated to form four indicators {Y
k
}, where k = 1,2,3,4 which represent
Economical, Technical, Social and Environmental indicators respectively. The intermediate
indicators {Y
k
} are placed at the medium level of the hierarchy and (Y
0
) is the overall
indicator to be estimated at the hierarchy top. The value of the overall indicator represents a
ranking score for the management scenario under investigation.


517


Table 1. Performance indicators [4, 5, 6, 7 and 8].




























Table 2. The Concept of Hierarchical Assessment
Performances Indicators Assessment
Ranking Score :
0
Y

Overall Aggregation
Intermediate Indicators:
{ }
k
Y Y Y ,... ,
2 1

Relative Rating &
Aggregation
Simulation Outputs :
{ }
ki
Y Y Y ,..... ,
12 11

Absolute Rating &
Aggregation


ANALYTICAL HIERARCHICAL PROCESS (AHP)
To deal with the second problem of the multiple participants, [9, 10] approach is
implemented to assign a weight for each of the decision makers. The following five criteria
have been used to estimate these weights: personal knowledge, performance measure,
personal interest and represented public. Let the DM
1
, DM
2
DM
J
be the set of involved
decision makers and C
1
, C
2,
C
5
are the mentioned set of evaluation criteria. Based on the
five criteria the weights of DMs will be obtained. The approach is summarized in four main
steps as follow:







I
NDICATORS
TERM
11

A
rea Based
Productivity
CO ) (
($) Pr
ha Area Irrigated
oduction

12

W
ater Based
Productivity
CO ) (
($) Pr
3
m byET consumed water of Volume
oduction

13

Fi
nancial-self-
sufficiency
CO
($) Re &
($)
quirements M O Total
Fees Collected Fund National +

21

Y
ield
Performance
EC
) ( arg
) (
ton Yield et T
ton Yield Actual

22

R
elative Water
Supply
EC
) ( inf
Percol Seepage piration Evapotrans
mm all Ra Irrigation
+ +
+

23

C
onveyance
Efficiency
EC
) (
) ( Re
3
3
m Water Irrigation Total
m quirement Irrigation Crop

31

In
terquartile
Ratio
OC
Re
Re
Quar Supplied Worst by ceived Water
Quart Supplied Best by ceived Water

31

E
mployment
Generation
OC
Days Working Official Number Annual
Schem in Labor ha Days Person Annual /

41

C
hange of EC NV
old
old new
EC
EC EC

42

P
otential of
Pollution
NV
) (
(
ton Nutrient Added of Amount
Farms the Leaving Nutrient of Amount

43

C
hange of GW-
table
NV ) (
) ( ) (
m depth old
m depth old m depth new

44

S
ediment
Yield
NV
) (
dim
ha Area Farm
Far the Leaving ent Se of Amount

518














Fig 2. The Importance
Hierarchy of DMs





Step 1: A three-level hierarchy has been constructed as shown in Fig 2. The involved
decision makers have been placed at the bottom level and the top goal is the DMs weight
{ }
DM
w
. In the intermediate level the five mentioned criteria have been placed.

Step 2: Pairwise comparison matrixes have been built to assess the importance of the
elements of this level with respect to each criteria of the upper level. The proper question in
the pairwise comparison is of the form: considering DM
1
and DM
2
of the bottom level, how
much more important is DM
1
compared to DM
2
with respect to their Personal Knowledge
(C
1
). Using this pairwise comparison approach, 5 matrices (J X J matrix) are constructed to
quantify the judgment of the comparison between the different involved decision makers with
respect to the five criteria of the level above (C
1
, C
2
, .. C
5
respectively).

Step 3: By repeating the same procedure of Step 2, another matrix is constructed to
compare the criteria themselves (C
1
, C
2
, , C
5
) with respect to the top-goal.

Step 4: Once all the six weighting vectors are estimated, they will be multiplied
appropriately through the branches of the hierarchy to determine the overall set of weights of
the bottom layer with respect to the top goal.
The described four steps were applied to assign weights for 8 fictitious decision
makers. For the pairwise comparison reasonable values were selected that reflect the opinion
of the author. The resulted weights are listed in Table (3).












Top
Goal
DM
1
DM
J
DM
2
1
1
w 1
2
w
1
3
w
1
4
w
1
5
w
1
w
1
C
2
C
3
C
4
C
5
C
2
w
3
w
4
w
5
w
{ }
DM
w
519


Table 3. The Weights of the DMs
CRITERIA

DM1 : Economical Manager
0
.151
DM2 : Technical Manager
0
.156
DM3 : Social Manager
0
.137
DM4 : Environmental Manager
0
.114
DM5 : First Farmer Representative
0
.110
DM6 : Second Farmer
Representative
0
.154
DM7 : Third Farmer
Representative
0
.100
DM8 : Forth Farmer
Representative
0
.077

NEUROFUZZY ASSESSMENT MODEL (NFAM)

A neurofuzzy model was constructed to assess the proposed scenarios. Fuzzy logic
was considered to overcome the third problem of uncertain and qualitative opinions of the
decision makers. The hierarchical structure of the performance indicators is horizontally
represented as show in Figure (3). The small blocks on the left side represent the two
absolute-rating layers. The large five blocks in the middle of the screen are the rule-blocks,
which perform relative rating and stepwise aggregation of the values of input- and
intermediate-indicators. The small block on the right side is the output interface of the overall
performance indicator. Each of the indicators was represented using fuzzy membership
functions with three linguistic terms. A voting system was implemented to formulate the
training sets for the model. The assigned weights of the decision makers were considered
while formulating the training sets. The model was trained with the aim to adjust the
membership functions and assign weights to the fuzzy rules. It can be noticed that the
membership functions of the indicator productivity/unit area have been adjusted after
training. Also each rule has a weight between (0-1) that reflects its importance. Once the
models components were adapted, it was tested to measure its performance through
assessing different decision scenarios.
520




CONCLUSION
Decision support systems are powerful tools to deal with the main problems
associated with planning of large water systems and managing their schemes. The
management of such systems faces many obstacles such as the interaction between
economical, social, technical and environmental aspects. The considerable conflict between
the involved interdisciplinary decision makers leads to deteriorating such schemes.
Developing DSS for managing water aspects faces also the difficulty of uncertainty and the
unavailability of crisp knowledge.
The knowledge management is a critical and important component of any DSS.
NFAM is a model that can be integrated to irrigation management DSSs as a knowledge
management subsystem. The NFAM is an attempt to integrate the Analytical Hierarchy
Process (AHP), Artificial Neural Networks (ANN) and Fuzzy Logic (FL) in a comprehensive
approach that overcomes the problems encountered in participatory planning. The developed
NFAM encourages the further development of DSSs for participatory management of water
systems. Automatic knowledge elicitation using ANN enables the fast generation of rule-base
of assessment models. Such an approach is also recommended to be implemented by the
ecologists to develop rule-based habitat prognoses models.

REFERENCES
[1] Pallas, P. (1993). Water and Sustainable Agricultural development: The Role
of Planning and Design of Irrigation and Drainage Systems. Trans. 15
th
Congress on
Irrigation and Drainage, Vol 1-J .pp.53-71.
[2] Sewilam, H.A.N. (2002) NeuroFuzzy Modelling for Conflict Resolution in
Irrigation Management. PhD Thesis, Aachen University of Technology, Germany.
[3] Adelman, L. (1992). Evaluating Decision Support and Expert Systems, J ohn
Wiley and Sons, New York.
[4] Bos, M.G., D.H. Murry Rust, D.J . Merry, H.G. J ohnson, and W.B. Snellen
(1994). Methodologies for assessing performance of irrigation and drainage
management. Irrigation and Drainage System, pp. 231-261.
Fig 3. The trained NeuroFuzzy Assessment Model
521


[5] Molden, D., R. Sakthivadivel, Ch.J . Perry, Ch. de Fraiture and W.H. Kloezen
(1998). Indicators for Comparing Performance of Irrigated Agricultural Systems.
International Water Management Institute, Research Report 20, Colombo, Sri Lanka.
[6] Bos M.G., W. Wolters, A. Drovandi and J .A. Morabito (1991). The Viejo
Retamo secondary canal performance evaluation case study: Mendoza, Argentina.
Irrigation and Drainage Systems Vol.5, pp. 77-88.
[7] Abernethy, C. (1986). Performance Measurement in Canal Water
Management: A discussion. ODI/IIMI Irrigation Management Network Paper 86/2d.
[8] Levine, G. (1982). Relative water supply: An explanatory variable for
irrigation systems. Technical Report No. 6. Cornell University, Ithaca, New York,
USA.
[9] Saaty, T.L. (1980). The Analytic Hierarchy Process. Mc Graw-Hill,
International Book Company.
[10] Zio , E. (1996). On the use of the analytic hierarchy process in the aggregation
of expert judgements. In: Reliability Engineering and System Safety, Vol. 53, pp.127-
138.



522

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


APPLIED INDICATORS FOR EVALUATION THE
SOCIO-ECONOMIC AND AGRICULTURAL
ASPECTS RELATED TO WATER RESOURCES IN
EGYPT

Assem Afify
1
, Khalef Maherzi
2
, Mohammed Abdel Motaleb
3
and Atef
Hamdy
4


1
Research Associate Professor, Strategic Research Unit, National Water Research Center,
Egypt.
2
Research Assistant, Mediterranean Agronomic Institiute of Bari, Italy.
3
Research Professor, Director, Water Resources Research Institute, National Water Research
Center, Egypt.
4
Director of Research, Mediterranean Agronomic Institute of Bari, Italy.


ABSTRACT
A multi-disciplinary approach is getting to be an essential pre-requisite in
any water resources study or project in many countries. Egypt is a good
example of a developing country, where many projects are currently
progressing steadily. Particularly, in the field of agricultural development,
national projects are getting more emphasize from the government to secure
the country. In addition, water is very essential for any development,
particularly for agriculture. It is well known that Egypt is using about 85%
of the available water resources in the agricultural sector. Hence, it is very
important to study the socio-economic and agricultural aspects as related to
water. Evaluation for the overall status of the current socio-economic,
environmental, and water resources system at present requires a massive and
detailed data collection. This may not be feasible for the current study to
achieve. The proposed study, however, will tackle only the issue of
agriculture and socio-economic aspects.
The study aims at the evaluation of the socio-economic and agriculture sub-
systems as related to the use of water in agriculture. The main objective of
523
the study is to develop agricultural and socio-economic indicators for the
country on the national and regional levels. Water indicators are also
developed but for a sub-regional level, where only the Delta region is
considered as a separate case study.
The results from the study are presented to compare between the different
sub-regions of the Delta concerning the agriculture water availability. The
study also recommended developing a GIS/software package to facilitate
presenting the results from all different categories of indicators visually in a
better way.
Keywords: indicators, socio-economic, agriculture, water resources.

1 INTRODUCTION
Sustainable development indicators measure sustainability or sustainable
development performance. Most environmental indicators have a sustainable
development framework in which environmental, economic and social indicators are
linked together. Similarly, sustainable development indicators in Egypt may cover the
following categories: economic indicators, socio-economic indicators, environmental
indicators and water indicators. An example of economic indicator on the national level
is the Gross Domestic Product. It is calculated as the total of all sectors value at factor
cost [1]. Socio-economic indicators category reflects the interaction between the society
and the economy. For example, the economy will have a direct effect on the society in
terms of education, health, and employment. Therefore, an important indicator in the
socio-economic category is the employment fraction. The indirect link between the
economy and the society is through the environment. For example, excessive use of
pesticides by farmers or absence of domestic sewerage in rural areas will lead to
contamination in surface and ground water.
Another socio-economic indicator, commonly used in Egypt, is the grain security. It is
the security of the rice, wheat, and maize. The grain security is the fraction or
percentage of the national production of any type of grains to the local consumption of
this type [1]. If the security is 100% or more, this means that the nation consumption is
completely satisfied by the production of this type. If it is lower than 100%, this means
that the nation consumption is not satisfied by production and the nation needs to import
the deficit part.
An example for environmental indicators is the integrated water quality index for
industrial wastewater, which is determined based on the standards given by law No. 48,
1989 for six major parameters and their actual concentrations [2]. The parameters are
selected to reflect the adverse effects of the pollutants. These parameters are the
biological oxygen demand (BOD), the chemical oxygen demand (COD), heavy metals
(HM), oil and grease, suspended solids (SS) and the total dissolved solids (TDS). The
integrated water quality index for industrial wastewater can be estimated by applying the
root-mean-square for the different parameters indices [3].
Obviously, water plays a key rule in any development project in Egypt.
Therefore, water indicators should measure in the first stand sufficiency of water for
development at present, and also measure the availability of water for future
developments. The first can be described as the water security indicator. This is defined
as the fraction or the percentage of all the available water resources to the sum of all the
524
water requirements. The second indicator, expressed as water sustainability indicator,
highlights on the sustainability of the potential water resources in the future [4].
Indicators for Egypt's sustainable development were introduced and applied on
the Southern Valley project (Toshka) [5], as one of the major development projects in the
country. However, and in order to investigate the impacts of any other
development project, the actual state of the water resources system should be
first defined, and an over view of the related agricultural and socio-economic
environment must also be considered. However, detailed view of such water
resources, socio-economic and agricultural systems in a country cannot be
figuratively induced due to the complexity and the high number of elements that
might be considered. It is then clear that decision-makers must elaborate some
indicators to have an idea about the impact of different elements on the system.
The main goal of the proposed study is to evaluate the current status of the water
resources system through developing integrated indicators measuring water, agriculture,
social, and economic dimensions.


2 DEVELOPMENT OF THE INDICATORS
A description for the different categories of indicators and their development are
going to be discussed in this section. Three major categories of indicators are discussed
thoroughly; agriculture, socio-economic and water indicators. Agriculture indicators
considered in this study are the cropping pattern, crop yield and return of major cereal crops.
Agricultural ownership and powerstock fraction, and land use indicators are also considered.
Among socio-economic indicators considered in this study are: population and household
densities, employement fraction, owner size fraction, drinking water service, and sewerage
service indicators. Water indicators considered in this study are agriculture water needed and
available, and the fraction of agriculture water availability. The detailed description of each
indicator is discussed in the following sub-sections.

2.1 Agriculture Indicators
2.1.1 Cropping Pattern
Cropping pattern is defined as the fraction of a specific area used for a specific
crop in respect to the total area used for all the crops. It is calculated for the six most
important crops in Egypt: Rice, Maize, Sorghum, Wheat, Sugar cane and Cotton. Eqn (1)
is used to obtain the cropping pattern for each Governorate.

=
=
n
i
i
CA
c
CA
c
CP
1
(1)

In which, CP
c
is the cropping pattern for a certain crop, CA
c
is the cropped area for the
same crop, CA
i
is the cropped area for any crop i, and n is the number of crops cultivated
during the year.

2.1.2 Crop Yield
Crop yield is defined as the quantity of a crop produced by the unit area of
agricultural land (feddan). It is obtained for the most important crops (Beans, Nile maize,
Summer Maize, Wheat, Rice and Barley). Crop yield is estimated in ton/fedan [6] and is
calculated by Eqn (2).

525
crop this for Used Area
crop a by Produced Quantity
Yield = (2)
2.1.3 Crop Return
The crop return or the production output from a certain crop is defined as the
average amount of money per feddan that the farmer will get it out from the cultivation
of that crop. It is calculated by Eqn (3) for each Governorate in Egypt.

i
Qp
i
Yield
i
Fo = (3)

In which, Fo
i
is the production output for a certain crop i and Qp
i
is the price for one ton
of the same crop.

2.1.4 Agricultural Ownership
The agricultural ownership is divided into three sub indicators, the first one is the
agriculture ownership for farm size between 0 and 5 feddan, it is defined as the fraction
of the area used by small farms, so that the farm area dose not exceed 5 feddan. It is
calculated by Eqn (4) for each Governorate [1].

+
+

50
S
50 5
S
5 0
S
5 0
S
5 0
AOS (4)
Where:
AOS
0-5
is the agriculture ownership indicator for farms between 0 and 5 feddan, and S
0-5

, S
5-50
and S
50+
are the total area of farms, in which their individual farm size is (0-5
feddan), (5-50 feddan), and more than 50 feddan, respectively.

2.1.5 Agricultural Power Stock
Agricultural power stock is defined as the total number of people ready to work in
the agricultural field. It can be estimated based on census for each Governorate [1] for
the total power stock less the employees in all non-agricultural fields, Eqn (5).

Employees Ag Non - Stock Power Total Stock Power e Agricultur = (5)

2.1.6 Land Use Indicators
Three land use indicators are developed. The first one is called the inhabited area
indicator, the second is the agricultural land indicator, and the third indicator is the house
utility indicator [7]. These indicators are calculated using equations (6, 7, and 8).
area Total
area d Uninhabite
indicator area d Uninhabite = (6)
area Total
land e Agricultur
indicator land griculture A = (7)
area nhabited I
area house Total
indicator utility House = (8)




526
2.2 Socio-Economic Indicators
2.2.1 Population Density
The population density is defined as the number of inhabitants by square
kilometer [1]. An example for the population density of some governorates is shown in
table 1. It is calculated using Eqn (9).

e governorat the of area Total
e governorat a in population Total
Density Population = (9)

Table 1: Population density
Name of the Governorate
Density
(1/Km2)
Alexandria 1181
Aswan 28
Damietta 889
Ismailiya 160

2.2.2 Average Household Density
Average household density is defined as the average number of households living
in the same building [1]. It is calculated by dividing the total number of households in a
Governorate by the total number of buildings in the same Governorate, Eqn (10).
number Building
number Household
density househlod Average = (10)

2.2.3 Employment Fraction
Employment fraction is defined as the number of the employed divided by the
number of employees and the unemployed summed together [7]. It is calculated using
Eqn (11).
UnEmp Emp
Emp
Empf
+
= (11)

In which, Emp
f
is the employment fraction indicator, Emp is the total number of
employees in the Governorate, and UnEmp is the total number of the unemployed
personnel in the same Governorate.

2.2.4 Agricultural Power Stock Fraction
Agricultural power stock fraction is defined as the fraction of the people ready to
work in the agricultural field in respect to the total power stock in the whole
Governorate. This indicator is calculated using Eqn (12) for each Governorate in Egypt.

Stock Power Total
Stock Power e Agricultur
Stock Power al Agricultur of Fraction = (12)

2.2.5 Agricultural Owner Size Fraction
Agricultural owner size fraction [1] is defined as the fraction of small, medium
and larges fields in respect to the total number of fields or owners. Small fields are fields
with area less than 5 feddan. Medium fields have a total area from five to fifty feddan,
527
and large fields are fields with area greater than fifty feddan. Equation (13) is used to
obtain the agricultural owner size fraction for small fields:

more] and Fields[50 50] - Fields[5 5] - Fields[0
5] - [0 Fields
5] AOS[0
+ +
= (13)
Where,
AOS[0-5] is the agricultural owner size fraction of fields less than 5 feddan.
Fields[0-5] is the number of fields with an area of less than 5 feddan.
Fields[5-50] is the number of fields with an area from 5 to 50 feddan.
Fields[50 and more] is the number of fields with an area of more than 50 feddan.

2.2.6 Drinking Water Service
Drinking water service is a group of indicators, which is divided into three sub
indicators according to the type of drinking water connection. Households can be
classified according to the type of drinking water services available, where tap water can
be either inside the apartment, or only inside the building, or public tap outside the
building. In a certain Governorate for example, the apartment-water-supply percentage
for an urban region is defined as the percentage of households, with a supply inside their
apartment, to the total number of households. Based on figures obtained for 1996 census
[8], the three sub indicators are presented in figure (1) below for some governorates.

0%
20%
40%
60%
80%
100%
Dakahleya Fayoum Gharbia Giza
F
r
a
c
t
i
o
n

(
%
)
other type of water
supply
building water
supply
apartment water
supply
Figure 1: Water service indicators for the urban region, after [8].

2.2.7 Sewerage Service
Sewerage service is a group of indicators, which is divided into three sub
indicators according to the type of service connection with the buildings. Buildings can
be classified according to the type of service connection into three groups; public
sewerage, other, and no sewer. In a certain Governorate for example, the fraction of
buildings with no sewerage in a rural area is defined as the percentage of buildings
without a sewerage system to the total number of buildings for the same area. Figure 2
presents this group of indicators based on the available census data [1].

528
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Assiut Beni Suef Damietta Qalubia
F
r
a
c
t
i
o
n

(
%
)
No sewer
Other
Public
Figure 2: Sewerage service indicators for the rural region.

2.3 Water Indicators
2.3.1 Agriculture Water Needed
Agriculture water needed for any crop, CW
i
is defined as the amount of water
needed per unit area (feddan) for the full growth of the crop. Similarly, agriculture water
needed for a certain Governorate can be obtained as the total quantity of water needed to
irrigate all the cultivated crops divided by the total cultivated area. This indicator was
estimated based on the amount of crop water requirements at the farm level, where it
dose not count for the conveyance losses due to seepage and evaporation [9]. By this
assumption, Eqn 14 is used to calculate the agriculture water needed for the whole
Governorate on monthly basis. Data used to calculate this indicator are obtained from
the Irrigation Sector for the 1997 year [10].

=
i
CAi
i
i W C CAi
Wn (14)

2.3.2 Agriculture Water Available
Water available is defined as the quantity of water that is delivered to the field,
which gives an estimate for the quantity of water that the farmer use every month to
irrigate his field. Due to the lake of data, this indicator will be calculated only for three
zones which represent the east, the west, and the middle Delta. To do this, first, three
different zones (see figure 3), with every canal that enters each zone were determined.
The quantity of water allocated to each zone is obtained by summation of the flows
passing trough these canals using the following formulas [11].

Rayah nasery - El Rayah behary - El
1 F F Fz + = (West Delta)
Ryah Twfeky Basousia Shrkawia Esmailia
2 F F F F Fz + + + = (East Delta)
Rayah abassy - El nagayl - El Rayah mounofy - El
3 F F F F Fz
Darwa
+ + + = (Middle Delta)

Where:
F
Z1
is the flow allocated to Zone 1, West Delta,
529
F
Z2
is the flow allocated to Zone 2, East Delta,
F
Z3
is the flow allocated to Zone 3, Middle Delta, and
F
any canal
is the flow allocated to any canal.

Figure 3: The delta region.


2.3.3 Fraction of Agriculture Water Availability
The fraction of agriculture water availability is defined as the fraction of the
available water in respect to the needed water for each Zone. This indicator is calculated
for each zone on monthly basis using Eqn (15).

m
Wn
m
Wa
m
WAF = (15)

In which, WAF
m
is the fraction of agriculture water availability for any month, Wa
m
is
the amount of water available, and Wn
m
is the amount of water needed for the same
month. Available water in each zone can be obtained as the amount of water allocated to
each zone, F
Z
less the amount of seepage losses, where seepage losses can roughly be
estimated for canals with large flows only [11].

3 RESULTS AND DISCUSSION
In the previous section, the development of the different categories of indicators
was illustrated. Results from selected indicators are demonstrated in this section for each
category of indicators.

3.1 Agriculture Indicators
3.1.1 Crop Yield and Crop Return Indicators
Among the different agriculture indicators developed are the crop yield and the
crop return. The results from these two indicators are presented in figures 4 and 5.
Yields of summer crops (rice and maize) are higher than yields for winter crops (wheat,
barley and beans). It is also clear from figure 4 that Fayoum Governorate has recorded a
fewer yield than the two other Governorates for all summer crops and beans. Regarding
crop return, rice has remarkably higher return than the rest of all other crops, reaching
530
about 2500 LE/feddan. For maize barley and wheat, Gharbia has indicated higher return
than other Governorates.

3.1.2 Land Use Indicators
Land use indicators defined by equations (6, 7, 8) are presented in figure (6). As
indicated by figure 6, the agricultural land fraction to the total area is the highest for both
Dakahleya and Gharbia governorates, where it reaches 0.82 and 0.86 respectively. These
two governorates have zero value for the uninhabited fraction with respect to the total
area. Aswan Governorate has the highest record for the uninhabited fraction of 0.97.
Regarding the house utility fraction with respect to the inhabited area, Cairo has the
highest record of 0.87 compared to the other Governorates.

3.2 Socio-Economic Indicators
3.2.1 Employment and Agricultural Power Stock Fractions
The results from applying Eqns (11, 12) are plotted in figure (7) for the
employment and the agricultural power stock. The denominator in both equations is
basically the same, which is the total power stock. It is obvious from the figure that there
is no big variations among the results for these two indicators. The employment fraction
is around 0.9 for all governorates, slightly above for Red Sea and below for Dakahlia.
Agricultural power stock is around 0.4 for Dakahlia, and slightly less for other
governorates.

3.2.2 Agricultural Owner Size Fraction
The results from applying Eqn (13) are plotted in figure (8) for the agricultural
owner size fraction of fields less than 5 feddan, AOS[0-5]. Similarly, the fractions for
other land size classes are obtained and plotted as well. The results as shown in figure
(8) indicate that the agricultural owner size fraction for the first class, fields of less than 5
feddan, is higher for the fractions of the second and third class except of Ismalia, where it
shows a fraction of about 60% of owners whose fields are more than 50 feddan (third
class).

3.3 Water Indicators
The results for the agricultural water availability, as obtained by Eqn (15) on
monthly basis, are plotted for the three zones (West, East, and Middle Delta). Figures (9,
10 and 11) demonstrate these results for West, East, and Middle Delta. The fraction of
agricultural water availability indicates shortage of water whenever its value is less than
one, and water sufficiency whenever its value is greater or equal one. It can be
concluded from these figures that both East Delta and Middle Delta suffers from water
shortage during summer months (J une through September). The Middle Delta suffers
from water shortage for longer period starting from March. The shortage during summer
may be due to excessive rice cultivations in these two zones.

4 CONCLUSIONS AND RECOMMENDATIONS
Evaluation of the socio-economic, agriculture, and water aspects as related to the
water resources system in Egypt is not an easy process. This is possible through the
development of some indicators in each category. Massive data collection for the raw
data was essential from different sources to evaluate the indicators. Bits of information
and data about the population, households, drinking water services, and sewerage
services were collected. In additions information about the distribution of land use,
531
water distribution, agricultural ownership and size, crop yield and cropping pattern were
all collected and analyzed.
This huge amount of information was analyzed in this paper by presenting the
results for a limited number of indicators for each category (socio-economic, agriculture,
and water). Sample of the results are shown for selected governorates only. Water
indicators are presented for the Delta region, where it is shown for three zones only
(West, East, and Middle Delta). It is recommended to build a GIS application to be able
to present the results for all the governorates of Egypt in an easy and user-friendly way.

REFERENCES
[1] CAMPAS, Statistical Yearly Book 1992-1996 and 1996-1997, Arab Republic of
Egypt central Agency for Public Mobilization and Statistic.
[2] DSS, 1999, Review of EWRSES Existing Indicators, Working Paper, Report NO
9901-15 DSS, Strategic Research Unit, National Water Research Center, 1999.
[3] Simonovic S. and Fahmy H. 1999, A New Modeling Approach for Water Resource
policy Analysis, Water Resources Research, Vol. 35, No 1, J an 1999.
[4] RNPDII, 1997, Nile Water Strategic Research Unit, Final Report, River Nile
Protection and Development Project, Phase II, National Water Research Center,
Report No A2090211-9705-01-FR.
[5] Kheireldin K., A. Afifi and I. El-Gafy 1999, Review of Applied Indicators for Egypt
Sustainable Development. International Conference on Integrated Management of
Water Resources in the 21
st
Century, Cairo, Egypt, Nov. 1999.
[6] AEB, Agro economic Bulletin. Ministry of Agriculture and Land Reclamation, Year Book
in Arabic, 1997.
[7] EDB, 1997, Egypt Description by Numbers. Information Center and Decision Support of
the Prime Minister Head Cabinet, Year Book in Arabic, 1997.
[8] METAP, 1997, Water supply and Sanitation in Rural Egypt 1997, Mediterranean
Environmental Technical Assistance Program (META/PPU). Final Report, Egypt
Nov. 1997.
[9] Afify, A. M. and Alaa E. Abdin. Crop-Water Management for Intensive Production
under Restricted Water Supply. Proceedings of the International Conference on
Water Resources Management in Arid Regions, Kuwait Institute for Scientific
Research, Kuwait, March 23-27, 2002.
[10] MWRI, personal communication, Head of Irrigation Sector, Ministry of Water
Resources and Irrigation, Sep. 2000.
[11] Maherzi, Khalef 2002, A GIS Platform for the Evaluation of a Water Resources
System (Case Study from Egypt). M. Sc. Thesis, International Center for Advanced
Mediterranean Agronomic Studies. CIHEAM, BARI, Italy.
532
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Wheat Barley Summer rice Summer
maize
Nilli maize Beans
Y
i
e
l
d

(
t
o
n
/
f
e
d
d
a
n
)
Fayoum
Gharbia
Kafr El-Sheick

Figure 4: Crop yield of different crops.

0
500
1000
1500
2000
2500
3000
C
r
o
p

R
e
t
u
r
n

(
L
E
/
f
e
d
d
a
n
)
Wheat Summer maize Rice Nile maize Barley
Aswan
Damietta
Gharbia
Suez

Figure 5: Crop return of different crops.

0
0.2
0.4
0.6
0.8
1
1.2
House utility Agr Land Uninhabited
Land Use Indicators
F
r
a
c
t
i
o
n
Aswan
Cairo
Dakahleya
Damietta
Fayoum
Gharbia

Figure 6: Land use indicators.

533
0
10
20
30
40
50
60
70
80
Assiut Beheira Ismailiya Menia
O
w
n
e
r

s
i
z
e

(
%
)
0 to 5
5 to 50
50 and more
Figure 7: Agricultural owner size fraction for different land classes.

0 0.2 0.4 0.6 0.8 1
Dakahleya
Giza
Ismailiya
Menia
Red Sea
F
r
a
c
t
i
o
n
Agricultural
power stock
fraction
Employment
Figure 8: Fractions of agricultural power stock and employment.

Zone 1: West Delta
0
0.5
1
1.5
2
2.5
3
3.5
J
a
n
u
a
r
y
F
e
b
r
u
a
r
y
M
a
r
c
h
A
p
r
i
l
M
a
y
J
u
n
e
J
u
l
y
A
u
g
u
s
t
S
e
p
t
e
m
b
e
r
O
c
t
o
b
e
r
N
o
v
e
m
b
e
r
D
e
c
e
m
b
e
r
Months
F
r
a
c
t
i
o
n

Figure 9: Fraction of agricultural water availability in the West Delta.

534
Zone 2: East Delta
0
0.5
1
1.5
2
2.5
3
3.5
J
a
n
u
a
r
y
F
e
b
r
u
a
r
y
M
a
r
c
h
A
p
r
i
l
M
a
y
J
u
n
e
J
u
l
y
A
u
g
u
s
t
S
e
p
t
e
m
b
e
r
O
c
t
o
b
e
r
N
o
v
e
m
b
e
r
D
e
c
e
m
b
e
r
Months
F
r
a
c
t
i
o
n


Figure 10: Fraction of agricultural water availability in the East Delta.

Zone 3: Middle Delta
0
0.5
1
1.5
2
2.5
J
a
n
u
a
r
y
F
e
b
r
u
a
r
y
M
a
r
c
h
A
p
r
i
l
M
a
y
J
u
n
e
J
u
l
y
A
u
g
u
s
t
S
e
p
t
e
m
b
e
r
O
c
t
o
b
e
r
N
o
v
e
m
b
e
r
D
e
c
e
m
b
e
r
Month
F
r
a
c
t
i
o
n

Figure 11: Fraction of agricultural water availability in the Middle Delta.
535

536

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005








THERMAL POLLUTION IN SHALLOW SEA

Y. Shawky
(1)
and S. Abdel- Fattah
(2)

(1) Researcher, Hydraulics Research Institute, National Water Research Center.
(2)Associate Prof., National Water Research Center, Ministry of Water
Resources and Irrigation, Egypt.
1. ABSTRACT
Ayoun Moussa Power Plant, AMPP, is located at the Suez Gulf in Egypt with a
capacity of 640 MW. The aim of this research work was to determine the optimal
intake/discharge structures configurations of the power plant, which will result in
minimizing the thermal pollution in the Sues Gulf due to the disposal of hot water,
resulted from the cooling process into it. For this purpose, two physical hydraulic
models representing the area of AMPP at the Suez Gulf were constructed at the
Hydraulics Research Institute, Delta Barrage, Egypt. The first model was
constructed in order to study the thermal plume behavior at the near field zone.
The model had undistorted scale of 1:25 and represented 840 m of the shoreline
and 650 m normal to it, together with the proposed discharge (outfall) structure.
The second model was constructed to study the thermal plume behavior at the far
field zone. The model had distorted scale of 1:100 horizontally and 1:20 vertically.


The results of the near field model investigations indicated that the dilution of the
thermal plume was achieved in a relatively short distance. Its maximum distance from
the discharge structure was 190 m for the counter line, which has 3
o
C increase above
the ambient temperature. The hydraulic tests of one or two units in operation showed
reasonable mixing characteristics, even with the low tide condition. The inclination of
the outfall structure resulted in minimizing the plume travel toward the intake
structure. It also caused traveling of the hot water more to the Northwest direction. In
order to protect the Gulf bed downstream the outfall structure from scour, a riprap
with D
50
= 0.32 m was recommended to place in an area, which extends 187.5m from
the outfall structure. With respect to the far field model, the hydraulic investigations
results indicated that, for the discharge structure, the maximum temperature rise
measured at the vicinity of the intake canal was 1
o
C. No hot water recirculation was
measured or observed in any tests, even with tests of temperature buildup due to tidal
537
cycles. No movement was observed in rip-rap either over the buried discharge pipes
or in front of the structure. For the intake canal, the canal inclination was designed in
such a way to avoid hot water withdrawal. Canal alignment showed a good flow
characteristic, as the velocity of the lower layer was much faster than the upper layer.
This also proves that the designed bed level was suitable to keep the stratified flow
and prevent shallow water effect. The distance between the intake canal and outfall
structure was far enough to prevent hot water recirculation. No riprap movement was
observed in any of the tested types.

2. INTRODUCTION
Ayoun Moussa Power Plant (AMPP) is located on the eastern-north coast of the Suez Gulf and it
uses the gulf water for its cooling system [8]. The plant consists of two power generation units,
with total capacity of 640 MW. It has once-through cooling system with an offshore discharge
structure. The cooling water discharge is 25.92 m
3
/s, and the heat rejected to the Gulf (Q*T*
C
p
), where C
p
is the specific heat of the water [3] is approximately 925 MW (3570 x 10
6
BTU/hr)
at a water temperature rise of 8.5
o
C.

3. PURPOSE OF STUDY
The main aim of this study is to determine the intake/discharge structures configurations, which
will result in minimizing the thermal pollution into the Suez Gulf due to the disposal of hot water
into it from the plant. The minimization of thermal pollution is essentially to fulfil the
requirements of the environmental criteria by the Ministry of Water Resources and Irrigation Law
No. 48, 1982. This law specifies that a maximum a temperature rise above ambient of 5
o
C, and a
maximum discharge temperature of 35
o
C. Also, it is required to prevent the recirculation of warm
water to the plant inlet and prevent the bed erosion at the vicinity of the plant.

The study enables the optimization of the discharge structure location and dimensions, the rip-
rap gradation and extension, intake canal alignment, and skimmer wall location and dimensions.
For this purpose two physical models were constructed. The near field model which has distorted
scale of 1:25, and the far field one which has distorted scale of 1:100 horizontally and 1:20
vertically.

4. SITE AND PLANT CHARACTERISTICS
4.1 Site Characteristics
According to Reference [8], the bathymetric condition in the vicinity of the site was developed
from the survey performed by Suez Canal Institute in July 1995. Tide levels are measured for the
period of Jan. 1994 through Sep. 1995. From these measurements, the following tide levels are
obtained:

Mean Sea Level Datum (MSLD) = 0.0 m, Mean Water Level (MSL) = 0.25 m MSLD, Minimum
Water Level = - 0.85 m MSLD, Mean Lower Low Water (NLLW) = + 0.65 m MSLD, Mean
Higher High Water (MHHW) = + 1. 10 m MSLD and Maximum Water Level = + 1.35 m MSLD

From the data presented in hydrographical survey report, the current velocity in the vicinity of
intake inlet and discharge outlet ranges from 0.04 to 0.12 m/s with an average of approximately
0.08 m/s. The measured maximum current velocity is approximately 0.2 m/s. The current
directions at intake and discharge structures location are toward northwest (NW) and southeast
538
(SE) with predominant direction being toward southeast. The monthly average sea water surface
temperatures in the Gulf of Suez range from 16.7
o
C in February to 27.2
o
C in August. The
maximum and minimum sea water temperatures are approximately 29
o
C and 15.4
o
C,
respectively.

4.2 Plant Characteristics
The proposed design concept consists of a dredged channel with dikes for the intake and buried
pipes for the discharge outfall. The intake is furnished with a skimmer wall at the shoreline and
the discharge outfall is furnished with a rectangular opening in a sheet pile wall at the outlet. The
following provides a description of the intake and discharge structures.

4.2.1 Description of the Cooling Water Intake
The intake consists of dredged channel with protective earth filled dikes and a skimmer wall
located at the shoreline. The flow from the inlet channel is conveyed to the plant by two buried
pipes each 3 m in diameter. The earth dikes are covered with rip-rap over a filter layer. The
dredged intake channel is trapezoidal in shape and extends offshore to a distance of about 1,600
m to a water depth of 5.50 m measured from mean sea level datum (MSLD). Natural sea bed
elevations along the alignment vary from + 0.5 m MSLD at the shoreline to - 1.0 m MSLD at a
distance of about 1,000 m from the shoreline. The height of the dikes varies from about 1.0 m at
the shoreline to about 2.5 m at a distance of about 1,000 m from shoreline.

To reduce the amount of heat recirculation, a skimmer wall is required. If the findings of the
thermal physical model study require the placing of skimmer wall at the channel inlet, the
protective dikes described above need to be extended further offshore to a suitable depth
determined from the model study.

4.2.2 Description of the Discharge Outfall
The discharge outfall consists of two buried discharge pipes extending offshore, approximately
1,200 m, to a location with sea bed elevation of - 3.25 m MSLI. The two pipes are connected to a
rectangular bow with opening made in a sheet piles wall at the outfall location. The discharge
pipes are buried below the sea bottom, and are protected by rip-rap as needed to prevent erosion
by tide currents and wave action.

The rectangular opening in the sheet piles wall is designed to provide a surface discharge with
uniform high exit velocity at all tide levels. This concept will create rapid initial mixing thus
reducing the size of the high surface temperature rise isotherms and also minimize local erosion
of the bottom sediment. The Gulf bed in the vicinity of the discharge weir box is protected by
riprap with filter layer.

Based on the above concept, the opening is 7.5 m wide and 1.5 m high, located in water depth of
3.25 m measured from mean sea level datum. The opening invert elevation is at 2.25 m (MSLI),
one meter above the sea bottom. The selected configuration provides a uniform exit velocity of
2.3 m/s for all tide levels. The estimated extent of the erosion was determined from preliminary
mathematical model study [8]. The exact extent of the riprap and opening size are to be
determined from the physical model study. The model study must be based on movable bed and
must consider the cross currents, and gradation of the bed bottom sediment.
539
5. PHYSICAL MODELS
5.1 General
For correct reproduction of the important hydraulic and thermal phenomena in a hydraulic
model, a number of requirements must be fulfilled when determining the model scales.
Unfortunately, it is not possible to find out a set of model scales, which allows reproduction of
all hydro-thermal processes simultaneously. The division of the area of investigation into two
zones often provides a practical solution of this problem.

- Near field mixing zone, which is the area close to the outfall structure, where free turbulence is
created by the shearing action between hot water discharge and the ambient water causing the jet
diffusion of the heated water. The entrainment of ambient water in the near field and the
buoyancy of the thermal discharge will force the hot water towards the water surface, resulting in
a stratified flow.

-Far field heat loss zone, which is the area where the heat is distributed through diffusion caused
by ambient turbulence and also where heat loss through the air-water interface is effective. Here
the heat transfer to the air is proportional to the local temperature difference between the water
surface and the natural water temperature, which is function of the air temperature, solar effect,
wind speed, and relative humidity. The transition between both zones is determined by the
geometry of the outfall structure, and characteristics of the receiving water as well as flow
conditions.

5.2 Near Field Model
The general test philosophy was to test the design of the discharge structure for a range of the
gulf tidal currents and levels for one and two units in operation. In the model study, not only the
possible combinations of tidal conditions were tested, but also some assumed conditions were
tested for better analysis of the model results. Figure (1) shows the near field model layout

The near field model test program is shown in Table (1). For every run, surface horizontal plume
temperature, vertical temperature distribution at three profiles, and the velocity distribution at the
discharge vicinity were measured. The average horizontal temperature at the model inlet was
used as a back ground temperature to calculate the temperature rise and recirculation. Dry air
temperature, relative humidity, and wind speed were measured at a point, which is 2.5 m above
the model surface.
Figure (1) The Near Field Model Layout

540
Table (1) Near Field Model Test Program

Current Test
No.
Season No. of
Units
Tidal Level
MSL (m) Coming From Speed
(m/s)
Comments
1 Winter 2 - 0.85 NW to SE 0.08
2 Winter 2 - 0.85 SE to NW 0.08
3 Winter 2 + 1.35 NW to SE 0.00
4 Winter 2 + 1.35 SE to NW 0.08
5 Winter 2 + 0.25 ---- 0.08
6 Summer 2 - 0.85 NW to SE 0.00
7 Winter 2 - 0.85 NW to SE 0.08
8 Summer 2 + 1.35 SE to NW 0.08
Tests Carried out to
design the rip-rap
protection
9 Summer 2 - 0.85 NW to SE 0.00
10 Summer 2 + 1.35 SE to NW 0.08
11 Summer 2 + 0.25 ---- 0.08
12 Summer 2 - 0.85 NW to SE 0.00
13 Summer 1 - 0.85 NW to SE 0.08
14 Winter 1 - 0.85 NW to SE 0.08

5.2.1 Model Results
5.2.1.1 Dilution of the Thermal Plume
The temperature rise T
x
at each point x is calculated using the following equation
[ ]
m b x
m i o
p i o
x
T T
T T
T T
T

=
) (
) (
(1)
Where: T
o
: The outfall temperature, T
i
: The intake temperature, T
x
: Temperature at point x, T
b
:
Back - ground temperature, P: Prototype, M: Model

Since the size of the thermal mixing zone is not specified in the Egyptian environmental criteria,
it dose not appear that the mixing zone size is the controlling factor in selecting an outfall
structure alternative. However, the study considered minimizing the size of the mixing zone.

The dilution of the thermal plume is achieved in a relatively short distance from the discharge
structure as indicated in Table (2). Its maximum distance from the discharge structure was 55m for
the counter line, which has 5
o
C increase above the ambient temperature (the mixing zone border)
and 190 m for the counter line, which has 3
o
C increase above the ambient temperature. These
distance are considered small and the value of 3
o
C excess does not have strong effect on the
water environment. The tests of one or two units in operation showed reasonable mixing
characteristics, even with the low tide condition. The inclination of the outfall structure results in
minimizing the plume travel toward the intake structure. It also caused travelling of the hot water
more to the North-West direction.

541
Table (2) Dilution of the Thermal Plume in the Different Tests

T
Max. Distance from the discharge
structure, m
Test no.
7.0 10 9
6.5 12 9
6.0 20 2
5.5 45 2
5.0 55 2
4.5 130 12
4.0 140 12
3.5 180 9
3.0 190 9
2.5 out of modelled area 13 & 14
2.0 out of modelled area 9& 11 &13
1.5 out of modelled area 3 & 5 & 9 & 11 & 13
1.0 out of modelled area All Tests
0.5 out of modelled area All Tests

5.2.1.2 Required Rip-rap for Bed Protection
In order to study the required rip-rap extension, the model bed in front of the discharge structure
was levelled 2.5 m lower than the existing bathymetry. This area was initially filled with sand
with D
50
= 0. 387 mm. A scour occurred in front of the structure and extended 60.0 m. The area
of the scour was filled with rip-rap with D
50
= 0.32 m. The rest of the lower area was filled with
light material equivalent to the bed material size to check the possibility of occurring other scour.

The scour extended to the end of the lower area. The lower area was extended for another 120 m.
The original lower area was filled with rip-pap and the new area with sand. Velocities along the
centerline of the plume were measured at 0.2 of the local depth-above the bottom. Measurements
extended until reach a value of 0.3 m/s. The maximum distance to reach this velocity was 187.5
m, (test No. 9). This area was filled with rip-rap, while the rest of the lower area was filled with
the light material. No movement of rip-rap or the light material was observed.

So, the discharge structure as designed by the Consultant is recommended. Rip-rap with D
50

= 0.32 m is recommended to protect an area, which extends 187.5 m from the discharge
structure.

5.3 Far Field Model
The general test philosophy was to test the design of the cooling system for a range of the
gulf tidal currents and levels for one and two units in operation. Figure (2) shows the near
field model layout. The far field model test program is shown in Table (3). For every run,
surface horizontal plume temperature, vertical temperature distribution at three profiles, and
the velocity distribution at the intake canal were measured.


542

Figure (2) The Far Field Model Layout.
Table (3) Far Field Model Test Program
Current Comments Test
No.
Season No. of
Units
Tidal Level
MSL (m) Coming From Speed
(m/s)

1 Summer 2 + 1.35 NW 0.08
2 Summer 2 - 0.85 SE 0.08
3 Summer 2 + 0.25 - 0.00
4 Winter 2 + 1.35 NW 0.08
5 Winter 2 - 0.85 SE 0.08
6 Winter 2 + 0.25 - 0.00
Original Skimmer Wall Design
7 Summer 2 + 1.35 NW 0.08
8 Summer 2 - 0.85 SE 0.08
9 Summer 2 + 0.25 - 0.00
First Modified Skimmer Wall
Design
10 Summer 2 + 1.35 NW 0.08
11 Summer 1 - 0.85 SE 0.08
12 Summer 2 + 0.25 - 0.00
13 Winter 2 + 1.35 NW 0.08
14 Winter 1 - 0.85 SE 0.08
15 Winter 2 + 0.25 - 0.00
Second Modified Skimmer Wall
Design
16 Summer 2 + 1.35 NW 0.08
17 Summer 1 - 0.85 SE 0.08
18 Summer 2 + 0.25 - 0.00
19 Winter 2 + 1.35 NW 0.08
20 Winter 1 - 0.85 SE 0.08
21 Winter 2 + 0.25 - 0.00
Final Design of Skimmer Wall
22 Summer 2 + 1.35 Three Successive tidal Cycles
Case of High Tide
23 Summer 1 - 0.85 Three Successive tidal Cycles
Case of Low Tide

543
5.3.1 Model Results
5.3.1.2 Converting Model Results to Prototype Data
For converting the measured temperatures in the model to its equivalent temperatures in
prototype, the following scheme was used.

For model temperature conversions, the one Dimensional conservation of energy
equation gives:

p
C
K
Q
BL
E
L o
e
T T
T T

.
0
1
(2)

Where:T
o
: Temperature at distance from the outfall structure = 0, T
L
: Temperature at distance
from the outfall structure = L, T
E
: Equilibrium temperature, B: Model width, K: Heat loss
coefficient, Q: Discharge

By applying Equation (2) to model and prototype, the following ratio can be obtained
p
p
C
m
p
C
K
Q
BL
K
Q
BL
r
e
e
T

1
1
(3)
Determine T
r
using model and prototype data, then model temperature data need to be adjusted
by T
r
, to obtain prototype T
L
at distance L using:

m
E
L o
r
P
E o
L o
T T
T T
T T T
T T

0
1
(4)
The above scheme of converting model results to prototype data was used to calculate the
correction factor of conversion at different lines of measurement (distance from discharge
structure towards the intake structure).
Using Equation (4), we can derived:

m r p
T C T = * (5)

m E o r
p E o
r
T T T
T T
C
) (
) ( 1


= (6)

Where: C
r
: correction factor, T
p
: (T
o
-T
L
)
p
, T
m
: (T
o
-T
L
)
m


The correction factor of the discharge structure vicinity (L= 0) is varied between 1.27 and 0.795.
This mean that for those tests where correction factors less than 1, the prototype temperatures
will be higher than the measured values with maximum factor of 1.26. But this area is mainly
near field and the results are significantly affected by the scale effect. So, it is much better to
depend on the results of the near field model study, when dealing with this area.


The correction factor of the far field zone (L > 1) is varied between 1.35 and 2.78. This means
544
that the surface water temperatures in prototype are always less than those measured in the
model. It is necessary to mention that the correction factor was only used, if the measured
temperature was greater than the model water inlet (background) temperature. The measured
temperature rise above the ambient temperature (inlet water temperature), within this zone, was
varied between 2.0 and 0.5
o
C. This means that the temperature rise in prototype will be less than
these values. This model temperature exaggeration was expected when model scale was selected.

5.3.1.2 Heat exchange coefficient
From Meyer equation and the meteorological measurements, the heat transfer coefficient of
the model area was calculated for all tests as follows:

The Heat exchange coefficient K= 15.7 + (0.26 + ) (a + b U) BTU / (RT
2
.

DAY. F
O
)

Where: K = heat loss coefficient in BTU/(FT / DAY.
O
F) Meyer formula, = vapour
pressure coefficient, a & b = constants of wind function, U = win
d velocity in miles per hour.

5.3.1.3 Data Presentation
The analysis of the conversion of model measurements to prototype data proves, that the
temperature rise over the background (ambient) temperature in the model is higher than its
equivalent values in prototype. So it was decided to present the measured data in form of
temperature rise above the background (model inlet) temperature. The correction factor of each
location (length) was calculated. This scheme was used to present surface water isothermal lines,
temperature profiles, and temperature records during tests of temperature build up due to tidal
cycle. Velocities were measured inside the intake channel. They were measured at one profile at
the centreline of the canal cross-section, which located 270 m upstream the pump house.
Velocities were measured at 0.2, 0.5, and 0.8 of the local depth. These heights were measured
from canal bed.

5.3.1.4 Skimmer wall
- Case of original skimmer wall design (Alternative l):
As shown in Figure (3), the skimmer wall was located at the end of the intake channel. The
opening width was 50 m, the bed level under the opening was (-6.5 m MSL) and the height
was 2.5 m. The float tests (floats were distributed over the shallow area between the intake
and the outlet) in the first six tests was indicated that 60 % of the floats were directed to the
intake channel and they entered the channel during changing the situation from low to high
tide. That means, that any pollutants may enter the intake channel. It can be seen that rather
high temperature rise is found in the vicinity of the intake channel. So, it was decided to
change the location of the skimmer wall. Figure (4) shows the surface isothermal map of the
temperature difference at the Skimmer wall and the intake channel location in case of high
tide in the summer with 2 units in operation (test No. 1).

545


Figure (3) Original Design of Skimmer Wall

Figure (4) Surface Isothermal Map of Temperature Difference- Original Design of
Skimmer wall (Case of high tide- summer- 2 units in operation)

- First Modification of the Skimmer Wall (Alternative 2):
Figure (5) shows the modified design of the skimmer wall. In this modification, the skimmer
wall was moved upstream direction along the intake channel where it is installed at the
beginning of dikes. The opening width was 40 m and the bed level at the opening site was (-
6.5 m MSL) but the skimmer wall level was changed from (- 4.0 m MSL) to (- 3.0 m MSL).
The flat distance under the wall was 30 m (at a level of - 6.5 m MSL). The transition slope
between the bed of the intake channel (3.5 m MSL) and the lowest level under the opening
(-6.5 m MSL) was gentle upstream the wall (1:75) and somewhat steeper downstream (1:73).
Tests (7) to (9) were conducted for the cases of high, low and slack in the summer season.
The results showed some improvements in the distribution of the temperature rise in the
546
vicinity of the intake channel. The die observations showed the presence of some eddies and
vortices upstream the skimmer wall. These eddies may cause mixing mechanism in front of
the skimmer wall which in turn contribute in the withdraw of the pollutants to the intake
channel. Figure (6) shows the surface isothermal map of the temperature difference at the
Skimmer wall and the intake channel location in case of high tide in the summer with 2 units
in operation (test No. 7). It is clear from the figure that some displacement happened to the
contour line of 0.5
o
C but still it is close to the skimmer wall and the inlet channel. Based on
the existence of eddies and heat, it was decided to do a modification to the skimmer wall.




Figure (5) First Modification of the Skimmer Wall

Figure (6) Surface Isothermal Map of Temperature Difference- First Modification of
the Skimmer wall (Case of high tide- summer- 2 units in operation)

547
- Second Modification of the Skimmer Wall (Alternative 3):
To prevent the eddy formation upstream the skimmer wall, the opening width was increased
from 40 m to 46 m while the opening dimensions still the same. Figure (7) shows the details
of the second modified design. The length of the flat area under the opening increased from
30 to 35 m. The transition slope downstream the skimmer wall was, modified to (1:30). Tests
(10) to (15) were carried out. The results of the tests showed a rather good distribution of the
water temperature rise in the vicinity of the intake channel. Figure (8) shows the surface
isothermal map of the temperature difference at the Skimmer wall and the intake channel
location in case of high tide in the summer with 2 units in operation (test No. 10). It is clear
from the figure that good displacement happened to the contour line of 0.5
o
C and became far
from the skimmer wall and the inlet channel. Although eddies in front of the skimmer wall
was decreased, but they still observed.

Figure (7) Second Modification of the Skimmer Wall


Figure (8) Surface Isothermal Map of Temperature Difference- Second Modification of
the Skimmer wall (Case of high tide- summer- 2 units in operation)

548
- Final Design of Skimmer Wall (Alternative 4):
To prevent the eddy formation completely, a new configuration was proposed. Figure (9) shows
the final design of the skimmer wall. A basin with an area of 100 m long and 100 m wide was
constructed in front of the wall with a bed level of (4.00 m). The opening width was kept the
same (46 m). The results of tests (18 to 23) showed that no change in temperature distribution in
the vicinity of the skimmer wall. All eddies and vortices were disappeared which means
improvement of the performance of the skimmer wall. The thermal contours became far away
from the skimmer wall and the inlet channel as shown in figure (10)



Figure (9) Final Design of the Skimmer Wall


Figure (10) Surface Isothermal Map of Temperature Difference- Final Design of the
Skimmer wall (Case of high tide- summer- 2 units in operation)

549
5.3.1.5 Temperature Build up Due to Tidal Cycle
Tests (22) and (23) were carried out to show the effect of successive tidal cycles on the
temperature rise in the vicinity of the intake channel. The temperature measurements were
conducted at a cross section located at 0. 8 Km from the outlet directed to the intake channel
(2.7 Km from the origin of the modelled area).

The recorded temperature rise at four locations which they have the highest temperature rise
was compared with the maximum temperature rise at the same location during tests (16, 17,
19, and 20). The main reason of this comparison was to determine the effect of tidal cycle on
prototype conditions. The results of the comparison are presented in Table (5)

Table (5) Effect of Successive Tidal Cycles on the Temperature Rise
in the Vicinity of the Intake Channel

Test No. Point No. T
o
C T
L

o
C T
H

o
C
1 1.00 0.75 0.95
2 0.90 0.40 0.60
22 3 1.80 0.80 1.30
4 0.65 -0.00 0.15
1 0.60 0.40 0.55
2 0.90 0.25 0.40
23 3 1.60 0.60 1.10
4 1.20 0.70 1.20
Where:
- T is the maximum-recorded temperature rise above background temperature
during the test at this point.
- T
L
is the excess of the temperature rise during build up tests over the
temperature rise during the tests of low tidal level, Tests 17 and 20.
- T
H
is the excess of the temperature rise during build up tests over the
temperature rise during the tests of high tidal level, Tests 16 and 19.

The maximum excess temperature rise was 0.8
o
C, point 3 Test 22. This value considered the
maximum temperature build up due to the tidal cycle. The maximum-recorded temperature
rise was 1.8
o
C, point 3 Test 22. The correction factor of this location is 1.95. This means that
the temperature rise over the ambient temperature in prototype will be less than 1.8
o
C.

5.3.1.6 Rip-rap Protection
The rip-rap used for protecting the dikes along the intake channel and the outlet was tested in
the model regarding to instability. It was found that the gradation of the rip-rap protection
was rather stable. Four types of rip-rap are specified as follows:
- Type (1): The D
50
of this type is 0. 15 m. It is located at the inner slope along the
complete length of the dike. It is also located at the outer slope of the dike to cover
a distance from the shoreline from 0 to 550 m.
- Type (2): The D
50
of this type is 0.2 m and is located at the outer slope of the dike
to cover a distance from the shoreline from 550 to 765 m.
- Type (3): The D
50
of this type is 0.4 m and is located at the outer slope of the dike
to cover a distance from the shoreline from 765 to 950 m.
- Type (4): The D
50
of this type is 0.5 m and is located at the outer slope of the dike
to cover a distance from the shoreline from 950 to 960 m.
550
6. CONCLUSIONS
Research Investigations were carried out to determine the optimal intake/discharge structures
configurations of the Ayoun Moussa Power Plant, located at the Suez Gulf in Egypt. These
optimal configurations will result in minimizing the thermal pollution in the Sues Gulf due to
the disposal of hot water, resulted from the cooling process into it. Based on the investigation
results, the following points were concluded:

6.1 Discharge Structure
- The dilution of the thermal plume is achieved in a relatively short distance from
the discharge structure, which measured in the near field model. Its maximum
distance from the discharge structure was 190 m for the counter line, which has 3
o
C
increase above the ambient temperature. This distance is considered small and the
value of 3
o
C excess does not have strong effect on the water environment.
- The inclination of the outfall structure results in minimizing the plume travel
toward the intake structure. It also caused travelling of the hot water more to the
North-West direction.
- No movement was observed in rip-rap either over the buried discharge pipes or in
front of the structure.

6.2 Intake Canal
- The maximum temperature rise, measured at the vicinity of the intake canal is 1
o
C (Tests 1, 2, and 3).
- Current directions SE and NW affected the plume behaviour, but they never create
any recirculation of the hot water even with low tidal level.
- Conversion of the model measurements to prototype data proves that the
temperature rise above the ambient temperature in prototype will be less than its
equivalent measured values in the model. This also confirms that no hot water
recirculation can be expected in prototype not only because no recirculation was
observed in all tests.
- Canal alignment showed a good flow characteristic, as the velocity of the lower
layer was much faster than the upper layer. This also proves that, the bed level of
53-m MSL, is enough to keep the stratified flow and prevent shallow water effect
(affected of the hot layer by the Gulf bed).
- No hot water recirculation was measured or observed in any tests, even with tests
of temperature build-up due to tidal cycles, although model temperature were
exaggerated by the effect of the scale. This again proves that no recirculation can
be expected in prototype.
- Distance between the intake canal and outfall structure is enough to prevent
recirculation of the hot water.
- No rip-rap movement was observed in any of the four types.

6.3 Skimmer Wall
- Alternative 1 of the skimmer wall prevented the recirculation of the hot
water. But pollutants and oil may enter the intake canal. This will increase
the cost of the maintenance.
- Alternatives 2 and 3 created vortex upstream the skimmer wall. No air
entering vortices were observed. But the vortices occurrence, created
mixing of the upper and lower layers in front of the, skimmer wall and
withdrawal pollutants and oil.
551
- Calculations as well as the model results, proved that Alternative 4
withdrawal only water from the lower layer. No vortices or eddies were
observed upstream the skimmer wall. Therefore, this alternative is
recommended.

6.4 Temperature Build up Due to Tidal Cycles
- Maximum temperature build up due to tidal cycles was 0.8
o
C.
- The build up temperature has no significant effect on prototype conditions, as its
magnitude is small, specially because the analysis of conversion of model
measurements to prototype data proves that the measured temperature rise in the
model is exaggerated.
- No hot water recirculation was measured during operation of both tests.

6.5 Rip-Rap Protection
- The rip-rap protection, at the dike of the intake canal, and over the buried
discharge pipes, as designed by Consultant is recommended. Also rip-rap
protection at the vicinity discharge structure (D
50
= 0.32 m) is recommended to
protect an area, which extends 187.5 m from the discharge structure.

7. ACKNOWLEDGEMENTS
Prof Dr. M.T.K. Gaweesh, Director, and Eng. I. El-Desouki, Deputy Director, of the
Hydraulics Research Institute, Delta Barrage, Egypt, are gratefully acknowledged for their
support and useful comments. Thanks are also due to the Institute staff for their effort and
time that was spent during the model construction, and measurements of this investigation.

8. REFERENCES
[1]: Acker, P., and White, W.R., Sediment Transport: New Approach and Analysis, Journal of
Hydraulics Division, ASCE, No. HY 11, USA, 1973.
[2]: Delft Hydraulics and Massachusetts Institute of Technology, European course on heat
disposal from power generation in the water environment, Delft, 1975
[3]: D.S. Miller & B.A. Brighouse Thermal Discharges, Published by British Hydro-mechanics
Research Association, England, 2001.
[4]: Helmut Kobus, Hydraulic Modelling German Association for Water Resources and Land
Improvement, Bulletin, 2000.
[5]: H.E. Sweers, A Monogram to Estimate the Heat Exchange Coefficient at the Air-Water
Interface as a Function of the Wind Speed and Temperature, a Critical Survey of Some
Literature, Journal of Hydrology 30 (1076) 375 - 401.
[6]: Hoogendoom, D. and Quist, A. "Effects of Model Distortion on the. Near Field Behaviour of
Thermal Discharges Third Waste Heat Management and Utilization Conference Miami,1981
[7]: Hydraulics Research Institute Ayoun Moussa Power Plant Near Field Model Study, Sep.
1996. Ayoun Moussa Power Plant Far Field Model Study, Jan. 1997.
[8]: PGESCo, Specifications for Thermal Physical Model Study for Once-Through Cooling
Intake and Discharge for Ayoun Moussa Project, 1995.
[9]: Sues Canal Authority, Dredging Department, Hydrographic Survey and Planning Department
Hydro graphic and Topographic Survey for Ayoun Moussa Power Plant Units l&2,1995.

552

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


AN EMPIRICAL EQUATION FOR PREDICTION OF
LONGSHORE CURRENT AT ROSETTA
PROMONTORY OF THE NILE DELTA

Y.M. Gewilli and A.I. Abo zed
Coastal Research Institute National Water Research Center.
Ministry of Water Resources and Irrigation
15 El-Pharanna st., El-shalalat, 21514, Alexandria, Egypt.


ABSTRACT

In general, the Nile delta is a wave-dominated coast and more precisely wave-
and longterm current-dominated coast. Littoral current is particularly
responsible for most of the longshore transport of sediment along the Nile delta.
Analysis of a dimensionless longshore current function is developed based on
the wave records and shoreline orientation at Rosetta promontory. Regression
analysis is applied using a 6-year wave dataset to transform the predicted
function to an empirical equation. The results obtained are statistically in good
agreement with the measured longshore currents. The computed longshore current
is used to calculate littoral drift rate which is found to be comparable with other
approaches at the same locality. The computed net and gross sediment transport
are 0.69 x 10
6
m
3
/year to the south southwest and 0.73 x10
6
m
3
/year, respectively.
The results indicate that 70 % of the total amount of the moved sediment takes
place during storm periods which span only 20 % of the whole year.

KEYWORDS: Longshore current, beach erosion, littoral drift rate, Nile delta
553

INTRODUCTION
Waves are responsible for driving the nearshore currents system which is, in turn, responsible
for the important process of longshore sediment transport. When waves approach the coast at
an angle, they cause sand to be transported along the shore i.e. longshore sediment transport
(CERC [1]). Sediment transport is divided into two general classes: transport parallel to the
shore (longshore transport) and transport perpendicular to the shore (cross-shore transport).
The sand movement in the alongshore direction is referred to as the longshore sediment
transport, while actual volumes of sand involved in the transport are termed the littoral
transport rate or littoral drift (Q). This sediment transport-related information is one of the
critical parameters used in designing protective structures. Therefore, devising of an equation
to predict longshore currents at any time based on wave data is essential to cover different sea
condition all over the day.
Several formulae have been developed to compute longshore current based on continuity of
water mass, energy flux, momentum flux and radiation stress approaches. These formulae are
developed by many authors, among them, Inman and Bagnold [2], Bruun [3], Galvin and
Eagleson [4], Butnam et al [5], Brebner and Kamphuis [6], Inman and Quinn [7], Nagai [8],
Longuet-Higgins [9], Komar and Inman [10], Komar [11] and the Coastal Engineering
Research Center, (CERC) [1]. These formulae depend on wave data at breaking, which is
difficult to obtain satisfactory, especially the wave direction. The longshore current is
measured twice daily, one in the morning and the second in the afternoon when the sea
condition is suitable. Practically, researchers and engineers cannot depend on these kinds of
periodical data to predict the longshore sediment transport accurately. This is due to the fact
that longshore current can not be measured during storms and also measurements never cover
the entire day.
The purpose of the present study is to develop an empirical equation to determine the
longshore current speed and direction on the western flank of Rosetta promontory based on
wave data and shoreline orientation. The derived data will be used to estimate rate of littoral
drift using the formula of Komar [12].

STUDY AREA
The Rosetta promontory is located on the northwestern coast of the Nile delta (Figure 1A).
The shoreline is oriented NNE to SSW and is composed of fine to very fine sand. The
promontory was formed by the discharged sediments from Rosetta branch during the
Holocene (7,000 years BP) and historical times (Said [13]).The Rosetta promontory among the
delta coastlines has been subjected to the worst severe erosion as discharges of the river began
to decrease in the late 19th century as a result of climate change and construction of dams and
water control works on the river (UNDP/UNESCO, [14]; Fanos et al. [15]). This erosion is
associated with considerable sedimentation in the Rosetta estuary.
The large-scale pattern of shoreline change along the Rosetta promontory has been studied by
many authors (Frihy et al. [16], Frihy et al. [17], Fanos, et al. [18] and Fanos, et al. [19]).
Among these studies is the study of Frihy et al. [20]. They reported extensive areas of erosion
and accretion along the length of the promontory over the period 1955 to 2000, with the most
severe erosion centred on the promontory tip along the western and eastern sides of the
distributary mouth (Figure 1C). This erosion diminishes further along the promontory flanks,
ultimately reverting to accretion and yielding two nodal points west and east of the river
mouth. To mitigate erosion at the outer margin of the Rosetta promontory, two seawalls
comprised of 4 to7 ton dolos were constructed to the west and east of the Rosetta mouth
during the period 1989 to 1991 (Figure 1C). The western and eastern sea walls extend 1.5 km
and 3.35 km, respectively, have heights of 6.75 m above MSL and widths of 48 to 70 m.
554

Following constructing this wall, shoreline retreat is diminished to ~10m/yr outflanking the
downdrift area of the western seawall.

METHODOLOGY
1. Measurements of wave and longshore current
In this study, measurements of wave parameters and longshore currents are required to
achieve the study objectives. Directional waves were measured by means of CAS (Cassette
Acquisition System) at Abu-Quir bay over ~5 years between 1985 to December 1990. This
instrument includes three pressure sensors array and they were installed at 18 m depth in
Abu-Quir bay (location in Figure 1A). Wave data are recorded for 34 min every 6 hours per
day. Data are statistically analyzed to obtain significant wave height, wave period and wave
direction.
At the western flank of Rosetta promontory longshore current measurements were measured
at two stations simultaneously with recording waves at Abu Quir Bay (see Figure 1C for
location). Measurements were obtained inside the surf zone at water depths ranging from 1.2
to 1.5 m by tracking the movement of a float (buoy), and measuring the time it took to travel
a distance of 20 m in the longshore direction. Measurements of current speed and direction
were taken twice a day, once in the morning and again in the afternoon. Littoral drift was also
determined each 6 hours using the estimated littoral current and wave data.
2. Dimensional Analysis
Longshore current velocity and direction characterized by the following hydraulic conditions:
1. deep water wave height (H
o
), 2. deep water wave direction (
o
), 3. Shoreline orientation
expressed by direction of the normal to the shoreline (
s
), 4. Gravitational acceleration (g),
and (5) other minor factors such as longshore variations of breaking wave heights are
neglected in the present study.
In accordance to the previous factors, longshore current velocity (v
c
), can be expressed as
follows:

( ) g H f v
s o o c
, , ,
1
=
(1)
Appling the Buckingham -theorem, function 1 can be written in the following dimensionless
form


{ }
o s
o
c
f
H g
v
=

( 2 sin
1
(2)
Function 2 expresses the dimensionless longshore current velocity in terms of wave
parameters in deep water (wave height and wave direction) and shoreline orientation.

3. Empirical prediction of longshore currents
In this study a refraction wave model developed by Gewilli [21] were used to transfer wave
data to deep water and then shoreward to the breaking zone. This model takes the effect of
shoaling and refraction and has two options (back refraction and forward refraction). The
resulted refraction pattern of waves arriving from the NNW is shown in Figure 1B. This
pattern reveals zones of high energy (wave convergence) off the promontory tip and low
energy wave (wave divergence) along Abu Quir and Abu Khashaba coasts that result in
strong longshore gradients of wave heights and breaker angles, and therefore of sand
transport rates (Figure 1D).
The shoreline orientation of the study area measured from high resolution map, the measured
longshore current and the corresponding wave data in deep water were used to transform
555

function 2 to an applicable equation. Applying regression analysis between the two non
dimensional parameters in function 2, a relationship, Figure 2, for longshore current
velocities and direction is obtained as follow:
[ ] ) ( 2 sin . 1766 . 0
o s sig o c
H g v =

(3)
Where: H
o-sig
is the significant wave height in deep water in (m), g is the gravitational
acceleration in (m/sec
2
),
s
is the direction of the normal to the shoreline in degree from the
north, and
o
is the deep water wave direction in degree from the north.
The resulted equation should be applied with under conditions similar to Rosetta area. The
average absolute percentage error for this equation is 3.4 % while the correlation coefficient
is 0.96. The positive value for the predicted longshore current, (v
c
), means that the current is
towards east or north northeast according to the shoreline orientation, but the negative value
means the opposite direction.

4. Estimation of littoral drift
Bagnold [22] proposed an equation for longshore sediment transport rate under the combined
action of waves and currents. Komar [12] simplified this equation to be as follows:
Per v H Q
c sig b s
. . . 026 . 0
2

=
(4)
Where: Q
s
is the volume of longshore sediment transport rate in m
3
/year, H
b-sig
is the
significant wave height at breaking in (m), v
c
is the average velocity of longshore current in
(m/sec), Per = 60 x 60 x 24 x 365 x %age, and %age is the percentage of occurrence
towards a certain direction.
Substituting the value of (Per) in equation (4), leads to:
age v H Q
c sig b s
% . . . 820000
2

=
(5)

RESULTS AND DISCUSSION
1. Wave inducing longshore current
Previous studies on the Nile delta has agreed that wave induced longshore currents are
responsible for coastline changes. Among those is the study of Frihy et al. [15] that related
net changes to the effect of wave-induced current due to wave direction versus local SW-NE
shoreline orientation. Analysis of incident waves versus shoreline orientation along the western
flank of the Rosetta promontory revealed that the N, NNW, NW, WNW and NNE (totaling 90

o
) waves are jointly acting to transport sediment toward the SSW along the western flank of
the Rosetta promontory. Conversely, small wave component approaching from the W (20
o
)
sediment moves to the NNE. Such wave sector diagram is schematically shown in Figure 1C.
Accordingly, the wave distributions measured in present study, Figure 1B, indicate that
around 92 % of wave components (blown from N, NNW, NW, WNW and NNE) generate
currents to the SSW while 5 % (blown from W) induce currents to the NNE direction.

2. Predicting longshore current
The devised Equation 3 cited above was used to predict the longshore currents on the western
side of Rosetta promontory using the wave dataset recorded in deep water. The predicted
longshore currents are subjected to statistical analysis and the results obtained are listed in
Table 1. Maximum predicted longshore currents are 115 cm/sec to the south southwest and
88 cm/sec to the north northeast; corresponding to an average longshore current and its
percentage of occurrence (44 cm/sec, 85%) to the south southwest and (28 cm/sec, 10%) to
the north northeast. In order to verify the devised Equation 3 values of longshore currents
556

calculated are statistically correlated with corresponding values measured in the field (Figure
2). It is seen that the regression line indicate perfect agreement with a considerable high
correlation coefficient of 0.96. The plot also presents current measurements trending NNE
(positive values) and others heading SSW (negative values).

Table (1): Predicted longshore current during the period from 1985 to 1990 using
equation (3).
Average velocity
(cm/sec.)
Maximum
velocity (cm/sec.)
% of occurrence
To north northeast 28.0 88 10.0 %
To south southwest 44.0 115 85.0 %

The calculated net and gross sediment transport rate estimated from equation (5) on west side
of Rosetta mouth is found to be 690000 m
3
/year to the south southwest and 730000 m
3
/year,
respectively (Table 2). Comparisons were made between the predicted longshore sediment
transport rate from equation (5) and the results from other techniques such as beach profiles
and LITPACK Package. Results obtained confirm that predicted sediment volumes are
comparable with those calculated from beach profiles during the period from 1985 to 1990
and also in agreement with the results yielded from LITPACK package for sediment transport
calculations (Danida fellowship Program [23] ).

Table 2. Comparison between predicted longshore sediment transport and other
techniques
Predicted
(10
3
m
3
/year)
Beach Profiles
(10
3
m
3
/year)
LITPACK Package
(10
3
m
3
/year)
Net 690
(To south southwest)
660
(To south southwest)
723
(To south southwest)
Gross 730 - 760

A relationship between accumulated littoral transport rates estimated in this study and
percentage of occurrence of available wave heights is established in Figure 3. The gradual
upward trend of the littoral transport rate and proportions of wave height revealed that wave
heights equal to or greater than 1.5 m are generated during 20 % of the whole year. During
this small proportion period, 70 % of the sediment moves as a result of storm action.


CONCLUSIONS
In the Nile delta, and particularly along the shore of Rosetta promontory, the longshore
current is the main factor responsible for the sediment transport causes the beach erosion.
This current is essential for designing effective protective measures along the coastal region. In
the present work, an empirical equation was devised to predict the longshore current velocity
and direction at the western flank of Rosetta promontory based on both wave height and
direction in deep water and shoreline orientation. The average absolute percentage error of
the devised equation is estimated to be 3.4 %. The littoral drift rates estimated from the
predicted longshore current seems to be comparable with those calculated from actual beach
profiles and the module of LITPACK Package. The total sediment transport rate on west side
of Rosetta mouth equals 730000 m
3
/year. Of this value 94.5% is directed to the SSW as net
transport, which is corresponding to wave components blown from N, NNW, NW, WNW and
NNE. Around 70% of the estimated sediment rate moves during winter storms. The
557

empirical equation devised in this study can be successfully used to predict continued
longshore currents in any time once wave data are available at the same locality.

ACKNOWLEDGEMENTS
We express our appreciation to Dr. Omran Frihy, whose review helped improve this
manuscript.

REFERENCES
[1] CERC, Shore Protection Manual, Coastal Engineering Research Center, U.S. Army
Corps of Engineering, U.S. Government Printing Office, Washington, D.C., (1973 and 1984).
[2] Inman, D.L. and Bagnold, R.A., "Beach and Nearshore Processes: Littoral Processes,
In the sea ed. M.N. Hill, v.3, Interscience New York, P. 529-553, (1963).
[3] Bruun, P., Longshore Currents and Longshore Troughs Journal of Geophysical
Research, V. 68, P. 1065-1078, (1963).
[4] Galvin, C.J. and Eagleson, P.S., Experimental Study of Longshore Currents on a
Plan Beach, U.S. Army, Coastal Engineering Center, Tech. Memo. No.10, P.1-80, (1965).
[5] Putnam, J.P., Munk, W.H. and Traylor, M.A. The Prediction of Longshore
Currents, Transaction Am. Geophys. Union, V. 30, P. 337-345, (1949).
[6] Brebner, A. and Kamphuis, J.W., Models Tests on the Relationship Between Deep
Water Wave Characteristics and Longshore Currents, Queens University, Kingston,
Ontario, Canada, Civil Engineering, Report no. 31, P. 1-25, (1963).
[7] Inman, D.L. and Quinn, W.H., Current in The Surf Zone, Proceeding of the 2
nd

Conference on Coastal Engineering (ASCE), P. 24-36, (1952).
[8] Nagai, S., On Coastal Groins, Proceeding of the 1
st
Conference on Coastal
Engineering (ASCE), Japan, P. 99-104, (1984).
[9] Longuet-Higgins, M.S., Longshore Currents Generated by Obliquely Incident
Waves, Journal of Geophysical Research, V. 75, P. 6778-6789, (1970a).
[10] Komar P.D. and Inman D.L., Longshore Sand Transport on Beaches Journal of
Geophysical Research, V. 75, P. 5914-5927, (1970).
[11] Komar P.D., Beach-Slope Dependence of Longshore Currents, Journal of Water
Ways, Ports, Coastal and Ocean Division, ASCE, V. 105, WW4, P. 460-464, (1979).
[12] Komar P.D., "Littoral Sediment Transport: Handbook of Coastal and Ocean
Engineering", J.B. Herbich (editor), v.2 chapter 11, pp. 681-715.
[13] Said, R., "The Geological Evolution of The River Nile", New York: Springs-Verlag, ,
1 15 51 1 p p. . ( (1 19 98 81 1) ). .
[14] UNDP/UNESCO, "Final Technical Report" Coastal Protection Studies, A.R.E.
FMR/SC/OPS/230, Pairs, Vol. 2, pp 206 483 (1978).
[15] Fanos A.M.; Khafagy A.A., and Dean, R.G., "Protective Works on The Nile Delta
Coast" Journal of Coastal Research, 11, 516-528 (1995).

[16] Frihy O. E., Nasr S. M. and El Raey M., "Temporal shoreline and bottom changes of
the inner continental shelf off the Nile Delta, Egypt". Journal of Coastal Research, 7, 2, 465-
475 (1991).
[17] Frihy O. E. and Komar, P. D.,"Long-term shoreline changes and the concentration of
heavy minerals in beach sands of the Nile Delta, Egypt". Marine Geology, 115:253-261
(1993).
[18] Fanos, A.M., Naffaa, M.G., Gewilli, Y.M. and Ali, M.A., "Long and Short Term
Changes of Rosetta Promontory, Egypt". MEDCOAST, The Second International Conference
of The Mediterranean Coastal Environment, Tarragona, Spain, October 24-27, (1995).
558

[19] Fanos, A.M., Naffaa, M.G., Gewilli, Y.M. and Ali, M.A., "Erosion of Rosetta
Promontory, The Nile Delta, Egypt". Fourth International Conference on Coastal and Port
Engineering in Developing Countries (COPEDEC IV), Rio de Janeiro, Brazil, 25-29
September (1995), pp. 470-485.
[20] Frihy, O.E., Debes, E. and El Sayed, W. "Processes reshaping the Nile delta
promontories of Egypt": pre- and post protection. Geomorphology, 5 53 3: : 2 26 63 3- -2 27 79 9. . (2003).
[21] Gewilli Y.M., The Longshore Currents and Sediment Transport along The Nile Delta
Coast. M.sc. Thesis Submitted to the Department of Hydraulic & Irrigation, Faculty of
Engineering, Alexandria University, Alexandria, Egypt, (1994).
[22] Bagnold, R.A., Beach and Nearshore Processes; Part I: Mechanics of Marine
Sedimentation, in the Sea: Ideas and Observations, V.3, M.N. Hill, ed. Interscience, New
York, P. 507-528, (1963).
[23] Danida Fellowship Program, Training Program for Coastal Engineers from Egypt,
"Numerical Modeling of Waves and Sediment Transport Conditions for The Nile Delta
Coast." Trainee Report at Danish Hydraulic Institute, Denmark, (1996).

NOMENCLATURE
g Acceleration of gravity.
Q
s
Volume of longshore sediment transport in m
3
/year.
H
b-sig
Significant wave height at breaking.
v
c
Average velocity of longshore current.
%age Percentage of occurrence towards a certain direction.
H
o-sig
Significant wave height in deep water.

o
Deep water wave direction.

s
Direction of the normal to the shoreline.


559



Figure 1. (A) The Nile delta. (B) Distribution of wave directions recorded off Abu Quir Bay.
(C) Shoreline net changes along the Rosetta promontory (1990 - 2000), and showing
sediment sources and Wave sector diagram for incident waves versus shoreline orientation.
(D) Wave refraction patterns computed for waves from the NNW direction.

560





-0.25
-0.20
-0.15
-0.10
-0.05
0.00
0.05
0.10
0.15
0.20
0.25
-1.5 -1 -0.5 0 0.5 1 1.5
Li ne of best agr eem ent


Figure 2. Relation between
o
c
H g
v

and { }
o s
( 2 sin .





0.00E+00
1.50E+05
3.00E+05
4.50E+05
6.00E+05
7.50E+05
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6
Wave Height (m)
A
c
c
.

S
e
d
i
m
e
n
t

T
r
a
n
s
p
o
r
t
(
m
3
/
y
e
a
r
)
0.00
0.20
0.40
0.60
0.80
1.00
A
c
c
.

%
a
g
e

o
f

O
c
c
u
r
r
e
n
c
e
Amount of sediment transport
%age of occurrence of wave height



Figure 3. Relationship between accumulated sediment transport rates calculated at the
western side of the Rosetta promontory and percentage of occurrence of wave heights classes.



561

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005












APPLICABILITY OF USING THE MULTIPLE
LATERAL DEMANDS FOR DESIGNING WATER
NETWORKS


Mohamed Shamrukh
Department of Civil Engineering, Faculty of Engineering,
ElMinia University
E-mail: mshamrukh@hotmail.com


Abstract
Many mathematical algorithms have been evolved for solving and analyzing
the water distribution system such as Hardy-Cross and linear theory methods. In
these traditional design methods, demands (consumption flow) are allocated to
the two end nodes of the system pipes. Grouping water usage at the end nodes
instead of at the actual locations where water is withdrawn from the system,
produces differences between computer-predicted and actual field performance.
In real-world water supply pipelines, the demands are withdrawn at different
distances not at the ends of the pipeline. A numerical analysis of flow through
the main pipes of water networks utilizing a hypothetical assumption of equate
distribution of multiple lateral withdrawal lines (demands) along the pipe length
has been presented. With the modified approach, head loss calculations in water
pipes should be modified in the situations with a limited number of intermediate
pipes withdrawing water from the pipe. Equations for friction head loss could be
reduced by a coefficient equals to 0.5 for intermediate water laterals less than
two points. This coefficient is applied in case of inflow rate which is less than
two times the withdrawal rate (Q<2Q
c
). The uniform withdraw method could be
used in situations when the number of lateral withdrawals more than 10 points.
Using this introduced approach (multiple and equate uniform lateral
withdrawals) does not need the process of demand allocation used in the
traditional methods for design of water pipes. Thus, the actual demand is
allocated to the pipe itself instead of its nodes. Therefore, the developed
approach is expected to improve the design and simulation processes of water
distribution systems to match its reality.
562


Keywords: Water supply, Pipeline network, Demands, Design

1. INTRODUCTION
Water Distribution Networks are very important, for us as they are life supporting
elements, for any town or city. Distribution System (or Network) is a grid formed by nodes,
links, and loops, if any. It can also have pumps, valves, tanks, and other accessories. Of the
total expenditure incurred on different facilities of water supply system, the expenditure
incurred on distribution network is quite large and may even exceed 60%. The cost of
operating a water distribution system is be substantial (due to maintenance, repair, water
treatment, energy costs, etc.), but still one of the main costs is that of the pipelines
themselves. Given a network layout and requirements (flow and pressure), the optimal design
problem of water distribution systems has been viewed as the selection of pipe sizes which
will minimize overall costs. Among the most effective methods to solve this problem is
simulating water distribution. Simulation of hydraulic behavior within a pressurized looped
pipe network is a complex task which effectively means solving a system of non-linear
equations.

Design of water network involves calculation of unknown parameters, which can be
combination of nodal heads and pipe flow. In fact, each pipe, fittings, and tank in the water
system has to be included in the system design. This process needs from the utilities to invest
a lot of money and time. In the pipe network, there are two processes known as
skeletonization and demand allocation which have significant impact on the design process.
As far as a water distribution system is concerned, skeletonization and demand allocation are
usually the most important and critical tasks in design. Skeletonization is the process of
selecting the parts of the network that have a significant impact on its hydraulic behavior to be
included in the design process. Skeletonization is a practical approach that allows the designer
to produce reliable and accurate results without investing unnecessary time and money.
However, skeletonization should not decrease the accuracy of the network design. The
consumption or use of water, demand, is the driving force behind the hydraulic capacity in
distribution systems. After consumption rates are determined, the water use is spatially
distributed as demands, assigned to model nodes which is called demand allocation. Thus, a
demand allocation is the process of assigning the demand of a zone to the model nodes.
Unaccounted-for-water is next added proportionally to demands.

Actually all the streets of any city include a water pipe and many lateral service
connections. Let us consider the proposed subdivision shown in Figure 1, which is tied into an
existing water system model. Figure 1 shows how the distributors and service connections for
domestic, industrial, and commercial demands are removed in the network design. Although a
designer would almost never include the nodes of distributors and individual connections,
demand of them are allocated to the appropriate nodes as shown in Figure 1. Under normal
demand situations, this skeletonization must not significantly affect calculated nodal pressure.
In real water system, such the one shown in Figure 1, there are distributors and service
connections that have a significant demand flow such as schools, commercial establishments,
and multi-story buildings. The nodes of those intersected pipes should not be omitted (i.e.,
skeletonized) in order to not affecting the accuracy of design process. This illustrated example
(Figure 1), which can be extrapolated into much larger networks, shows how lateral demand
pipes are removed in all literature methods of water network design. Design methods in the
literature, demand of the distributors and service connections are allocated to the end nodes of
main pipes. In fact, this process of demand allocation to end nodes of main pipes decreases
563

the accuracy of calculated flow rate and nodal head. The methodology of demand allocation
to end nodes of each main pipe needs to be modified to match the real-world water networks.


Fig 1. Schematic of real pipeline network with demand allocations.

The intent of this work is to investigate the applicability of using the multi-lateral demand
withdrawal from the network pipeline in the network design process. The energy equations
i.e. friction head loss, of the network have been modified to account for the reality of lateral
demand in the water network.

2. DESIGN OF WATER NETWORKS
Real water distribution systems do not consist of a single pipe and cannot be described by
a single set of continuity and energy equations. Instead, one continuity equation must be
developed for each node in the system, and one energy equation must be developed for each
pipe (or loop), depending on the method used. For real water systems, these equations can
number up to thousands. For Static Analysis, these rules are used for solving the networks
unknown parameters. For network design, we generally use either of the following methods:
Hardy Cross Method, Linear Theory Method, Newton Raphson Method, and Gradient
Method. In recent years, a number of significant improvements have been introduced in the
methods and techniques used in the analyses of water networks (Bhave 1991; Haestad; 2003).

The first systematic approach for solving these equations was developed by Hardy Cross
(1936). This method solves for flow as basic variable parameter, other parameters are solved
afterwards. The invention of digital computers, however, allowed more powerful numerical
techniques to be developed. These techniques set up and solve the system of equations
describing the hydraulics of the network in a matrix form. Because the energy equations are
nonlinear in terms of flow and head, they cannot be solved directly. Instead, these techniques
estimate a solution and then iteratively improve it until the difference between solutions falls
within a specified tolerance. At this point, the hydraulic equations are considered solved.

564

In the aforementioned design techniques, network equations are based on the simplifying
assumption that water demands are allocated to end nodes of each main network's pipe.
Nevertheless, any elements can potentially influence the system hydraulic behavior should be
included in the design process. For most water networks, this criterion includes large water
intermediate consumers. However, water usage in reality occurs along the entire length of a
pipe (Figure 1). In this case, it should be an intermediate node at each significant lateral
distribution pipe. A numerical analysis of flow through pipe networks utilizing a hypothetical
assumption of uniformity of outflows (demands) along the entire length of the pipe has been
studied (Ali 1994 and 1999). However, in real distribution systems, water is withdrawn from
the system where the distributors are connected to the main pipes at intermediate nodes.
These locations of intermediate intersections with main pipes are certainly not distributed
along every length unit allover the entire length of the main pipe (i.e. every 1.0m for 1000m
pipe length).

3. NETWORK HYDRAULICS
In networks of interconnected hydraulic elements, every element is influenced by each of
its neighbors. the entire system is interrelated in such a way that the condition of one element
must be consistent with the condition of all other elements. Two concepts define these
interconnections: Conservation of mass and Conservation of energy.

The principle of conservation of mass dictates that the fluid mass entering any pipe will
be equal to the mass leaving the pipe. In network modeling, all demands are lumped at the
nodes of each pipe without any intermediate withdrawal. The equation is:

0
in out
node
Q Q =
(1)
where
in
Q
= the inflow into the node;
out
Q = the outflow from node including the withdrawn demand at the node
c
Q
.

The principle of conservation of energy dictates that the difference in energy between two
points must be the same regardless of the path that is taken (Bernoulli, 1738). This means that
the sum of head losses around a pipe loop is equal to zero. For convenience within a hydraulic
analysis, the equation is written in terms of head as follows:
L
loop
h
= 0 (2)
where
L
h = the head loss in every pipe. The frictional head loss can be estimated using any of
the two following formula (Haestad; 2003):
2
. .
2
L
L v
h f
D g
=
Darcy-Weisbach (Colebrook-White) (3a)

87 . 4 852 . 1
852 . 1

69 . 10
D C
Q L
h
HW
L
=
Hazen-Williams formula (3b)

Equation (3) can be rewritten as follows

565

2
Q L K h
L
=
(4)

where
2
2
1
.
gDA
f K =
for Darcy-Weisbach, as example.
in previous equations, L = the pipe length, D = the diameter, v = the flow velocity, g = the
acceleration of gravity, A = pipe sectional area, Q = flow rate in pipe, C
HW
= Hazen-Williams
C-factor, and f is the friction coefficient which may be generally given from Moody Diagram
or from the following equation, turbulent flow (Swamee and Jain 1976):
2
9 . 0
74 . 5
7 . 3
log 325 . 1

+ =
e
R D
f

(5)
where R
e
= Reynolds number = vD/, where being the kinematic viscosity of water and is
the roughness height of the inner surface of the pipe.

The head loss equation with uniformly withdrawal demand was presented by Abdel-Lah
(1989) and Ali (1994 and 1999) which could be written as follows;

2

m L
Q L K h =
(6)
in which
2
3
1
2 2
c c i i m
Q Q Q Q Q + =
(7)
where Q
i
=the inflow rate into node i, and Q
c
=the withdrawal or consumed flow rate.

Uniform Lateral Demand for Hardy-Cross Method
Hardy-Cross method uses the system of the previous equations from (1) to (5) for
determination of the flow rates in pipes and nodal head. For network with multiple lateral
demands, Hardy-Cross method has been generalized and summarized as follows:
First: An initial flow rate in each pipeline for all the network loops is determined such that
conservation of mass at each node is satisfied.
Second: The head loss (h) around each loop of the network keeping track of signs is
calculated. Then pipe flows are adjusted iteratively using the following equation for Q;

= =

=
N
i
i
i
N
i
i i
Q
h
n h Q
1 1
(8)
in which N=number of pipes and n=2.0 for Darcy Weisbach losses or 1.85 for Hazen
Williams losses. The head loss in this equation Q was calculated using the uniform
withdrawal from the pipes.
Third: with repeating the second step until the change in flow Q in each pipe is less than the
convergence criteria.

In the previous method, demands were assumed to withdraw uniformly from the main
pipelines. This approach has many advantages over the old method of allocating the demand
to the end nodes of the main pipes. However, there still some disadvantage of the modified
method of assuming the uniform lateral demand. The assumption of the uniformity of the
566

lateral demand along the entire length of the pipeline is not a real case. In reality, the
intersection of lateral withdrawals are not distributed every unit length of the main pipes of
the network (i.e. each 1.0 m for 1000m pipe length). In reality, lateral demands are allocated
at arbitrary distances allover the pipeline depending on the urban planning of the streets and
houses block. According to current situation of our cities planning, the lateral demand may be
every 100 m for a pipe of 1000m length or at different distances but certainly not each 1.0 m.

In the next sections a new modification for the uniform lateral withdraw is provided. The
current modified methodology is assumed to overcome shortcomings of the uniform lateral
withdrawal. Head loss equation used in Hardy-Cross method for network solution is modified
to match the reality of water distribution systems.

4. MULTIPLE LATERAL DEMANDS
To match the reality of urban planning of streets, Figure 1, several scenarios of pipes
were studied and the lateral demands were assumed to withdraw at different intermediate
locations. Lateral intersections were assumed to range from 2 points to 10 points along the
entire length of the pipeline. This assumption was assumed to match the reality of urban
planning for the intersected streets. The assumed arrangements of the main pipeline and
intermediate intersections of lateral distributors for water use are shown in the schematic real
main water pipe, Figure 2.











Figure 2. Intermediate intersections of lateral distributors with main pipeline

The velocity within the pipes is assumed to be in the allowable limits, about 1.0 m/s. The
studied cases of the main pipes are (36) cases for the main pipeline as shown in Figure 3.
Multiple laterals of withdrawal are varied from one to two, three, six, and ten intermediate
intersections to cover most regualr urban planning streets. The length of the pipe was changed
three times: 500m, 800m, and 1000m. The upstream flow rate was multiplied four times i.e.,
Q
c
, 2Q
c
, 4Q
c
, and 8Q
c
for each pipe length. The diameter of the main pipe was changed to
keep the velocity under the allowable range for each flow rate. Lateral intermediate demands
were kept constant for each flow rate. Effluent from the end node is equal to the input flow
minus lateral demand for each case. Thus, the total numbers of studied pipe scenarios are 36.
Head loss for every case is calculated and is given as a percentage from the original head loss
using the traditional scenario, without any lateral intersections demand (i.e. demand only
allocated at end node). The results of the studied cases are illustrated in Table 1. The first
study case of the main pipe was assigned an upstream flow rate to equal the consumed water
Q
i
School
Houses
Houses Houses Houses Houses
Length, L
Q
e
Hospital
J-1 J-2
567

from the pipe (i.e. Q
in
=Q
c
) with no effluent flow from the end node. This is the case when the
network main pipe is located in network downstream, i.e. network end pipe. In the rest of pipe
cases, there is effluent flow from the end node of the main pipe as shown in Figure 3.










Figure 3. Scenarios of multiple laterals, case of three intermediate demand points.



Table 1. Head loss for different intermediate lateral withdrawal.
Scenario Pipe
length (m)
No. of
intermediate
lateral pipes
Length of
intermediate pipe
segments (m)
Head loss
percentage
(%)


Q
in
=Q
c


5
0
0
,

8
0
0
,

1
0
0
0


End node
1
3
6
9
uniform demand
No laterals
L /2
L /4
L /7
L /10
1.0m
100%
50%
40%
36%
35%
33%


Q
in
=2Q
c


5
0
0
,

8
0
0
,

1
0
0
0


End node
1
3
6
9
uniform demand
No laterals
L /2
L /4
L /7
L /10
1.0m
100%
63%
60%
59%
58%
58%


Q
in
=4Q
c


5
0
0
,

8
0
0
,

1
0
0
0


End node
1
3
6
9
uniform demand
No laterals
L /2
L /4
L /7
L /10
1.0m
100%
78%
77.6%
77.4%
77%
77%
Q
in
Length, L
Q
e
Upstream
node
End node
Intermediate
demand
point
Intermediate
demand
point
Intermediate
demand
point
568



Q
in
=8Q
c


5
0
0
,

8
0
0
,

1
0
0
0


End node
1
3
6
9
uniform demand
No laterals
L /2
L /4
L /7
L /10
1.0m
100%
88.3%
88.2%
88.1%
88%
88%



5. RESULTS AND DISCUSSION
The results of the studied assumed pipe cases show that there are differences among the
traditional method of demand allocation, uniform withdrawals, and multi-intermediate
laterals. From Table 1, the traditional method gives the highest values of head loss and the
uniform withdrawals give the lowest. The presented method herein which is the closest
scenario to the reality of water networks gives results much smaller than the traditional
method of demand allocating into the pipe end nodes. The results indicate that the percentage
of head loss does not change with increasing the pipe length from 500m, to 800m, and to
1000m with the same number of intermediate nodes.

The estimated head loss for the case of pipe inflow equals to pipe demand (i.e. Q
in
=Q
c
)
with no effluent flow from the end node is shown in Figure 4. The shown results are given
regardless what is the length of the main pipe. Head loss values in Figure 4 show that when
the laterals are one intermediate distributor or service pipe, the head loss is decreased to be
50% of the traditional method (demand at end junctions only). Certainly, this is a significant
difference between traditional method and the suggested method with intermediate
withdrawals. It is clear from this study that the number of intermediate withdrawals is affects
significantly greatly the total head loss of the network pipe. As the number of the intermediate
lateral withdrawals increases to 10 intersection points, the head loss decreases to 35% of the
traditional method. The head loss of 10 laterals which is 35% is almost the same head loss
using the uniform method introduced by Abdel-Lah (1989) and Ali (1994) which is 33.3%.





















0
10
20
30
40
50
60
70
80
90
100
0 2 4 6 8 10 12
No. of lateral intermediate withdrawals
H
e
a
d

l
o
s
s

p
e
r
c
e
n
t
a
g
e
569



Figure 4. Head loss vs. No. of lateral intersected pipes

It is clear that the difference between the suggested method and the method of uniform
withdrawal could be neglected if the intermediate points are more than five as shown in Table
1. The difference between the two methods in this case is less than 4%. Figure 4 shows that
when the intermediate withdrawals are more than 10 points, the designer can use the method
of uniform withdrawals. In this case, equations 6 and 7 could be used to calculate the head
loss in network pipes. If the number of intermediate withdrawals is less than three, the
suggested method could be used to match the reality of pipe network planning. A coefficient
of 0.5 could be used to modify the calculations of head loss used in network design.

Most of the water pipe networks consist of upstream and downstream group of pipes.
Upstream pipe mean that the pipe which convey the water for both its demand and feed the
downstream pipe. Thus, the inflow into the studied pipes is increased four times to be 8Q
c
.
Figure 5 shows the head loss for the input flow rates changing from Q
c
to 8Q
c
. It is clear that,
as the inflow increases the head loss percentage of suggested method increases. The head loss
equals to 88% from the traditional method for inflow=8Q
c
. Thus, the head loss difference
between the traditional, uniform, and multiple methods is about 12% for high inflow rates.


























Figure 5. Head loss for different inflows and intermediate laterals.



0
10
20
30
40
50
60
70
80
90
100
0 2 4 6 8 10
No. of intermediate laterals
H
e
a
d

l
o
s
s

p
e
r
c
e
n
t
a
g
e

%
inflow=Qc
inflow=2Qc
inflow=4Qc
inflow=8Qc
570

When the inflow increases more than 2Qc, the head loss difference between the
suggested multiple method and the uniform method is about 4%. As shown in Figure 6, this
small difference could be neglected in reality. Furthermore, the head loss values at inflow
more than 4Q
c
are almost the same for two methods as shown in Figure 6. It can be noticed
that at inflow more than 2Q
c
, previous equations (6) and (7) could be used to calculate the
head loss regardless of the number of intermediate withdrawal points. Although the method of
uniform demand assume that the water is withdrawn every unit length, the values obtained
from it is equivalent to the suggested method for inflow more than 2Q
c
.




















Figure 6. Difference in head loss for the uniform and suggested methods.

6. CONCLUSIONS
Water distribution systems are designed using demand allocation and traditional methods
for friction head loss calculations. An improved methodology for head loss calculations in
comparison with both traditional method and uniform withdrawals from the main pipes of
water networks was presented. The introduced method assumes a multiple lateral withdrawals
from the water pipes. This assumption was developed to match the reality of urban planning
of water distribution systems. More realistic empirical coefficient is introduced for solving the
looped or branched water pipes depending on the number of lateral withdrawals. Head loss
calculations in water pipes should be modified in case of a limited number of intermediate
intersection lateral pipes with the main pipe. Darcy-Weisbach and Hazen-Williams equations
for head loss calculations could be reduced by a coefficient equals to 0.5 for intermediate
water laterals less than two points. This coefficient is applied in case of inflow to the pipe
which is less than two times the withdrawal from the pipe (Q<2Q
c
).

In water distribution systems, there are upstream pipes in which the inflow for each pipe
is greater than two times the demand from it. For these pipes, the uniform withdrawals
equations 6 and (7) could be used to calculate head loss due to friction. Furthermore, the
uniform method could be used in situations when the number of lateral withdrawals is more
than 10 points regardless the pipe inflow to demand ratio. Using this introduced approach
(multiple and equate uniform lateral withdrawals) is expected to overcome the disadvantages
0
10
20
30
40
50
60
70
80
90
100
0 1 2 3 4 5 6 7 8 9 10
Pipe inflow
H
e
a
d

l
o
s
s

p
e
r
c
e
n
t
a
g
e
%
suggested method
uniform use method
571

of demand allocation to only the end two nodes of the pipe. Grouping water usage at the end
nodes instead of at the actual locations where water is withdrawn from the system produces
differences between computer-predicted and actual field performance. By applying this new
approach in the design of water networks, there is no need to the process of demand allocation
used in the literature design methods. The actual field demands are allocated to the pipe itself
instead of its end two nodes. Thus, the developed approach is expected to improve the design
and simulation processes of water distribution systems to match its reality.

REFERENCES
Abdel-Lah, A.K. (1989). Study of the Pipe Networks for Water Distribution. M.Sc Thesis,
Assiut University.
AbdelSadek, F. I., Ali, N. A. and Abdel-Lah, A. K. (1991). Hydraulic and Geometric
Refinements for Designing Water Pipe Networks. Bulletin of the faculty of Engineering,
Assiut University, 19(2): 13-20.
Ali, N.A. (1994). Improved Hydraulic Algorithm for Analysis of Pipe Network Applying
Uniform Lateral Demands. Bulletin of the Faculty of Engineering, Assiut University,
22(2): 39-47.
Ali, N.A. (1999). Hydraulic and Economical Criticism for Water Supply Networks. Bulletin
of Faculty of Engineering, Assiut University, 27(1): 33-40.
Bhave, P. R. (1991). Analysis of Flow in Water Distribution Networks. Technomics,
Lancaster, Pennsylvania.
Charades, K. M. and Stewart, K. H. (1975). Spar city Oriented Analysis of Large Pipe
Network. Journal of Hydraulics, ASCE, 101: 341-355.
Colebrook, C. F. (1939). Turbulent Flow in Pipes with Particular Reference to the Transition
Region between Smooth and Rough Pipe Laws. Journal of Inst. of Civil Engineers. 11:
133-156.
Colebrook, C. F. and White, C. M. (1937). Experiment with Fluid Friction in Roughened
Pipes. Proc. Roy. Sec., Series A, 161: 367-381.
Collins, A.G. and Johnson, R.L. (1975). Finite Element Method for Water Distribution
Networks, Journal of AWWA, pp. 385-389.
Cross, H. (1936). "Analysis of Flow in Networks of Conduits or Conductors." University of
Illinois Experiment Station Bulletin No. 286, Department of Civil Engineering,
University of Illinois, Champaign Urbana, Illinois.
Haestad Methods, T. M. Walski, D. V. Chase, D. A. Savic, W. Grayman, S. Beckwith, and E.
Koelle. (2003). Advanced Water Distribution Modeling and Management. Haestad
Press. Waterbury, Connecticut.
Jappson, R.W. (1981). Analysis of Flow in Pipe Network. ANN Arbor Science Publishers
Inc., pp. 44-46
Larock, B. E., Jeppson, R. W., and Watters, G. Z. (1999). Handbook of Pipeline Systems.
CRC Press, Boca Raton, Florida.
Moody, L. F. (1944). Friction Factors for Pipe Flow. Transactions of the American Society of
Mechanical Engineers, Vol. 66.
Muss, D. L. (1960). Friction Losses in Lines with Service Connections. Journal of Hydraulics
Division, ASCE, 86(4): 35.
Rossman, L.A. (2000). EPANET Users Manual. Risk Reduction Engineering Laboratory,
U.S. Environmental Protection Agency, Cincinnati, Ohio.
Swamee, P. K., and Jain, A. K. (1976). Explicit Equations for Pipe Flow Problems. Journal of
Hydraulic Engineering, ASCE, 102(5): 657.
Wood, D.J . and Races, A.G. (1981). Reliability of Algorithms for Pipe Network Analysis. Journal
of Hydraulic Engineering, ASCE, 107: 1145-1161
572

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005












ASSESSMENT AND MODELLING OF SEA LEVEL
RISE AND METROLOGICAL CHANGES IN EGYPT

Gomaa M. Dawod
1
, Mamdouh M. Meligy
2
, and Hoda F. Mohamed
1


1
Survey Research Institute, National Water Research Center, Cairo
TelFax: 02-5867174 E-Mail: dawod_gomaa@yahoo.com
2
Marin Survey Department, Egyptian Naval Forces, Alexandria


ABSTRACT

Evaluating and formulating the rising rates of sea level and metrological
parameters is a fundamental demand from engineering and environmental points
of view. Daily tidal records and various metrological observations, at
Alexandria, have been collected and analyzed covering the period from 1985 to
2003. The attained results show that the main metrological factors affecting the
sea level rise are the air temperature, air pressure, and wind speed with
correlation factors equal 0.60, -0.39, and -0.23 respectively. A relatively small
relationship has been found between the sea level variations and the solar
activities, as represented by the sun spot number. Additionally, it has been
concluded that the 6
th
-order polynomial function, among five investigated types
of mathematical formulas, is the optimum model describing the relationship
between the sea level and the metrological parameters. Moreover, the annual
rising rates of the studied quantities have been estimated as 0.20 cm/y for sea
level, 0.04
o
C/y for air temperature, -0.03 mb/y for air pressure, 0.05 knot/y for
wind speed, and 1.07 y
-1
for the number of sun spots. The sea level rise is one of
the main elements that cause shoreline retreat and coastal erosion. Hence, it is
recommended that the sea level rise should be taken into consideration in
national development planes, particularly for coastal regions.

KEYWORDS: Sea level rise, Metrological changes, Tide gauge


573
INTRODUCTION

Measurements of sea level have a long history in several places all over the world, and
have been utilized in a wide variety of scientific applications. Changes in the mean sea level
are related to the changes in the volume of water in the oceans and to variations in the ocean
currents. Sea level studies, in nowadays, concern with coastal erosion and protection,
geological processes, and climate changes. Glacial melting, expansion of water due to
heating, and changes in sea surface gradients are some factors leading to sea level change.
This paper aims to investigate the relationship between the sea level changes and metrological
variations in Egypt from both environmental and geodetic points of view.

The sea level rise is one of the main factors that cause shoreline retreat, coastal
erosion, lowlands overflow, and increase the salinity of lakes and aquifers. Consequently,
there exists a critical demand to explore, on geodetic and environmental basis, the
consequences of sea level changes in Egypt. Moreover, an accurate multi-purpose monitoring
system has to be established, over the Egyptian shores, to measure precisely the sea level rise.

GLOBAL SEA LEVEL RISE

The combination of sea level changes measured by several techniques shows a relatively
rapid rise of sea level from 20,000 years ago, gradually slowing down 8,000 years ago when
levels were some 15 meters below those of today. The increase then proceeded more
gradually until present levels were reached some 4,000 years ago. Since that time, the changes
have consisted of oscillations of small amplitude. Recent long-period sea level measurements
at several stations show that there is a general increase of Mean Sea Level (MSL) of about
0.15-0.25 meter per century [e.g. IOC, 2002]. MSL is the average level of the sea, usually
based on hourly values taken over a period of at least a year. The major effective causes for
long-periodic sea level oscillations are [Vanicek and Krakiwsky, 1986]:

- Glacial melt is the main constitute of the secular water rise.
- Changes due to atmospheric pressure that may reach up to several decimeters.
- Changes due to wind stress, which could reach a few decimeters in some places.
- Thermal structure of sea water is an important source of long periodic changes.
- River discharge fluctuations can contribute significantly to long-term variations.

The issue of sea level rise has attained more concerns on a global base in the last few
years. From a developmental point of view, the harmful effects of sea rise, especially in
coastal areas, raise the attention to carry out detailed studies on this subject. Long-term tidal
records may contribute significantly to precise determination of sea rise rate. It is evident that
the sea rise rate on a global scale during the twentieth century has an average of 1.5-2.5
mm/year [IOC, 2002, and NOAA, 2001]. These real estimates are derived from analysis of
actual long-term tide gauge data all over the world. On the other hand, other studies claim that
the rising rate will be much higher in the next few decades and will lead to hazardous
economical troubles [IPCC, 2001]. Most of these investigations are based on different
scenarios and estimates of the air temperature rise due to the global warming phenomena. For
example, recent environmental researches, based on business-as-usual scenario of greenhouse
gas emission, expect that the sea rise on a global scale would be of 12 cm by the year 2030
and about 50 cm by the year 2100. This implies that the sea level rising rate would be doubled
in the twenty-first century compared to the real rising rate of the twentieth century. Other
expectations lead to a sea level rise of about 2 meters by the year 2100 when taken into
574
account the subsidence because of land movements and removal of ground water. However, it
worth mentioning that many of these estimate has a great deal of uncertainties due to their
extrapolation nature and suspected assumptions [Houghton, 1997]. Moreover, the sea level
rise would differ from an area to another on the globe due to the different nature and amount
of the leading factors such as the human and industrial activities. Therefore, the optimum
realistic estimates of sea rise should result from analysis of real data sets, particularly long-
term tidal records.

Meteorological residuals are the non-tidal part that remains after removing the tides by
analysis. They are irregular, as are the variations in the weather. The term surge residuals is
commonly used to describe non-tidal variations in the observed sea level. Physically, the
atmosphere acts on the sea in two different manners. Changes in atmospheric pressure
produce changes in the pressure acting vertically on the sea surface. One-millibar increase of
atmospheric pressure decreases sea level by one centimeter. The drag of the wind on the sea
surface increases as the square of the wind speed, to a first approximation. This drag sets the
water in motion, where in shallow water the flow is in the direction of the wind, and in deeper
water the transport is at right angles to the right in the Northern Hemisphere.

RECENT SEA LEVEL OBSERVING SYSTEM IN EGYPT

Although there are several agencies in Egypt operate measuring equipment for
determining sea level changes, most of them suffer from several disadvantages in terms of
structure, accuracy, and compatibility. For example, the installed out-of-date floating devices
are not accurate enough to produce reliable results or estimates for the sea level rise rate.
Additionally, there is no distinguishing between the relative and the absolute sea-level rise in
these projects. Moreover, most of these activities are designed for a specific goal serving a
specific area, with no national interest in mind.

The Survey Research Institute (SRI) has installed the recent state-of-the-art Sea Level
Observing System (SLOS), inside the headquarter of the Naval Forces in Alexandria. That
tide gauge station has been constructed and operated, by the hydrographic survey department,
since 1993 for military marine applications. The gauge contained an old mechanical float-
gauge device and several separated sensors for measuring meteorological conditions. This
gauge station is located in the Western harbor, 800 meters north-west of the old tide gauge
station operated by the Alexandria harbor authority. A precise levelling line has been carried
out to connect the gauge staffs in both stations. Consequently, the constant datum shift was
obtained and used, in later processing stages, to relate the tide readings of the new station to
the old historical one. The research project has several objectives to fulfill both geodetic and
environmental goals including among others (SRI, 2002):

Precise monitoring of the sea-level rise in Egypt.
Accurate realization of the vertical geodetic datum in Egypt.
Studying the meteorological effects on the sea-level data.
Development of computer strategy and models to re-define the vertical datum.
Develop optimum models for meteorological effects on levelling measurements.

The new installed state-of-the-art SLOS consists of three devices integrated together
in a unified scheme: a tide gauge, a meteorological unit, and a satellite-based GPS geodetic
receiver (Figure 1). The utilized tide gauge is a high-accuracy temperature-compensated self-
contained instrument for measuring and recording tide and wave data. This state-of-the-art
575
device, called Wave and Tide Gauge WTG904 Series 3, is a semi-conductor strain gauge
pressure type that measures instantaneously the water column pressure, and converts it to
height of water with respect to a specific user-defined location. The WTG904' measuring
precision is 0.2 cm, and the tide is calculated continuously and recorded automatically every
10 minutes. The second device in this integrated system is the WMS-14 from Omega Inc.,
which is a microprocessor-based weather station. It features a combination wind vane and
three-cup anemometer, a temperature and relative humidity sensor with sun shield, a solid
state barometric pressure sensor, and, a self-emptying rain gauge. Hence, measurements of
wind speed, wind direction, temperature, humidity, and precipitation are collected and
processed by the control module of the station. In order to account for the land deformation at
the tide gauge site, a geodetic instrumentation must be installed as part of the integrated
SLOS. Hence, the Lieca GPS system 500 is installed which comprises GPS receiver, terminal
and post-processing software. The SR520 dual-frequency GPS receiver has 12 L1 and 12 L2
continuous data tracking channels that track both codes and carrier phases of the transmitted
satellite signals. It tracks up to 12 simultaneously satellites. Also, a high-precision AT504
choke-ring antenna is utilized to ensure high-quality reception of the satellites signals.

The new SLOS has been designed and installed based on the recent international
specifications of such environmental and engineering systems. The collected precise
homogenous data sets will enable an accurate monitoring of the sea level variations and the
re-definition of the national geodetic vertical datum of Egypt. Hence, it is recommended that
all tide gauge stations in Egypt should be upgraded to meet these objectives. Additionally, it
is recommended that the SLOS installed at Alexandria must be incorporated in the Global Sea
Level Observing System "GLOSS" program [Woodworth and Aarup, 2003] as well as in the
proposed sea level observing system over Africa, which aims to establish an integrated coastal
area management and early warning system [PSMSL, 2002].



Figure 1: Modern Sea-Level Observing System




576
LONG-TERM SEA LEVEL RISE IN EGYPT

Previous local investigations revealed that the MSL in Egypt has increased by 12 cm
from 1906 to 1980, which means a rate of approximately 1.5 mm per year [Nassar et al,
1998]. Also, Sharaf El-Din et al [1989] have estimated the MSL rise rate by 1.6 mm per year
at Alexandria in the period from 1958 to 1988. Moreover, El Fishawi, et al [1991] studied the
sea level characteristics along the Nile Delta Coast at Alexandria, Rosetta, Damietta, and Port
Said. It has concluded that the mean sea level displays a significant lateral change along the
coast; there is general increase from Alexandria to Damietta and marked decrease between
Damietta and Port Said. Dawod [2001] has realized that the MSL at Alexandria, from 1944 to
1999, varies from 1.6 cm to 23.5 cm above the current national vertical geodetic datum, with
a mean value of 11.1 cm. At Port Said, MSL changes between 1.7 cm to 24.0 cm, with an
average of 9.0 cm. Also, it has been concluded that the sea level is rising by a rate of 1.7
mm/year at Alexandria, and 2.4 mm/year at Port Said.

AVAILABLE DATA

The daily tide averages have been collected at Alexandria for an eighteen-year period
of time from 1985 to 2003. Additionally, monthly air pressure and air temperature values
have been obtained for Alexandria covering the same period. Their statistics, as summarized
in Table 1, reveal that the air pressure ranges from 1006.4 to 1022.7 with a mean of 1014.4,
while the air temperature varies between 13.5
o
C and 27.6
o
C with an average equals 20.6
o
C.
Moreover, the wind speed varies from 2.6 knot to 11.0 knot, with a mean of 5.6 knot. For the
same time interval, the monthly tide values range from 22.7 cm to 70.6 cm with a mean of
47.9 cm.

Recent research studies [e.g. Pararas-Carayannis, 2003] have shown that oscillations
in global mean temperature are correlated with corresponding oscillations in solar activity.
The traditional direct way to observe the Sun's activity is to look at the sunspot number
(SSN). Sunspots are manifestations of magnetically disturbed conditions at the Sun's visible
surface. The 11-year cycle of solar activities is the dominant pattern, where the current cycle
has started in 1996 and has reached its maximum in 2000. The data of SSN can be obtained
from several sources from the Internet [e.g. http://science.msfc.nasa.gov/]. Hence, SSN
occurred from 1985 to 2003 have been collected. These numbers range from 1 spot to 200
spots with a mean of 73 spots.

Table 1: Statistics of tide and metrological data from 1985 to 2003


Tide
(cm)
Air Temperature
(
o
C)
Air Pressure
(mb)
Wind Speed
(knot)
Sun Spot
Number

Minimum 22.7 13.5 1006.4
2.6 1
Maximum 70.6 27.6 1022.7
11.0 200
Mean 49.4 20.6 1014.4
5.6 73


577
TIDAL-WEATHER INTERRELATIONSHIP MODELLING

Although the tide usually dominates the spectrum of water level and current
fluctuations, non-tidal variations are important for navigation. Atmospheric pressure and wind
produce non-tidal rise in the sea water, usually known as surge. Therefore, the analysis of the
interrelation between the tide and the meteorological conditions is a crucial procedure to
estimate and more important to predict surges for navigation purposes. The first step is to
describe the tide-metrological relationships by a simple linear mathematical model. The
attained equations are:

T = 24.42 + 1.216 AT (1)
T = 1045.9 0.982 AP (2)
T = 49.08 + 0.051 WS (3)
T = 50.31 0.012 SSN (4)

where T is the tide in cm, AT is the air temperature in
o
C, AP is the air pressure in mb, WS is
the wind speed in knot, and SSN is the sun spot number.

In addition, the annual rising rates of the studied quantities have been calculated. Their
computed annual estimates are:
* 0.20 cm/year for sea level,
* 0.04
o
C/year for air temperature,
* -0.03 mb/year for air pressure,
* 0.05 knot/year for wind speed, and
* 1.07 year
-1
for the number of sun spots.

In order to judge the quality of those formulas, the coefficient of determination (R
2
)
statistical factor has been chosen as the basis of comparison. R
2
is a ratio (or percentage)
between the explained variation and the total variation of the regression model. It is an
important indicator of how well the resulting regression equation explains the relationship
among the variables. The bigger the value of R
2
, the stronger the linear relationship between
the observed and estimated values. The coefficient of determination is computed as:

R
2
= Explained variation / Total variation
= (Y
est
Y
m
)
2
/ (Y Y
m
)
2
(5)

where Y is the observed dependant variable, Y
est
is the estimated dependant variable; and Y
m

is the mean of observed dependant variable

Based on the primary achieved findings, it has been noticed that the linear trend does
not precisely fit the MSL variations of the collected data at Alexandria. Therefore, other
regression polynomials have been tried out. Those mathematical models include exponential,
logarithmic, power, and 6
th
-order polynomial. The whole attained results are summarized in
Table 2. From this table, it can be concluded that the 6
th
order polynomial produce the highest
coefficient of determination (86%) even though the improvements over the linear equation
(82%) is not greatly significant. This result emphasizes that the 6
th
order regression formula
may be considered as the optimum mathematical model describing the variations of MSL at
Alexandria.


578
Table 2: Statistical measures (R
2
) of attained mathematical models

Accomplished functions Tested parameters
Linear Exponential logarithmic Power 6
th
order
Air temperature 35 34 35 33 39
Air Pressure 16 15 16 15 25
Wind speed 0.03 0.1 0.1 0.1 11
SSN 0.5 0.9 0.3 0.6 6

CORRELATION ANALYSIS

Computing the correlation factors between the tested parameters is another significant
analysis step to investigate the tide-metrological relationships. The attained results have been
given in Table 3. From this table, it can be seen that the air temperature has a 60% positive
correlation impact on the recorded tide, while the air pressure has a negative correlation of
40%. Also, it has been found that the wind speed and the SSN give negative correlations of
23% and 7% respectively. Figures 2 to 5 depicts the relationships between the tide and each
element of the metrological parameters, where the correlation interactions can be recognized.

Table 3: Correlation matrix of tide-metrological parameters

Tide Air Temperature Air Pressure Wind Speed SSN
Tide 1
Air Temperature 60 1
Air Pressure -40 -80 1
Wind Speed -23 -49 28 1
SSN -7 5 2 17 1

0
10
20
30
40
50
60
70
80
1 31 61 91 121 151 181 211
Month
T
i
d
e
0
5
10
15
20
25
30
A
i
r

T
e
m
p
r
a
t
u
r
e
Tide Temp.


Figure 2: Inter-relationship between the tide and the air temperature
579
0
10
20
30
40
50
60
70
80
1 31 61 91 121 151 181 211
Month
T
i
d
e
995
1000
1005
1010
1015
1020
1025
A
i
r

P
r
e
s
s
u
r
e
Tide Pressure


Figure 3: Inter-relationship between the tide and the air pressure

0
2
4
6
8
10
12
14
16
18
1 31 61 91 121 151 181 211
Month
W
i
n
d

S
p
e
e
d
0
10
20
30
40
50
60
70
80
T
i
d
e
Wind S. Tide

Figure 4: Inter-relationship between the tide and the wind speed
0
10
20
30
40
50
60
70
80
1 31 61 91 121 151 181 211
Month
T
i
d
e
0
50
100
150
200
250
S
S
N
Tide SSN

Figure 5: Inter-relationship between the tide and the sun spot number
580
CONCLUSIONS

Due to the global weather variations, the investigations of the rise of sea level and
metrological parameters become an essential request from engineering and environmental
perspectives. Daily tidal records and various metrological observations, at Alexandria, have
been collected and analyzed covering the period from 1985 to 2003. The attained results show
that the main metrological factors affecting the sea level rise are the air temperature, air
pressure, and wind speed with correlation factors equal 0.60, -0.39, and -0.23 respectively.
Additionally, the solar activities produce a relatively small relationship with the sea level
variations. Among five investigated types of mathematical formulas, it has been concluded
that the 6
th
-order polynomial function is the optimum model describing the relationship
between the mean sea level and the metrological parameters. Moreover, the annual rising rates
of the studied quantities have been estimated as 0.20 cm/y for sea level, 0.04
o
C/y for air
temperature, -0.03 mb/y for air pressure, 0.05 knot/y for wind speed, and 1.07 y
-1
for the
number of sun spots.

A new state-of-the-art sea level observing system has been installed at Alexandria in
2001. Due to its reliability, integrity, and accuracy, it is recommended that all tide gauge
stations in Egypt being upgraded by such a high-technology system in order to monitor the
sea level changes in Egypt with a better level of accuracy. Additionally, it is recommended
that the tide gauges in Egypt be incorporated in the international GLOSS program and the
proposed regional sea level monitoring program in Africa. Since the sea level rise is one of
the main elements that cause shoreline retreat and coastal erosion, it is recommended that the
sea level rise should be taken into consideration in national development planes, particularly
for coastal regions.

REFERENCES

El-Fishawi, N., Khafagy, A., 1991, Sea level characteristics along the Nile delta coast, MBSS
Newsletter, No. 13, pp. 29-37.
Dawod, G., 2001, The magnitude and significance of long-term sea level rise in Egypt from a
geodetic perspective, Proceedings of the Eleventh International Conference on
Environmental Protection, Alexandria University, Alexandria, May 8-10, pp. 207-215
IOC (Intergovernmental Oceanographic Commission), 2002, Manual on sea level
measurement and interpretation, Volume 3: Reappraisals and recommendations as of
the year 2000, UNESCO.
IPCC (Intergovernmental Panel on Climate Change), 2001, Climate change 2001: The
scientific basis, UN Technical report.
Houghton, J., 1997, Global warming: The complete briefing, Cambridge university press.
Pararas-Carayannis, G., 2003, Climate changes, natural and man-made disasters - Assessment
of risks, preparedness and mitigation, Presented at the Climate change - disaster
preparedness workshop, Kiev, Ukraine, October 26-30.
PSMSL, 2002, Enhancing sea level monitoring in Africa as a component of integrated coastal
area management and early warning system, Proposal submitted to African
Partnership Conference, Johannesburg, South Africa, August.
Nassar, M., Baraka, M., and El-Shazly, A., 1998, Modeling local sea surface topography in
Egypt based on zero frequency response analysis, Scientific Bulletin of Faculty of
Engineering, Ain Shams University, V. 33, No. 1.
NOAA (U.S. National Oceanic and Atmospheric Administration), 2001, Sea level variations
of the United States 1854-1999, NOAA Technical Report NOS Co-OPS 36, 80 pp.
581
Sharaf El-Din, S., Ahmed, K., Khafagy, A., Fanos., A., and Ibrahim, A., 1989, Extreme sea
level values on the Egyptian Medetteranean coast for the next 50 years, Proceedings of
the International Seminar on Climate Fluctuations and Water Management, December
11-14, Cairo, Paper No. II-6.
SRI (Survey Research Institute), 2002, The establishment of a modern sea level monitoring
system towards the re-definition of the geodetic vertical datum in Egypt, SRI
Technical Report 2.
Vanek, P. and Kraakiwsky, E., 1986, Geodesy: The concept, Elsevier Science Publishers B.
V., The Netherlands.
Woodworth. P., and Aarup, T., 2003, A report on the status of the GLOSS program and a
proposal for taking the program forward, Technical report No. IOC/INF 1190, 41 pp.




582

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


SEDIMENT TRANSPORT IN APPROACH CHANNELS
TO HARBORS


Abdel Fatah El-Fiky
1
, Mohamed A. Gad
2
*, Mootaz M. Khaled
3
, Ayman A.
Saad
4



1 Professor, Civil Engineering Dept., Ain Shams University
2& 3 Assistant Professor, Civil Engineering Dept., Ain Shams University
4 Ph.D. Candidate, Civil Engineering Dept., Ain Shams University

* Corresponding author. Email: hydroshams@yahoo.CA

ABSTRACT

Accurate simulation of the two-dimensional hydrodynamics and sediment
transport is essential for the planning and design of new navigation channels to
harbors and its protecting breakwaters or the rehabilitation of existing
configurations suffering from excessive sedimentation. In this study, a new
multi-component technique for simulating the 2D depth averaged
hydrodynamics and sediment transport in navigation channels to harbors was
developed. The developed technique requires the application of three numerical
models: NMLONG model (1D depth averaged finite difference wave model),
RMA2 (2D depth averaged finite element hydrodynamic model), and SED2D
(2D depth averaged finite element sediment transport model). The 2D
hydrodynamic model (RMA2) was forced at the flow boundary using longshore
velocity profiles generated by the incoming waves via the application of
NMLONG model. Output of NMLONG provides also the boundary condition
for SED2D which takes a velocity field input from RMA2. The sensitivity tests
showed that the eddy viscosity coefficient E is a sensitive stability parameter in
(RMA2) and both of settling velocity V
sett
and sand grain size D
50

are the most
sensitive parameters in (SED2D).

583
Data collected in the period from 1989 and 1997 for the navigation channel of
Damietta harbor and the surrounding area was used to calibrate the muli-
component technique. The parameters were optimized by minimizing an
objective function (the RMSE between the calculated and measured bed
changes). The results showed that bed changes calculated by the calibrated
technique compare well to the measured data.


KEYWORDS Navigation Channels, Coastal, Hydrodynamics, Sediment Transport

1. INTRODUCTION

Ships having draft more than available surrounding sea depth need an approach channel
for entrance to the coastal ports. Depending upon the depths of the channel and sea bed slope,
the length of the channel varies from a few hundred meters to kilometers. Due to excessive
sedimentation in the navigation channels, maintenance dredging significantly shares the
maintenance cost of the port. Sedimentation in an access channel depends upon the geometry
of the approach channel, in addition to the hydrodynamic factors like waves and currents
prevailing at the site. Sediment characteristics play also a dominant role in the sediment
transport process. For example, to recover the sedimentation problem at the navigation
channel of Damietta harbor (Egypt), periodic annual or less frequent dredging of the channel
has to be carried out (with an annual average of 1.18 * 10
6
m
3
). Chronic shoaling of this
channel has threatening the vessels and cargo ships and the cost of annual dredging is
significantly high. The sedimentation process there is complex and is influenced by the
temporal variability in the direction and intensity of the incoming waves, currents, orientation
of the coastline, and seafloor morphology. Sediments are transported to the sink area in the
navigation channel from adjacent coastal sources at Burullus, Ras-Elbar (El- Asmar, 1995; El-
Asmar and White 1997; El-Asmar and White 2002; Frihy et al. 2002; and Frihy et al., 2004;
and others).
The longshore sediment transport rate constitutes a critical factor in most coastal
engineering designs. The longshore current generated by obliquely incident breaking waves
plays the most important role in transporting sediment in the surf zone. The longshore current
velocity varies across the surf zone, reaching a maximum value close to the wave-breaking
point (Galvin, 1967; Basco, 1982).
Mathematical modeling of sediment transport in coastal areas under the influence of
structures is of great practical interest. Depending on the problem, two different approaches
can be applied to model the nearshore morpho-hydrodynamics: the one-line approach and the
complex approach. The first one takes into account only the integrated result of morph-
hydrodynamic processes and does not include the detailed analyses of the acting mechanisms.
The beach profile in this case is assumed to be close to its equilibrium state and beach
development results from changes in the bulk longshore sediment flux. This approach forms
the basis of the so-called one-line models aimed at the prediction of shoreline evolution under
the influence of structures (Hanson, 1989; Leontyev, 1997). Such approach can poorly
describe the spatial 2D morphological changes in the surf and especially in the off-surf zones
(deeper shallow areas) due to a complicated 2D configuration of coastal structures (such as
those involved in navigation channels to harbors).
An alternative approach involves the modeling of the whole suite of elementary processes
responsible for the local morphological changes in a given area. A typical coastal area model
consists of several modules describing the wave field, the spatial distributions of wave-
584
induced currents and the associated sediment transport, and finally the resulting spatial and
temporal changes of the bed level. Using a grid with relatively small cells (i.e., the finite
difference method), the model of this kind can resolve the detailed patterns of local bed
deformations. Such an approach is employed, for example, in the model complexes developed
by Delft Hydraulics: (De Vriend et al., 1993; Roelvink et al., 1995), Danish Hydraulic
Institute (Broker, 1995; and Broker et al., 1995). The complex approach can be implemented
on a finite element mesh instead of a constant grid, this allows for significant reduction in
computations burden due to the ability of the finite element method to represent the shape of a
complicated 2D configuration with the least number of nodes and vertices (Blumberg, Alan
F., 1996). Attempts to evaluate the morpho-hydrodynamic effect of structures by using these
models are not yet numerous, however the results obtained look encouraging. In addition,
problems arise due to the lack of complete understanding and mathematical representation of
the interactions between the elementary processes (such as wave current interaction for
example). Besides, it is a problem to fulfill the requirement of accurate modeling and, at the
same time, to provide the computational efficiency necessary for practical use.
In the present paper, a multi-component technique capable of simulating the 2D
hydrodynamic and sediment transport in navigation channels to harbors is proposed in a trial
to fill the above mentioned gaps.
In the following sections, the problem of 2D hydrodynamics and sediment transport in
navigation channels is formulated, the components of the developed technique are described,
the functionality of the technique (including its data pre-processing) is demonstrated on an
example case, and finally sensitivity and calibration analysis is presented using data from
Dameitta area.

2. PROBLEM FORMULATION

The following objectives formulate the problems tackled in this paper:

Develop a multi-component technique for simulating the 2D hydrodynamics and
sediment transport in navigation channels.
Collect, geo-reference, and pre-process data from Dameitta area using GIS technology
to facilitate data input streams to the multi-component technique.
Demonstrate the functionality of the technique on example runs.
Perform a sensitivity and calibration analysis using the GIS-processed data.


3. THE MULTICOMPONENT TECHNIQUE

The conceptual model is a multi-component technique that requires the application of three
numerical models: (1) NMLONG, (2) RMA2, and (3) SED2D. The 2D hydrodynamic model
(RMA2) is forced at the inflow boundary using longshore velocity profiles generated by the
incoming waves via the application of NMLONG model. Output of NMLONG provides also
the boundary condition for SED2D which takes a velocity field input from RMA2. Figure (1)
presents a schematic flow chart for the procedure. A brief description of each model is shown
below.




585
3.1 NMLONG model

NMLONG (Kraus and Larson 1991) is a one dimensional depth averaged finite difference
numerical model that simulates wave transformation and longshore current over a bar and
trough beach profile. The wave model equations include the wave energy flux equation,
dispersion relation, cross-shore momentum equation, and the wave number equation. The
longshore current component of NMLONG implements a vertically integrated time-averaged
momentum equation. Details on NMLONG can be found in NMLONG user manual.

INPUT
Beach profile data.
Wave data.
Wind data.

OUTPUT
Cross-shore distribution of depth averaged wave-induced longshore current.
Cross-shore distribution of longshore sediment transport rate.

Note that NMLONG calculated velocity profile is adjusted using a coefficient K
v.
(should
be calibrated). In addition, NMLONG calculated sediment transport rates at the profile nodes
(i.e., boundary condition of SED2D) are adjusted using an adjustment parameter K
sed
(should
be optimized using calibration also) so that:


v NMLONG i
K V V
i
=
.......................(1).
sed NMLONG i
K Q Q
i
=
(2)
Where:

V
NMLONGi
= Longshore current velocity at node i calculated
from NMLONG. [m/s]
V
i
= Actual (i.e., corrected) longshore current velocity. [m/s]
Q
NMLONGi
= Sediment transport rate at node i on the profile
calculated from NMLONG. [m3/s/m]
Q
i
= Actual (i.e., corrected) sediment transport. [m3/s/m]


The reason for these adjustments is to try to account for any local-scale effects or
secondary processes that may not be accounted for in NMLONG empirical formulas. Since
SED2D model takes its boundary condition as concentration values at the boundary nodes, the
sediment corrected sediments discharges calculated from NMLONG (Q
i
) are converted to
depth averaged suspended sediment concentration using:

s
i i
i
s
water
i
i b
d V
Q
Q
Q
C

= = (3)

where:

C
bi
= depth averaged suspended sediment concentration at node i. [kg/ m
3
]
586
K
sed
= sediment correction factor (treated as calibration parameter) [-]

V
i
= longshore current velocity at depth d
i

(calculated from NMLONG) [m/ sec]
d
i
= water depth from (MSWL) [m]

s
= sediment particles density [kg/ m
3
]

Hence, the cross-shore sediment concentration, i.e., boundary condition of SED2D, is
determined using equation (2). Figure (2) shows diagram presenting a schematic NMLONG
output.

3.3 RMA2 model

RMA2 is a two dimensional depth averaged finite element hydrodynamic numerical
model. The model is based on the continuity and the fluid momentum equations. For details
on the model equations, assumptions, and numerical solution schemes, refer to King, 1997.
Additional details can be found in RMA2 user manual. A quick summary of the model input
and output follows.

Input
Study area as a group of feature objects (arcs, nodes, vertices, and polygons)
Finite element mesh
Bathymetry data
Material data (Spatial Manning roughness coefficient (n) and spatial Eddy viscosity
coefficient (E)
Hydrodynamic Boundary conditions

Output
Velocity magnitude and direction at each node
Water surface elevation at each node

3.4 SED2D model

Is a two dimensional depth averaged finite element sediment transport numerical model.
The model solves the convective-diffusion continuity equation of sediment mass and uses a
2D velocity field as input to calculate bottom shear stresses using different shear stress
formulas. In addition to critical shear stress formulas, the model incorporates a bed model to
perform material exchange. Full details of the model theoretical background and its numerical
solution scheme can be found in letter et al. 1998.

Input
2D hydrodynamic velocity field (RMA2 output)
Boundary conditions (nodal U.S. suspended sediment concentrations)
Bed type (sand or clay)
Sediment particle properties (D
50
, V
sett.
, specific gravity, and grain roughness)
Other parameters (diffusion coefficients, fluid density, and gravity, simulation time,
and simulation time step).


587
Output
Sediment concentration at each node at each time step
Bed change at each node at each time step


4. STUDY AREA AND DATA PRE-PROCESSING

Damietta Harbor is located on the northeastern coast of the Nile delta. It was constructed
in 1982 in order to increase the trade potential along the Mediterranean coast. A location was
selected in a coastal embayment at about 9.7 km west of the Damietta Nile mouth with
minimal effects from waves and currents, (Sogreah 1982). The harbor basin was erected
inland and its entrance was protected by two breakwaters. The western breakwater extends
about 1500m parallel to the navigation channel, attaining the 7 m-depth outside sea contour.
The eastern one is about 500m long, perpendicular to the shoreline, and extends to about the 3
m-water depth contour. The navigation channel extends offshore about 11.3 km with an
average water depth of about 15 m, while the width of the channel is 200m from km 0.0 to
km1.45 and 300m from km 1.80 until its offshore end at km 11.30. Since January 1984, the
channel has suffered from excessive sedimentation and subsequently threatening the
navigation activities

The field data collected for the study area includes:

Oceanographic (hydrodynamic) data (waves, currents, and littoral drift).
Bathymetric and bed change data (bed levels in the navigation channel and
surrounding area).
Wind and hydrodynamic data.
Sediment grain size data.

4.1 Waves

Wave action along the Mediterranean coast of Egypt is seasonal in intensity and direction,
and is strongly related to large-scale pressure system over the Mediterranean and the north
Atlantic. Knowledge of the wave climate on different parts of the coastline is essential for the
understanding of the present and future coastline development. Information of local wave
statistics is also essential for proper planning and design of coastal protection works, inlets,
harborsetc. The wave data file was obtained from field measurements carried out by the
Coastal Research Institute (CRI) from January 1991 to December 1991. They measured the
waves by using a pressure S4DW wave/current gauge installed approximately 1200 m at the
western side of the navigation channel, i.e. about 12 m water depth. The wave gauge recorded
the directional wave and current spectrum for 20 minute every 4 hours. These measurements
were corrected for pressure, refraction, and shoaling effects in order to obtain the deep-water
conditions.

Analysis of the Wave Data File

The deep-water wave climate describes the characteristics of waves offshore of the
coastline, i.e. before they are transformed by refraction and other bottom effects. The raw
wave data file is listed in terms of significant wave height, wave period, and direction with
respect to the north direction. Note that the raw wave angels with the north should undergo a
588
rotation transformation to be relative to the shoreline perpendicular before being inputted to
NMLONG.
The obtained wave data file was statistically analyzed using spreadsheets to get the
percentage of occurrence of wave heights with various directions in order to sub-classify the
year into period of approximately constant wave conditions (wave height and angel). the
results are as following:

Total number of stormy days = 37 day with percentage (10.14 %).
Number of stormy days in which the waves generate a left-to-right longshore current
into the study area = 21 day with percentage (5.75 %).
Number of stormy days in which the waves generate a right-to-left longshore current
into the study area = 16 day with percentage (4.39 %).
Total number of normal days = 259 day with percentage (70.96 %).
Number of normal days in which the waves generate a left-to-right longshore current
into the study area = 116 day with percentage (31.78 %).
Number of normal days in which the waves generate a right-to-left longshore current
into the study area = 143 day with percentage (39.18 %).
Number of calm days = 69 day with percentage (18.9 %)

From the above summary, we can divide the wave data file into 7 configurations as shown
in Table (1), which will be used in subsequent analysis in this study.


Table (1) Predominant Wave Configurations in Dameitta Area


4.2 Bathymetric and bed change data

Bathymetry and Bed Changes in the Navigation Channel
For the purpose of dredging works and for checking the navigability of the access channel,
soundings are made periodically before and after dredging works. During the period from
1987 to 1993, Suez Canal Authority (SCA) made soundings for the bed and from 1995 until
present Damietta Port Authority (DPA) made other soundings. The (x , y) coordinates of the
sounding points are referenced in a local coordinate system in which the Y-axis coincides
with the center line of the navigation channel and the origin lies at the harbor entrance.
Actually, the raw data collected by (SCA) are extended to the km5 .00 while the total length
of the navigation channel is 11.30km. That is because the offshore part of the channel does
not suffer significant sedimentation. The raw data are pre-processed using Geographical
Left-right longshore current Right-left longshore current
Wave
characteristics
(1) (2) (3) (4) (5) (6) (7)
Wave height
H
0
(m)
0.75 1.5 2.75 4.5 0.75 1.5 2.4
Wave period
T (sec)
5 6 8 10 5 6 7
Wave angle
(Deg.)
20 35 35 35 -44 -85 -55
Duration
(Days)
86 30 14 7 100 43 16
589
Information Systems (GIS). Figures from (3) to (5) show example cases of pre-processing the
raw data in the navigation channel and its local coordinate system. Each figure shows the
bathymetry in the channel just after a certain dredging and the bathymetry before the next
dredging. The bed change grided-based values are obtained by subtracting the two
bathymetric grids as following:
Using the raster calculator functions of ArcInfo GRID, the amount of bed change can be
obtained in ESRI grided format by the following raster equation:
) t ( ) t t ( t
Bathymetry Bathymetry d =
+
......................(4)
Where:
t = Period (days)
d
t
= Bed change grid during the period t (m)
Bathymetry
(t)
= Bathymetric grid after a certain dredging (m)

Bathymetry
(t+t)
= Bathymetric grid before the next dredging (m)
Thus an average bed change grid can be obtained from:

365 )
t
d
n
1
( d
n
1 i i
t
i

=

=

.......................(5)
Where:
d = Average bed change grid (m)
n = Number of periods studied

Figure (6) shows the average bathymetry and bed change in the navigation channel in the
SCA coordinate system

Bathymetry and Bed Changes in the Surrounding Area

The bathymetry in the surrounding area of the navigation channel was digitized using GIS
from bathymetric maps developed by the Coastal Research Institute (CRI), in an image
coordinate system. The bed change was also digitized from bed change maps in another
image coordinate system.

The Selected Common Coordinate System and Coordinate Systems Transformation

In order to properly overlay and merge the data obtained from the SCA (in local coordinate
system) with those obtained from the CRI (in images coordinate systems) to get the
bathymetry and bed change in the study area, both the (SCA) data and (CRI) data should lie in
the same projection and should use the same coordinate system. Hence a common coordinate
system had to be selected and all the data had to go through a coordinate system conversion to
the common system. The selected common coordinate system is constructed so that that the y-
axis is taken parallel to the local geographic north and the origin is located at the harbor
entrance. The conversions from the different coordinate systems to the common coordinate
system involve only coordinate transformations as following

1- From (SCA) system to the common coordinate system (CCS)

Since the selected coordinate system origin is the same as the origin of the (SCA) data,
the bathymetric and bed change grids were converted to Arc-Info point coverages and
590
the (x , y) coordinates of points features were rotated by an angle of (11.015
0
)
anti-clockwise.

2- From image to the common coordinate system (CCS)

Since the extent of the study area is very small and the distortions due to difference in
projections appear in larger areas, we assumed that the maps of (CRI) lie in the same
projection as the (CCS). Accordingly, the maps are required only to go through "Map-
To-Real world" registration. In order to do the coordinate conversion from image
coordinate system to the (CCS), a set of control (reference) points was selected in both
the image and (CCS). A least square regression analysis (with a 2
nd
degree
polynomial) was done to determine the conversion parameters in order to minimize
the location error.
Figure (7) shows the transformed locations of the bathymetry data sets from both sources in
the common coordinate system whereas Figure (8) shows the locations of the bed change
points after transforming to the common coordinate system.

The Merged Data Set

The bathymetry data from the two different coordinate systems (SCA data and CRI data)
were converted to the (CCS) as described above. The converted data were then merged in one
scatter data set to be used in subsequent analysis in this study. The same procedure was done
to the bed change data to provide one scatter data set for the bed change all over the entire
portion studied of the navigation channel and the surrounding area. The bathymetry scatter
data set was then interpolated using the IDW method and the same was done to the bed
change scatter data set, as shown in Figure (9).


5. EXAMPLE CASES

In this section, the multi-component conceptual model described above is applied in two
stormy example cases on the study area (Damietta harbor navigation channel and adjacent
areas) to demonstrate the functionality of the technique in simulating the 2D hydrodynamics
and sediment transport in navigation channels. The first example case for waves (H
o
= 4m, =
30, T = 9 sec) and wind ( = 45, speed = 15 m/s) causing left-to-right longshore currents.
The second example is for waves and winds from the opposite direction and having the same
wave and wind parameters (i.e., causing right-to-left longshore currents).
The study area was introduced as a set of feature arcs and feature polygons as shown in
Figure (10). The feature polygons were then used to generate the finite element mesh, shown
in Figure (11), and the bathymetric data was then interpolated to the mesh nodes using IDW
spatial interpolation.

Hydrodynamic Boundary Conditions

The nodal velocity vectors B.C. (obtained from NMLONG) were applied at the U.S.
boundary (left boundary in case of left-to-right current and right boundary in case of right-to-
left currents). Figure (12) shows wave-induced longshore current profile and the wave height
profile in the left-to-right case Figure (14) shows wave-induced longshore current profile and
the wave height distribution in the right-to-left case. In both cases, a head boundary condition
of the mean sea water level (100 m) is specified at the D.S B.C.
591
Sediment Concentration Boundary Conditions

The inflow nodal suspended sediment concentration B.C. (obtained from NMLONG) was
applied at the inflow boundary (left boundary in left-to-right case and the right boundary in
the other case). The suspended sediment concentration profile was calculated from equation
(3) with K
sed
= 1 , as shown in Figures (12 e) and (14 e).

Output of the Example Runs

Figures (13) and (15) show the hydrodynamic output (velocity vectors and velocity
magnitude contours) for both example cases. It is clear that the high flow velocity values in
the study area exist in the wave breaking zone (defined at the location of max. longshore
current at the boundary), while low values exist in the navigation channel due to the increase
of water depth. Also, note the circulation pattern (eddy current) between the two breakwaters
is very obvious in both cases.

Figures (13) and (15) show also the results for a 2-day SED2D simulation (bed change and
sediment concentration contours) for both example cases. It can be clearly seen that the max
sedimentation for both cases always occurs at the center of the eddy current which formed
midway between the breakwater tips (a result consistent with the measured data). From the
sedimentation pattern in the navigation channel, it can be seen that sediments can pass around
the tips of the breakwaters from different directions, entering directly into the navigation
channel. The reason for this pattern is that the harbor breakwaters are not long enough to
protect the relatively (relative to the sea floor in the adjacent area outside the channel) deep
part of the navigation channel from sand passing from different directions. The situation at the
eastern breakwater is more dangerous since it constitutes an "open window" to receive sand
approaching from right, and sand re-circulated from left through the eddy current which is
formed between the two breakwaters.


6. SENSITIVITY ANALYSIS AND CALIBRATION

Prior to the calibration process, the sensitivity of the model output response to variations in
different key input parameters was examined. The sensitivity was performed using the typical
"change and see" sequence. Parameters were changed in a systematic manner and the
response was observed. The results are consistent with what is reported in the literature of the
models.
It was found that the eddy viscosity coefficient E is a sensitive stability parameter in
(RMA2). RMA2 solution tends to diverge with small values of E. Increasing E to the stable
eddy viscosity leads to a converging solution. Further increase of the E value above the stable
eddy viscosity has no effect on the velocity field.

In SED2D, both the settling velocity V
sett.
and sand grain size D
50

are the most sensitive
parameters. The sand grain size highly interacts with the settling velocity affecting the
erosion/accretion pattern and the amount of sedimentation in the navigation channel.

Data collected in the period from 1989 and 1997 for the Dameitta harbor navigation
channel and the surrounding area was then used to calibrate the sensitive parameters of the
muli-component conceptual model.

592
6.1 RMA2 check

The CRI conducted a field survey to measure average current velocity in the study area
during April 1997 and January 1998. Since the measured current velocities are only month
averages (not event-based) and since the corresponding average wave conditions were
unknown, the objective of this part was only to check how far NMLONG and RAMA2 output
would be from the average measured velocity values if the models were run on average wave
conditions. This would enable any necessary adjustment of RMA2 parameters. Mannings n
values were assigned based on the physical properties of the bed material, water depth, and
known vegetation following the guidelines specified by (Chow 1959). Eddy viscosity values
were assigned based on previous studies as an initial guess for the stable E, which was
subsequently increased if necessary. Average wave conditions were used in both April 1997
and January 1998. Accordingly, the following parameters were subject to adjustment:

K
v
= Velocity profile correction factor.
n
1
= Manning roughness for shallow areas adjacent to the shoreline
n
2
= Manning roughness for deep area offshore including the harbor area.
E
1
= Eddy viscosity coefficient for shallow areas adjacent to the shoreline.
E
2
= Eddy viscosity coefficient for deep area offshore including the harbor
area.

Table (2) shows the adjusted values of RMA2 parameters. It should be noted again that
any differences are attributed to the poor current data which is only in the form of month-
averages and not event-based and hence was excluded for the presentation here.

Table (2) RMA2 adjusted parameters





6.2 SED2D Calibration

This calibration task constitutes the main calibration of the technique since it highly controls
the erosion/accretion pattern and the amount of sedimentation in the navigation channel which
is the main focus of this study. The parameters involved in the calibration of SED2D are:

K
sed1
= Left suspended sediment profile correction factor
K
sed2
= Right suspended sediment profile correction factor
D
50
= Sediment particle grain size
V
sett.
= Sediment particle settling velocity

The SED2D parameters are optimized by minimizing an objective function. The objective
function is the root mean square error (RMSE) between the calculated and measured bed
changes and is defined as following:

=
=
N
i
meas calc
d d
N
RMSE
1
2 / 1 2
. .
] ) (
1
[ .......................(6)

Parameter K
v
n
1
n
2
E
1
E
2
Value 1.18 0.043 0.03 6000 4500
593
where:

N = The number of mesh points in the navigation channel and adjacent area
d
calc.
= Annual calculated bed change [m]
d
mea
= Annual measured bed change [m]

Table (3) shows the lower and upper bounds as well as the incremental steps used for the
parameters in the global optimization heretical ssearch. The sediment grain size bounds were
specified based on the sediment grab sampler data from the surveying carried out by CRI
(1990-2002). Settling velocity bounds were computed from Gibbs et al. 1971.

Table (3) Limits of the SED2D calibration parameters space

Parameter Lower bound Upper bound Increment
K
sed1
0.5000 1.5000 0.05
K
sed2
0.5000 1.5000 0.05
D
50
0.063 0.113 0.01
V
sett.
0.003 0.008 0.001


Calculation of d
calc


The simulation time chosen for a SED2D run is a "hot start" of 24 hours (i.e., 24 hours
after a steady state of concentration or rate of bed change has been reached). Thus, for a given
run and a given parameters combination, SED2D output is in the form of bed change/day.
Since we are seeking annual calculated bed changes and since wave data was sub-classified
into 7 conditions over the whole year (refer to Table 1), SED2D had to be run for the 7
conditions in order to calculate annual bed change for a given combination of the parameters.
In other words, each wave condition contributes in the calculated bed change at any node
according to the configurations of the wave. Thus, the annual calculated bed change at any
node will be obtained by using the following equation:
(d
calc
) =
i i
i
cal
t d

=
) (
7
1
.(6)
where:
(d
calc
) = annual calculated bed change at a node [m]
(d
cal
)
I
= calculated bed change/day at the node from
wave configuration No. i [m]
t
i
= duration of wave configuration No. i [days]


Results of SED2D Calibration

The Optimum values of the calibrated SED2D parameters are listed in Table (4). Figure
(16) shows the comparison between the annual calculated bed changes and the annual
measured data. It was concluded that there is a good agreement in the spatial patterns and
values between bed changes calculated by the calibrated multi-component technique and the
measured bed changes.


594
Table (4) SED2D optimum parameters

Parameter K
sed1
K
sed2
D
50
V
sett.

Value 1.05 1.1 0.083 0.005


DISCUSSION

The developed technique is computationally not very demanding and with the current
advance in computers, it is recommended for practical engineering usage.
GIS is an essential pre-processing environment for any 2D coastal modeling.
The multi-component technique developed in this study can simulate the 2D
hydrodynamics and sediment transport in Navigation channel. It should be noted that
this methodology is appropriate only to be applied in deeper areas where wave effect
on the re-suspension of sediments is not significantly pronounced. For coastal
structures, near the swash zone or the breaking zone (such as detached break waters)
the technique described here is not suitable in its current conceptual design and
additional modifications should be incorporated.
Based on the analysis, it was found that stormy conditions are the main factor
influencing the amount of sediments transported along the shoreline.
The technique described can be used generally in any navigation channel. However, a
similar calibration analysis of the technique using data from other areas must be
performed before taking decisions based on the application of the technique.
The multi-component technique developed in this study was used to examine different
solutions for shoal reduction in the vicinity of Dameitta navigation channel. PART-II
of this study, which is included in the same proceeding, presents this analysis.


REFERENCES

Basco, D.R., 1982. Surf zone currents. Miscellaneous Report No. 82-7(1). Coastal
Engineering Research Center, U.S. Army Corps of Engineers.

Blumberg, Alan F., 1996. "An Estuarine and Coastal Ocean Version of POM". Proceedings of
the Princeton Ocean Model Users Meeting (POM96), Princeton, NJ.

Broker, I., 1995. Coastal area modelling. MAST 68-M Final Workshop, Gdansk, pp. 2-86-2-
90.

Broker, I., Johnson, H.K., Zyserman, J.A., Ronberg, J.K., Pedersen, C., Deigaard, R., Fredsoe,
J., 1995. Coastal profile and coastal area morphodynamic modelling. MAST 68-M Final
Workshop, Gdansk, pp. 7-12-7-16.

Chow, V.T., 1959. Open Channel Hydraulics. McGraw-Hill Inc., New York, NY.

De Vriend, H.J., Zyserman, J., Nicholson, J., Roelvink, J.A., Pechon, P., Southgate, H.N.,
1993. "Medium-term 2DH coastal area modelling". Coastal Engineering 21, 193-224.
Effects on the construction of the Port of Damietta on the evolution of the littoral drift.
Consultancy Report, number 35/1202/R9, 35 pp.

595
El-Asmar, H.M., White, K., 1997. "Rapid updating of maps of dynamic coastal landforms by
segmentation of Thematic Mapper imagery, example from the Nile Delta, Egypt".
Proceedings of 23rd Annual Conference of the Remote Sensing Society, The University of
Reading. Remote Sensing Society, Nottingham, pp. 515-520.

El-Asmar, H.M., 1995. "Impact of protection structures on physical and sedimentary
parameters along the Damietta coastal area, Nile Delta, Egypt". Egyptian Journal of
Sedimentology 3, 111 - 124.

El-Asmar H.M., and White K., 2002. "Changes in coastal sediment transport processes due to
the construction of New Damietta Harbor, Nile Delta, Egypt". Journal of Coastal Engineering,
46: pp 127-138.

Frihy O.E., El Banna M.M., and El Kolfat A.I., 2004. "Environmental impacts of Baltim and
Ras El Bar shore-parallel breakwater systems on the Nile delta littoral zone, Egypt". Journal
of Environmental Geology, 45: pp 381-390.

Frihy et al., 2002 Frihy, O.E., Badr, A.A., and Hassan , M.S., 2002. "Sedimentation Processes
at the Navigation Channel of the Damietta Harbour on the Northeastern Nile Delta Coast of
Egypt," Journal of Coastal Research, 18 (3), pp 459-469.

Galvin, C.J., 1967. "Longshore current velocity: a review of theory and data. Reviews of
Geophysics". 5 (3), 287- 304.

Gibbs, R.J., M.D.Mathews, and D.A. Link, 1971. "The relationship between sphere size and
settling velocity". J. of Sed. Petrology, Vol 41, pp 7-18.

Hanson, H., 1989. "GENESIS: a generalized shoreline change numerical model". J. Coastal
Res. 5 1 , 1-27.

Joseph V. Letter, Jr. , L. C. Roig, B. P. Donnell, W. A. Thomas, W. H. McAnally, and S. A.
Adamec, Jr. 1998. A user's manual for SED2D-WES version 4.3. US Army Corps of
Engineers-Waterways Experiment Station Hydraulics Laboratory

King, I.P., 1997. RMA2-WES version 4.3 User Manual. US Army Corps of Engineers-
Waterways Experiment Station Hydraulics Laboratory

Leont'yev, I.O., 1997. "Short-term shoreline changes due to cross-shore structures: a one-line
numerical model".Coastal Engineering 31, 59-75.

Nicholas C. Kraus and Magnus Larson 1991. NUMERICAL MODEL FOR SIMULATING
THE LONGSHORE CURRENT. US Army Corps of Engineers. Report 1.

Price, D.M., Chesher, T.J., Southgate, H.N., 1995. PISCES. A morphological coastal area
model. HR Wallindford. Report SR 411.

Roelvink, J.A., Reniers, A.J.H.M., Walstra, D.J.R., 1995. Medium-term morphodynamic
modelling. MAST 68-M Final Workshop, Gdansk, pp. 7-3-7-6.


596

597
N
Depth (m)
Depth (m)
BedChange (m)
18 - 17
17 - 16
16 - 15
15 - 14
14 - 13
13 - 12
12 - 11
11 - 10
10 - 09
09 - 08
08 - 07
07 - 06
06 - 05
05 - 04
18 - 17
17 - 16
16 - 15
15 - 14
14 - 13
13 - 12
12 - 11
11 - 10
10 - 09
09 - 08
08 - 07
07 - 06
06 - 05
05 - 04
-2.0- -1.5
-1.5 - -1.5
-1.0- -0.5
-0.5 - 0.0
0.0- 0.5
0.5 - 1.0
1.0 - 1.5
1.5 - 2.0
2.0 - 2.5
2.5 - 3.0
3.0 - 3.5
3.5 - 4.0
Sept. 30,
1990
May 4,
1991
Period of
219 days
0 500
Meters
# X
Y
Aug. 15,
1992
June 25,
1991
-2.0- -1.5
-1.5 - -1.5
-1.0- -0.5
-0.5 - 0.0
0.0- 0.5
0.5 - 1.0
1.0 - 1.5
1.5 - 2.0
2.0 - 2.5
2.5 - 3.0
3.0 - 3.5
3.5 - 4.0
18 - 17
17 - 16
16 - 15
15 - 14
14 - 13
13 - 12
12 - 11
11 - 10
10 - 09
09 - 08
08 - 07
07 - 06
06 - 05
05 - 04
18 - 17
17 - 16
16 - 15
15 - 14
14 - 13
13 - 12
12 - 11
11 - 10
10 - 09
09 - 08
08 - 07
07 - 06
06 - 05
05 - 04
BedChange (m) Depth (m) Depth (m)
N
Period of
417 days
0 500
Meters
# X
Y


















































Figure (3) Bathymetry and bed change in the navigation channel (Sept.1990- May1991)
Figure (4) Bathymetry and bed change in the navigation channel (June1991- Aug.1992)
598
N
Depth (m) Depth (m) BedChange (m)
18 - 17
17 - 16
16 - 15
15 - 14
14 - 13
13 - 12
12 - 11
11 - 10
10 - 09
09 - 08
08 - 07
07 - 06
06 - 05
05 - 04
18 - 17
17 - 16
16 - 15
15 - 14
14 - 13
13 - 12
12 - 11
11 - 10
10 - 09
09 - 08
08 - 07
07 - 06
06 - 05
05 - 04
-2.0- -1.5
-1.5 - -1.5
-1.0- -0.5
-0.5 - 0.0
0.0- 0.5
0.5 - 1.0
1.0 - 1.5
1.5 - 2.0
2.0 - 2.5
2.5 - 3.0
3.0 - 3.5
3.5 - 4.0
Nov. 19,
1992
July 1,
1993
Period of
224 days
0 500
Meters
# X
Y
-2.0- -1.5
-1.5 - -1.5
-1.0- -0.5
-0.5 - 0.0
0.0- 0.5
0.5 - 1.0
1.0 - 1.5
1.5 - 2.0
2.0 - 2.5
2.5 - 3.0
3.0 - 3.5
3.5 - 4.0
18 - 17
17 - 16
16 - 15
15 - 14
14 - 13
13 - 12
12 - 11
11 - 10
10 - 09
09 - 08
08 - 07
07 - 06
06 - 05
05 - 04
N
Depth (m) Bed Change (m)
(a) Typical after- dredgingbathymetry (m) (b) Average bed changeper year (m)
# X
Y
0 500
Met ers


















































Figure (5) Bathymetry and bed change in the navigation channel (Nov.1992- July1993)
Figure (6) Results of the pre-processing analysis of the bathymetry and bed change data
599
###
# #
#
# #
#
# #
#
#
#
# #
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # #
#
#
# # #
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # #
#
#
# # # # #
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
# # # #
#
# #
# # # #
# # #
# # #
#
#
#
#
#
#
# #
#
# # # # #
#
#
# #
#
#
#
# #
#
#
#
#
# # #
# # #
#
# # #
#
#
#
#
#
# #
#
#
#
#
# # # #
# #
# #
#
#
# # # # # # # #
# # # # # # # # # #
#
#
#
#
#
# # # # # #
# # #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # # # # # # # # #
#
#
# #
# # # # # # # # # # #
#
# #
#
#
#
#
# #
# #
# # #
#
#
# #
#
#
#
#
##
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
# # # # # # # # #
# #
#
#
#
# # # # # #
# # # # #
#
#
#
#
#
# # ## #
# #
#
#
#
# #
#
#
#
#
#
# #
#
#
#
#
# #
#
#
#
#
#
# # #
#
#
#
#
#
#
# # # # # #
#
#
#
#
# #
#
#
#
# # # #
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # #
#
#
#
#
#
# # # # #
#
#
# # # # # # # # # #
# # #
#
#
#
#
# # # # #
#
#
#
# #
#
#
# # # #
# # #
# #
# #
# # #
# #
#
# # # # # #
#
# #
#
#
# #
#
#
#
#
#
#
#
#
# # # #
# # ##
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # # # #
#
# #
#
# #
# #
#
#
# # # # #
#
# # # #
# #
#
# #
# #
# #
#
# # # # # #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # #
#
#
# #
#
#
#
#
#
# # # # # # # # #
#
# # #
###
# #
#
## # # # # # # # # # # #
# # # ## # #
#
#
#
#
#
#
# # # #
# # #
# # # #
# #
#
#
# # # # # # # # #
# # #
# #
#
# # # # # #
#
#
#
#
#
# # # #
#
#
#
# # #
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
# #
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
# # #
# # #
# # # # #
#
#
#
# #
# #
# # # # # #
# # # #
#
#
#
# # #
# ##
#
#
# #
#
# # # # #
# #
#
#
#
# #
## # #
#
#
# #
#
#
#
# #
#
#
#
#
# # # # # # #
#
# # #
#
# # # # # # # # # # # # # #
# # # # # # # # # # # # #
## # # # #
#
#
#
# # # # # # # # #
#
#
#
#
# # #
#
#
# #
#
#
# # # # # #
#
#
# # # # # # # # # #
# ## # # # #
#
#
#
# # # # #
#
# # # #
#
#
#
# # # #
#
#
#
#
#
#
#
# # #
# # # #
#
# #
#
#
#
#
#
# #
#
#
#
#
#
#
# # # # # # # # # # # # #
#
#
#
# # # # # #
##
# # # # # # # # # # # # # # # # # # # # # #
#
# # #
#
#
# # # # #
# # #
#
#
# # # # # # # # # # # # # # # # # # # # # # # # # #
#
#
# #
# # #
# # # # #
# # # # # # # # # # # # # # # # # # # # # #
# # # # # # # # #
#
# # # # # # # # # # # # # # # # # ## # # # # # # # #
#
#
# # # # # #
0 1 2 3 Ki lometers
N
#
X
Y
# ##### ##### ###### ##### ##### ##### # # ##### ##### ##### ##### ##### ##### ## ## ##### #### ##### ##### #### ##### ### ##### ##### ##### #### ##### ## # ##### ##### #### ##### ##### ##### ### # ###### ###### ##### ###### ###### ##### ### #### #### ##### ##### #### ##### ##### # ## ##### ##### ##### ##### ##### ##### # ##### #### #### #### ##### #### ## #### ##### #### ##### ##### ##### # #### ## ##### #### ##### ##### ##### ##### ##### ## ## ##### #### ##### #### ##### #### # ### #### ##### ##### ##### ##### ##### ### ## ###### ###### ###### ###### ###### ###### ### ### ##### ##### ##### ###### ##### ##### # #### ###### ###### ##### ###### ###### ##### ## ###### ###### ###### ####### ###### ###### ##### ## ###### ###### ###### ###### ###### ###### ###### ## ## ###### ###### ###### ####### ###### ###### ###### #### ###### ###### ###### ###### ###### ###### ### ##### ####### ###### ####### ###### ###### ####### ##### # ###### ###### ###### ###### ####### ###### ###### ## ## #### #### ##### #### #### #### #### #### # #### ###### ###### ###### ####### ###### ###### ###### ### ##### ##### ###### ##### ###### ##### ###### ##### ## # #### ##### ##### #### ##### #### ##### # ## ##### #### ##### ##### #### ##### ##### #### # ## #### ##### #### #### #### ##### #### ## #### #### #### #### ##### #### #### ## ##### ###### ###### ##### ###### ###### ###### ##### ## #### ####### ###### ###### ####### ###### ####### ###### ###### # ###### ###### ###### ###### ###### ####### ###### ###### ###### ##### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ##### ###### ###### ###### ###### ###### ##### ###### ###### ###### ###### ###### ### ###### ###### ####### ###### ###### ####### ###### ####### ###### ###### ####### ###### ### ### ##### ###### ###### ###### ###### ##### ###### ###### ###### ###### ##### ###### # ## ###### ##### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ## ###### ##### ###### ##### ###### ##### ###### ##### ###### ###### ##### ### ##### ##### ##### ###### ##### ###### ##### ###### ##### ###### ##### ##### # ##### ###### ###### ##### ###### ###### ##### ###### ###### ##### ###### ###### ### ##### ###### ##### ###### ###### ##### ###### ###### ##### ###### ##### ##### ###### ###### ###### ###### ##### ###### ###### ###### ###### ###### ###### ##### #### ##### ###### ###### ##### ###### ##### ###### ###### ##### ###### ###### ##### ## #### ##### #### ##### #### #### ##### #### ##### #### ##### # ##### ###### ##### ###### ##### ###### ##### ##### ###### ##### ###### ##### ###### ###### ###### ####### ###### ###### ####### ###### ###### ####### ###### ##### ### ##### ###### ##### ##### ###### ##### ###### ##### ###### ##### ##### ###### # ### ###### ####### ###### ####### ####### ###### ####### ###### ####### ####### ###### ## ##### ##### ###### ##### ###### ###### ##### ###### ##### ###### ##### ###### ### ###### ####### ###### ####### ###### ####### ###### ####### ###### ###### ####### ###### # ###### ##### ###### ###### ##### ###### ##### ###### ###### ##### ###### ##### ## ## ###### ##### ##### ###### ##### ##### ##### ###### ##### ##### ##### ###### #### ###### ###### ####### ###### ###### ###### ###### ####### ###### ###### ###### ##### #### ###### ##### ##### ###### ##### ###### ##### ##### ###### ##### ##### #### ###### ####### ###### ###### ####### ###### ####### ###### ####### ###### ###### ####### ### # ##### ###### ###### ##### ###### ##### ###### ###### ##### ###### ##### ###### ## ### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ### ##### ##### ###### ##### ##### ###### ##### ###### ##### ##### ###### ##### ##### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ##### # ##### ###### ##### ##### ###### ##### ##### ###### ##### ##### ###### ##### ## ###### ###### ###### ####### ###### ###### ###### ###### ###### ###### ###### ###### # # ##### ###### ##### ##### ###### ##### ##### ##### ###### ##### ##### ###### ### ###### ###### ###### ###### ###### ##### ###### ###### ###### ###### ###### ###### #### ###### ##### ##### ##### ##### ###### ##### ##### ##### ###### ##### # ##### ####### ###### ###### ####### ###### ###### ####### ###### ###### ###### ####### ### #### ###### ##### ##### ##### ##### ##### ##### ##### ##### ##### ##### ## ## ###### ###### ###### ###### ####### ###### ###### ###### ####### ###### ###### ###### # ## ##### ##### ###### ##### ##### ###### ##### ##### ###### ##### ##### #### ### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### #### ##### ##### ##### ###### ##### ##### ##### ##### ###### ##### ##### ##### #### # ###### ###### ##### ###### ###### ###### ###### ###### ###### ##### ###### ###### ### # ###### ####### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ## ### ##### ##### ##### ##### ###### ##### ##### ##### ##### ###### ##### ##### #### ###### ###### ###### ####### ###### ###### ###### ###### ###### ###### ###### # # ##### ##### ###### ##### ##### ###### ##### ##### ###### ##### ##### ##### ### ## ###### ###### ###### ####### ###### ###### ###### ###### ###### ###### ###### ###### # ## ##### ##### ##### ###### ##### ##### ###### ##### ##### ##### ###### #### ### ###### ###### ###### ###### ####### ###### ###### ###### ###### ###### ####### ###### #### ###### ##### ###### ##### ###### ##### ###### ##### ###### ##### ###### #### ##### ###### ####### ###### ###### ###### ###### ####### ###### ###### ###### ###### #### # ##### ###### ##### ###### ##### ###### ##### ###### ##### ###### ##### ###### ## ## #### #### #### #### #### #### #### #### #### #### #### #### # ## ##### ###### ##### ##### ###### ##### ###### ##### ###### ##### ###### ##### ### ###### ####### ###### ####### ###### ###### ####### ###### ####### ###### ###### ### #### ###### ###### ###### ###### ####### ###### ###### ###### ###### ###### ###### ## ###### ###### ####### ###### ####### ###### ####### ####### ###### ####### ###### ####### ### ##### ###### ##### ###### ##### ###### ##### ###### ##### ###### ##### #### ### ###### ###### ###### ####### ###### ###### ###### ###### ####### ###### ###### ###### # ### ###### ##### ##### ###### ##### ##### ##### ###### ##### ##### ###### ## ###### ###### ###### ###### ####### ###### ###### ###### ###### ####### ###### ###### ## ##### ##### ###### ##### ##### ###### ##### ###### ##### ##### ###### ##### ## # ###### ###### ###### ####### ###### ###### ###### ###### ####### ###### ###### ###### # # ##### ##### ###### ##### ##### ###### ##### ##### ###### ##### ##### #### ### ###### ###### ###### ####### ###### ###### ####### ###### ###### ###### ####### #### ### ###### ##### ##### ###### ##### ##### ###### ##### ##### ###### ##### ###### ###### ###### ####### ###### ###### ###### ###### ###### ###### ###### ####### #### ###### ##### ##### ###### ##### ##### ###### ##### ##### ###### ## # # ##### ###### ###### ###### ###### ###### ###### ##### ###### ###### ###### ###### ### ##### ###### ###### ###### ##### ###### ###### ###### ##### ###### ###### ##### # #### ### #### #### ### #### #### #### ### #### #### ### #### ##### ###### ##### ###### ##### ###### ##### ###### ##### ##### ###### #####
#### ##### ##### ###### ##### ##### ### #### ##### ##### ##### ##### ##### #### ##### ##### #### ##### #### ##### ## # ##### ##### ##### #### ##### #### #### #### ##### ##### ##### ##### ##### #### ###### ###### ###### ##### ###### ##### ## #### ##### ##### #### ##### ##### ### ##### ##### ##### ##### ##### ##### ## #### #### #### ##### #### #### #### ## #### ##### ##### ##### ##### ##### #### ### #### ##### ##### ##### ##### ##### #### ##### #### ##### #### ##### #### ### ##### ##### ##### ##### ##### ##### ##### ###### ###### ##### ###### ###### ###### ###### ###### ##### ##### ##### ###### ##### ### ## ###### ##### ###### ###### ##### ###### #### ### ###### ####### ###### ###### ####### ###### # ###### ###### ###### ###### ###### ###### ###### #### ##### ####### ###### ###### ###### ####### ###### ## ## ###### ###### ###### ###### ###### ###### ##### ### ###### ####### ###### ###### ####### ###### ####### # #### ####### ###### ###### ###### ###### ###### ##### ##### #### #### #### #### #### #### ##### ## # ###### ####### ###### ###### ###### ###### ####### ##### ## ###### ##### ###### ###### ##### ###### ##### #### #### #### ##### #### ##### #### ##### #### # ### #### ##### ##### #### ##### ##### #### ### #### ##### #### #### ##### #### #### ## #### #### #### ##### #### #### #### #### ### ##### ###### ###### ###### ##### ###### ###### #### # ####### ###### ####### ###### ###### ####### ###### ####### ## ##### ###### ###### ###### ###### ###### ###### ###### ###### ### ## ###### ###### ###### ###### ###### ###### ###### ###### ###### ####### ###### ###### ## ### ###### ##### ###### ###### ###### ###### ###### ###### ##### ###### ###### ###### ### ####### ###### ###### ####### ###### ###### ####### ###### ####### ###### ###### ###### ###### ###### ##### ###### ###### ###### ###### ##### ###### ###### ###### ###### ### ##### ###### ###### ###### ###### ###### ###### ###### ###### ##### ###### ###### ### ##### ###### ##### ###### ###### ##### ###### ##### ###### ##### ###### ##### ## ###### ##### ###### ##### ##### ###### ##### ###### ##### ###### ##### ## #### ###### ##### ###### ###### ##### ###### ###### ##### ###### ###### ##### ###### ### ##### ###### ###### ##### ###### ##### ###### ###### ##### ###### ###### # ### ###### ###### ###### ###### ###### ###### ##### ###### ###### ###### ###### ## # ###### ##### ###### ###### ##### ###### ###### ##### ###### ###### ##### ###### ## #### ##### #### ##### #### ##### #### #### ##### #### ##### ## #### ##### ###### ##### ###### ##### ###### ##### ###### ##### ###### ##### ## ### ###### ####### ###### ###### ####### ###### ###### ####### ###### ###### ###### ## ###### ##### ###### ##### ##### ###### ##### ###### ##### ###### ##### ###### ### ###### ####### ####### ###### ####### ###### ####### ####### ###### ####### ###### ##### ## ###### ##### ###### ###### ##### ###### ##### ###### ##### ###### ##### ###### #### ###### ###### ####### ###### ####### ###### ####### ###### ####### ###### ####### ### ### ###### ###### ##### ###### ##### ###### ###### ##### ###### ##### ###### #### ##### ###### ##### ##### ##### ###### ##### ##### ###### ##### ##### ##### ### # ###### ####### ###### ###### ###### ###### ####### ###### ###### ###### ###### ####### # ## ##### ###### ##### ##### ###### ##### ###### ##### ##### ###### ##### ###### ### ####### ###### ####### ###### ###### ####### ###### ####### ###### ####### ###### ###### #### ##### ###### ###### ##### ###### ###### ##### ###### ##### ###### ###### #### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ### ###### ##### ##### ###### ##### ###### ##### ##### ###### ##### ##### ###### ## ## ###### ###### ###### ###### ###### ###### ###### ####### ###### ###### ###### ###### # #### ##### ##### ###### ##### ###### ##### ##### ###### ##### ##### ###### #### #### ###### ###### ###### ###### ###### ###### ###### ###### ###### ####### ###### ### #### ##### ###### ##### ##### ###### ##### ##### ##### ###### ##### ##### ### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ## ## ##### ##### ###### ##### ##### ##### ###### ##### ##### ##### ##### #### ### ###### ###### ####### ###### ###### ###### ####### ###### ###### ####### ###### ###### ## ##### ##### ##### ##### ###### ##### ##### ##### ##### ##### ##### #### ##### ###### ###### ####### ###### ###### ###### ####### ###### ###### ###### ####### ### ##### ##### ##### ###### ##### ##### ###### ##### ##### ###### ##### ##### # # ###### ###### ###### ###### ###### ###### ###### ##### ###### ###### ###### ###### # ### ##### ##### ##### ###### ##### ##### ##### ##### ###### ##### ##### ##### # #### ###### ###### ###### ###### ##### ###### ###### ###### ###### ###### ###### ##### ##### ###### ###### ###### ###### ###### ###### ###### ###### ###### ####### ###### #### ###### ##### ##### ##### ##### ###### ##### ##### ##### ##### ###### ##### ## # ####### ###### ###### ###### ###### ###### ###### ###### ###### ####### ###### #### ### ##### ##### ###### ##### ##### ###### ##### ##### ###### ##### ##### ##### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### #### ##### ##### ##### ##### ###### ##### ##### ##### ###### ##### ##### ###### # ###### ###### ####### ###### ###### ###### ###### ###### ####### ###### ###### ###### ### ## ##### ###### ##### ###### ##### ###### ##### ###### ##### ###### ##### ###### # ## ####### ###### ###### ###### ###### ####### ###### ###### ###### ###### ####### ###### #### ##### ###### ##### ###### ##### ###### ##### ###### ##### ###### ##### ##### #### #### #### #### #### #### #### #### #### #### #### #### ### ##### ##### ###### ##### ##### ###### ##### ###### ##### ###### ##### ###### ## ###### ####### ###### ####### ###### ####### ###### ###### ####### ###### ####### ##### # ###### ####### ###### ###### ###### ###### ###### ###### ###### ###### ###### ##### ### ###### ####### ###### ####### ####### ###### ####### ###### ####### ###### ####### ###### ### ##### ###### ##### ###### ##### ###### ##### ###### ##### ###### ##### # ###### ###### ###### ####### ###### ###### ###### ####### ###### ###### ###### ####### ### # ##### ##### ###### ##### ##### ###### ##### ##### ###### ##### ##### ##### ### ###### ###### ####### ###### ###### ###### ###### ####### ###### ###### ###### ##### ## ###### ##### ###### ##### ##### ###### ##### ###### ##### ##### ###### #### #### ###### ####### ###### ###### ###### ###### ####### ###### ###### ###### ###### #### #### ##### ##### ###### ##### ##### ###### ##### ##### ###### ##### ##### # ###### ###### ####### ###### ###### ###### ####### ###### ###### ####### ###### ###### # # ##### ###### ##### ##### ###### ##### ##### ###### ##### ##### ###### ## ### ####### ###### ###### ###### ###### ###### ###### ####### ###### ###### ###### ### ## ##### ###### ##### ##### ###### ##### ##### ###### ##### ##### ##### #### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ###### ## ###### ###### ##### ###### ###### ###### ##### ###### ###### ###### ##### ###### ## ### #### ### #### #### ### #### #### #### ### #### #### ### ## ### ##### ###### ##### ###### ##### ###### ##### ###### ##### ##### ###### ##
###
# #
# #
#
# #
#
#
#
# #
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # #
#
#
# # #
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # #
#
#
# # # # #
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
# # # #
#
# #
# # # #
# # #
# # #
#
#
#
#
#
#
# #
#
# # # # #
#
#
# #
#
#
#
# #
#
#
#
#
# # #
# # #
#
# # #
#
#
#
#
#
# #
#
#
#
#
# # # #
# #
# #
#
#
# # # # # # # #
# # # # # # # # # #
#
#
#
#
#
# # # # # #
# # #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # # # # # # # # #
#
#
# #
# # # # # # # # # # #
#
# #
#
#
#
#
# #
# #
# # #
#
#
# #
#
#
#
#
##
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
# # # # # # # # #
# #
#
#
#
# # # # # #
# # # # #
#
#
#
#
#
# # ## #
# #
#
#
#
# #
#
#
#
#
#
# #
#
#
#
#
# #
#
#
#
#
#
# # #
#
#
#
#
#
#
# # # # # #
#
#
#
#
# #
#
#
#
# # # #
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # #
#
#
#
#
#
# # # # #
#
#
# # # # # # # # # #
# # #
#
#
#
#
# # # # #
#
#
#
# #
#
#
# # # #
# # #
# #
# #
# # #
# #
#
# # # # # #
#
# #
#
#
# #
#
#
#
#
#
#
#
#
# # # #
# # ##
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # # # #
#
# #
#
# #
# #
#
#
# # # # #
#
# # # #
# #
#
# #
# #
# #
#
# # # # # #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# # #
#
#
# #
#
#
#
#
#
# # # # # # # # #
#
# # #
###
# #
#
## # # # # # # # # # # #
# # # ## # #
#
#
#
#
#
#
# # # #
# # #
# # # #
# #
#
#
# # # # # # # # #
# # #
# #
#
# # # # # #
#
#
#
#
#
# # # #
#
#
#
# # #
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
# #
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
# # #
# # #
# # # # #
#
#
#
# #
# #
# # # # # #
# # # #
#
#
#
# # #
# ##
#
#
# #
#
# # # # #
# #
#
#
#
# #
## # #
#
#
# #
#
#
#
# #
#
#
#
#
# # # # # # #
#
# # #
#
# # # # # # # # # # # # # #
# # # # # # # # # # # # #
## # # # #
#
#
#
# # # # # # # # #
#
#
#
#
# # #
#
#
# #
#
#
# # # # # #
#
#
# # # # # # # # # #
# ## # # # #
#
#
#
# # # # #
#
# # # #
#
#
#
# # # #
#
#
#
#
#
#
#
# # #
# # # #
#
# #
#
#
#
#
#
# #
#
#
#
#
#
#
# # # # # # # # # # # # #
#
#
#
# # # # # #
##
# # # # # # # # # # # # # # # # # # # # # #
#
# # #
#
#
# # # # #
# # #
#
#
# # # # # # # # # # # # # # # # # # # # # # # # # #
#
#
# #
# # #
# # # # #
# # # # # # # # # # # # # # # # # # # # # #
# # # # # # # # #
#
# # # # # # # # # # # # # # # # # ## # # # # # # # #
#
#
# # # # # #
#
# #
#
#
# # #
# #
#
#
#
#
#
# #
# #
#
#
#
#
#
#
# #
#
#
#
#
#
(a) Bathymetry points digitized
from the Shore Protection
Authority Maps after coordinate
transformation.
(b) Bathymetry points obtained
from Suez Canal Authority after
coordinates transformation.
(c) The complete merged bathymetry
points in the selected common
coordinate system.
#
##
#
#
#
# # #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
##
#
#
#
#
#
#
# #
#
# #
#
#
#
#
# ##
#
#
#
#
#
#
#
#
#
#
#
##
#
#
#
#
#
#
#
# #
#
#
#
#
#
# #
#
#
# #
#
#
#
# #
#
#
# #
# #
#
# #
#
#
#
#
# #
#
#
# #
#
##
#
#
#
#
#
#
####
#
#
#
#
#
#
#
#
#
#
#
#
#
#
## ##
#
#
#
#
#
#
#
#
# #
# #
# # #
#
#
#
# # #
# #
# #
##
#
#
# #
# #
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
# #
#
##
#
#
#
#
#
#
#
#
#
#
###
#
##
#
#
#
##
#
# #
#
#
#
#
##
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
######
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
##########
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
# #
#
#
#
# # #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
# # #
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
# #
#
#
#
#
#
# #
#
#
# #
#
#
#
# #
#
#
# #
# #
#
# #
#
#
#
#
# #
#
#
# #
#
# #
#
#
#
#
#
#
### #
#
#
#
#
#
#
#
#
# ##
#
#
#
# # # #
#
#
#
#
#
#
#
#
# #
# #
# # #
#
#
#
# #
#
# #
# #
# #
#
#
# #
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
# #
#
# #
#
#
#
#
#
#
#
#
#
#
# # #
#
# #
#
#
#
# #
#
# #
#
#
#
#
##
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
#
#
#
#
# #
(c) The complete merged bed
change points in the selected
common coordinate system.
(b) Bed change points obtained
from Suez Canal Authority after
coordinates transformation.
(a) Bed change points digitized
from the Shore Protection
Authority Maps after coordinate
transformation.
N
#
X
Y
0 1 2 Kilomet er s


















































Figure (7) The bathymetry scatter data sets obtained from the different sources and
the slected common coordinate system
Figure (8) The bed change scatter data sets obtained from the different sources and
the selected common coordinate system
600


















































Bed levels relative to MSWL of (100.00) m
(a) IDW interpolated BATHYMETRY from the complete
merged bathymetry scatter data set
Bed change (m)
(b) IDW interpolated BED CHANGE values
from the complete merged bed change scatter
data set
Figure (9) Bathymetry and measured bed change in the navigation channel and
surrounding area
601


















































FEM
Detailed FEM around
breakwaters
Figure (11) Finite Element Mesh (FEM)
Figure (10) Map of the study area showing feature objects
602
0 500 1000 1500 2000 2500 3000 3500 4000 4500
Offshore Distance (m)
0
0.1
0.2
0.3
0.4
V
e
l
o
c
i
t
y
(
m
/
s
)
(b)
0 500 1000 1500 2000 2500 3000 3500 4000 4500
Offshore Distance (m)
-12
-10
-8
-6
-4
-2
0
D
e
p
t
h
(
m
)
(a)
0 500 1000 1500 2000 2500 3000 3500 4000 4500
Offshore Distance (m)
0
0.5
1
1.5
2
2.5
3
3.5
W
a
v
e
H
e
i
g
h
t
(
m
)
(c)
0 500 1000 1500 2000 2500 3000 3500 4000 4500
Offshore Distance (m)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
S
e
d
i
m
e
n
t
T
r
a
n
s
p
o
r
t
R
a
t
e
(
m
3
/
s
/
m
)
(d)
0 500 1000 1500 2000 2500 3000 3500 4000 4500
Offshore Distance (m)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
C
o
n
c
e
n
t
r
a
t
i
o
n
(
k
g
/
m
3
)
(e)


















































K
sed
= 1.0
Figure (12) U.S. boundary conditions for the left-to-right example case
603


















































Velocity vectors
Detailed velocity vectors around breakwaters
Velocity contours (m/sec)
Bed change (m) Sediment concentration (kg/m
3
)
Figure (13) RMA2 and SED2D output for the left-to-right example case
604
0 500 1000 1500 2000 2500 3000 3500
Offshore Distance (m)
-12
-10
-8
-6
-4
-2
0
D
e
p
t
h
(
m
)
(a)
0 500 1000 1500 2000 2500 3000 3500
Offshore Distance (m)
0
0.1
0.2
0.3
0.4
V
e
l
o
c
i
t
y
(
m
/
s
)
(b)
0 500 1000 1500 2000 2500 3000 3500
Offshore Distance (m)
0
0.5
1
1.5
2
2.5
3
3.5
W
a
v
e
H
e
i
g
h
t
(
m
)
(c)
0 500 1000 1500 2000 2500 3000 3500
Offshore Distance (m)
0.6
0.8
1
1.2
1.4
1.6
1.8
C
o
n
c
e
n
t
r
a
t
i
o
n
(
k
g
/
m
3
)
(e)
0 500 1000 1500 2000 2500 3000 3500
Offshore Distance (m)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
S
e
d
i
m
e
n
t
T
r
a
n
s
p
o
r
t
R
a
t
e
(
m
3
/
s
/
m
)
(d)


















































Figure (14) U.S. boundary conditions for the right-to-left example case
K
sed
= 1.0
605


















































Velocity vectors
Detailed velocity vectors around breakwaters
Bed change (m) Sediment concentration (kg/m
3
)
Figure (15) RMA2 and SED2D output for the right-to-left example case
Velocity contours (m/sec)
606












































Bed change (m) Bed change (m)
(a) Annual measured bed changes (b) Annual calculated bed changes
Figure (16) Comparison between calculated bed changes and measured data in
Damietta harbor navigation channel and surrounding area
607

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


INFLUENCE OF THERMAL TREATMENT ON THE
PROPRIETIES OF
KIESELGUHR USED AS AN AID FILTER


ATOUI Lhadi
1
, AMRI Rdha
2

1: Professor at the metallurgy department and genius of materials of the
University of Annaba
2: Researcher at the metallurgy department and genius of materials of the University of
Annaba

Abstract

Kieselguhr or diatomaceous earth consists of the skeletons of diatoms [18].
Instead of carbon, these algae deposit silicates from the water in order to form
their skeletons in a precise and symmetrical shape [19].
This study shows the influence of thermal treatment on the physical and
chemical proprieties of kieselguhr.
The effect of the thermal treatment on the size distribution, morphology of the
particles, hydraulic resistance and ability of keeping is remarkable.
This study allowed drawing information in order to improve the quality of the
kieselguhr to be used as an aid filter.


1. Introduction

Filtration is an operation which consists with the separation of two phases one continues and
the other dispersed using a filter medium, of which it is permeable that for only one phase [2].
In the chemical and agro alimentary industry one finds different types of filtration however
precoat filtration is used [14]. This technique calls upon the use of materials pulverizing,
called the additives of filtration (aid filter). The purpose of these materials are to improve the
operation of filtration [10],[11] (to have a good flow of filtration, a cycle of filtration more
prolonged and a quite clear filtrate). Kieselguhr is the one of the most important additives of
filtration and most used, the use of the rough kieselguhr led to bad results, therefore it must
undergo heat treatments and/or chemical for that it can be used like an aid filter. In this work
one treated rough kieselguhr for several lengths of time with 1000C. The products obtained
are used in the filtration of water. The results obtained are presented in this work.
608

2. Experimental Details

2.1. Preparation of the samples

he studied samples have sudden thermals treatments under air with 1000C according to the
thermal cycle presented in Fig 1.











Fig1. Cycle of heat treatment


The processing time is respectively 30-60-90-120 min, after having to finish the treatment,
one puts the samples treated in a dissector to cool and so that it is prevented that they absorb
moisture

Note: One indicates by E1, E2, E3 and E4 the samples treated during 30, 60, 90 and 120
minutes respectively, and by E0 the rough sample. One indicates the fraction lower than
75m of the E2 sample by E5.

2.2. Apparent density

The apparent density is given using a graduated test-tube. One fixes a weight and one
determines.
Its volume by the test-tube, then we determine the density [9].


2.3. Granulometry

Distribution For the fractions higher than 75m we use sieves whose their openings are
respectively: 75m-150m-250m-500m. The granulometry of the fractions lower than
75m is to determine using optical microscope [1].

2.4. Morphology of the grains

The morphology of the grains is determined by optical microscope and MEB. The observation
by MEB requires a metallization of the powder [1], [23].

2.5. Hydraulic resistance

To determine hydraulic resistance, we carries out a filtration with constant pressure of a
suspension formed by kieselguhr and distilled water (the filter is supplied by gravity) [10],
S
a
m
p
l
e

(
1
)
S
a
m
p
l
e

(
2
)

S
a
m
p
l
e

(
3
)
S
a
m
p
l
e

(
4
)

TC
1000C
0 30 60 90 120 Time of
treatment
609

[20]. One thus records time the volume of recovered filtrate. We plot the curves (t/v, v) which
present rights.
Hydraulic resistance is given by the following formula:


=2.p..
2
tan/.w


p: the difference in pressures between the upstream and the downstream of the filter:
:section of the filter.
Tan: the slope of the right (t/v, v).
:dynamic viscosity of water.

W=.S/1-ms

m: coefficient of moisture, it is expressed as follows:

m=1+./ (1-).
s


:density of the filtrate.

s
:density of kieselguhr.
S: mass content of the suspension of solid matter.

S=M
s
/M
p


M
s
: mass of dry solid of the suspension.
M
p
: mass of suspension.

2.6. Capacity of retention

We prepare kieselguhr and worn water suspensions whose mass contents are respectively:
0.01-0.025-0.05-0.1, and we mix the suspensions during 15min, and filters on a filter paper,
using an optical turbidimeter [22], [25]. The turbidity of each suspension is determined.

3. Results and discussions

Rough kieselguhr presents the value of the most important density, 0.29 g/ml, the presence of
moisture and the organic matter is the origin of this value, the E1 sample presents the lowest
value 0.17 g/ml, therefore has elimination of moisture and the organic matter.

The density increases value for the samples which underwent more prolonged processing
times, for E4 the density equalizes 0.23 g/ml, therefore we can say that the apparent packing
is due to the agglomeration of the particles.

Table 1 Density connects samples

Sample E0 E1 E2 E3 E4
Densit
(g/ml)
0.29 0.17 0.21 0.22 0.23

610


The crude contains very fine particles, the mode for the crude is 2.5m, and as an aid
filter should have in preference only the particles having a diameter higher than 10m [
],therefore such a granulometry leads to problems in filtration. The average is 32m, the
difference between the mode and the average is important thus the average does not represent
well the granulometry of the sample, granulometry changes according to the processing time,
the mode for E4 is 32m, the average is 132m. we can as note as the difference between the
mode and the average of all the samples is important, which implies a bad representation of
the average. The relative curves are presented in figures 2, 3, 4, 5 and 6.
According to the images of the optical and electronic microscope, the grains have porous
structure photographs 1 and 3. The shape of healthy grain is round; they have the aspect of a
hive of bees.
The images of the optical microscope do not show well the effect of the treatment on
porosity, but they show the effect on the form and granulometry. According to photograph 2,
we can notice the change of the form the structure and the size of the grains.
The images by MEB allowed us to note the effect of the heat treatment on porosity.
The pores have undergoes a change photographs 4 and 5.



Table 2 Capacity of retention for the various samples










It is noted here that the turbidity of distilled water is 0 NTU.
The higher limit of the reading of turbidimeter used is 400 NTU.



The crude kieselgur have a hydraulic resistance equalizes with 6.34.10
-10
m/Kg, it is
more important than those of the sample is treated E1, E2, E3, E4 whose values are 0.31.10
-10
,
0.38.10
-10
, 0.26.10
-10
and 0.26.10-
10
respectively, and that is due to their granulometric
distribution, the presence of the fine particles increases the specific surface in an important
way what implies the increase in hydraulic resistance. The curves of filtration are presented in
figures 7, 8, 9, 10 and 11.

The crude kieselgur presents the better Capacity of retention of all the samples. The
moisure-holding capacity decreases according to the processing time, plus the treatment is
prolonged more the kieselgur obtained has a weaker moisure-holding capacity. This fact there
is not related to the size of the grains like showed it the E5 sample of which it presented a
weak moisure-holding capacity, then one can say that this property is related to the structure
of the grains of kieselgur because the latter is very affected by the heat treatment, in more the
particles agglomerated will lose their porous structure by the effect of sintering.
S.100 0 1.25 2.5 5 10
E
0
Higher than 400 253 18.1 17.7 5.4
E
1
Higher than 400 Higher than 400 102 22.5 12.5
E
2
Higher than 400 Higher than 400 189 29.3 15.8
E
3
Higher than 400 Higher than 400 220 33.5 24
E
4
Higher than 400 Higher than 400 223 35 26
T
u
r
b
i
d
i
t

N
T
U

E
5
Higher than 400 Higher than 400 Higher than 400 192 46

611


Conclusion

The heat treatment thus the time of maintenance has a great influence on the
physicochemical properties of kieselguhr, the experiments carried out made it possible to
draw the results following: - granulometry and granulometric distribution: granulometry
is affected by the heat treatment, a treatment of 120 min with 1000C is adequate to have
the better granulometry so that produces it obtained either used in the filtration whose
diameter average correspondent is equal to 132.44 %m, moreover the granulometric
distribution better is that weakest comparatively with the other samples.
Density and apparent porosity: a treatment of 30 min with 1000C gives the density
connect weakest .0.178 g/ml, thus porosity connect strongest 0.92.
Hydraulic resistance: more the treatment is prolonged more hydraulic resistance of the
products is weak, a treatment of 120 min with 1000C gives a product with a hydraulic
resistance equalizes with 0.26.10
-10
m/Kg. The crude has a hydraulic resistance equalizes
with 6.34.10
-10
m/Kg.
Capacity of retention: the better Capacity of retention is that of the crude then products
which underwent the shortest treatment.
The morphology of the grains: morphology is also affected by the heat treatment and the
time of maintenance. Micrographies enabled us to observe different the changes on
morphology from the grains.
The chemical behavior: the crude thus contains calcite it presents a bad inertia a treatment
of 60 min ensures the disappearance of calcite.

And as a good aid filter must have the following properties:

A good granulometric distribution and a good granulometry an important porosity, low
hydraulic resistance, good capacite of retention and good chemical inertia.
Thus the time of maintenance has positive effects and the other negative ones on the
quality of the kieselguhr intended for filtration, which leads us to make a compromise in the
choice of the processing time and consequence to draw several varieties from kieselguhr
according to the processing time, and of this fact we can make the following
recommendations:

The use of rough kieselguhr like an aid filter led to bad results: a low flow of filtration, a
bad chemical inertia, difficulty of formation of the cake on the medium of filtration.
A processing time of 60 min is sufficient for the treatment of kieselguhr destinies to the
filtration of the suspensions of low or average viscosity and turbidity; for the suspensions
of low turbidity and viscosity, we use the fractions higher than 75 m kieselguhr treated
with 60 min and if we work with low flow and for the strongly turbid and viscous
suspensions, we use the kieselguhr treated during 30 min (the same variety is used in the
prelayer and proportioning).
If we work with high flow, one uses in the prouche fraction higher than 75m and for the
proportioning of the kieselguhr treated during 30 min.

References

1. Particule size measurment powder simpling and particule size measurment,
Terence Allen, (5th edition)-V1. Champan and hall-USA, (1997), (44-187).
612

2. Filters and filtration. Hand Book. Christopher Dickenson, (4th edition), Elsevier
advanced technology-British, (1997), (324-329).
3. Encyclopedia of chemical technology. J ohn Wiley and son. V5-(3rd edition).
Ceramics (thermic treatment), (1997), (206-66).
4. Encyclopedia of chemical technology. J ohn Wiley and son, (3rd edition)-V7-
USA,(1997), (603-614).
5. Etude de ladsorption de composs organiques sur des alumino-silicates
naturelles traites. Zabat Nacera, Mmoire de magistre, dpartement de gnie
chimie. Universit de Annaba, (2000).
6. Powder hand book of science and technologie. Muhammede Fayd. (2nd edition),
champan and hall-New York, (1997), (1-90), (983-723).
7. Influence de la temprature de traitement sur les proprits physico-chimiques du
systme kieselguhr kaolin. Tabi Yasmina, mmoire de magister, dpartement de
mtallurgie et gnie des matriaux. Universit de Annaba, (2001).
8. Influence des paramtres de dpts sur les caractristiques dun transistor effet
de champ grille isole en couches minces base de silicium amorphe
hydrogn. Masbah Ammar, Mmoire de magistre, Institut de sciences exactes,
Universit de Batna, (1999).
9. Perrys Chemical engineers Hand Book. Robert Perry, (6th edition), Mc graw H
internationl edition-USA, (1987).
10. Technique de lingnieur J 3. Filtration introduction, J 3500-1-J 3600-4
11. Technique de lingnieur J 3. Filtration sur support, J 3501-1-J 350515.
12. Ullmanns encyclopedia of industrial chemistry. (6th edition)-VA23. Silica
Kieselguhr,(2003), ( 607-614)
13. Ullmanns encyclopedia of industrial chemistry. (6th edition)-VB2. Solid liquid
separation, introduction, (2003), (9.1-9.7)
14. Ullmanns encyclopedia of industrial chemistry. (6th edition)-VB2. Filtration,
(2003), (10-1, 10-14)
15. Ullmanns encyclpedia of industrial chemistry. (6th edition)-VB2. Particle size
Analysis characterization of classification processm 2003m (2,1 2,24)
16. Analyse quantitative, V. Alexev. Edition mir, (1981).
17. Particule measurment surface area and pore size determination. Terence Allen.
Champan a hall, (4th edition)-V2-USA, (1997), (1-6).
18. Materials Hand Book. Georges Brady. (3rd edition), Clauser, MC graw Hill INC,
(1991), (451-452) Kieselguhr, 270: Diatomaceous earth.
19. Trait de cramiques et matriau minraux. Ca jouenne, dition septime 14 Rue
falgiere 75015 paris, (1990).
20. Unit operation of chemical engineering. Warren.Mccabe. (4th edition). By graw
Hill-Singapore, (1985), (869-878).
21. Powder technology. Koichi linaya Lohn keith -HEMISPHERE Publisting-USA,
(1984).
22. LAnalyse de leau. J ean Rodier (8me dition), DUNDO-Paris, (1996) (67-71).
23. Mthodes physiques dtude des minraux et matires solides. J -perberhart.
Doin-Paris, (1976), (449-451, 467).
24. Size enlargement by agglomeration. Wolf Gring Pietsch. Wileg, England, (1991).
25. Solid liquid separation. Ernest J .Henleg. J ohn Wiley and son-Canada, (1981).
26. lessentiel de la chimie. Pascal CLAVIER. Ellipses-Paris (1999).


613

Internet

1. www.cmeba.univ-rennesl.fr
2. www.educnet.education.fr
3. www.haleco.fr
4. www.begerow.de
5. www.minerals.usgs.gov
6. www.epa.gov/ttn/chief/ap42/ch11
7. www.state.ok.us















Photo1. Microstructure of grains de E
0
















Photo 2. Microstructures of grains of size higher than 75 m of E4









50 m
100 m
614

















Photo 3. Microstructure of a grain de E
0


















Photo 4 . Microstructure of a grain of E4














Photo 5 .

615


















Photo 6. Microstructure of the pores of a grain














Figure 2. Relative curve of E
0
Figure 3. Relative curve of E
1















Figure 4. Relative curve of E
2
Figure 5. Relative curve of E
3

courbe relative de E 0
0
0.5
1
1.5
2
2.5
3
0 20 40 60 80 100 120
Diametre moyen (m)
p
o
u
r
c
e
n
t
a
g
e

P
o
u
r
c
e
n
t
a
g
e
/

m

Diamtremoyen (m)
courbe relative de E 1
0
0.5
1
1.5
2
2.5
0 100 200 300 400 500 600 700
P
o
u
r
c
e
n
t
a
g
e
/

m

Diamtremoyen (m)
courbe relative de 2
0
0.5
1
1.5
2
2.5
0 100 200 300 400 500 600 700
P
o
u
r
c
e
n
t
a
g
e
/

m

Diamtremoyen (m)
courbe relative de 3
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
0 100 200 300 400 500 600 700
P
o
u
r
c
e
n
t
a
g
e
/

m

Diamtremoyen (m)
616















Figure 6. Relative curve of E
4
Figure 7. Curve of filtration of. E
0















Figure 8. Curve de filtration de E
1
Figure 9. courve de filtration de E
2















Figure 10. Curve of filtration of E
3
Figure 11.curve of filtration de E
4



courbe relative de E 4
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
0 100 200 300 400 500 600 700
P
o
u
r
c
e
n
t
a
g
e
/

m

Diamtremoyen (m)
Courbe de filtration de E 0
y=0.0096x+0.0595
R
2
=0.9983
0.000
0.100
0.200
0.300
0.400
0.500
0.600
0.000 10.000 20.000 30.000 40.000 50.000 60.000
V (ml)
t/V

(m
in
t
/
v

(
m
i
n
/
m
l
)

V (ml)
Courbe de filtration de E 1
y=0.001x+0.0713
R
2
=0.6722
0.000
0.020
0.040
0.060
0.080
0.100
0.120
0.140
0.000 10.000 20.000 30.000 40.000 50.000 60.000
V (ml)
t/V

(m
in
t
/
v

(
m
i
n
/
m
l
)

V (ml)
Courbe de filtration de E 2
y=0.0006x+0.0114
R
2
=0.9635
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.000 10.000 20.000 30.000 40.000 50.000 60.000 70.000 80.000
V (ml)
t/V

(m
in
t
/
v

(
m
i
n
/
m
l
)

V (ml)
Courbe de filtration de E 3
y=0.0004x+0.0111
R
2
=0.9867
0.000
0.005
0.010
0.015
0.020
0.025
0.030
0.035
0.000 10.000 20.000 30.000 40.000 50.000 60.000
V (ml)
t/V

(m
in
t
/
v

(
m
i
n
/
m
l
)

V (ml)
Courbe de filtration de E 4
y=0.0004x+0.0321
R
2
=0.9734
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0.000 10.000 20.000 30.000 40.000 50.000 60.000 70.000 80.000
V (ml)
t/V

(m
in
t
/
v

(
m
i
n
/
m
l
)

V (ml)
617

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005









THE IMPACT OF CASPIAN SEA LEVEL VARIATION
ON SHORELINES


Prof. Y. Hassanzadeh
Faculty of Civil Engineering, University of Tabriz,
Tabriz, Iran,
Fax: 00984113344287
email : yhassanzadeh @ tabrizu. ac. ir


Abstract
Several simulation of the earths climate indicates that the planet may undergo
significant warming over the next century. Consequently, in Caspian Sea basin
we can expect a rise in mean annual temperature of about 3
oc
while rainfall will
vary but generally increase across the whole watershed. As the climate warms,
the Caspian Sea will expand, some of the polar ice will melt and sea water
surface level will rise. Now, with a solid scientific consensus on the subject of
global warming, the hard questions are about what the resultant effects will be
on Caspian Sea shoreline populations and environments.
By using different simulations of global weather patterns, we will be able to
make detailed estimations about crop response, soil erosion, changes in income
and unemployment, which will stem from a warmer planet.
The present examination of the effect of Caspian Sea level variation on Iranian
shoreline is intended to provide quantitative results and explore the extent and
nature of such changes.



Ked words: Sea level change, Caspian Sea, Greenhouse effect, Shoreline



635

1- Introduction

The enhance of greenhouse effect has been entrained the certain unfavorable
conditions in the Caspian Sea level. In the northern area of Iran, the rise of
Caspian Sea level has affected the shoreline, economic use of the coast, natural
habitats, tourist resorts and the loss of livelihood. Increasing of peak rivers on
one hand and rising of the Caspian Sea level on the other, results in the
inundated farms, and under the hydrodynamic and wave actions, the unprotected
shoreline are eroded and deformed to the state of dynamic equilibrium. Thus, it
is recommended to design and realize the transverse and longitudinal structures
along the coast to stop the erosion phenomenon, littoral drift and protection of
the coast from mainly frontal wave attacks.
Though prediction about the extent of the greenhouse effect are notoriously
unreliable, various scenarios suggest a global atmospheric warming of 1-3
oc
over
the next 50 years. This could produce a rise in mean sea level (M.S.L.) of
between 0.3 and 1.5m together with an increase in rainfall mainly at higher
latitudes. Increase in M.S.L. of this magnitude are unlikely to produce
significant change in shelf sea tides.
Thus while tides at estuary mouths may remain unchanged, the change in M.S.L.
may significantly affect tidal propagation within shallow estuaries. Such
changes may extend not only to elevations and currents but also to the mixing
and sedimentary regime and be especially important for increased intrusion of
salt water. Accompanying changes in rainfall, and hence river flow, may further
complicate predictions of the latter effect [5,6].
The tidal conditions, including Caspian Sea Level (C.S.L.), could be analyzed
by means of mean tide measured data. Data from various continuously and
temporarily recording station along the Russia , Kasakhestan, Turkmanistan,
Iran and Azerbaijan coasts have been used. Further details are still under
investigation in the different Coastal Research Centers of the related countries.
The present examination of the effect of a rise in C.S.L. on tides and currents in
estuaries is not intended to provide quantitative results but more to explore the
extent and nature of such changes.
In this paper the environmental impacts of the rising level of Caspian Sea on
the shoreline of the northern region of Iran by means of the observed and
collected field data has been described.

2- Flood climatology

The climate along the north coast of Iran is subtropical, with humid seasons.
The average annual precipitation of northern basin of Iran is about 488,8mm,
and its watershed area is 177000 km
2
with an average annual Runoff of 14865
MCM. Runoff volume and rate are highly sensitive to precipitation volume and
636
intensity, and because of the low soil moisture capacity all of the additional
hurricane-induced precipitation due to climate change becomes runoff. As a
result, the relative increase in runoff volume will exceed the relative increase in
precipitation volume according to the following relationship:

r
p
p
p
r
r
V
V
V
V
V
V
.

Where V
r
and V
p
represent respectively the original volumes of runoff and
precipitation, and V
r
and V
p
show the incremental increases in runoff and
precipitation. As an example if 80% of the original precipitation becomes runoff
) 25 . 1 8 . 0 ( = =
r
p
p
r
V
V
or
V
V
, with a 40% increase in precipitation ) 4 . 0 ( =

p
p
V
V
, runoff
volume ) (
r
r
V
V
will increase by 40% . 1.25=50%.
The increased precipitation provokes substantially greater flooding that could
negate most of the benefits of present water-management practices in this basin.
The hydrologic problem is to determine how the design storm for a given
recurrence interval and duration may change. Design storm data are usually
complied by analysis of historical precipitation data.
An underlying assumption of the design-storm method is that the climate is
steady and precipitation statistics are stationary, i.e., that historical events may
be used with a high degree of confidence to characterize future events. Climate
change violates this assumption. Emanuel [2,3] postulates that the
thermodynamics of a mature hurricane can be described in terms of Carnot cycle
driven by the temperature difference between the sea surface and the upper
atmosphere. The hurricane-fueling potential is latent heat from the sea surface,
realized by the system when water vapor condenses to fall out as rain. Using a
model based on this approach, he can compute the maximum sustainable
difference in surface pressure between the center and outside of a hurricane, p
s
,
which is a measure of hurricane. The change in p
s
under global warming can
be estimated using output from general circulation models of the earths climate
that have been subjected to a doubling of atmospheric CO
2
. As discussed by
Emanuel [3], there are three principal factors of the background environment in
which the hurricane is embedded that determine p
s
: Sea surface temperature,
atmospheric relative humidity near the sea surface, and the temperature
difference between the surface and the lower stratosphere. Using general
circulation model predictions and historical experience, he argues that his model
is relatively insensitive to the changes in relative humidity and surface-
stratosphere temperature difference expected to occur under a doubling Co
2
. As
consequence, the greatest impact on p
s
under global warming would come
from warmer sea surface temperature.
637

3- Hydrodynamic and Hydrology problem

In northern shoreline of Iran, the rise of Caspian Sea-level (Fig. 1) is
affected the economic use of the coast and natural habitats. The major effect on
people has been loss of livelihoods [Fig.4]. The tourist resorts are also swamped
or lose their sandy beaches. The process involves not only a shift in the
shoreline, but as increase in erosion as deeper waters allow bigger waves to hit
the previous shoreline. The deeper water and more violent wave action will also
destroy complex and extremely productive rice fields. There may also be
increase in salinity in some areas as sea level rise (about 3gr/lit), but the possible
interaction of fresh and salt water flows are too complexes and must be
considered as the subjects of future researches. The higher rainfall will mean
more water for irrigation in the Caspian Sea neighbor countries. This may be
considered as a favorable condition to increase the irrigable area. But the
increased rainfall will increase erosion in the whole watershed area. The
increased rainfall will also accelerate the leaching of nutrients from the soil,
further lowering production. The higher temperature will suppress rice yield.
The length of Caspian sea (Fig. 1) from north to south is about 1200 km and its
width is in order of 200-450 Km. The length of Caspian sea shore line reach out
7000 Km with the length of Iranian shore line of 950 Km.
The most important river of the Caspian Sea basin is Volga river with a
discharge capacity of 75% of its total inflow. The surface and volume of
Caspian Sea is respectively about 430000 km
2
and 79000 km
3
.
From the view point of topography, hydrographic and environment situation, the
Caspian Sea is divided on three zones, so called northern, central and southern
area (Fig. 1). The characteristics of each of them are indicated in table 1. Table 2
and also Fig. 2 show the centurial variation of the Caspian Sea level [4].
The variation of Caspian Sea area (Km
2
) with respect of the altitude measured
from mean sea level (m) is presented in Fig 3. Two typical examples for the
southern part of the Caspian Sea level variation effects on the Sea ward
boundary are given in Fig. 4.


Table 1. Characteristics of the Caspian Sea

Caspian
Sea
Northern
Zone
Central
Zone
Southey
Zone
Qarabugaz
Gulf
Average depth (m)
181 6 175 320 7
Maximum depth (m)
946 20 768 946 15
Surface area (km
2
)
430000 120000 140000 155000 15000
Volume (km
3
)
79000 750 28000 50000 105

638

Table 2. balance of Caspian Sea (1880 1990)

Annual average
inflow (included
subsurface flow
(5000 Mm3/Year))
Annual average
outflow to
Qarabugaz gulf

Year Variation
of
Caspian
sea level
(cm) (cm) Mm3 (cm) Mm3
Evapor
ation
(cm)
1880-1913 -92 75.5 306000 6.7 24700 71.5
1914-1932 16 80.4 321000 4.9 19700 74.7
1933-1940 -172 58.5 229000 2.7 10500 77.3
1941-1956 -61 77.5 292000 3.1 11600 78.2
1957-1970 -4 75.5 281000 2.6 9500 72
1971-1977 -61 65.2 236000 1.9 6900 72
1978-1990 166 81.9 305000 0.5 1800 68.4
1880-1990 -206 75.3 293000 4.2 16600 73.5


4- Shore- line protection

The design strength of offshore structures is generally based upon a design
wave approach. Under the wave actions the unprotected shores are usually
deformed to the state of dynamic equilibrium. Design methods for mounds break
waters are generally based on laboratory tests. Extrapolation of results to field is
always affected by some indetermination both for scale effects and for inability
of laboratory instrumentation in reproducing prototype conditions. A field
analysis of breakwater behavior is necessary to evaluate capability of empirical
design in predicting the damage level due to Sea level rising and wave attack.
It is very urgent to establish different design condition for different types of
water- control structures. Dams or structures affecting public safety must satisfy
more strict design criteria.

5- Results and Discussion

In this paper, it is given a rather detailed picture of the hydrodynamic and
hydrologic situation of the Caspian Sea-level fluctuation and then using the
outflow runoff balance. It is proved that the main causes of the Caspian Sea-
level rising are due to the following factors:
- The enhanced greenhouse effect will lead to global warming, higher
rainfall, greater erosion and lower crop production in Caspian Sea basin.
For the Caspian Sea littoral of Iran, Specially Astara, Anzali, Ramsar,
Nowshahr, Babolsar, the Caspian Sea has particular importance.
Therefore, it is necessary that rising of the Caspian Sea-level during the
639
decade must be considered to study. Presently, the Caspian Sea-level,
compared to 1978, is approximately 2,5m higher. Thus, it is 1,5m above
the mark previously approved by the Caspian littoral countries as norm
and standard for the construction of littoral installations.
- The increasing of rivers runoff, being controlled by 75% by the Volga
river as well as the tectonic activities of the earth.
- The Co
2
content of the atmosphere has been rising at the rate of about
0.4% per year for at least two decades. Much of this increase, is the result
of burning fuels and cutting down and burning forests. The researches
show that plants play a role about equal to the seas and oceans in
balancing the So-called greenhouse effect.
- According to estimates based on water balance (Fig.3), the water level
mark will decrease about - 24, 8m in the year 2005.
- The Caspian sea-level fluctuation affects the quality of surface and
subsurface water.
- Neither a prediction of the mean sea level rise nor of the storm surge
numbers and heights can be given merely by interpretation of historical
data. The presented conditions show our actual lack of background
knowledge and the complexity of nature.

References

[1] Hansen, J. Climate sensitivity American Geophysical Union, Vol. 29,
Washington, D. C. 130-163, 1984.
[2] Emanuel, K. An air-sea interaction theory for tropical cyclones J.
atmos. Sci, 43, 585-604-1986.
[3] Emanuel, K. The dependence of hurricane intensity on climate Nature,
320, 483-485, 1987.
[4] Caplin, P. The first meeting of the regional task team on the implications
of climate change in the Caspian sea region, Moscow, 20-2 may 1994,
report of meeting, UNEP.
[5] Siefert, W. Mean Sea Level changes and storm surge probability,
XXIII Congress, IAHR, Ottawa, Canada, 1989
[6] Prandle, D. The impact of Mean Sea Level change on estuarine
dynamics, XXIII Congress, IAHR, Ottawa, Canada, 1989.









640
Fig. 1. Location map of Caspian Sea with related Rivers


641

























Fig. 2. Hypsometric Curve of Caspian Sea
















Fig. 3. Variation of the Caspian Sea Level with respect of the time (Year)
642










Fig. 4. Inundated region due to rising level of Caspian Sea (Tonkabon; June 1995)
643

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005












CHLORINATION AND OPTIMAL
CHLORINE DOSAGE FOR NILE WATER

Mohamed Shamrukh *
Department of Civil Engineering, Faculty of Engineering,
ElMinia University, EGYPT
E-mail: mshamrukh@hotmail.com

Yahia Hassan
Department of Chemical Engineering, Faculty of Engineering,
ElMinia University, EGYPT

Abstract
Chlorine is the most widely used disinfectant in Egyptian drinking water
because it is effective at low concentration, is cheap, and forms a residual if
applied in sufficient dosage. Chlorine dosage is applied at the treatment plant
and is very much dependent upon raw water characteristics. With traveling
through the distribution system, chlorine decay Nevertheless, Egyptian water
suppliers attempt to maintain a detectable chlorine residual (0.1 to 0.2 mg/L free
Cl
2
) within the distribution system to minimize the potential for microbial
growth. In Upper Egypt, rural residents are supplied with treated Nile water
throughout a compact treatment plants. However, Egyptian drinking water
supplies recommend a primary and secondary dose of 6 and 4 mg/L,
respectively to be applied in the treatment plants. Samples were collected at the
intake of selected treatment plants located in AbuTisht, Upper Egypt. Different
chlorine dosages and residuals for these samples were measured and analyzed
laboratory in May 2004. The breakpoint curve of chlorine was delineated and
the obtained results are used to determine the suitable chlorine dosage that added
into Nile water. Furthermore, the coefficient of bulk chlorine decay was
determined. The obtained optimal chlorine dosage and decay coefficient could
be used to simulate chlorine concentration in the distribution systems in Upper
Egypt using any water quality code such as EPANET developed by USEPA.
From this study, the optimal primary dose and decay coefficient that might be
apply for Nile water treatment in Upper Egypt are 3.5mg/L and 0.034 /hour,
respectively.

Keywords: Chlorine, Disinfection, Drinking water, Water supply, Nile.
644

INTRODUCTION
A major objective of drinking water treatment is to provide microbiologically safe
drinking water. The combination of conventional drinking water treatment and disinfection
has proved to be one of the major public health advances in modern times. The most
important use of disinfectants in water treatment is to limit waterborne disease and inactivate
pathogenic organisms in water supplies. Chlorine is, by far, the most commonly used
disinfectant in the drinking water treatment industry (White 1992; Haested 2003). In Egypt,
chlorine is most often the primary and the only final disinfectant added to treated water for
microbiological protection before it is discharged into a drinking water distribution system.

However, disinfectants, especially chlorine, react with natural organic matter (NOM) to
form disinfection by-products (DBPs), which are considered to be of concern from a chronic
exposure point of view. These DBPs include halogenated organics, such as THMs, and
organic oxidation byproducts such as aldehydes, and inorganics such as chlorate and chlorite.
A high NOM concentration, exceeds 2.0 mg/L, indicates a high potential for DBPs formation.
Therefore, the applied disinfectant dose must be sufficient to meet the inherent demand in the
treated water, to provide sufficient protection against microbial infection, and at the same
time minimize DBPs. Drinking water disinfection, therefore, poses the dilemma of a risk
tradeoff. Chemical disinfection reduces risk of infectious disease, but the interaction between
chemical disinfectants and precursor materials in source water results in the formation of
DBPs. In fact, disinfection of drinking water has dramatically reduced outbreaks of diseases
attributable to waterborne pathogens. However, the identification of DBP such as
trihalomethanes (THM), in drinking water (USEPA 1990, 1999) raised questions about
possible health risks posed by these DBPs. Trihalomethanes and other DBPs have been shown
to be carcinogenic, mutagenic, etc. Since 1974, additional DBPs have been identified, and
concerns have intensified about health risks resulting from exposures to DBPs.

The most common disinfection method used by water systems in Egypt is chlorination.
Chlorine may be added to drinking water as a gas or liquid. The advantages of gas chlorine
are the realtive cost of the gas chlorine and the minimal effect that gas chlorine has on pH.
The safety issues are considered the negative aspects of its useThe gas form or chlorination
system is most common for water systems in Egypt and therefore is the focus of this work. In
addition to chlorine gas, chlorine is also available in hypochlorite form as both aqueous
solutions and dry solids. The most common forms are sodium hypochlorite as aqueous and
calcium hypochlorite (White, 1992). Sodium hypochlorite solution typically contains 12.5%
available chlorine. Granular calcium hypochlorite commercially available typically contains
65% available chlorine. The powder is mixed with water and fed with the chemical feed pump
used for the calcium or sodium hypochlorite.

To maintain biological stability in the distribution system, the standards requires that
treated water have a residual disinfectant of 0.2 mg/L when entering the distribution system.
A measurable disinfectant residual (0.1 to 0.2 mg/L) must be maintained in the distribution
system, or the utility must show through monitoring that the heterotrophic plate count (HPC)
remains less than 500/100mL. Several environmental factors influence the inactivation
efficiency of chlorine, including water temperature, pH, contact time, mixing, turbidity,
interfering substances, and the concentration of available chlorine. In general, the highest
levels of pathogen inactivation are achieved with high chlorine residuals, long contact times,
high water temperature, and good mixing, combined with a low pH, low turbidity, and the
absence of interfering substances. One aspect of this research is the assessment of
chlorination of Nile water in Upper Egypt areas. The results of this study will assist utilities in
645

selecting optimal chlorine dose for microbial inactivation which ensures protection against
pathogens while minimizing the formation of chlorine dioxide-related DBPs. Chemical
kinetics of chlorine to define the bulk decay coefficient is presented which could be used in
chlorine simulation in water distribution systems in Egypt.

CHLORINATION PROCESS
Chlorination is used in water treatment facilities primarily for disinfection. Because of
chlorines oxidizing powers, it has been found to serve other useful purposes in water
treatment, such as iron and manganese oxidation. In Egyptian treatment plants, liquefied
chlorine gas is typically evaporated to gaseous chlorine prior to metering. Once the
compressed liquid chlorine is evaporated, chlorine gas is typically fed under vacuum
conditions. Either an injector or a vacuum induction mixer usually creates the required
vacuum. The injector uses water flowing through a venturi to draw the chlorine gas into a
side stream of carrier water to form a concentrated chlorine solution. This solution is then
introduced into the process water through a diffuser or mixed with a mechanical mixer. A
vacuum induction mixer uses the motive forces of the mixer to create a vacuum and draws the
chlorine gas directly into the process water at the mixer (USEPA 1990).

Chlorine Dosage & Residual
The total amount of chlorine fed into a volume of water by the chlorine feed equipment is
referred to as dosage and is calculated in mg/L. Chlorine is a very active chemical oxidizing
agent. When injected into water, it combines readily with certain inorganic substances that are
oxidizable (e.g., hydrogen sulfide, ferrous iron, etc.) and with organic impurities including
microorganisms and nitrogen compounds (e.g. animal wastes, ammonia from fertilizers, etc.).
These reactions consume some of the chlorine. The amount of chlorine added is called the
chlorine dose and the amount of chlorine consumed is called chlorine demand. Then chlorine
residual is estimated from the following equation:
Chlorine Residual = Chlorine Dose - Chlorine Demand (1)

Chlorine Reaction in Water
Regardless of the form of chlorine used, the reaction in water is basically the same. The
chlorine mixed with water will produce hypochlorous acid (HOCl). The following equation
presents the hydrolysis reaction (USEPA 1999; White 1992):
Cl
2(g)
+H
2
O HOCl +H
+

Cl (2)
The addition of chlorine gas to water reduces the pH of the water due to the production of
hydrogen ion. Hypochlorous acid is a weak acid, meaning it dissociates slightly into hydrogen
and hypochlorite ions as noted in Equation 3:
HOCl H
+
+

OCl (3)
The measurement of hypochlorous acid (HOCl) and hypochlorite ion (

OCl ) is called free
chlorine residual (Figure 1). Between a pH of 6.5 and 8.5 this dissociation is incomplete and
both HOCl and OCl
-
species are present to some extent (White, 1992). Below a pH of 6.5, no
dissociation of HOCl occurs, while above a pH of 8.5, complete dissociation to OCl
-
occurs.
As the germicidal effects of HOCl is much higher than that of OCl
-
, chlorination at a lower
pH is preferred. This is due to the more facile movement of the uncharged acid through the
bacterial cell wall compared to OCl
-
. For this reason, careful regulation of pH is necessary to
assure maximum disinfection with minimum chlorine dose. In the case of potable water,
646

however, it is also necessary that waters with relatively low pH (below 7.5) tend to be
corrosive.


Figure 1. Chlorine reactions with water

If organic or inorganic compounds are available in the water, particularly nitrogen
compounds, the hypochlorous acid will combine with these compounds to produce
chloramines and/or chloro-organic compounds. The measurement of the presence of these
compounds is called combined chlorine residual. Free chlorine has proven to be a more
effective disinfectant than combined chlorine compounds. The measurement of both the free
and combined chlorine residuals is called total residual as following:
Free Residual Chlorine + Combined Residual Chlorine = Total Chlorine Residual (4)
If there are organic and inorganic compounds in the water, then enough chlorine has to be
added to complete the reactions with these compounds before an adequate free residual can be
produced. This process is called breakpoint chlorination which represented from plotting of
applied chlorine dose versus total chlorine. This curve evolves four distinct areas as shown in
Figure 2. The first area represents the oxidation of more reactive compounds than NH
3
by
chlorine and is an inorganic demand phase. The second area represents an increase in
combined chlorine residual (chloramines). The residual begins to drop because of destruction
of the combined chloramines. At this point the nitrogen is given off as a gas and is lost to the
atmosphere. Phase three continues until breakpoint. After breakpoint, all NH
3
products have
been fully oxidized and free chlorine residual begins to form phase four. As a rule of thumb,
the free residual should be at least 85% of the total residual in order to prevent chlorine taste
and odor problems and insure an adequate free residual for effective disinfection (USEPA
1990).

Disinfection with chlorine is not instantaneous. Time is required in order that any
pathogens present in the water are inactivated. The time taken for different types of microbes
to be killed varies widely, therefore need an adequate contact time; this is usually 30 minutes.
One of the most important factors for determining or predicting the germicidal efficiency of
any disinfectant is the CT factor. The CT factor is defined as the product of the residual
647

disinfectant concentration, C, in mg/L, and the contact time, T, in minutes that residual
disinfectant is in contact with the water.





Figure 2. Breakpoint chlorination curve, (EPA, 1990)


MATERIALS AND METHODS
Study Area
As concern over disinfection by products and disinfectant decay within the distribution
systems has increased, more attention has been paid to the chlorination. To provide water
suppliers in Upper Egypt, raw water at three water treatment's intakes located at AbuTisht
County in Qena Governorate were chosen to carry out this investigation. The locations of the
sampling points could represent the area from Aswan to Assiut which is 500 Km long of
River Nile. The treatment plants serve a population of 20 thousands (Figure 1). River Nile and
its main streams in the study area were the source of raw water that has been studied. The
treatment processes used are the conventional method which consists of screening,
coagulation/flocculation by alum, sedimentation, and sand filtration.

There are a primary and secondary disinfection in treatment plants which are applied
before coagulation and before reservoir, respectively. The detention time in the plant reservoir
before distribution is 30 min. Disinfection in the plants is applied using chlorine gas contained
in pressurized cylinders. The primary and secondary chlorine dosage is 5.0 to 3.0 mg/L,
respectively. These doses are constants all year seasons no matter is the quality of raw water.
Egyptian regulations of drinking water specify that residual chlorine must be in range of 0.1
to0.2 mg/l to prevent microbial growth in the water distribution system.

Sampling and Analytical Program
During the period from April to May 2004, many water samples of raw water from River
Nile and its main streams were collected at the three water treatment plants. Samples of Nile
water were collected from the depth of 1.5m from surface near the plants' water intake points.
Chlorine
Reducing
Compounds
648

Collected samples were analyzed using a Hach DR2000 Spectrophotometer for most of water
measurements. Meters for pH and TDS measurements were used for other water quality.
Instrument startup and analysis were carried out as detailed in the operating manual. Each
measurement was made in duplicate and the absorbance due to background was subtracted
from the mean of the two values. DPD indicator (N, N, diethyl-p-phenylenediamine) was
used as chemical reagent for chlorine measurements. Free available chlorine (Cl
2
, HOCl,
OCl) oxidizes DPD to produce a red color. This color is then measured using DR2000.
Quality characteristics of Nile water at the sampled locations were analyzed and given in
Table 1.

Chlorine dosages were created using a granular calcium hypochlorite which contains 65%
available chlorine was used to prepare the different values of chlorine dosages .About 12
flasks have been prepared for each sampling site of the three plant intakes. The chlorine
dosage were as following: 0.5, 1.0, 1.5, 2.0, 3.0, 4.0, 6.0, 8.0, 10, 12, 14, 16 mg/L.
Measurements of free chlorine and combined chlorine were taken 30 minutes after dosing.
Three flasks were continued in measurements of free and combined chlorine for a period of
12 hours along to estimate the coefficient of chlorine bulk reaction.


Aswan
High Dam
Aswan
Qena
Nile
V
a
l
l
e
y
Red Sea
Mediterranean Sea
Sinai
Cairo
Study site
EGYPT


Figure 3. Study site location and sampled area
RESULTS AND DISCUSSION
Nile Water Quality
Important water quality parameters of raw Nile water at the study area in Qena
governorate are listed in Table 1. Most of quality parameters are under the allowable limits
for drinking purposes. There was a relatively high concentration of organic matter (4.50
mg/L) that may cause a problem for forming DBP in the distribution systems upon using
649

excess chlorine dosage. In addition there nil concentration of ammonia and very small
amounts of reduced iron and manganese that will consume chlorine for oxidation. Actually it
is anticipated that the DBP especially THM will be formed due to high concentration of
organic matter 3.0 to 4.0 mg/L which exceeds 2.0 mg/L. It is recommended to carry out an
investigation to measure these DBP in the study site to confirm this anticipation.

Table 1. General characteristics of raw River Nile water at the study site

Parameter pH TDS DO NOM Fe Mn NO
3
SO
4
alkalinity hardness
Value
mg/L
7.5
7.7
135
150
5.5
6.5
3.0
4.0
0.1
0.2
0.05
0.15
0.7
1.5
35
40
105
115
100
110
-NOM: natural organic matter
-alkalinity and hardness given as CaCO
3


Breakpoint Chlorination Curve
In the study area, Egyptian drinking water supplies recommend a primary and secondary
dose of 6 and 4 mg/L, respectively to be applied in the treatment plants. In fact these values
are constant for all the treatment plants in the study area all times without change. Up to
knowledge of the authors, there is no breakpoint curves are carried out experimentally for the
chlorination of the treatment pants in Upper Egypt. In order to effectively treat drinking water
with chlorine, the chlorine demand and breakpoint of the water to be treated needs to be
calculated. Results of the bottle analysis for chlorine doses and residuals are presented in
Table 2. Free residuals were measured at some points of chlorine doses. Chlorine demands
have been calculated according to Equations 1. It is clear from Table 1 that the total residuals
(combined and free), free, and demand are increasing with the increasing of chlorine added to
the water samples.

Table 1. Chlorine doses and residuals measurements

Chlorine dose (mg/L) 0.5 1.0 2.0 3.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0
Chlorine residual (mg/L) 0.20 0.35 0.61 1.07 1.30 2.10 2.80 3.51 4.42 5.11 6.60
Free chlorine (mg/L) - - 0.33 0.45 0.56 0.72 1.21 - 2.1 - 2.95
Chlorine demand (mg/L) 0.3 0.65 1.39 1.93 2.7 3.9 5.2 6.5 7.6 8.9 9.4


The results of chlorine measurements were taken to graph the breakpoint chlorination as
shown in Figure 4. Dash lines represent the trendlines of total available chlorine and free
available chlorine for the sampling water. At the beginning of graph there is zone one in
which added chlorine start to oxidize the reduced compounds in sampled Nile water such as
iron, manganese, and nitrate. As listed in Table 1, these reduced constituents are found in
small amounts in Nile water in the study area. Therefore, the amount of chlorine consumed in
this process is about 0.2mg/L. The important notice from Figure 4, is the breakpoint of
chlorine. Figure 4 shows that there is no clear breakpoint in this graph where the chlorine
residual s the least. This is may be due to the nil existence of ammonia-nitrogen in sampled
Nile water. Therefore, no show up for zones two and three of the typical breakpoint curve
(zones just before breakpoint, Figure 2). Chloro-organic compounds and chloramines
formation in zone two and its destruction in zone three did not clearly formed for Nile water
650

in the sampled area. However, zones from one to three shown in Figure 2 of breakpoint curve
need more investigations with more small amounts of chlorine doses in order to highlight its
details.

Thus, in Nile raw waters that do not contain ammonia, the chlorination curve begins at
the point where chlorine additions provide free and available chlorine in the system (at
0.2mg/L chlorine dose). Chlorine dose added to drinking water must satisfy demand, give a
free residual at customers tape equal to 0.1~0.2mg/L, and minimize the cost and DBP
formation (Rodriguez, 1995).. Hence, the most suitable chlorine dose could be obtained from
Figure 4 which gives free chlorine of 0.5 mg/l at entrance of the distribution systems is
3.5mg/L. This dose might be divided to two parts for primary and to secondary disinfection in
the treatment plants in Upper Egypt. This chlorine dose which operators need to add order to
maintain 0.20mg/L free chlorine residual in their systems.
























Figure 4. Breakpoint chlorination curve for Nile water in Upper Egypt


Bulk Reaction Coefficient
For water distribution systems, chlorine is shown reacting in the bulk fluid with natural
organic matter (NOM), and at the pipe wall, where oxidation reactions with biofilms and the
pipe material can occur (Clark 1994). For chlorine decay analysis, reaction dynamics can be
specified using bulk and wall reaction coefficients. Bulk reaction coefficients associated with
individual pipes and storage tanks. Wall reaction coefficients only associated with individual
pipes, applied globally, or assigned to groups of pipes with similar characteristics. Bulk
reaction coefficients can be determined through laboratory testing. Wall reaction coefficients
must be measured using field tests or determined as part of the calibration process of
0
1
2
3
4
5
6
7
8
0 2 4 6 8 10 12 14 16 18
Chlorine dose mg/L
C
h
l
o
r
i
n
e

r
e
s
i
d
u
a
l

m
g
/
L
total residual
free residual
651

distribution systems. The most commonly used reaction model is the first order decay model
(Haested, 2003). This has been applied to chlorine decay, and other decay processes. A first
order decay is equivalent to an exponential decay, represented by Equation 5:
kt
o t
e C C

=
or
kt C C
t o
= ln ln
(5)
where C
t
is chlorine concentration at time t, C
o
is the initial concentration (at time zero), and k
is reaction rate. The values of bulk decay, k, generally expressed 1/hr or 1/day.

Two bottles of chlorine measurements were kept for 24 hours to measure chlorine
concentrations along time. 24 hours represent the longest residence time in most of the water
distribution systems in the studied area. Figure 5 show the graphing of the first order decay of
chorine. The chlorine concentrations are charted along the y-axis logarithmic(the dependent
variable), and the time is charted along the x-axis. Reaction rate coefficient, k, can be found
using linear regression techniques. A best-fit straight line is drawn through the data collected
from the two bottles test as illustrated in Figure 5. The slope of the line for the drawn data in
Figure 5, 0.034 hr
-1
, becomes the bulk reaction coefficient. Note that the reaction coefficient
is negative since the chlorine concentration decays over time.

Bulk reaction coefficients are associated with pipes for purposes of a simulation, and are
assumed to remain constant throughout the simulation for a particular pipe. Since the bulk
reaction coefficient is, in reality, associated with the fluid itself, the bulk reaction rate can
change throughout the actual system as water from different sources (Nile and groundwater)
becomes mixed at nodes. When assigning bulk reaction coefficients for pipes, the mixing can
be considered by designing and conducting bottle tests with representative source mixtures. A
source tracing analysis can assist in determining the degree of mixing in a system. Source
blending can change over the course of the day for a particular pipe, thus the predominant
source or mix of sources should be used in assigning the bulk reaction coefficient.




















Figure 5. First order equation of chlorine bulk decay
Ct/Co =3.64e
-0.034t
0.1
1
10
0 5 10 15 20 25
Time, hr
C
h
l
o
r
i
n
e

c
o
n
c
e
n
t
r
a
t
i
o
n
,

m
g
/
l
652


CONCLUSIONS
Egyptian drinking water supplies recommend a chlorine dose for drinking water to be
10mg/L which applied in the treatment plants. Samples were collected at the intake of
selected treatment plants located in AbuTisht, Upper Egypt in April and May 2004.
Experimental work has been carried out to determine the suitable dose and reaction
coefficient of chlorine applied in the treatment plants in Upper Egypt. The suitable chlorine
dose that could be added to drinking water in Upper Egypt is 3.5mg/L. This chlorine dose will
satisfy demand, give a free residual at customers tape equal to 0.1~0.2mg/L, and minimize
both chlorine cost and DBP formation. This dose could be divided to two parts, primary and
secondary disinfections. Chlorination breakpoint curve has been delineated for Nile water in
the studied area. Results show that there is no breakpoint where the chlorine residual is the
least. This may be due to the nil content of ammonia-N in Nile water in the site. In fact more
investigation is needed to carry out to give more details for small chlorine doses. In addition,
the coefficient of bulk chlorine decay was determined. The obtained bulk decay coefficient
could be used to simulate chlorine concentration in the distribution systems in Upper Egypt
using any water quality code such as EPANET. From this study, the coefficient of chlorine
bulk decay for Nile water in the study area in Upper Egypt is 0.034 /hour.

REFERENCES
AWWA and ASCE (American Water Works Association and American Society of Civil
Engineers). (1997). Water Treatment Plant Design. McGraw-Hill, New York, NY.
Clark, R. M., Grayman, W. M., Goodrich, J . A., Deininger, R. A., and Skov, K. (1994).
Measuring and Modeling Chlorine Propagation in Water Distribution Systems. J. Water
Resour. Plang. and Mgmt., ASCE, 120(6) 871-887.
Clark, R. M., Grayman, W. M., Males, R. M., and Hess, A. F. (1993). Modeling Contaminant
Propagation in Drinking-Water Distribution Systems. J. Envir. Eng., ASCE, 119(2), 349-
364.
Clark, R. M., Smalley, G., Goodrich, J ., Tull, R., Rossman, L. A., Vasconcelos, J . J ., and
Boulos, P. F (1993). Managing Water quality in Distribution Systems: Minimizing
Chemical and Microbial Risk. AWWA Annual Conference Proc., Texas, pp 419-470.
Elton, A., Brammer, L. F., and Tansley, N. S. (1995). Water Quality Modeling in Distribution
Networks. J. Am. Water Works Assoc., 87(7) 44-52.
Hach Company (1996). DR-2000 Users Manual. Hach Company, Ohio, USA.
Haestad Methods, T. M. Walski, D. V. Chase, D. A. Savic, W. Grayman, S. Beckwith, and E.
Koelle. (2003). Advanced Water Distribution Modeling and Management. Haestad Press.
Waterbury, Connecticut.
Hart, F. L., Meader, J . L., and Chiang, S. (1986). CLNET- A Simulation Model for Tracing
Chlorine Residuals in a Potable Water Distribution Network. AWWA Distribution
System Symp. Proc., Minneapolis, MN, pp 193-203.
Kennedy, M. S., Moegling, S., Sarikelle, S., and Suravallop, K. (1993). Effects of Storage
Tank Design On Water Quality. AWWA Annual Conference Proc., Texas, pp 83-111.
Lu, C., Biswas, P., and Clark, R. M. (1993). Modeling of breakpoint reaction in drinking
water distribution pipes. Environmental International, 19(6), 543-560.
Males, R. M., Clark, R. M., Grayman,W. M., and Coyle, J . A.. (1990). Operation of Water
Distribution Systems to Improve Water Quality. Risk Reduction Engineering Laboratory,
Office of Research and Development, U. S. Environmental Agency, Cincinnati, Ohio.
Rossman, L. A. (2001). EPANET-Users Manual. U.S. Environmental Protection Agency,
Risk Reduction Engineering Laboratory, Cincinnati, Ohio.
653

Rossman, L. A., Clark, R. M., and Grayman, W. M. (1994) Modeling Chlorine Residuals in
Drinking-Water Distribution Systems. J. Envir. Eng., ASCE, 120(4), 803-820.
Skov, K. R., and Hess, A. F. (1991). Field Monitoring Procedures to Calibrate a Water
Distribution Systems Water Quality Model. AWWA Annual Conference Proc., pp 157-
163.
Standard Methods (1995). Standard Methods for the Examination of Water and Wastewater
19th Edition, American Public Health Association, AWWA, Water Environment
Federation, Washington, D.C.
USEPA (1990). Technologies for upgrading Existing or Designing New Drinking Water
Treatment Facilities. United States Environmental Protection Agency, Office of Drinking
Water, Cincinnati, Ohio, EPA/625/4-89/023.
USEPA (1999). Guidance Manual Alternative Disinfectants and Oxidants. United States
Environmental Protection Agency, Office of Drinking Water, Cincinnati, Ohio
White, G.C. (1992). Handbook of Chlorination and Alternative Disinfectants. Vol. 3. Van
Nostrand Reinhold Co. New York, NY.
White, G.C. (1986). Handbook of Chlorination. Van Nostrand Reinhold Company, New
York, NY
World Health Organization (1993). Guidelines for drinking-water quality, Second Edition,
Volume 1, Recommendations, p. 135.

654

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005



WATER QUALITY ASSESSMENT OF IZMIT BAY BY
TWO DIMENSIONAL MODELLING


Aye EDNLLER
1*
Enis MORKO
2
Kaan DENGL
3


1
Assist. Prof., Boazii University, Kandilli Observatory and Earthquake Research Institute,
Earthquake Engineering Department, Istanbul - Turkey
2
Assoct. Prof., Marmara University, Department of Chemical Engineering, Istanbul -Turkey
3
PhD Candidate, Boazii University, Department of Civil Engineering, Istanbul-Turkey


ABSTRACT

In Turkey, rapid urbanization and industrialization since 1950 caused severe
environmental pollution problems in water bodies, especially in regions where
industry and population are intense. Water bodies that are affected most are
closed systems like Izmit Bay in the Marmara Sea. Izmit Bay, which is in the
northeast region of the Marmara Sea, has an assimilation capacity that is much
lower than wastewater discharged. Therefore, in the last 40 years, ecosystem has
been severely damaged in Izmit Bay. The aim of this paper is to mention the
full aspects of environmental pollution of Izmit Bay through experimental data
and a two-dimensional numerical model. In the models, data obtained both from
the industrial point sources and sampling stations at the sea were employed. The
most important pollutant agents taken into account in this study are TOC, SS,
TP, and TN. A comparison between the calculated and measured concentration
values of the pollutants at the station points are presented to illustrate the
compatibility of the results and to understand the possible critical conditions that
may take place. From the results of the numerical models, it is concluded that
two dimensional numerical modeling of the transportation of four different
pollutant agents in the bay provides satisfactory results. It is recommended that
immediate precautions should be undertaken for the Izmit Bay.

Keywords: zmit Bay, Water pollution, Nutrients, Modeling, Waste water.


655

1. INTRODUCTION

In Turkey, rapid urbanization and industrialization since 1950 caused severe
environmental problems in water bodies, especially in regions where industry and population
are intense Mostly effected water bodies are semi enclosed systems like Izmit Bay in the
Marmara Sea. Izmit Bay, which is in the northeast of Marmara Sea, has an assimilation
capacity that is much lower than the wastewater discharged. Therefore, in the last 40 years,
ecosystem has been severely damaged in Izmit bay waters. Today, most of the discharge from
the industries, located in the vicinity of Izmit Bay, treated either by the industry or the large
treatment plant owned by local authorities. But there is no study in the removal of the
nutrients and the toxic material present in flows. A significant portion of the discharge
coming from the residential areas is not treated. Only the treatment plant owned by Izmit
Municipality partly treats the residential wastewater produced in the municipality borders
together with some industrial discharges which decreased concentration to some certain level.
Therefore, no desired recovery in the ecosystem of the bay is observed. Two water masses
exist in Izmit Bay. These water masses are separated by thermocline and halocine. Water in
the upper layer has large variations in temperature because of the contact with the atmosphere,
but has a low salinity.
The brackish water which originated in the Black Sea forms a 10 to 50 m thick surface
layer. Its salinity and temperature vary from 22-24 ppt and 20-24
o
C in summer, to 26 ppt and
7
o
Cin winter. Marine water, which arrived from the Mediterranean, underlies the permanent
pycnocline and has much more stable properties: salinity and temperature ranges are 37.5
38.5 ppt and 14-15
o
C, respectively (Figure 1).



Figure 1. Monthly variations of temperature and salinity in the surface waters of Izmit Bay
between 1984 and 1988

In this paper, selected results of examination of physical and biochemical data obtained
from 9 stations during 2 cruises between February and April 1999 presented together with the
results of the modeling to observe the distribution of nutrients organic matter and suspended
solids (Figure 2) (Dengili, 1999).

656


Figure 2. Industrials and Station Points in Izmit Bay


2. BACKGROUND

Eutrophication that results from the inflow of nutrients discharged by land based sources
has been a major problem in many coastal waters. The most conspicuous manifestations are
the occurance of algal blooms, mucilage (Zambianchi et al., 1992) and hypoxia in lower
layers (Justic and Legovic, 1987; Legovic et al., 1987). Practically the same discharges may
affect the trophic state differentliy; this depends on coastal circulation and the existence of
stratification (Cruzado and Legovic, 1994). Permanent stratification reduces horoixontal
advection in the lower layer (Sorokin, 1988; Kullenber, 1983). In addition, it blocks vertical
transport of oxygen to the bottom layer (Legovic et al., 1991).
Existence of large nutrient inputs from land-based sources discharging into the Izmit
Bay is believed to be the major cause eutrophication while permanent halocine that limits
vertical mixing and induces a slow water exchange between segments of the bay and the
adjacent Marmara Sea, are believed to be the major conditions favouring eutrohication
(Akkaya et al., 1983).
Although severaly studies have discussed treatment alternatives of waste effluents to the
zmit Bay, very limited work has been published based on the results of systematic
measurements of oceanographic characteristics and limiting nutrients (Akkaya et al, 1983;
Timur, et al., 1982; DAMOC, 1971; SWECO and BMB, 1976).
Transparency of the bay water as secchi disk decreased from 11 m. to 1-3 m from 1970
to 1999. DO resulted to a critical value at lower layers from 3.5 mg/year in 1950 to 0.5 mg/l
in 1999 at the end of summer. Nutrient concentrations showed high values in eastern section
by the flows which increased the productivity and accelerated the algal bloom. Organic matter
contents were inclined from western to eastern sections. Pollutant matters both from inputs
and primary production have been resulted in increase, especially in the eastern section by
broken down 30 mg/l.
657

3. EXPERIMENTAL STUDY

Industrial based point sources are generally located at the northern part of the bay
(Figure 2). In south, a small amount of wastewater (2 per cent of total) originating from the
industry is discharged around Yalova. Since this amount is too small, it is neglected. Point
sources, in which samples were taken, are given in Figure 2. Four of these major industrials
are petrochemical facilities (Tpra, Petkim, Shell, Petrol Office). Eastern Channel is the exit
of the treatment plant treating only industrial wastewater. Seka is a paper producing factory.
Koruma Tarm is a pesticide producing factory. Mannesmann is producing pipes. Dil River is
discharging the wastewater coming from the dye industry, and the detergent industry
The most important part of an environmental projects is collecting samples. In this
study, project samples were collected both from point sources and sampling stations at sea.
Winter and spring samples were collected and analysed. Nine station points were chosen.
While choosing these stations their distance to the point sources, water currents were
considered. These sample stations are given in the Figure 2. Water samples for TP, TN, TOC,
SS were taken. Also light penetration measurements, temperature measurements, and
dissolved oxygen measurements were conducted (Dengili, 1999).
Seawater samples were taken from different depths in the station points by using nansen
type water sampler bottles. The water samples were taken from 0.5m, 5 m., 10m., 20m.,30m.,
40m., and 50m., depth according to the water depth at the stations. The deepest sample was
taken from the depth of 50 m. Composite samples were taken from the point sources for two
hours at 15 minutes intervals and carried to the laboratory in the ice boxes.


4. RESULTS AND DISCUSSION

Wastewater discharge from the point sources is given in Figure 3. It is obvious that
highest amount of discharge (82 per cent of all) is coming from the Dil River. TOC is the
highest amount of pollutant discharged from the land based sources. This is mainly because of
Dil River since that river also discharges the residential loads dorm the residences around Dil
River. This is clearly indicates that, more than 90 per cent of TOC load is coming from Dil
River.

4.1 Wastewater Discharge from Land Sources (February)

Highest TOC concentration measured is at the station 4 (4.81 mg/l), at 5 m. depth.
Lowest TOC concentration measured is, at the station 5 (1.32 mg/l), at 50m. depth.
Highest TN concentration is 7.11 g/l at the station4, at 0.4 m. depth. The lowest TN
concentration is at the station 2 (2.5 g/l), at 20 m. depth.
Is is found that the highest SS concentration is at the station 5 an it is 30.8 mg/l at 10
m. depth. Also the lowest concentration of SS is at the station 4 and it is 17.6 mg/l, at the
depth of 10 m.
Highest TP concentration is measured at the station 3 at the depth of 5 m., and it is an
extremely high concentration (75.8 g/l ). The lowest TP concentration is at the station 6 (3.1
g/l), at the depth of 20 m. Note that there are great variations between the concentrations of
TP measured at different stations, even at different depths in the same station.

658

10800
2328,36
35500
5340
24645,7
277271,6
87,5 125 97,5 112 251,45
0
50000
100000
150000
200000
250000
300000
S
e
k
a

T
r
e
a
t
m
e
n
t

P
l
a
n
t
S
e
k
a

C
l
o
r

A
l
.
E
a
s
t
e
r
n

C
h
a
n
n
e
l
T
u
p
r
a
s
P
e
t
k
i
m

C
.
C
D
i
l
d
e
r
e
s
i
P
e
t
r
o
l

O
f
i
s
i

C
.
C
.
P
e
t
r
o
l

O
f
i
s
i

I
.
E
.
S
h
e
l
l

I
.
E
.
K
o
r
u
m
a

T
a
r
i
m
M
a
n
n
e
s
m
a
n
Dischar ge Point
F
l
o
w

R
a
t
e

(
m
3
/
d
a
y
)


Figure 3. Average Amount of Wastewater discharge from land sources (Feb. and April data)

4.2. Wastewater Discharge from Land Sources (April)

Highest TOC concentration measured is, at the station five (4.9 g/l), at 0.5m. depth.
Lowest TOC concentration measured is, at the stations eight and nine (0.8 mg/l), at 50 m
depth. Highest TN concentration is 692 g/l at the station seven, at 5 m. depth. The lowest TN
concentration is at the station seven again (57 g/l), at 25 m. depth. It is found that the highest
SS concentration is at the station one it is 32.4 mg/l at 0.5 depth. Also, the lowest
concentration of SS is at the station one and it is 32.4 mg/l at 0.5 m. depth. Also the lowest
concentration of SS is at the station three and it is 19.6 mg/l, at the depth of 10 m. Highest TP
concentration is measured at the station one at the depth of 5 m., and it is an extremelu high
concentration (372 g/l). The lowest TP concentration is at the station four (8g/l), at the
depth of 20 m. Note that there are great variations between the concentrations of TP measured
at different stations, even at different stations, even at different depths in the same station.
Average SS, TN, TP, and TOC Load from land sources from February and April data are
summarized in the Figures 4 and 7.

151,2349,85 622 128,15
2537,3
62660,45
0,55 1,45 0,75 0,5 2,2
0
10000
20000
30000
40000
50000
60000
70000
S
e
k
a

T
r
e
a
t
m
e
n
t

P
l
a
n
t
S
e
k
a

C
l
o
r

A
l
.
E
a
s
t
e
r
n

C
h
a
n
n
e
l
T
u
p
r
a
s
P
e
t
k
i
m

C
.
C
D
i
l
d
e
r
e
s
i
P
e
t
r
o
l

O
f
i
s
i

C
.
C
.
P
e
t
r
o
l

O
f
i
s
i

I
.
E
.
S
h
e
l
l

I
.
E
.
K
o
r
u
m
a

T
a
r
i
m
M
a
n
n
e
s
m
a
n
Dischar ge Point
S
S

L
o
a
d

(
k
g

/

d
a
y
)


Figure 4. Average SS Load from land sources from February and April Data

659

175,3991,095
1639,24
197,05752,845
70426,99
0,245 1,82 0,445 1,415 3,175
0
10000
20000
30000
40000
50000
60000
70000
80000
S
e
k
a

T
r
e
a
t
m
e
n
t

P
l
a
n
t
S
e
k
a

C
l
o
r

A
l
.
E
a
s
t
e
r
n

C
h
a
n
n
e
l
T
u
p
r
a
s
P
e
t
k
i
m

C
.
C
D
i
l
d
e
r
e
s
i
P
e
t
r
o
l

O
f
i
s
i

C
.
C
.
P
e
t
r
o
l

O
f
i
s
i

I
.
E
.
S
h
e
l
l

I
.
E
.
K
o
r
u
m
a

T
a
r
i
m
M
a
n
n
e
s
m
a
n
Discharge Point
T
N

L
o
a
d

(
k
g

/

d
a
y
)


Figure 5. Average TN Load from land sources from February and April Data

29,05
0,124
70,31
1,5
50,07
386,875
0,02750,0615 0,117 0,079 0,174
0
50
100
150
200
250
300
350
400
450
S
e
k
a

T
r
e
a
t
m
e
n
t

P
l
a
n
t
S
e
k
a

C
l
o
r

A
l
.
E
a
s
t
e
r
n

C
h
a
n
n
e
l
T
u
p
r
a
s
P
e
t
k
i
m

C
.
C
D
i
l
d
e
r
e
s
i
P
e
t
r
o
l

O
f
i
s
i

C
.
C
.
P
e
t
r
o
l

O
f
i
s
i

I
.
E
.
S
h
e
l
l

I
.
E
.
K
o
r
u
m
a

T
a
r
i
m
M
a
n
n
e
s
m
a
n
Dischar ge Point
T
P

L
o
a
d

(
k
g

/

d
a
y
)


Figure 6. Average TP Load from land sources from February and April Data

117,7 15,1
1762,7
88,64640,15
144284,2
1,1 2,65 2,55 0,45 2,9
0
20000
40000
60000
80000
100000
120000
140000
160000
S
e
k
a

T
r
e
a
t
m
e
n
t

P
l
a
n
t
S
e
k
a

C
l
o
r

A
l
.
E
a
s
t
e
r
n

C
h
a
n
n
e
l
T
u
p
r
a
s
P
e
t
k
i
m

C
.
C
D
i
l
d
e
r
e
s
i
P
e
t
r
o
l

O
f
i
s
i

C
.
C
.
P
e
t
r
o
l

O
f
i
s
i

I
.
E
.
S
h
e
l
l

I
.
E
.
K
o
r
u
m
a

T
a
r
i
m
M
a
n
n
e
s
m
a
n
Discharge Point
T
O
C

L
o
a
d

(
k
g
/
d
a
y
)


Figure 7. Average TOC Load from land sources from February and April Data
660

5. NUMERICAL MODEL RESULTS
Two dimensional dispersion model for water bodies was use in the numerical analysis.
The most important pollutant agents taken into account in analysis are TOC, SS, TP, and TN.
The data required for the concentration field of a pollutant include the natural concentration,
the extinction half life of the pollutant, concentration measurements inside the region and
sources on the mainland body are obtained from the TUBTAK-MAM (The Scientific and
Technical Research Council of Turkey Marmara Research Center) Research Report
(Morkoc, et al., 2000; Dengili, 1999). The results of measurements of the pollutant agents
named as TOC, SS, TP and TN are summarized below:
TOC February

In the results of the modeling, a high concentration of pollutant is observed around Dil
River Region and Eastern Channel exit (Figure 8). But the site measurements indicate that
there is no such high level of pollutant concentration around Dil River. Therefore there is a
high difference between the measured and the calculated values in satation 7. This is because
of the great amount of TOC load coming from Dil River. Other stations seem to have the
differences in measured and calculated values within the percentage error. The western part of
the western section and the middle section seem to have not much problem with TOC
pollution. At station five calculated value deviates from the measured value and after that
there is an upward trend until the concentration reaches its peak value in station seven (4.13
mg/l).. After that this time concentration level shows a downward trend where the measured
values show nearly constant trend. At station seven the error is 52 per cent and at station nine
calculated and measured values are closely matching. This is likely to occur because of the
biological activity taking place around Dil River and weak currents. If biological activity is
high and the currents are weak high amounts of TOCs consumed quickly without going long
distances. Measured values support this. But calculated values show that biodegradation is
low close to Dil River but it is high at station five and eight.


Figure 8. Model Output for TOC in February 1999.
661

TOC April

Model indicates that there is an increase in the concentration level of TOC around Dil
River (Figure 9). But the measured values indicate that there is no such increase in pollution
level. Also model shows that in stations five, six, eight and nine calculated vales are smaller
than the measured values. As measured concentrations of TOC in February and April are
compared, there is no significant change in the concentrations at stations. April concentration
of TOC is highest (6 mg/l) at station seven. The preceding two station points (station five and
six) and the following two stations (station eight and nine) have low calculated concentration
level than the measured concentration levels. But measured values of concentration starting
from station five are not changing very much. This shows that extinction rate of TOC taken in
the model is lower than actual in the waters close to Dil River.




Figure 9. Model Output for TOC in April 1999.


SS February

Distribution of the SS is homogenous in the middle and eastern sections (Figure 10). But
there is an increase in the concentration level close to Dil River. To the west the model
indicates that SS concentration is low which indicates that a big difference in the values of
measured and calculated concentrations in station eight and station nine. The increase of SS
concentration around Dil river is an indication of high amount of erosion. SS concentration
reaches its peak value at station seven (28.51 mg/l). After that station calculated values drop
below the measured values. While the measured values show an increasing trend the
calculated values show a decreasing trend starting from station eight. This may be because of
the possibility of strong westward water currents taking place which is not present in the
current data used for modeling.

662


Figure 10. Model Output for SS in February 1999.

SS April

Model output shows that there is no major change in the concentration of SS in the bay as
compared to February results (Figure 11). Concentration of SS is higher around Dil river than
other locations. Also the measured and the calculated concentrations of SS at the stations are
closely matching except station seven. The situation at station seven is likely to occur because
of the high amount of load coming from Dil River. The calculated concentration at station
seven is exceeding the measured concentration by 18 per cent. This deviation is because of
the precipitation of SS, which was not taken into account in the model. Because the model is
two-dimensional that ignores the vertical transportation of pollutants. The pollution close to
the land sources is because of the heavy rainfalls in the spring season. As a result of these
rainfalls erosion may occur and large amount of soil both precipitated and suspended is
carried to the bay via point and non point land sources.


663



Figure 11. Model Output for SS in April 1999.

TP February

Western and the middle section except Dil River region have small concentrations of
phosphates (Figure 12). Pollution is accumulated around the land sources especially Tupras
and Dil River. Highest concentration values is at station seven (0.03 mg/l). This is three times
the measured value at station seven which is 0.01 mg/l. The reason for such measured value
may be because of high biological activity and weak currents around Dil River. High
biological activity may consume the TP in a short transportation distance. But in the model,
extinction rate of the pollutant must be taken constant throughout the bay.


Figure 12. Model Output for TP in February 1999.

664

TP April

The easternmost parts of the bay seems to be less polluted than the rest of the bay
(Figure 13). Pollution is highest close to Dil River and Eastern Channel. Comparison between
the calculated and measure values show that except stations five, six and seven the values are
matching. Measured concentration values decrease starting from station one to station four,
but there is an upward trend at stations five and six., but again more inclined decrease starts
from the seventh station to ninth station. But he calculated values show a different trend. At
station seven the calculated value exceeds the measured value nearly 60 per cent. It must be
first pointed out that the pollutant concentrations are calculated taking the land sources basis.
Since the measurement at stations and land sources are conducted in April (start of the rainy
season) the effect of the high amount of phosphate coming from the non-point sources
(agricultural run-off) may be the source of high measured TP concentration at stations five
and six. Also TP load coming from Dil River may be consumed because of increased
biological activity so TP pollution may be restricted to a small area around Dil River. This
may be the reason for the lower measured concentration of TP the calculated concentration of
TP.



Figure 13. Model Output for TP in April 1999.

TN February

Model indicates that there is a severe TN pollution throughout the bay (Figure 14). TN
concentration exceeds 0.87 mg/l in some areas. Nearly at every station (except station six) the
calculated and the measured concentrations of TN are closely matching. At station six
calculated value exceeds the measured value by 56 per cent. There may be an accumulation of
TN.
665


Figure 14. Model Output for TN in February 1999.

TN April

As compared to February results there is a slight increase in TN concentration
throughout the bay (Figure 15). There is an increase in the concentration of TN closer to Dil
River. Also it has the highest concentration in the vicinity of Dil River. A comparison
between the measured and the calculated values at the stations clearly shows that the model is
successful for TN. The calculated and measured values are closely matching. As the measured
concentrations of TN in February and April are compared, a slight decrease in the
concentrations at all stations in April is observed. This is because of the seasonal increase of
the biological activity.


666



Figure 15. Model Output for TN in April 1999.


6. CONCLUSIONS

From the results of the modeling, it can be concluded that the two dimensional
modeling of the transportation of four different pollutants in the bay is relatively satisfactory.
Since the modeling is done in two dimensions, it is clear that the third dimension (the depth)
is omitted. In this study, only the pollutant loads discharged from the industrials located at the
northern part of Izmit Bay and from Dil River, which also discharges residential loads, were
taken into consideration. But these are not the only sources of pollution. There are also non-
point sources. Such environmental studies require at least one year to give reliable results. But
even within the limited scope of this study one can easily observe that high amount of
pollutant load is coming from Dil River. An extensive research and immediate precautions
must be done for that region.


REFERENCES

Akkaya, M., Soybay, S, Karsan, S. and Orhon, D., "revention and removal of the water
pollution in the zmit Bay: Determination of technological aspects. Third Progress Report,
Istanbul Technical University, Istanbul, pp.76-1983.
Cruzado, A., and Legovic, t., Proc. of the 2
nd
;Oceanographic Conf. Lisbon, November 14-19,
1994.
DAMOC, Istanbul Region Drinking Water and Sewerage Master Plan and Feasibility Studies.
UNDP-WHO Project Rep., Los Angeles, pp.385-1971
Dengili, K., Water Quality Assessment of zmit Bay by two-dimensional Modelling,
Boazii University Msc Thesis, !999.
Justic, D., Legovic, T., and Rottini-Sandrini, L., Estuarine, Coastal and Shelf Science 25, pp.
435-445, 1987.
667

Kullenberg, G. The Baltic Sea. In: Ecosystem of the World, B.H. Ketchum (Ed.) Elsevier,
New York, 1983, pp.309-335.
Kullenberg, G. The Baltic Sea. In: Ecosystem of the World, B.H. Ketchum (Ed.) Elsevier,
New York, 1983, pp.309-335.
Kor, N, The Control of Pollution in zmit Bay, TBTAK, Marmara Res. Center, Publ. No.
MAG 211/A, Gebze, pp.12, 1974.
Legovic, T., and Justic, D., MAP Technical Reports 78, pp.1-26, 1994.
Legovic,T., D. Petricioli, and V. Zutic, Marine Chemistry 32, pp.347-359, 1991.
Morkoc, E., Edinliler, A., Tufekci, H., Tolun, L., Tufekci, V., Karakoc, F.T., Okay, O.S.
Dengili, K., Aktan, Y. zmit Krfezinin iyiletirilmesi. Sonu Raporu, TUBITAK-MAM
Yayn, Gebze-Kocaeli (2000).
SWECO and BMB, zmit Sewerage Project: Master Plan, Kocaeli, p.301, 1976.
Timur, A., Dumlu, G., Timur, H., iler, M. and Balka, T. Prevention and removal of water
pollution in the zmit Bay. Determination of technological aspects TBTAK, Marmara
Res. Institute, Gebze, p.383, 1982.
Yu, I Sorokin, The Black Sea, In. Ecosytems of the World, B. H. Ketchum, (Ed.), Elsevier,
New York, pp.253-292.
Zambianchi, E., Calvitti, C., Cecamore, P., Damion, F., Ferulano, E., and Lanciano, xxxx,
Science of Total Environment, Suppl, pp.581-598, 1992.
668

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005








SUITABILITY OF MAIN CANALS WATER
QUALITY FOR IRRIGATION IN EGYPT

By
Dr. AMAL EL SHERBINI
Associate Professor, Nile Research Institute, NWRC Building, El Qanater P.O 13621
Email: amalsherb@Yahoo.Com

ABSTRACT
The quantity of freshwater used for irrigation in Egypt represents 85 % of the
released water from the Aswan Dam. The main irrigation canals receive water
from the Nile River. They also receive domestic, industrial, and agricultural
drainage wastewater. The quality of irrigation water is of particular importance
in arid and semi arid zones where high temperature and low humidity. Water
used for irrigation can vary greatly in quality depending upon type and quantity
of dissolved salts. A number of different irrigation water quality criteria have
been proposed. Potential management problems are associated with irrigation
water quality such as salinity hazard, sodium hazard, pH value, alkalinity,
specific ions, and other potential irrigation water contaminants such as heavy
metals and microbial contaminants. Irrigation water quality criteria for the main
canals and rayahas in Egypt are discussed in this paper using a group of samples
collected in both high and low flow seasons in the period between years 2000 to
2004. The main canals and rayahas water quality varies between class I
(excellent) to class II (good) using sodium and salinity hazards as a measure
respectively. In addition to that, the irrigation waters quality constituents such as
pH value, alkalinity, chloride, nitrate, sulfate, and other potential contaminants
such as trace metals are within the suggested limits for irrigation. Although the
main canals receive different wastes, it is shown in this paper the suitability of
their water quality for irrigation in both seasons for growing wide range of
crops. However, some precautions for monitoring and conserving this suitability
are recommended.

Key Words:
Irrigation Water Quality Criteria, Sodium Adsorption Ratio (SAR), Irrigation Water
Classification.
669
INTRODUCTION
Egypt is facing water scarcity as the demand is increased due to rapidly growing
population, increased urbanization, higher standard of living and an agricultural expansion.
Agriculture is strongly influenced by the availability of water. Irrigation has taken place in the
Nile Valley for nearly 5000 years. Egyptian irrigation system is extensive and complex. It
consists of the Aswan High Dam, eight main barrages, approximately 30,000 km of public
canals, 17,000 km of public drains, 80,000 km of private canals (mesqas or field ditches) and
farm drains, 450,000 private water lifiting devices (sakaias or pumps), 22,000 public water
control structures , and 670 large pumping stations for irrigation. In Egypt the water is
transferred from the Nile River to the fields through a system of main canals and rayahs,
secondary canals, third order and meskas. Four rayahs and three main canals are branched
from the Nile at the Delta Barrage. Tawfiki, and Menoufi rayahs and the following main
irrigation canals in Upper Egypt are considered in this paper:
Asfoun and Kalabia canals are branched from the Nile upstream Esna Barrage,
East and west Naga Hamadi canals are bifurcated upstream Naga Hammadi
Barrage,
Intake of the Ibrahimia canal is upstream Assiut Barrage, and
Bahr Yousef canal is branched from the Ibrahimia canal.
In this paper suitability of these irrigation system waters for irrigation is discussed.

IRRIGATION WATER QUALITY MONITORING
Since 1999, the main irrigation canals, and rayahas are monitored as a part of the
Egyptian National Water Quality Monitoring Program within the activities of the National
Water Quality and Availability Management (NAWQAM) project. The objective of the
monitoring for the main irrigation canals is to measure the seasonal variations and identify
water quality changes between the Nile River and the main irrigation canals.
Water quality samples are continuously collected from 11 main irrigation canals and
rayahs. The monitoring program designed to be implemented twice a year during the low and
high flow conditions (i.e. February/ August). Eight water quality sampling campaigns have
been carried so far. A group of physical, chemical, and microbiological parameters are
measured in each water sample. In this paper the water quality data of the main canals and
rayahas are discussed during the high (August/September) and low (February/March) flow
conditions in the period from years 2000 to 2004 with respect to Irrigation Water Quality
Criteria.

IRRIGATION WATER QUALITY CRITERIA
Although irrigation has been practiced throughout the world since very ancient time
for several centuries, it is only in this century that the importance of the quality of irrigation
water has been recognized. The quality of irrigation water is of particular importance for arid
and semi arid. Recently a number of different irrigation water quality guidelines have been
proposed. The following categories of potential management problems associated with
irrigation water quality are:
o Salinity Hazard is the most significant water quality criterion for crop
productivity which is measured by Total Dissolved Solids (TDS) and Electrical
Conductivity (EC).
o Sodium Hazard is defined by its specific detrimental effects on soil physical
properties. A deterioration of the physical condition of a soil is indirect effect of
high sodium content. Sodium hazard is expressed as Sodium Adsorption Ratio
(SAR). This index quantifies the relative proportion of sodium (Na
+
) to calcium
670
(Ca
++
) and magnesium (Mg
++
) ions in a sample. The Sodium Adsorption Ratio
(SAR) can be calculated as follows:

2 / ) ( / Mg Ca Na SAR + =
(1)

Where:
Na
+
, Ca
++
, and Mg
++
are expressed in meq/l.
The adjusted sodium adsorption ratio (SAR
adj
) is a recent modification of Eq (1) it reflects
the changes in calcium in the soil water more accurately due to receiving the irrigation
canals to municipal wastewaters.

2 / ) ( / Mg Cax Na SARadj + =
(2)

Where:
Na
+
and Mg
++
are expressed in meq/l., the value Ca
x
also expressed in meq/l and obtained
from Table according to EC in ds/m (Metcalf & Eddy, 1991).

o pH value and alkalinity. The acidity or basicity of irrigation water is expressed as
pH value (<7.0 acidic; >basic). The normal pH range for irrigation water is from
(6.5 to 8.4).

o Specific ions:
Chloride (Cl) is an anion (ve charged ion) frequently occurring in
irrigation water. A Cl determination is used mainly in irrigation water to
indicate possible toxicities to sensitive crops. Although, Cl is essential
element to plants in low concentrations, it can cause toxicity to sensitive
crops at high concentrations.
Nitrogen in irrigation water (N) is largely a fertility issue, and nitrate
(NO
3
) ion often occurs at higher concentration than ammonium in
irrigation water. Waters high in N can cause quality problems in crops such
as sugar beets and barley.
Sulfate (SO
4
)

ion is a major contributor to salinity in many irrigation
waters. However, toxicity is an issue at very high concentrations where
high sulfate can interfere with uptake of other nutrients.
Boron (B) is essential element in low concentrations, but toxic at higher
concentrations.

o Heavy metals such as aluminum (Al), cadmium (Cd), chromium (Cd), copper
(Cu), iron (Fe), manganese (Mn), lead (pb), vanadium (V), zinc (Zn), arsenic
(As), cobalt (Co), nickel (Ni), and selenium (Se).

o Microbiological contaminants. Contaminated water has always been an
important agent in the spread of diseases. Microbiological criteria are presently
undergoing re-evaluation throughout the world, and the historical dependence
upon total and fecal coliforms is being supplemented by more specific,
epidemiologically derived indicators of water quality. Fecal coliform is used as
microbiological indicators for irrigation water quality.

671
RESULTS AND DISCUSSIONS
This research discussed the water quality data collected from the 11 main irrigation
canals & rayahas for four water quality sampling campaigns collected in each season in the
period from year 2000 to 2004. The quality of these main canals and rayahas are discussed in
this paper with respect to the irrigation water quality criteria (Bauder et al, 2004) as follows:

Salinity Hazard
Salinity hazard is measured by Total Dissolved Solids (TDS) and Electric
Conductivity (EC) in the water quality samples collected form the main canals and rayahas. In
Table (1) during the high flows, the average TDS and EC varied from 186 mg/l and 0.286
ds/m at Asfoun canal to 317 mg/l and 0.492 ds/m at Baher Yuuesf (Sakoula) respectively.
However, in the low flows the average TDS and EC varied from 198 mg/l and 0.307 ds/m at
Asfoun canal to 400 mg/l and 0.476 ds/m at Baher Yuuesf (Sakoula) respectively in the study
period. The water quality of the main canals and rayahas with respect to salinity hazard is
classified in class II as (Good), as values of TDS and EC range between 175 to 525 mg/l and
between 0.25 to 0.75 ds/m respectively according to the irrigation water quality criteria. Table
(1) also indicates that the salinity in the low flows is higher than in the high flows; however,
the quality of water in both seasons is very good as far as salinity is concerned.

Sodium Hazard
Sodium hazard is measured by calculating the average of SAR and SAR
adj
for the
main canals over the study period for each season as shown in Table (1). During the high
flows average SAR and SAR
adj
varied from 0.806 and 0.795 at Ibrahimia canal (Minia) to
1.016 and 1.071 at Bahr Yuesf (Sakoula) respectively. However, during the low flows the
SAR and SAR
adj
varied from 0.806 and 0.891 at Ibrahimia canal (Dairout) to 1.612 and 1.996
at Ibrahimia canal (Beni Suef). The quality of the main canals with respect to sodium hazard
lays in class I as (Excellent) where average either SAR or SAR
adj
are less than 3 according to
the irrigation water quality criteria. Table (1) also shows that the average SAR and SAR
adj
in
the low flows are higher than in the high flows. In addition to that there is no sodium hazard
detected in the main canals water quality in both seasons.

pH value and Alkalinity
The pH values are indicators of the acidity or basicity for the irrigation water. Table
(2) shows that pH values for the main canals are within the normal range (6.5 8.4) for
irrigation water in both flow conditions over the study period. pHs Lower than 6.5 may
accelerate irrigation system corrosion. However, pHs values above 8.5 indicate high
carbonate and bicarbonates, known as alkalinity. High carbonates cause alkaline water which
intensify sodic soil condition. In this study the alkalinity consists of bicarbonates only
because detected carbonates are zero. This confirmed that the main canals and rayhas water
quality has no sodium hazard as indicated by adjusted SAR.

Specific Ions:
Chlorides (Cl) as shown in Table (2) varied as average concentrations for the main
canals and rayhas from 9.27 mg/l at Asfoun canal to 34.53 mg/l at Bahr Yusef (Lahoun) canal
during the high flows. However, during the low flows the average Cl varied from 11.41 mg/l
at West Naga Hammadie canal to 48.80 mg/l at Ibrahimia (Beni Suef) canal as shown in
Table (2). The average Cl fluctuates in both flow conditions and generally safe for all plants
because their concentrations in the main canals are below 70 mg/l as recommended by
irrigation water quality guidelines.
672
The average Nitrates (NO
3
) altered from 0.20 mg/l at Tawfiki and Mnoufi rayahas to
2.44 mg/l at Bahr yusef (Sakoula) in the high flows. However, in the low flow the average
NO
3
concentration altered from 1.12 mg/l at Asfoun canal to 10.13 mg/l at Ibrahimia canal
(Minia). It is also shown from Table (2) that average NO
3
during the low flows is higher than
in the high flows at most of the canals. Nitrate is lower than 45 mg/l the recommended level
to overcome by good fertilizers and irrigation management.
The average sulfates (SO
4
) for the main canals and rayahas during high flows
fluctuated from 19 mg/l at Kalabia canal to 52.63 mg/l at Bahr Yuesf (Sakoula). However, in
the low flows the average (SO
4
) varied 22.88 mg/l at Ibrahimia canal (Minia) to 63.78 mg/l at
Ibrahimia canal (Beni Suef) as indicated in the Table (2). Also, it is shown from Table (2) that
the average concentrations (SO
4
) in the low flows are higher than in the high flows at almost
all canals under study.

Boron
Boron (B) was not measured over the study period for the main canals and rayahas.
However, reference values for boron were measured in the Nile waters varied from 0.01 mg/l
to 0.275 mg/l. Therefore, it is expected that boron level may not reach the toxic level and the
quality of water will be applicable to even sensitive crops where boron concentrations varied
from 0.50 to 1.00 mg/l.

Heavy Metals
The heavy metals are specific ion toxicity, which is difficult to correct without
changing the crop or the water supply. In sever cases water used that contains trace elements
concentration above the recommended levels, may accumulate in plants and soils and can
cause human and animal health hazards or phytotoxicity in plants . The following trace metals
are discussed for the main irrigation canals and rayahas in both high and low conditions over
the study period from 2000 to 2004 as shown in Tables (3&4) with respect to the maximum
recommended levels for irrigation water (Metcalf & Eddy, 1991).
Average aluminum (Al) concentrations fluctuated from 0.092 mg/l at Ibarhimia canal
(Dairout) to 0.807 mg/l at Tawfiki rayah during the high flows. On the other hand, the average
Al during the low flows varied from 0.155 mg/l at Menoufia rayaha to 0.501 mg/l at Bahr
Yusef canal (Sakoula). The average Al in most of these canals in low flows is lower than in
high flows. However, in both flow conditions the average Al concentrations are less than 5
mg/l the maximum recommended level.
Average cadmium (Cd) varied from 0.0005 mg/l at Kelabia and Baher Yusef canals to
0.001 mg/l at Asfoun canal during the high flows. However, during the low flows the average
Cd varied from 0.0005 mg/l at Kelabia canal to 0.0040 mg/l at Tawfiki rayah. Average Cd
concentrations in both flow conditions are lower than 0.01 mg/l the maximum recommended
level. Conservative limits recommended because of its potential for accumulation in plants
and soils to concentrations that may be harmful to humans.
The average chromium (Cr) in the high flows varied from 0.002 mg/l at Ibarhimia
canal (Beni suef) to 0.006 mg/l at East Naga Hammadie canal, however, in the low flows the
average Cr varied from 0.002 mg/l at Ibarhimia canal (Minia) to 0.099 mg/l at Tawfiki rayah.
In both flow conditions the average Cr concentrations are lower than 0.1 mg/l the maximum
recommended level.
The average copper (Cu) concentrations varied from 0.002 mg/l at East Naga
Hammadie canal to 0.023 mg/l at Tawfiki rayah in the high flows. In the low flows the
average Cu fluctuated from 0.002 mg/l at Ibarhimia canal (Dairout) to 0.02 mg/l at at Tawfiki
673
rayah. The Cu as average concentrations in both flow conditions are below 0.2 mg/l the
maximum recommended level.
Iron (Fe) concentrations as average fluctuated from 0.163 mg/l at Ibarhimia canal
(Dariout) to 1.386 mg/l at Tawfiki rayah in the high flows. While, in the low flows the
average Fe varied from 0.211 mg/l at Ibarhimia canal (Dariout) to 1.497 mg/l at Bahr Yusef
(Sakoula). The average Fe in both flow seasons is lower than 5 mg/l the maximum
recommended level.
The average manganese (Mn) in the high flows varied from 0.030 mg/l at East Naga
Hammadie canal to 0.190 mg/l at Tawfiki rayah. In the low flows, the average Mn fluctuated
from 0.041 mg/l at Kalabia canal to 0.136 mg/l at Ibarhimia canal (Beni Suef). The average
Mn concentration is still below 0.20 the maximum recommended level.
Average lead (pb) concentrations fluctuated from 0.013 mg/l at Ibarhimia canal (Beni
Suef) to 0.025 mg/l at West Naga Hammadie canal in the high flows. While, in the low flows
average pb concentrations varied from 0.005 mg/l at West and East Naga Hammadie canals to
0.201 mg/l Bahr Yusef (Sakoula). In both flow conditions the average lead concentrations is
very low and blow 5 mg/l the maximum recommended level. It can inhabit plant cell growth
at very high concentration.
Average vanadium (V) level is 0.005 mg/l at all canals except West Naga Hammadie,
where V reached 0.12 mg/l in the high flows. On the other hand, the average V varied from
0.005 mg/l at Ibarhimia canal (Dairout & Minia), East Naga Hammadie and Asfoun canals to
0.660 mg/l at Bahr Yusef (Sakoula) in the low flows. In both flow conditions very rare cases
exceed 0.10 mg /l the maximum recommended level. This element is very essential because it
is toxic to many plants at relatively low concentrations.
The average Zinc (Zn) concentrations altered from 0.005 mg/l at Tawfiki rayah to
0.102 mg/l at Asfoun canal in the high flows. While, in the low flows the average Zn levels
altered from 0.009 mg/l at Asfoun canal to 0.112 mg/l at Tawfiki rayah. Although, the
average Zn varied from canal to canal and from season to the other in the same canal, their
concentrations still lower than 2 mg/l the maximum recommended level.
Arsenic (As), cobalt (Co), nickel (Ni), and selenium (Se) concentrations in both flow
conditions are constant at all canals and equal to 0.01, 0.0005, 0.005, and 0.03 in mg/l
respectively which are lower than the detection levels of measurements. The concentrations of
these elements are also lower than 0.1, 0.05, 0.2, and 0.02 mg/l respectively the maximum
recommended level for each element.

Fecal Coliform
Detection of fecal coliform as microbiological indicator is selected to protect the
ultimate consumers of the crops, grazing animals, farm workers, and in the case of non-farm
irrigation, the general public. These criteria depend upon the method of applying the irrigation
water, the type of crop grown, and the way the crop is used.
Table (5) shows the average fecal coliform varied from 2585 CFU/100ml at Tawfiki
rayah to 43675 CFU/100ml at Bahr Yusef (Lahoun) in the high flows. However, in the low
flows fluctuated from 713 CFU/100ml at Tawfiki rayah to 17775 CFU/100ml at Bahr Ysef
(Sakoula). Fecal coliform counts in high flows which correspond to summer season are higher
than in the low flows. The average fecal coilform for all the main canals and rayahs exceeded
200/100ml the maximum recommended level for the irrigation for crops eaten raw in both
flow seasons (Johnson, 2001) due to receiving the canals different wastes specifically raw or
inadequate treatment municipal wastes.



674
CONCLUSION
This research presents the suitability of the main canals and rayahas water quality for
irrigation over the study period from years 2000 to 2004 during high and low flow seasons.
This paper discussed the suitability by comparing the different water quality criteria with
respect to the suggested levels for irrigation. The irrigation water quality criteria discussed
here are: salinity and sodium hazards, pH, alkalinity, chloride, nitrate, sulfate, heavy metals,
and fecal coliform.
The nutrient content of irrigation water can be highly variable from one location to
another and from season to the other in the same location. This study indicates that the waters
of the main canals and rayahas are suitable as far as salinity or sodium hazard are concerned.
The quality of water in these canals changes between class I (excellent) to class II (good)
using sodium and salinity hazards as measure respectively. Furthermore, the other water
quality criteria such as pH, alkalinity, chlorides, sulfates, nitrates are high in the low flows
than in the high flows in most of the location due to low discharge. In both flow conditions,
these quality criteria of the main canals and rayahas under study are lower than the maximum
recommended levels for irrigation.
The study also reveals that the heavy metals concentrations such as aluminum,
cadmium, chromium, copper, iron, manganese, lead, zinc, arsenic, cobalt, nickel, and
selenium are below the maximum recommended levels for irrigation except vanadium, where
in very rare locations exceed the maximum recommended level. The concentrations of the
heavy metals during the high flows are high than in the low flows, may be due to sediment
water interactions and sediment deposition in the low flows.
Fecal coliform count of the main canals and rayahas is microbiological indicator in
irrigation water quality criteria. Average fecal coliform counts in the high flows are higher
than in the low flows over the study period for the main canals and rayahas may be due to the
increase in the municipal wastes in summer season. The average fecal coliform counts in both
flow seasons exceeded the suggested levels for irrigation (200/100ml) for crop eaten raw due
to municipal wastes either treated or raw received by the canals.

RECOMMENDATIONS
The following are recommended:
o Continuous maintenance and monitoring for the irrigation systems,
o Villages are to be provided with sewerage systems parallel to the water supply
facilities,
o A Program for pollution control should be established to the pollution sources by
introducing the most appropriate technology of municipal waste water treatment,
o Agricultural practice must be improved to reduce the uncontrolled use of pesticide,
herbicides, and fertilizers,
o Minimizing the open water areas by using the more advanced irrigation systems,
o Good drainage to allow a continuous movement of water and salt below root zone,
o Quality problems in crops by high nitrogen can be overcome by good fertilizer
and irrigation management, and
o Laws and regulations should be modified for practicing the water bodies against
pollution to avoid their applications difficulties.

REFERENCES
Bauder, T.A., Cardon,G.E., Waskom, R.M., and Davis, J.G., " Irrigation Water Quality
Criteria", Colorado State University Cooperative Extension Agriculture, August 2004.
675
El - Ganzori, A., Abdel Gawad, S.T., El Sayed, A., and Hashim, A., "Assessment of the
Drainage Water Reuse in the Irrigation Improvement Areas, Kafer Sheikh Governorate of
Egypt", ICEHM2000, Cairo University, September 2000, page 333 344.
El Sherbini, A.M, M. M. Radwan, and "Quality of Irrigation and Drainage Water in Egypt",
Agricultural and Food in the Arab World: Constrains and Future Plans, 1
st
Egyptian and
Syrian Conference (1
st
ESC), 8-11 December 2003, El Minia-Egypt.
El Sherbini, A.M., "Impacts of Different Factors Affecting River Nile Water Quality", Arab
Water 2004, Second Regional Conference on Action Plans for Integrated Development, 12-15
April Cairo-Egypt.
El-Sherbini, A. M., " Design of Monitoring Networks for Water Quality Management in the
river Nile", ph.D thesis, Faculty of Engineering, Ain Shams University, Cairo - Egypt,
October 1996.
Frank, K.D., and Delymy Hay, "Testing Irrigation Water", Water Resources Management,
NebGuid, May 1993.
Johnson, T.R., "Water Quality Criteria for Microbiological Indicators", Overview Report,
Ministry of Water, Land, and Air Protection, Government of British Colombia, August 2001.
Martin Hvidt, "Water Resources Planning in Egypt", Odense University, Denmark, The
Middle Eastern Environment, St Malo Press, J anuary 2000.
Metcalf and Eddy, Inc., "Wastewater Engineering: Treatment, Disposal, Reuse", 3
rd
ed.,
McGraw-Hill, New York, 1991.
Rosenzweing C., and Hillel, D., "Potential Imacts of Climate Change on Agriculture and Food
Supply", CONESQUENCES, Volume 1, Number 2, 1995.




























676
T
a
b
l
e

(
1
)

A
v
e
r
a
g
e

S
a
l
i
n
i
t
y

a
n
d

S
o
d
i
u
m

H
a
z
a
r
d
s

f
r
o
m

Y
e
a
r

2
0
0
0

t
o

2
0
0
4

f
o
r

H
i
g
h

a
n
d

L
o
w

F
l
o
w

C
o
n
d
i
t
i
o
n
s

N
a
m
e

o
f

C
a
n
a
l
s

T
D
S

(
m
g
/
l
)

E
C

(
d
s
/
m
)

Q
u
a
l
i
t
y

C
l
a
s
s

S
A
R

S
A
R
a
d
j

Q
u
a
l
i
t
y

C
l
a
s
s


H
i
g
h

L
o
w

H
i
g
h

L
o
w


H
i
g
h

L
o
w

H
i
g
h

L
o
w


T
a
w
f
i
k
i

r
a
y
a
h
.

2
3
0

2
4
3

0
.
3
6
0

0
.
3
8
5

G
o
o
d

0
.
8
7
2

1
.
1
0
8

0
.
8
8
0

1
.
0
3
2

E
x
c
e
l
l
e
n
t

M
o
n
o
u
f
i
a

r
a
y
a
h
.

2
3
9

2
4
7

0
.
3
5
6

0
.
3
9
1

G
o
o
d

0
.
9
0
7

1
.
0
4
2

0
.
9
3
7

0
.
9
5
5

E
x
c
e
l
l
e
n
t

I
b
r
a
h
i
m
i
a

c
a
n
.

(
D
a
i
r
o
u
t
)

1
9
5

2
0
5

0
.
3
0
3

0
.
3
1
8

G
o
o
d

0
.
8
6
6

0
.
8
9
1

0
.
8
5
2

0
.
8
0
6

E
x
c
e
l
l
e
n
t

I
b
r
a
h
i
m
i
a

c
a
n
.
(
M
i
n
i
a
)

1
9
4

2
0
5

0
.
3
0
4

0
.
3
2
0

G
o
o
d

0
.
8
0
6

0
.
9
0
1

0
.
7
9
5

0
.
8
0
4

E
x
c
e
l
l
e
n
t

I
b
r
a
h
i
m
i
a

c
a
n
.
(
B
e
n
i

S
u
e
f
)

2
2
0

4
0
0

0
.
3
4
3

0
.
6
2
6

G
o
o
d

0
.
8
2
3

1
.
5
9
6

0
.
8
4
1

1
.
6
1
2

E
x
c
e
l
l
e
n
t

B
a
h
r

Y
u
e
s
f

(
S
a
k
o
u
l
a
)

c
a
n
.

3
1
7

3
0
4

0
.
4
9
3

0
.
4
7
6

G
o
o
d

1
.
0
1
6

1
.
4
2
5

1
.
0
7
1

1
.
3
1
4

E
x
c
e
l
l
e
n
t

B
a
h
r

Y
u
e
s
f

(
L
a
h
o
u
n
)

c
a
n
.

3
0
8

3
0
7

0
.
4
7
8

0
.
4
7
7

G
o
o
d

0
.
9
7
7

1
.
4
7
2

1
.
0
2
7

1
.
3
3
5

E
x
c
e
l
l
e
n
t

W
e
s
t

N
a
g
a

H
a
m
m
a
d
i
e

c
a
n
.

1
9
6

2
0
1

0
.
3
0
3

0
.
3
1
6

G
o
o
d

0
.
8
9
3

0
.
9
5
1

0
.
8
7
5

0
.
8
0
0

E
x
c
e
l
l
e
n
t

E
a
s
t

N
a
g
a

H
a
m
m
a
d
i
e

c
a
n
.

2
0
3

2
1
0

0
.
3
1
7

0
.
3
2
2

G
o
o
d

0
.
9
1
4

1
.
2
0
7

0
.
8
9
8

1
.
0
8
0

E
x
c
e
l
l
e
n
t

A
s
f
o
u
n

c
a
n
.

1
8
6

1
9
8

0
.
2
8
6

0
.
3
0
7

G
o
o
d

0
.
8
3
3

0
.
9
5
5

0
.
8
0
6

0
.
8
7
8

E
x
c
e
l
l
e
n
t

K
e
l
a
b
i
a

c
a
n
.

1
9
8

2
1
6

0
.
3
0
6

0
.
3
3
3

G
o
o
d

0
.
8
8
2

1
.
0
4
7

0
.
8
7
7

0
.
9
4
6

E
x
c
e
l
l
e
n
t

T
a
b
l
e

(
2
)

A
v
e
r
a
g
e

p
H
,

A
l
k
a
l
i
n
i
t
y
,

C
h
l
o
r
i
d
e
,

N
i
t
r
a
t
e
,

a
n
d

S
u
l
f
a
t
e

f
r
o
m

Y
e
a
r

2
0
0
0

t
o

2
0
0
4

f
o
r

H
i
g
h

a
n
d

L
o
w

F
l
o
w

C
o
n
d
i
t
i
o
n
s

N
a
m
e

o
f

C
a
n
a
l
s

p
H

v
a
l
u
e

A
l
k
a
l
i
n
i
t
y

(
m
g
/
l
)

N
i
t
r
a
t
e

(
m
g
/
l
)

C
h
l
o
r
i
d
e

(
m
g
/
l
)

S
u
l
f
a
t
e

(
m
g
/
l
)


H
i
g
h


L
o
w


H
i
g
h


L
o
w

H
i
g
h


L
o
w

H
i
g
h


L
o
w


H
i
g
h


L
o
w


T
a
w
f
i
k
i

r
a
y
a
h
.

7
.
9
7
0

7
.
8
7
5

1
4
1
.
5
0

1
3
9
.
2
5

0
.
2
0
0

1
.
3
0

1
9
.
9
5

1
8
.
5
0

2
2
.
4
0

2
7
.
8
3

M
o
n
o
u
f
i
a

r
a
y
a
h
.

7
.
8
3
0

7
.
7
7
5

1
4
6
.
0
0

1
3
5
.
0
0

0
.
2
0
0

1
.
4
0

1
7
.
5
9

2
0
.
7
8

2
4
.
8
0

2
8
.
4
8

I
b
r
a
h
i
m
i
a

c
a
n
.

(
D
a
i
r
o
u
t
)

8
.
2
8
0

8
.
3
9
3

1
3
4
.
0
0

1
2
9
.
0
0

1
.
0
0

1
.
0
0

1
4
.
7
0

1
2
.
1
4

2
3
.
0
0

2
8
.
5
0

I
b
r
a
h
i
m
i
a

c
a
n
.
(
M
i
n
i
a
)

7
.
9
2
8

7
.
8
6
8

1
2
9
.
5
0

1
3
3
.
0
0

3
.
2
0

1
.
2
0
0

1
4
.
7
0

1
3
.
0
9

2
1
.
7
3

2
2
.
8
8

I
b
r
a
h
i
m
i
a

c
a
n
.
(
B
e
n
i

S
u
e
f
)

7
.
8
1
0

7
.
6
1
8

1
3
4
.
0
0

1
7
0
.
0
0

0
.
7
0
3

1
0
.
1
3

1
4
.
4
8

4
8
.
5
6

3
7
.
1
8

6
3
.
7
8

B
a
h
r

Y
u
e
s
f

(
S
a
k
o
u
l
a
)

c
a
n
.

7
.
9
4
0

7
.
6
3
0

1
4
7
.
2
5

1
4
4
.
0
0

2
.
4
4

2
.
2
0
0

3
3
.
8
6

3
1
.
4
5

5
2
.
6
3

4
9
.
8
3

B
a
h
r

Y
u
e
s
f

(
L
a
h
o
u
n
)

c
a
n
.

7
.
8
6
3

7
.
6
9
0

1
4
6
.
5
0

1
4
4
.
5
0

2
.
0
0

2
.
6
0
7

3
4
.
5
3

2
9
.
7
5

4
1
.
8
7

5
3
.
0
0

W
e
s
t

N
a
g
a

H
a
m
m
a
d
i
e

c
a
n
.

7
.
9
4
3

7
.
9
5

1
2
9
.
0
0

1
2
7
.
5
0

1
.
1
1

1
.
1
1

1
2
.
9
3

1
1
.
4
1

2
3
.
3
8

2
7
.
0
0

E
a
s
t

N
a
g
a

H
a
m
m
a
d
i
e

c
a
n
.

8
.
2
1
3

8
.
1
8
5

1
2
8
.
5
0

1
3
1
.
5
0

1
.
6
0

1
.
1
9

1
2
.
9
2

1
2
.
8
7

2
3
.
8
8

2
7
.
7
5

A
s
f
o
u
n

c
a
n
.

7
.
7
7
3

7
.
9
8
9

1
2
5
.
0
0

1
2
9
.
5
0

1
.
1
0

1
.
1
2

9
.
2
7

1
2
.
3
8

1
9
.
2
5

2
3
.
5
0

K
e
l
a
b
i
a

c
a
n
.

7
.
6
9
5

7
.
7
8
8

1
3
5
.
0
0

1
3
2
.
0
0

1
.
2
8

1
.
2
5

1
1
.
7
1

1
4
.
2
0

1
9
.
0
0

3
0
.
0
0


677
T
a
b
l
e

(
3
)

A
v
e
r
a
g
e

A
l
u
m
i
n
u
m
,

C
a
d
m
i
u
m
,

C
h
r
o
m
i
u
m
,

a
n
d

C
o
p
p
e
r

f
r
o
m

Y
e
a
r

2
0
0
0

t
o

2
0
0
4

f
o
r

H
i
g
h

a
n
d

L
o
w

F
l
o
w

C
o
n
d
i
t
i
o
n
s

N
a
m
e

o
f

C
a
n
a
l
s

A
l
u
m
i
n
u
m

(
m
g
/
l
)

C
a
d
m
i
u
m
(
m
g
/
l
)

C
h
r
o
m
i
u
m
(
m
g
/
l
)

C
o
p
p
e
r

(
m
g
/
)


H
i
g
h

L
o
w

H
i
g
h

L
o
w

H
i
g
h

L
o
w

H
i
g
h

L
o
w

T
a
w
f
i
k
i

r
a
y
a
h
.

0
.
8
0
7

0
.
2
0
7

0
.
0
0
0
5

0
.
0
0
4

0
.
0
0
6

0
.
0
9
9

0
.
0
2
3

0
.
0
2

M
o
n
o
u
f
i
a

r
a
y
a
h
.

0
.
0
9
3

0
.
1
5
5

0
.
0
0
0
5

0
.
0
0
2

0
.
0
0
4

0
.
0
2
5

0
.
0
2
0

0
.
0
2

I
b
r
a
h
i
m
i
a

c
a
n
.

(
D
a
i
r
o
u
t
)

0
.
0
9
2

0
.
4
8
2

0
.
0
0
0
7

0
.
0
0
0
5

0
.
0
0
2

0
.
0
1

0
.
0
1
7

0
.
0
0
2

I
b
r
a
h
i
m
i
a

c
a
n
.
(
M
i
n
i
a
)

0
.
1
7
9

0
.
3
1
1

0
.
0
0
0
7

0
.
0
0
2

0
.
0
0
2

0
.
0
0
2

0
.
0
0
2

0
.
0
1
0

I
b
r
a
h
i
m
i
a

c
a
n
.
(
B
e
n
i

S
u
e
f
)

0
.
1
2
2

0
.
4
4
5

0
.
0
0
0
7

0
.
0
0
0
7

0
.
0
0
2

0
.
0
4
6

0
.
0
0
2

0
.
0
4
1

B
a
h
r

Y
u
e
s
f

(
S
a
k
o
u
l
a
)

c
a
n
.

0
.
7
0
2

0
.
5
0
1

0
.
0
0
0
5

0
.
0
0
1

0
.
0
0
2

0
.
0
0
2

0
.
0
0
2

0
.
0
3
0

B
a
h
r

Y
u
e
s
f

(
L
a
h
o
u
n
)

c
a
n
.

0
.
1
9
2

0
.
3
2
8

0
.
0
0
5

0
.
0
0
3

0
.
0
0
2

0
.
0
1
3

0
.
0
0
2

0
.
0
3
1

W
e
s
t

N
a
g
a

H
a
m
m
a
d
i
e

c
a
n
.

0
.
3
5
0

0
.
3
3
4

0
.
0
0
2

0
.
0
0
1

0
.
0
0
5

0
.
0
0
9

0
.
0
1
8

0
.
0
2
3

E
a
s
t

N
a
g
a

H
a
m
m
a
d
i
e

c
a
n
.

0
.
1
0
8

0
.
2
3
5

0
.
0
0
2

0
.
0
0
0
7

0
.
0
0
6

0
.
0
1

0
.
0
0
2

0
.
0
2
1

A
s
f
o
u
n

c
a
n
.

0
.
1
2
9

0
.
2
9
2

0
.
0
0
1

0
.
0
0
0
7

0
.
0
0
3

0
.
0
0
9

0
.
0
2
2

0
.
0
1
8

K
e
l
a
b
i
a

c
a
n
.

0
.
1
4
8

0
.
1
7
7

0
.
0
0
0
5

0
.
0
0
0
5

0
.
0
0
3

0
.
0
0
9

0
.
0
1
7

0
.
0
2
3

T
a
b
l
e

(
4
)

A
v
e
r
a
g
e

I
r
o
n
,

M
a
n
g
a
n
e
s
e
,

L
e
a
d
,

V
a
n
a
d
i
u
m
,

a
n
d

Z
i
n
c

f
r
o
m

Y
e
a
r

2
0
0
0

t
o

2
0
0
4

f
o
r

H
i
g
h

a
n
d

L
o
w

F
l
o
w

C
o
n
d
i
t
i
o
n
s

N
a
m
e

o
f

C
a
n
a
l
s

I
r
o
n

(
m
g
/
)

M
a
n
g
a
n
e
s
e

(
m
g
/
)

L
e
a
d

(
m
g
/
l
)

V
a
n
a
d
i
u
m

(
m
g
/
l
)
Z
i
n
c

(
m
g
/
l
)


H
i
g
h


L
o
w


H
i
g
h


L
o
w


H
i
g
h


L
o
w


H
i
g
h


L
o
w


H
i
g
h


L
o
w


T
a
w
f
i
k
i

r
a
y
a
h
.

1
.
3
8
6

0
.
3
0
0

1
.
0
1
0

0
.
0
6
4

0
.
0
1
0

0
.
0
2
7

0
.
0
0
5

0
.
0
9
5

0
.
0
0
5

0
.
1
1
2

M
o
n
o
u
f
i
a

r
a
y
a
h
.

0
.
2
6
5

0
.
2
6
5

0
.
0
5
4

0
.
1
1
7

0
.
0
1
4

0
.
0
1
5

0
.
0
0
5

0
.
0
1
0

0
.
0
0
5

0
.
0
3
3

I
b
r
a
h
i
m
i
a

c
a
n
.

(
D
a
i
r
o
u
t
)

0
.
1
6
3

0
.
2
1
1

0
.
0
4
4

0
.
0
7
7

0
.
0
1
6

0
.
0
0
5

0
.
0
0
5

0
.
0
0
5

0
.
0
8
3

0
.
0
1
0

I
b
r
a
h
i
m
i
a

c
a
n
.
(
M
i
n
i
a
)

0
.
2
7
9

0
.
2
8
5

0
.
0
5
2

0
.
0
8
0

0
.
0
1
7

0
.
0
2
6

0
.
0
0
5

0
.
0
0
5

0
.
0
6
9

0
.
0
1
0

I
b
r
a
h
i
m
i
a

c
a
n
.
(
B
e
n
i

S
u
e
f
)

0
.
1
7
2

0
.
7
0
0

0
.
0
4
4

0
.
1
3
6

0
.
0
1
3

0
.
0
1
2

0
.
0
0
5

0
.
0
8
5

0
.
0
4
6

0
.
0
4
8

B
a
h
r

Y
u
e
s
f

(
S
a
k
o
u
l
a
)

c
a
n
.

0
.
7
5
4

1
.
4
9
7

0
.
0
9
4

0
.
1
2
4

0
.
0
1
7

0
.
2
0
1

0
.
0
0
5

0
.
6
6
0

0
.
0
4
8

0
.
0
9
4

B
a
h
r

Y
u
e
s
f

(
L
a
h
o
u
n
)

c
a
n
.

0
.
2
7
7

0
.
6
7
2

0
.
0
5
5

0
.
1
2
0

0
.
0
1
8

0
.
0
1
2

0
.
0
0
5

0
.
1
8
9

0
.
0
8
9

0
.
0
4
4

W
e
s
t

N
a
g
a

H
a
m
m
a
d
i
e

c
a
n
.

0
.
4
0
2

0
.
1
6
2

0
.
0
4
7

0
.
0
1
0

0
.
0
2
5

0
.
0
0
5

0
.
1
2
0

0
.
0
0
6

0
.
0
7
9

0
.
1
8
7

E
a
s
t

N
a
g
a

H
a
m
m
a
d
i
e

c
a
n
.

0
.
1
2
7

0
.
2
1
5

0
.
0
3
0

0
.
0
4
7

0
.
0
2
0

0
.
0
0
5

0
.
0
0
5

0
.
0
0
5

0
.
0
7
8

0
.
0
1
1

A
s
f
o
u
n

c
a
n
.

0
.
1
7
3

0
.
1
3
8

0
.
0
4
2

0
.
0
5
0

0
.
0
1
5

0
.
0
0
5

0
.
0
0
5

0
.
0
0
5

0
.
1
0
2

0
.
0
0
9

K
e
l
a
b
i
a

c
a
n
.

0
.
2
5
6

0
.
2
8
2

0
.
0
7
5

0
.
0
4
1

0
.
0
2
0

0
.
0
0
9

0
.
0
0
5

0
.
0
0
6

0
.
0
9
1

0
.
0
1
7


678
T
a
b
l
e

(
5
)
:

A
v
e
r
a
g
e

F
e
c
a
l

C
o
l
i
f
o
r
m

f
r
o
m

2
0
0
0

t
o

2
0
0
4


N
a
m
e

o
f

C
a
n
a
l
s

F
e
c
a
l

C
o
l
i
f
o
r
m

(
C
F
U
/
1
0
0
m
l
)

Q
u
a
l
i
t
y

S
t
a
t
u
s

a
n
d

R
e
m
a
r
k
s


H
i
g
h

F
l
o
w

L
o
w

F
l
o
w


T
a
w
f
i
k
i

r
a
y
a
h
.

2
5
8
5

7
1
3

T
h
e

q
u
a
l
i
t
y

o
f

w
a
t
e
r

i
s

n
o
t

a
p
p
l
i
c
a
b
l
e

f
o
r

c
r
o
p

e
a
t
e
n

r
a
w
.

T
h
e

R
e
c
o
m
m
e
n
d
e
d

I
r
r
i
g
a
t
i
o
n

W
a
t
e
r

Q
u
a
l
i
t
y

C
r
i
t
e
r
i
a

f
o
r

C
r
o
p
s

E
a
t
e
n

R
a
w

i
s

2
0
0

C
F
U
/
1
0
0
m
l
.

M
o
n
o
u
f
i
a

r
a
y
a
h
.

1
6
3
0
0

1
4
1
7
5


I
b
r
a
h
i
m
i
a

c
a
n
.

(
D
a
i
r
o
u
t
)

5
2
7
5

1
4
6
3


I
b
r
a
h
i
m
i
a

c
a
n
.
(
M
i
n
i
a
)

4
7
1
3

1
1
7
5


I
b
r
a
h
i
m
i
a

c
a
n
.
(
B
e
n
i

S
u
e
f
)

3
8
0
5
0

4
1
2
5


B
a
h
r

Y
u
e
s
f

(
S
a
k
o
u
l
a
)

c
a
n
.

1
0
1
0
0

1
7
7
7
5


B
a
h
r

Y
u
e
s
f

(
L
a
h
o
u
n
)

c
a
n
.

4
3
6
7
5

5
8
7
5


W
e
s
t

N
a
g
a

H
a
m
m
a
d
i
e

c
a
n
.

1
7
5
0
0

2
4
3
8


E
a
s
t

N
a
g
a

H
a
m
m
a
d
i
e

c
a
n
.

7
2
5
0

1
6
2
5


A
s
f
o
u
n

c
a
n
.

3
1
7
5

2
2
1
3


K
e
l
a
b
i
a

c
a
n
.

2
7
5
0

1
9
7
5




679

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


IMPACT ASSESSMENT OF INDUSTRIAL
WASTEWATER DIAPOSAL INTO IRRIGATION
DRAINAGE SYSTEM
(CASE STUDY)


Elazizy
1
, I., Khalifa
1
, A., Abdel Nasser
2
, G. and El Daly
1
, M.
1. Ain Shams University, Faculty of Engineering, Irrigation and Hydraulics
Dept.
2. Director of the Drainage



ABSTRACT
Non-conventional water resources management techniques are becoming
vital to maintain the current levels of water consumption in the Egypt. These
waters are suffering from industrial and domestic wastewater pollution.
The industrial effluent from the Gharbia governorate has been evaluated.
Most of these effluents are disposed into the Gharbia main drain without any
treatment. The drain was closed by a weir to prevent seawater intrusion and use
the excess water for serving the horizontal expansion plan. Recognizing the fact
that the drain serves as a main source of water for reclamation lands,
Environmental Impact Assessment (EIA) has been conducted in order to identify
any expected negative impacts of these effluents and suggest the necessary
mitigation actions to alleviate them.
Moreover, WASP model was calibrated and used to simulate different
scenarios and to study the effect of varying the pollution loads. Predicted values
showed that concentrations exceeded the legislative standards given in the
National Law 48 of 1982. The simulations conducted indicated that primary
treatment is ample to alleviate the detrimental effect of discharged effluents. The
study recommended setting incentives to encourage industries to treat their
effluents prior to discharge.
Key word: pollution- drainage system-WASAP model- impact assessment

680
1.0 INTRODUCTION

Conventional water resources in Egypt are very limited. Egypt relies on the Nile River
as a major source for fresh water and in addition the small amount of scattered seasonal
rainfall in the Mediterranean coastal strip. With the construction of new industrial cities,
agricultural expansion and the daily increasing population conventional water resources in
Egypt will not be enough to fulfill the countrys needs in the near future. Therefore, non-
conventional water resources management techniques are becoming vital to maintain the
current levels of water consumption in the country.
Environmental Assessment is one of the most valuable, and interdisciplinary decision-
making process of evaluation of potential impacts of proposed actions, and its practical
alternatives on physical, biological, cultural and socioeconomics attributes of particular region
(Datta, A. K. 2000). The methodology adopted for Environmental assessment could be shown
as in Fig. (1).


Fig. (1) Graphic description of the methodology applied for the
Environmental Assessment for project


The Environmental Assessment has been applied on the El Gharbia governorate as case study.
This Governorate has many industrial factories most of them disposed their effluent into the
River Nile or to the drainage network without treatment. This receiving water is facing a threat
of being unsuitable for household use due to the huge amount of pollutants being discharged
daily in it. El Gharbia Main Drain system is located in Middle Nile Delta region. It is
considered as one of the largest drainage systems in the Nile Delta. The drain receives
industrial effluents, without any treatment, from different industries located in Tanta and El
Mahala El Kobra and discharges 0.85 BCM annually to the sea. The drain was closed by a
weir at the beginning of 1990 to prevent seawater intrusion and use the excess water for
serving the horizontal expansion plan. Recognizing the fact that the drain serves as a main
source of water for reclamation lands, and noting that both official and non-official reuse are
found along the drain.
Moreover, the numerical model WASP6 produced by EPA (Wool, T.A., et al., 2002)
was calibrated and used as an integral part in the EIA process to simulate different scenarios
and study the effect of varying the pollution loads.

681
2.0 SITE DESCRIPTION

El-Gharbia governorate, is located in the middle of the delta. It extends from Damietta
branch in the east to Rossetta branch in the west. It is bounded on the north by Kafr El-Shiek
governorate and on the south by Al-Mounoufiya governorate. It covers an area of 1943.3 km
2

inhabited by roughly 3 million people. Administratively, the governorate consists of 8 centers
which are (Tanta, EL Mahala EL Kobra, Zefta, Kafr El Zayat, El Santa, Samanoud, Katour,
and Basyoun), 8 cities, 316 villages and 1188 sub-villages (kafr). Tanta, the capital, is located
in the heart of the governorate, 94 km north of Cairo, and 130 km southeast of Alexandria. It
is Egypts fifth largest city and inhabited by almost 335,000 people.
Agriculture is the means of living for most of the people of the governorate. The
cultivated land is approximately 394,900 feddans, covering 85% of the total area. Diversity in
crops makes the governorate productive all over the year, especially at the end of October
when its the time for cotton harvesting. Cultivated crops include cotton, rice, wheat, fruits,
maze and potatoes to mention a few. Not only do people work in agriculture, but a
considerable amount is bound to develop industries in the governorate.
El Gharbia governorate comprises many industrial establishments. The most crucial
sectors are chemical, textile and food sectors. There are 92 chemical industries in the
governorate, 587 textile industries and 224 food industries. This gives us a clear view of how
much the effect of this industry is on the environment. Most industries discharge their
effluents into the water bodies without any treatment. The textile industry discharges 14.6
MCM annually, ranking first, followed by the food industry that discharge 11.4 MCM
annually then the chemical industry that discharge 2.5 MCM annually (NWRP, 2001).
The main source of fresh water is the Nile River. This single source is facing a threat
of being unsuitable for household use due to the huge amount of pollutants being discharged
daily in it. Moreover, no restrictions or legislations are implemented in order to avert the
deterioration of the water quality. Drains receive 22.6 MCM annually of wastewater. This
poses a great threat on the quality of the surface water in drains that it used further
downstream for horizontal expansion. The governmental network receives 3.6 MCM annually
while the Nile River receives 2.2 MCM annually.
Water is used in different industrial sectors for processing, cooling and domestic use.
Processing ranks first and counts for 55.4% of all the discharged quantities, followed by
cooling water systems (40.6%) and finally domestic wastewater (4%). Table (1) lists the
disposal locations of industrial wastewater to the drains.

Table (1): Disposal locations of industrial wastewater to the drains
Center
Name
Factory Name Sector Disposal Location
Tanta Tanta Oil & Soap Co. Meet Hebesh Oil & Soap Mahlet Rouh Drain
Tanta Tanta Oil & Soap Co. Al-Ahlia Oil & Soap Genag Drain
Tanta El Delta Co. for Spinning & Weaving Textile Mahlet Rouh Drain
Kafr El Zayat Alex. For Oil & Soap Co. Oil & Soap Genag Drain
El Mahala Tanta Oil & Soap Co. Oil & Soap Drain No. 5
El Mahala Tanta Oil & Soap Co. Oil & Soap Agricultural Drain
El Mahala Misr Spinning, Weaving & Garments Co. Textile Drain No. 5
El Mahala Nasr Co. for Spinning, Weaving & Dyeing Textile Drain No. 5
(Source: NWRP, 2001)


682

2.1 Gharbia main drain

The drainage network consists of many drains that receive the excess water from agriculture,
industry and municipal run-off and discharge it into main drains or directly into the Nile River.
El Gharbia Main Drain, passing through Tanta and El Mahala El Kobra, collects all the
discharged effluents before it discharges 0.85 BCM annually in the Mediterranean Sea (Abd
El Rashid, 2002). Most of the factories apply primary treatment, yet, a considerable number
still discharge without treatment. In the Gharbia Drain system, all catchments are drained by
pumping Fig. (2). The system consists of eleven pumping stations of which four pumping
stations lift water to irrigation canals. The East Menufeya (MG-1), the Mahalla El Kubra
(MM-l) and the Mahallet Ruh P.S (MG-3) in the upstream part of the system pump their
drainage water almost completely into the irrigation system. The pump stations Samatay
(MG-4), Segaaya (MG-3), No. 5 (MG-5) and No. 6 (MG-7) discharge directly, or via a side
branch, into the Drain. Downstream of the delivery side of P.S. No. 6, the Hamul P.S. (MG-8)
lifts water from the Gharbia Drain into Tira Irrigation System.

Further downstream a side branch conveys the input of the pumping stations No. 3 (MG-10)
and 4 (MG-9) into the Drain. The last drainage pumping station is Hafir Shehab El Din (MG-
12), whereafter the drain continues to the Mediterranean Sea. Both official and non-official
reuse intakes are found along the main part of the Drain. The Gharbia Drain was closed by
weir with four orifices at the beginning of 1990 to prevent the sea water intrusion. Those
orifices are completely opened to the sea for 60 days only and closed for the rest of the year.
Almost all the drainage water was delivered to Wallda canal, El Nile canal, Khashaa fish farm,
Hamul reuse P.S., and unofficially reused by the farmers. Since then, the measurements at
Ezbet El Nile Bridge (MG-15), Bridge No.7 (MG-11) and New Gharbia Outfall were
cancelled (DRI, 2002).





Fig. (2): Schema represents Gharbia Main Drain System

3.0 COLLECTED AND MEASURED DATA

The water quality data was collected at 10 sampling locations over stretch of 70 km length of
the Gharbia main drain. Each sample is collected in plastic container of 1 liter and sent to the
chemical laboratory for BOD, COD, TDS, HM. The DO and pH is measured in the field.

M
E
D
I
T
E
R
R
A
N
E
A
N

S
E
A

P.S. No. 5
(MG-5)
SEGAAYA P.S.
(MG-2)
P.S. No. 6
(MG-7)
P.S. No. 3
(MG-10)
P.S. No. 4
(MG-9)
SHEHAB EL DIN P.S.
(MG-12)
HAMUL P.S.
(MG-8)
SAMATAY P.S.
(MG-4)
(MG-15)
(MG-14)
(MG-11)
GHARBIA WEIR
(MG-13)
(MG-28)
TIRA IRRIGATION SYSTEM
683
The wastewater out-fall being discharged into the drain was from ten factories comprising
three different industrial sectors, namely, food, chemical and textile. Average value of
wastewater flow was found to be 6213.2 m
3
/day. The values of the water quality parameters
are present in Table (2).

Table (2): Values of water quality parameters for different factories discharging into El Gharbia Main
Drain
Name Location Production COD BOD TDS SS OIL HM pH
Delta Textile Tanta Fabrics 220 192 165 40 12 7 9
Misr Textile El Mahala Fabrics 108 64 1200 60 8 1 7.5
Nasr Textile El Mahala Threading 760 480 1600 235 80 0 8
Coca Cola Tanta Beverages 37 37 121 3 0 0 8
Oils & Soap Tanta Oils & Soap 49 27 372 66 188 0 7.3
Tobacco Tanta Tobacco 36 20 420 7 1 0.2 7.5
Oils & Soap El Mahala Animal food 49 27 372 66 188 0 7.3
Misr Ahleia Tanta Animal food 2537 200 1078 2232 1649 0 10
Narobin Tanta Paintings 1020 330 430 22 2 0.8 7.6
Oil & Lint Tanta Paintings, Oil 1324 963 0 0 361 0 7.3


The water quality of the disposal points from the factory is compared by the Law 48
for the disposal of industrial effluent to the drainage water. These comparison are showed
from Figures (3) to (9).
The pH values (Fig. 3) of discharged effluents is ranged from 5 to 9 which are
acceptable. Values above or beyond the normal limits are unacceptable. This is due to the
various washing operations within the factory.
Figure (4) clearly shows that HM doesnt exist in the effluents of most factories
discharging on El-Gharbia Main Drain except the Delta textile is highly above the limit. This
maybe contributed to the excessive use of dyes in the textile industry and the lack of proper
treatment for the wastewater.
Figuer (5) shows Oil and Grease in most cases are extremely higher than the allowable
limits. A closer look to the type of industries gives us a clue to the reason. Textile factories
use dyes excessively, while chemical industries produce soap, oil and paintings with
noticeably high levels of oil in its wastewater effluent. In some cases Oil and Grease can be
seen floating on the surface by naked eye.
Figuer(6) represents SS in the wastewater discharged from different factories is
usually within the normal limits, with only two factories slightly above the normal. However,
in one case, the SS value reached 2232 mg/l which is extremely high compared to the
required limit. This is due to the aggravated amounts of fibers loaded in the wastewater as a
by product of industry. In another case, it reached 235 mg/l due to the excessive use of salts as
a raw material and its production as a by product of industry.
TDS values (Fig. 7) are plotted with respect to the law. Apparently, all values are
beyond the limit due to the fact that TDS can be easily removed from wastewater prior to
discharge.
Fig. (8) represents the BOD concentation. It is clear that wastewater containing oil,
grease and soap residues exert a strong biochemical load on the receiving water. In some
cases the BOD level reached 963 mg/l which is highly above legislative limits. Dyes and
dyeing wastes reportedly contribute to high BOD values. In one case BOD values reached
480 mg/l, while in the other it was 200 mg/l. These high levels of BOD have a detrimental
effect on the water quality and can cause depletion of the DO in the water body.
684
The COD in Figure (9) shows the COD concentration in wastewater of different
industries. As mentioned previously, BOD levels rise due to excessive discharge of
wastewater containing dyeing wastes, oil and grease. COD and BOD are related. When one
rises, the other rises too. This explains the high values of COD in the three factories listed
here. These are the same factories that contributed to high values of BOD.





















Figure (3): pH values for different factories on Gharbia Main Drain





















Figure (4): HM values for different factories on Gharbia Main Drain

7
1
0 0 0
0.2
0 0
0.8
0
0
3
6
9
D
e
l
t
a

T
e
x
t
i
l
e
M
i
s
r

T
e
x
t
i
l
e
N
a
s
r

T
e
x
t
i
l
e
C
o
c
a

C
o
l
a
O
i
l
s

&

S
o
a
p
C
i
g
a
r
e
t
t
e
s

&
T
o
b
a
c
c
o
O
i
l
s

&

S
o
a
p
M
i
s
r

A
h
l
e
i
a
N
a
r
o
b
i
n
O
i
l

&

L
i
n
t
H
M

(
m
g
/
l
)
Law 48
9
7.5
8 8
7.3
7.5
7.3
10
7.6
7.3
0
2
4
6
8
10
12
D
e
l
t
a

T
e
x
t
i
l
e
M
i
s
r

T
e
x
t
i
l
e
N
a
s
r

T
e
x
t
i
l
e
C
o
c
a

C
o
l
a
O
i
l
s

&

S
o
a
p
C
i
g
a
r
e
t
t
e
s

&
T
o
b
a
c
c
o
O
i
l
s

&

S
o
a
p
M
i
s
r

A
h
l
e
i
a
N
a
r
o
b
i
n
O
i
l

&

L
i
n
t
p
H
Max.
Min.
685
























Figure (5): Oil values for different factories on Gharbia Main Drain























Figure (6): SS values for different factories on Gharbia Main Drain

12 8
80
0
188
1
188
1649
2
361
0
200
400
600
800
1000
1200
1400
1600
1800
D
e
l
t
a

T
e
x
t
i
l
e
M
i
s
r

T
e
x
t
i
l
e
N
a
s
r

T
e
x
t
i
l
e
C
o
c
a

C
o
l
a
O
i
l
s

&

S
o
a
p
C
i
g
a
r
e
t
t
e
s

&
T
o
b
a
c
c
o
O
i
l
s

&

S
o
a
p
M
i
s
r

A
h
l
e
i
a
N
a
r
o
b
i
n
O
i
l

&

L
i
n
t
O
i
l

(
m
g
/
l
)
Law 48
40
60
235
3
66
7
66
2232
22
0
0
400
800
1200
1600
2000
2400
D
e
l
t
a

T
e
x
t
i
l
e
M
i
s
r

T
e
x
t
i
l
e
N
a
s
r

T
e
x
t
i
l
e
C
o
c
a

C
o
l
a
O
i
l
s

&

S
o
a
p
C
i
g
a
r
e
t
t
e
s

&
T
o
b
a
c
c
o
O
i
l
s

&

S
o
a
p
M
i
s
r

A
h
l
e
i
a
N
a
r
o
b
i
n
O
i
l

&

L
i
n
t
S
S

(
m
g
/
l
)
Law 48
686























Figure (7): TDS values for different factories on Gharbia Main Drain






















Figure (8): BOD values for different factories on Gharbia Main Drain



165
1200
1600
121
372
420
372
1078
430
0
0
300
600
900
1200
1500
1800
2100
D
e
l
t
a

T
e
x
t
i
l
e
M
i
s
r

T
e
x
t
i
l
e
N
a
s
r

T
e
x
t
i
l
e
C
o
c
a

C
o
l
a
O
i
l
s

&

S
o
a
p
C
i
g
a
r
e
t
t
e
s

&
T
o
b
a
c
c
o
O
i
l
s

&

S
o
a
p
M
i
s
r

A
h
l
e
i
a
N
a
r
o
b
i
n
O
i
l

&

L
i
n
t
T
D
S
(
m
g
/
l
)
Law 48
192
64
480
37
27 20 27
200
330
963
0
200
400
600
800
1000
1200
D
e
l
t
a

T
e
x
t
i
l
e
M
i
s
r

T
e
x
t
i
l
e
N
a
s
r

T
e
x
t
i
l
e
C
o
c
a

C
o
l
a
O
i
l
s

&

S
o
a
p
C
i
g
a
r
e
t
t
e
s

&
T
o
b
a
c
c
o
O
i
l
s

&

S
o
a
p
M
i
s
r

A
h
l
e
i
a
N
a
r
o
b
i
n
O
i
l

&

L
i
n
t
B
O
D

(
m
g
/
l
)
Law 48
687

























Figure (9): COD values for different factories on Gharbia Main Drain


4.0 WASP Model

WASP is a water quality program used by United States Environmental Protection Agency
(USEPA). It is a generalized modeling framework that is used to simulate the movement and
interaction of pollutants within the water.

WASP is supplied with two kinetic sub-models to simulate two of the major classes of water
quality problems; conventional pollution involving dissolved oxygen, biochemical oxygen
demand, nutrients and eutrophication and toxic pollution involving organic chemicals, metals, and
sediment.

4.1 Model Implementation

4.1.1Hydrodynamic model

Discharge, velocity, and cross sectional details at different locations of the stream are
the first data to be collected. There are several options for specifying water velocity and depth
to WASP6. Depth and velocity can be held constant by entering their values in the Depth and
Velocity multiplier field and setting the exponent to zero. One may also allow depth and
velocity to vary as a function of flow.
b
aQ v = (1)
220
108
760
37 49 36 49
2537
1020
1324
0
600
1200
1800
2400
3000
D
e
l
t
a

T
e
x
t
i
l
e
M
i
s
r

T
e
x
t
i
l
e
N
a
s
r

T
e
x
t
i
l
e
C
o
c
a

C
o
l
a
O
i
l
s

&

S
o
a
p
C
i
g
a
r
e
t
t
e
s

&
T
o
b
a
c
c
o
O
i
l
s

&

S
o
a
p
M
i
s
r

A
h
l
e
i
a
N
a
r
o
b
i
n
O
i
l

&

L
i
n
t
C
O
D

(
m
g
/
l
)
Law 48
688
Where;
V =Water velocity within the segment (m/s)
a =Hydrologic coefficient of velocity (value from input file)
Q =Discharge from segment (cms)
b =Hydrologic exponent for velocity (value from input file)

A relation is sketched between v and Q on log-log scale and the coefficients a and b are
calculated as follows:

=
2
1
2
1
ln
ln
Q
Q
v
v
b (2)

b
Q
v
a
1
1
=
(3)

Where v
1
, Q
1
and v
2
, Q
2
are the velocities and discharges between upstream and down
stream respectively.
4.1.2 Water quality model

Dissolved oxygen (DO) is one of the most important variables in water quality
analysis. Low concentrations directly affect fish and alter a healthy ecological balance.
Because DO is affected by many other water quality parameters, it is a sensitive indicator of
the health of the aquatic system. The Streeter-Phelps Oxygen Model is used for DO
calibration.
Using in first instance, a non-dispersive flow system (river) and only the components
re-aeration and aerobic BOD removal (i.e. without ammonia), we can make a simple oxygen
balance model for e.g. a river subject to a single point source organic waste pollutant input
(e.g. municipal sewage or waste). This simple model is also known as the Streeter-Phelps
oxygen balance model for streams (Chapra, 1997).
The basic differential equation for the oxygen deficit can be written as:
D K e L K
dt
dD
a
t K
o d
r
=

(4)
Where;
D =Oxygen deficit, (mg/l)
K
d
=BOD removal rate, (day
-1
)
L
o
=Initial BOD concentration in (kg/m
3
) or (mg/l) in the river at the outfall point of the
pollution source
K
r
=BOD removal rate in the river, in (day
-1
)
K
a
=Re-aeration rate in (day
-1
);

Equation (4) represents a non-homogeneous, ordinary linear first order differential
equation and expresses the balance between transport of oxygen via the river velocity
689
(advection), BOD uptake by micro-organisms for removal of the organic pollution and re-
aeration with the atmosphere. The solution of the Streeter-Phelps oxygen balance model of a
river is,

( )
t K
j
t K t K
r a
o d a a r
e D e e
K K
L K
t D

+

= ) ( (5)


4.2 Model calibration

Water quality of Gharbia Main Drain is very important due to the wide use of its water
to supplement the deficit in irrigation water in the governorate. A model for El Gharbia Main
Drain was set to describe and analyze the water quality of the drain. Discharge, velocity, and
cross sectional details at different locations of the drain were first collected. Afterwards, the
drain was divided into 16 surface segments. The geometric properties of each segment were
then determined. Water quality data was collected at 10 sampling locations over a stretch of
70 Km. Average values of DO varied from 2 - 12 mg/l. Calibration of the hydrodynamic part
of the model was first carried out using data of year 1999/2000. Fig. (10) and Fig. (11) show
the measured discharge versus the observed one and predicted results closely matches at two
different locations. Figure (11) shows that there is difference between the measured and the
predicted at MG 15 starting from March to September this is due to the unofficial re-use from
the drain. After calibration the values of coefficient a in Equ. (2) range from 0.0053 to
0.0081 while that of b in Equ. (4) ranges from 0.975 to 1.57.
Calibration of DO for the Gharbia drain is showed in Fig. (12). The calibration of the
process compartment of the model was carried out sequentially by using transformation
kinetic parameters. Values of DO appear to vary slightly from observed data. The value of the
aeration rate of Gharbia drain is found to be equal to 0.507 day
-1
.





















Figure (10): Predicted Flow versus Observed values at MG 28 (cms) (1999/2000)
0
1
2
3
4
5
6
7
8
9
10
A
u
g
u
s
t
S
e
p
t
e
m
b
e
r
O
c
t
o
b
e
r
N
o
v
e
m
b
e
r
D
e
c
e
m
b
e
r
J
a
n
u
a
r
y
F
e
b
r
u
a
r
y
M
a
r
c
h
A
p
r
i
l
M
a
y
J
u
n
e
J
u
l
y
Q

(
c
m
s
)
predicted
observed
690











Figure (11): Predicted Flow versus Observed values at MG 15 (cms) (1999/2000)
0.00
2.00
4.00
6.00
A
u
g
u
s
t
S
e
p
t
e
m
b
e
r
O
c
t
o
b
e
r
N
o
v
e
m
b
e
r
D
e
c
e
m
b
e
r
J
a
n
u
a
r
y
F
e
b
r
u
a
r
y
M
a
r
c
h
A
p
r
i
l
M
a
y
J
u
n
e
J
u
l
y
Month
D
O

m
g
/
l
predicted
observed



Figure (12): Predicted DO (mg/l) versus Observed values at MG 15 (1999/2000)

0.00
5.00
10.00
15.00
20.00
25.00
A
u
g
u
s
t
S
e
p
t
e
m
b
e
r
O
c
t
o
b
e
r
N
o
v
e
m
b
e
r
D
e
c
e
m
b
e
r
J
a
n
u
a
r
y
F
e
b
r
u
a
r
y
M
a
r
c
h
A
p
r
i
l
M
a
y
J
u
n
e
J
u
l
y
Q

(
c
m
s
)
predicted
observed
691
4.3 Water Quality Control Scenarios

After calibration water quality control scenarios are simulated using the pre-calibrated model.
The main objective of this simulation is to test and evaluate the different scenarios for
improving the water quality of El Gharbia Main Drain. Two main category of treatment are
performed for each factory. The first one by is primary treatment which means by
sedimentation process. The second category is secondary treatment which is biologically
treated by aeration. For each category different scenarios are performed to choose the best
treatment methodology for each factory. The scenarios depend on reducing the BOD
concentration and the WASP6 model calculate the DO concentration after it was calibrated.

First: Primary Treatment for wastewater
Primary treatment allows reduction of BOD of (40-50)% adopted from Hof, A. and Schuurmans,
W. (2000). Three scenarios are simulated allowing a reduction of BOD of 50%.
Scenario 1 Reduction of BOD of El Mahala El Kobra wastewater only
Scenario 2 Reduction of BOD of Tanta wastewater only
Scenario 3 Reduction of BOD of both Tanta and El Mahala El Kobra wastewaters.

Figure (13) compares between the different scenarios for primary treatment for Gharbia drain. It is
obvious that Scenario 1 indicated an increase of 4.3% in DO concentrations due to primary
treatment of El Mahala El Kobra wastewater. Scenario 2 indicated an increase of about 44% in
DO concentrations due to primary treatment of Tanta wastewater. Scenario 3 indicated an increase
of about 46% in DO concentrations due to primary treatment of all wastewater which is preferred.

Second: Secondary Treatment for wastewater
Secondary treatment allows a reduction of BOD of (85-95)%. Three scenarios are simulated
allowing a reduction of BOD of 90%.
Scenario 4 Reduction of BOD of El Mahala El Kobra wastewater only
Scenario 5 Reduction of BOD of Tanta wastewater only
Scenario 6 Reduction of BOD of both Tanta and El Mahala El Kobra wastewaters

Figure (14) compares between the different scenarios for secondary treatment for Gharbia drain.
Scenario 4 indicated an increase of 6% in DO concentrations due to secondary treatment of El
Mahala El Kobra wastewater. Scenario 5 indicated an increase of about 78% in DO
concentrations due to secondary treatment of Tanta wastewater. Scenario 6 indicated an increase
of about 82% in DO concentrations due to secondary treatment of all wastewater.









692
Month
2.00
3.00
4.00
5.00
6.00
7.00
8.00
A
u
g
u
s
t
S
e
p
t
e
m
b
e
r
O
c
t
o
b
e
r
N
o
v
e
m
b
e
r
D
e
c
e
m
b
e
r
J
a
n
u
a
r
y
F
e
b
r
u
a
r
y
M
a
r
c
h
A
p
r
i
l
M
a
y
J
u
n
e
J
u
l
y
D
O

(
m
g
/
l
)
Scenario 1
Scenario 2
Scenario 3
Without Treatment


Fig. (13) DO concentration before and after primary treatment at MG15

0.00
2.00
4.00
6.00
8.00
10.00
123456789
1
0
1
1
1
2
Time
D
O

(
m
g
/
l
)
Scenario 4
Scenario 5
Scenario 6
Without Treatment

Fig. (14) DO concentration before and after secondary treatment at MG15

693

CONCLUSIONS AND RECOMMENDATIONS

Unofficial re-use is present along El Gharbia Main Drain, hindering a proper
monitoring system for the drain.
The drain is receiving massive pollution loads from the different industries along the
drain, depleting its DO concentrations and setting it unsuitable for use although it is
used in as water resource fro the expansion projects.
Wasp model is calibrated and used for simulating Gharbia main drain successfully.
Different scenarios for wastewater treatment process are performed to minimize the
pollution
It is found that Primary treatment is ample to set the DO concentrations above the
standard regulatory limits given in the National Law 48 of 1982. and secondary
treatment can raise DO concentrations by 82%.
Also it is recommended to that the different industrial sectors should be encouraged to
recycle cooling water and processing wastewater whenever feasible.
A system of incentives has to be developed to encourage the different industrial
sectors to minimize emission of pollutants and to treat their effluents prior to
discharge.

REFERENCES

Abdel Rasheed A., (2002), Water Quality Control In Open Drains and its effect on Soil
Properties, Unpublished Ph.D. Thesis, Faculty of Agriculture, Ain Shams University
Chapra S.C., (1997), Surface Water Quality Modeling, McGraw-Hill
Covar A.P., (1976), Selecting the Proper Rearation Coefficient for Use in Water Quality
Models. Presented at the USEPA Conference on Environmental Simulation Modeling,
April 19-22, 1976. Cincinnati, Ohio.
Datta, K. A. (2000). "Introduction to Environmental Science and Engineering". Oxford & Ibh
Publishing Co. PV. LTD. New Delhi.
DRI, (2002), Drainage Water Status in the Nile Delta, Yearbook 1999/2000, Technical Report
No.64
Hof, A. and Schuurmans, W. (2000)"Water Quality Control in Open Channels". Water
Science and Technology J ournal, Vol. 42 No. 1-2, pp 153-159
NWRP, (2001), Industrial water and wastewater, EGYPT Needs Assessment. National Water
Resources Plan, Ministry of Water Resources and Irrigation (MWRI).
Wool T.A., et al., (2002), Water Analysis Simulation Program, WASP-Version 6.0,Users
Manual, US EPA





694

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


BEST MANAGEMENT PRACTICES FOR THE DRAINAGE
WATER REUSE

A. El-Hawary
Researcher, Drainage Research Institute, National Water Research Center, Cairo, Egypt

ABSTRACT

The objective of the study is to evaluate and suggest the Best Management
Practices for the drainage water reuse in irrigation under the Egyptian conditions.

The simulation model DRAINMOD was used to study different irrigation
management practices when drainage water is used in irrigation. The simulations
tested the impact of irrigation water quantity, the irrigation interval and the role of
agricultural drainage. The simulations tested the impact of different practices on the
soil salinity and the crop yield.

The simulations were conducted for four quantities of irrigation water and three
irrigation intervals. Also, the need for the agricultural drainage was tested. The
resulted soil salinity and crop yield were compared to select the best irrigation
management practices that minimize the resulted soil salinity and save the crop
yield.

The simulations results show that increase of irrigation water will prevent
accumulation of pollutants in the soil and keep soil and crops in good conditions.
Also, reducing the irrigation interval is suitable for irrigation using drainage water.
The need for good agricultural drainage is particularly important when drainage
water is used in irrigation.

Key words: water reuse, irrigation, crop, soil, pollution, management
695


INTRODUCTION

Drainage water reuse for irrigation may be hazardous to the environment, since the irrigation
water contains pollutants such as salinity, pathogens, trace and heavy metals, nutrients and
pesticides. In addition, use of marginal-quality water has the potential of causing serious
problems of soil degradation and reduction in crop productivity. Other problems such as human
health hazards and quality degradation of groundwater are also possible.

The safe use of the drainage water in irrigation needs to apply appropriate management to reduce
the negative impacts. The objective of the recommended management practices involve
managing, limiting, or removing pollutants such as salinity, nitrates, phosphates, sediments,
pesticides, pathogens, heavy metals and other chemicals.

The simulation model, DRAINMOD, has been used successfully in many areas for predicting the
effect of water management systems on crop yield and watershed hydrology over large period of
climatological record (Skaggs et al. 1982; Ravelo et al. 1982). The model was first developed in
the early 1970s by Skaggs (Skaggs, 1978). Several validation studies for the model that used
field data have been done by several researchers (Skaggs 1980, 1982; Chang et al. 1983).
Kandil, et al. (1992) modified DRAINMOD to include the effect of irrigation water quality on
soil salinity. The modified version, DRAINMOD-S, is capable of predicting the long-term
effects of stresses due to soil salinity stresses on crop yields in humid and arid irrigated cropland
(Kandil et al 1993). The reliability of the modified version DRAINMOD-S was evaluated using
data collected from the Zankalon Pilot Area (ZPA), in Egypt. The evaluation was performed by
comparing the observed and simulated salt concentration in the soil profile and other
hydrological parameters (Kandil et al. 1992).

Oster (1994) mentioned that using poor quality waters requires two changes: (1) suitable crop
rotation where salt-tolerant crops could be included; and (2) improvement in water management
system especially in irrigation and drainage. Using saline water of 1000 ppm for irrigation, soil
salinity and crop yield was simulated for different water management systems. Water
management changes include the irrigation water quantity and the frequency of irrigation
(irrigation interval). Also, the impact of the drainage will be studied as recommended by
Shalhevet (1994).

The objective of this study is to specify the best management practices for saline water irrigation.
Saline water will be applied for crop irrigation instead of fresh water. In all cases the simulation
period was 10 years.
MATERIALS AND METHODS
The Simulation Model DRAINMOD-S
The model is based on a water balance for a section of soil of unit surface area that extends from
the impermeable layer to the surface. The water balance method is an alternate approach of
characterizing flow in drained soils. Given an initial moisture content distribution, a volumetric
696


input of water, and an evapotranspiration rate, then a drainage rate, a water table elevation, and a
volumetric moisture distribution can be computed by making an appropriate set of assumptions.
The water balance for a time increment of t may be expressed as:

V
a
=D +ET +DS - F [1]

where V
a
is the change in the air volume or water free pore space (cm) in the section, D is
drainage (cm) from (or sub-irrigation into) this section, ET is evapotranspiration (cm), DS is
deep seepage (cm), and F is infiltration (cm) entering the section in t.

The terms on the right hand side of Equation [1] are computed as functions of the water table
elevation, soil water content, soil properties, site and drainage system parameters, crop and stage
of growth, and atmospheric conditions. The amount of runoff, and storage on the surface is
computed from a water balance at the soil surface for each time increment

P =F +S +RO [2]

where P is the precipitation (cm), S is the change in volume of water stored on the surface (cm),
and RO is runoff (cm) during t.

DRAINMOD-S uses the standard deterministic approach for modeling solute transport in the
unsaturated zone based on the adjective-dispersive equation. For one-dimensional system
comprised of non-reactive solutes, the equation takes the following form:

( )

C
t z
D
C
z
q
C
z
c
hz z
=

+
[3]

where C is the volume-average solute concentration, is the volumetric water content, it is the
time, D
hz
is the hydrodynamic dispersion coefficient in the vertical direction, q
z
is the flux and
c represents the solute source/sink term used to represent additional processes
(precipitation/dissolution).

The general crop response model used in DRAINMOD-S can be written as:

RY =RY
w
* RY
d
* RY
p
* RY
s
[4]

where RY is the overall relative yield for a given year. RY
w
, RY
d
, RY
p
, RY
s
are the relative
yield that would be obtained if only wet stress (excessive soil water), drought stress (shortage of
water), planting delay stress, and salinity stress occurred respectively. The overall relative yield,
(RY =Y/Yo) is the ratio of the measured or observed yield for a given year (Y) to the long-term
average yield that would result from a combination of adequate irrigation, efficient drainage,
favorable root zone salinity and good trafficability so that planting and other farm operations are
done in time.
697


Simulation Area

The suggested best management practices of drainage water irrigation will be developed using
the water management simulation model DRAINMOD. The model will run using data from the
Zankalon Pilot Area (ZPA) in Sharkia Governorate, Egypt. The ZPA is located 84 km north east
of Cairo, Egypt. The experimental field has a total area of 3.78 ha. It is relatively flat with
average elevation of 9.5 m above mean sea level. ZPA is divided into three units (1 - 3) with five
treatments each (A - E). The water management system consists of lined irrigation canals and
tile drains spaced 20 m apart, buried approximately 1 m deep. Each canal is provided with gates
to control the amount of irrigation which is applied by the surface basin method. Drainage
laterals discharge in sub-collectors which in turn empty into a collector leading to an open drain.
The experimental site is provided with facilities for measuring the rate of subsurface drainage,
water table depths and irrigation amounts.

The soil of ZPA consists of heavy clays down to a relatively great depth. The hydraulic
conductivity was measured in 2.0 m deep holes using the auger hole method. The average
hydraulic conductivity is 0.05 m/day, the average bulk density is 1.32 gm/cm
3
and the average
porosity is 50.5%. The mechanical analysis for the soil shows that it consists of 46.0% clay,
39.0% silt and nearly 15.0% sand (Kandil et al 1993). Climatological data were obtained from
the weather station installed in the ZPA (rainfall, pan evaporation, wind speed, and solar
radiation). Hydrological parameters measured on the area were, volume of irrigation and
drainage water, drainage water salinity and water table depths. Soil salinity measured down to
150 cm below the soil surface at 15 cm increments.

Input Data for DRAINMOD-S
Climatological data
Using DRAINMOD-S to simulate on-farm water management system requires daily
evapotranspiration and hourly precipitation as a weather data. The weather data of two years
were used in the simulation and repeated to cover the whole simulation period. Precipitation in
ZPA is rare as verified by the precipitation records of the weather station installed at the area. In
the simulations, irrigation water was considered as precipitation and entered into the model as
hourly precipitation. Potential evapotranspiration (PET) depends on climatological factors which
include net radiation, temperature, humidity, and wind velocity. Daily PET is computed using the
Modified Penman Equation.

Soil

The lateral saturated hydraulic conductivity (K) was determined in the field with the auger hole
method and geometric average is shown in Table 1.



698


Table 1 Summary of the Zankalon Pilot Area Soil Properties
Parameter Value
Depth to restricting layer
Average Hydraulic conductivity
Volumetric water content at lower limit (Wilting point)
The critical water depth for the wet and drought stresses calculations
Initial Water Table Depth
160 cm
0.5 cm/hr
0.3
30 cm
85 cm

Soil water characteristics for 10 cm increments of the soil profile down to the drain depth were
obtained in laboratory using undisturbed soil cores.
Coefficients of the Green-Ampt infiltration equation and maximum rate of upward water
movement as a function of water table depth were determined from the lateral saturated hydraulic
conductivity and soil water characteristic relationships using the SOILPREP program of
DRAINMOD (El-Refaie, 1994).

Soil samples were taken for chemical analysis to determine the initial soil salinity. The samples
were taken at 15 cm increment of the soil profile down to 150 cm below the soil surface.

Crop
The suggested crops for the study area are cotton in the summer and berseem in the winter.
Cotton represents the salt tolerant crops and berseem is a moderately salt sensitive crop. This
crop rotation takes one year to be completed (Figure1).
60 75 270 300 60 75
270
300
COTTON BERSEEM BERSEEM
day of crop season end day of crop season begin
Schematic diagram of the simulated crop rotation.
BERSEEM COTTON

Figure 1 Schematic diagram of the simulated crop rotation

699



DRAINMOD can simulate crop rotation comprising more than one crop. The input data of each
crop consists of the effective root depth and the major dates of planting, harvesting and stresses
counting. The change in the effective root depth with time was estimated by crop growth stage
coefficients K
c
and given by Kandil et al 1992.The effective root depths as a function of days
after planting are presented for both crops in Table 2. A value of 3 cm was used for the fallow
periods to reflect the soil depth from which water will be evaporated in the absence of a crop.
Parameters used for evaluating the yield using of each crop are given in Table 3.

Table 2 Root Depths of Crops Included in the Simulated Crop Rotation.
Berseem Cotton
Dates Depth (cm) Dates Depth (cm)
01/01
03/02
20/02
01/10
31/10
15/11
30/11
18/12
31/12
25.6
30
3
3
3
14.2
18.9
22.3
25.6
01/01
01/04
20/04
19/06
18/07
18/08
02/09
10/09
31/12
3
3
3
15
25
33
35
36
3

Table 3 Crop yield functions Parameters used for evaluating the yields of different crops.
Item Berseem Cotton
Soil salinity threshold (mg/1)
Slope for higher salinity values (% per mg/l)
Relative yield without deficit stress (%)
Slope for yield versus deficit stress-day-index
Relative yield without wet stress (%)
Slope for yield versus wet stress-day-index
960
0.01
100
1.22
100
0.55
6160
0.00650
100
1.22
108
0.65

Irrigation

The simulated area is irrigated by basin irrigation. The simulated irrigation water salinity is 1000
ppm. The irrigation water quantity for berseem is 1134 m3/feddan with an irrigation interval of
30 days while cotton is irrigated by 3696 m3/feddan and irrigation interval of 15 days as shown
in Table 4.




700


Table 4 Irrigation Water Quantities of the Simulated Crops
Berseem Cotton
Date Irrigation Amount (cm) Date Irrigation Amount (cm)
1/11
01/12
01/01
01/02

6.75
6.75
6.75
6.75

15/03
01/04
15/04
01/05
15/05
01/06
15/06
01/07
15/07
01/08
15/08
01/09
7.33
7.33
7.33
7.33
7.33
7.33
7.33
7.33
7.33
7.33
7.33
7.33
Total 27 Total 88

Agricultural drainage
Agricultural drainage is very important when the saline water is used in irrigation. The role of the
drainage in saving the soil and crop yield was shown by simulating the area without the drainage.
A summary of the simulated drainage system input parameters is given in Table 5.

Table 5 Summary of the Simulated Drainage System Input Parameters.
Parameter Value
Drain spacing
Drain depth
Maximum depth of surface storage
Drain radius
Effective drain radius
Drain pipe capacity
20 m
1.0 m
20 mm
100 mm
40 mm
2.5 cm/day

SIMULATIONS AND RESULTS
Irrigation
Different suggested changes in the irrigation practices were input to the model to simulate them.
Those changes are in the quantity and schedule of irrigation water.

In this study cotton and berseem will be irrigated using saline water having a salinity of 1000
ppm. Two irrigation intervals will be used separately in the simulations. First irrigation interval
is, 15 days for cotton and 30 days for berseem. Second irrigation interval is 10 days for each
crop. Also, the quantity of the irrigation water was increased by 10, 15, 25, 35 and 50% for soil
leaching.
701


Irrigation water quantity (leaching requirements)

The recommended amounts of irrigation water under the Egyptian conditions for cotton and
berseem are 3696 m
3
/fed and 1134 m
3
/fed respectively. This quantity is recommended when
fresh water (300 ppm) is used for irrigation. More water is obviously needed for leaching when
saline water is used for irrigation in order to prevent salinization of soil profile.

DRAINMOD simulated the irrigation of cotton and berseem applying different leaching
fractions. The applied leaching fractions are ranged from zero to 50% of the original crop water
duties. From the results, applying a leaching ratio between 25% and 35% will mitigate the soil
salinization without loss of much water for soil leaching (Fig. 2).

The results show that cotton yield was slightly affected by the change in the volume of leaching
water while berseem yield is highly affected by the increase in the irrigation water. The excessive
water stress condition decreases the yield of berseem and accelerates the soil salinization.
However, the use of extra water for leaching purpose is needed to prevent salt accumulation
which affects salt sensitive crops in the rotation. This extra irrigation water can be harmful to the
crops in the same crop rotation as shown for berseem yield (Fig. 3) unless narrow drain spacing
is used.
0
20
40
60
80
100
120
140
160
0 1000 2000 3000 4000 5000 6000
soil salinity ppm


s
o
i
l

d
e
p
t
h

c
m
initial soil salinity
L.F=0 %
L.F=15%
L.F=25%
L.F=35%
L.F=50%

Figure 2 Simulation results for soil salinity after 10 years of cotton and berseem irrigation using
saline water (1000 ppm) and different leaching fractions
702


0
10
20
30
40
50
60
70
80
90
100
BERSEEM COTTON
crop
R
e
l
a
t
i
v
e

c
r
o
p

y
i
e
l
d

L. F. =0
L. F. =15%
L. F. =50%

Figure 3 Simulation results for the crop yield of cotton and berseem using different leaching
fraction of irrigation water

Irrigation interval (irrigation frequency)

The length of irrigation interval is critical when saline water is used in irrigation. Frequent
irrigation prevents the salinization of the root zone as the upward flux from the water table
decreased. Results of simulations show that short irrigation intervals decrease the salinity
accumulation (Fig. 4). Reducing the irrigation interval, i.e., increasing the number of irrigations
is suitable for irrigation using saline water but this practice should be practical and economical.
The irrigation interval of cotton will be reduced from 15 days to 10 days while the irrigation
interval of berseem will be reduced from 30 days to 20 days.
703


0
20
40
60
80
100
120
140
160
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
soil salinity ppm


s
o
i
l

d
e
p
t
h

c
m
initial soil salinity
Low Frequency
High Frequency

Figure 4 Simulation results for soil salinity after 10 years of cotton and berseem irrigation using
saline water (1000ppm) and different Irrigation Frequency
Agricultural Drainage

The need for an effective agricultural drainage is particularly important when salt water is used in
irrigation. Leaching water for soil will increase, so, drainage intensity should increase. Narrow
drain spacing helps in leaching excessive salt from the root zone especially when saline water is
used for irrigation. On the other hand, wide drain spacing may cause wet stresses due to water
logging and salinity stresses can be expected.

DRAINMOD evaluated the change in soil salinity without agricultural drainage. The resulted soil
salinity, compared with the soil salinity with drainage, is significantly high especially in the top
of the soil (fig 5).
704


0
20
40
60
80
100
120
140
160
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
soil salinity ppm


s
o
i
l

d
e
p
t
h

c
m
initial soil salinity
with drainage
without drainage
Figure 5 Simulation results for soil salinity after 10 years of cotton and berseem irrigation using
saline water (1000ppm) with and without subsurface drainage

SUMMARY AND CONCLUSIONS
The principal objective of this research was to estimate the best water management system for
using saline water in irrigation. The simulation model DRAINMOD was used to study different
irrigation management practices that can be apply with drainage water irrigation. The objective
of the simulations is to predicate the impact of the tested irrigation practices on the soil salinity
and the crop yield.

Data collected from the experimental site ZPA, together with data available from other sources in
the literature formed the basis for model inputs. Under the same climatic conditions and soil type
and quality; the suggested one-year crop rotation was tested. The suggested crop rotation
includes berseem and cotton.

The simulations were conducted for four quantities of irrigation water and three irrigation
intervals. Also, the need for the agricultural drainage was tested. The resulted soil salinity and
crop yield were compared to select the best irrigation management practices for the drainage
water reuse in irrigation under the Egyptian conditions. Hence, the selected irrigation practices
will minimize the resulted soil salinity and save the crop yield.



705


The results of the simulations led to the following conclusions for the irrigation of berseem and
cotton using saline water:

- Increasing of irrigation water with the ratio between 25% and 35% will prevent
accumulation of salinity in the soil and keep soil and crops in good conditions.

- Reducing the irrigation interval, i.e., increasing the number of irrigations is suitable for
irrigation using drainage water. The irrigation interval of cotton will be reduced from 15
days to 10 days while the irrigation interval of berseem will be reduced from 30 days to
20 days.

- The need for good agricultural drainage is particularly important when drainage water is
used in irrigation. The leaching requirements for soil will increase so drain spacing will
decrease.
REFERENCES
Chang, A.C., R.W. Skaggs, L.F. Hermsmeier and W.R. J ohnston. 1983. Evaluation of a Water
Management Model for Irrigated Agriculture. TRANSACTIONS of the ASAE 26(2):
412-418, 422.
Kandil, H.M, R.W. Skaggs, S.M. Abdel Dayem, Y. Aiad and J .W. Gilliam. 1993. DRAINMOD-
S: Water Management Model for Irrigated Arid Lands, 2. Crop Yield and Applications.
Workshop on Subsurface Drainage Simulation Model ICID September 1993. Holland.
Kandil, H.M., Miller, C.T., and Skaggs, R.W. 1992. "Modeling long-term solute transport in
drained, unsaturated zones", Water Resources Research, Vol. 28, 2797-2809.
Kandil, H.M., R.W. Skaggs, S.M. Abdel Dayem, Y. Aiad and J .W. Gilliam. 1992.
DRAINMOD-S: Water Management Model for Irrigated Arid Lands, 1. Theories and
Tests. ASAE Paper No 922566, ASAE, St. J oseph, MI 49085.
Oster, J . D. 1994. Irrigation with Poor Quality Water. Agricultural Water Management 25 (1994)
pp 233-269.
Ravelo, J . L. Reddel, A. Hiler and R. W. Skaggs. 1982. Incorporating Crop Needs into Drainage
System Design. Transactions ASAE 78 (2540):623-629.
Shalhevet, J.1994. Using Water of Marginal Quality for Production: Major Issues. Agricultural
Water Management 25 (1994) pp 233-269.
Skaggs, R. W., S. Hardjoamidjojo, E. H. Wiser and E. A. Hiler. 1982. Simulation of Crop
Response to Surface and Subsurface Drainage Systems. Transactions ASAE 79 (2555):
1673-1678.
Skaggs, R.W. 1978. "A water management model for shallow water table soils", Water
Resources Research Institute, University of North Carolina Report No. 134, Raleigh, NC.
Skaggs, R.W. 1980. DRAINMOD-S reference report; Methods for design and evaluation of
drainage-water management systems for soils with high water tables. USDA-SCS, South
National Technical Center, Fort Warth, TX. 330p.
Skaggs, R.W. 1982. Field evaluation of a water management simulation model. Transactions of
the ASAE 25 (3): 666 - 674.

706

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005








STATUS OF WETLANDS IN TURKEY AND SEYFE
LAKE ECOLOGICAL PRESERVATION PROJECT


M. Sait TAHMSCOLU
General Directorate of State Hydraulic Works (DS)
Environmental Division Manager
06100 Ycetepe / Ankara / TURKEY
Fax: +90 (312) 417 13 78
E-mail: msaitt@yahoo.com

Prof. Dr. Veysel EROLU

General Director
State Hydraulic Works
06100 Ycetepe / Ankara / TURKEY


ABSTRACT
Malaria epidemic spreading fast all over the country was resulted in
drying all the wetlands 30 years ago; however, this practice as we all witnessed
threatened the existence of the endangered spices.
Lake Seyfe, one of the most important wetlands of Turkey, has been taken
under protection by the Ministry of Environment and Forestry and given the
status of Nature Preservation Area status which is one of the strictest status
exist. In addition, this area was declared to be the first degree Natural Sit Area
by Preservation of Cultural and Natural Resources Commission and mentioned
to be one of the five areas to be preserved by the authorities according to the
international RAMSAR agreement signed by Turkey.
In 1987, Seyfe Mucur Basin Reclamation Planning Report covering
23 226 ha of land was prepared. In 07.20.1990, with the initiatives of
Environment Undersecretary and other institutions, DS was given the authority
to develop physical reports. With the approval of feasibility report prepared by
DS, Seyfe Ecology Preservation Project was put in to the investment plan with
30 million USD first appraisals according to year 1990 consumer price index
and the very same year the work was started. With the completion of the project,
707

15 792 ha of land will have been reclaimed to farming and some endangered
bird population will have been protected.

KEYWORDS
Wetland, Seyfe Lake, Ecological preservation, DS (General Directorate of State Hydraulic
Works)

1- WETLANDS IN TURKEY

Parallel to the global destruction and devastation of wetlands, a total wetland area of about
200 000 ha has been dried up for the purpose of preventing the disease of malaria and gaining
land and area for agriculture and settlement; it has been damaged and devastated as a result of
industrial pollution, illegal hunting and uncontrolled reed cutting.
At present, Turkey is still the richest country in Europe and the Middle East in view of
wetlands; the country has presently about 1000 wetlands having a total area of more than 3
million hectares. It has been established in the light of international criteria as a result of
assessments by the Ministry of Environment and Forestry that 200 of these wetlands have
international significance.
The underlying reason for the fact that the wetlands in the country are of international
significance is that the two most important ones of the bird migration routes in the Western
Palearctic Region pass via Turkey. By means of the migration route of oruh Valley which
enters Turkey from the Eastern Black Sea Region, more than 200 000 predatory birds fly over
the River of oruh, spreading to the wetlands in the Eastern Anatolia Region. This migration
over Turkey is the biggest migration by the predatory birds in the Western Palearctic Region.
The migration route of Boazii, which enters the country via Thrace in Western Black
Sea and then passes to Anatolia via the Strait, becomes a stage for the eye-catching crossings
of more than 250 000 strokes in groups of 200 to 700. Therefore, the existence of the
wetlands in Turkey bears much more importance so that the birds can have this long journey
safely.
Following the realization of the significance of the wetlands by the country like the global
cases, Turkey became a signatory for RAMSAR Convention on May 17, 1994.
9 wetlands including Seyfe Lake has been included in Ramsar List so far. Parallel to a
decision adopted by the National Wetlands Commission in 2004, studies have been initiated
to add 20 more wetlands in Ramsar List. Turkey has been working with great efforts to
preserve, develop and restore all the wetlands and ensure logical utilization by means of
studies aiming at ensuring enactment of legislation and formation of an inventory and data
base with regard to the preservation of wetlands, promoting awareness building and providing
training for capacity expansion as regards the issues concerning wetlands and preparation and
implementation of management plans for wetlands. It also hosts many national and
international meetings addressing the issues related to wetlands.

2- SEYFE LAKE

Coordinates : 39 12 North
34 25 East
Elevation : 1110 meters
Covering area : 10700 hectares

708

It is located in the tectonic depression north-east of the Krehir province in the Inner
Anatolia. Administratively, it is within the border of the Mucur district of the Krehir
province. It is named after the Seyfe village west of it.
It is 220 km from Ankara, and 30 km from Krehir province.
The lake, which is at the southern tip of the 152.200 hectares Seyfe closed reservoir, has
occurred as a result of post Oligosen faults developed in various directions. The lake, which is
at the base layer (1110 meters) of the reservoir, is also natural drainage area of the reservoir.
It is bordered by Glkaya Hills from the west (1167), Kzlda from the north (1335
meters), and by the little hills whose highest reach up to 1135 meters from the east.
Feeding of the Lake Seyfe is by the springs at the west and north-west of the lake,
underground sources and surface currents at drainage area and falling rain to the Lake area.
Main springs feeding the Lake are Seyfe, Horla and Yenidoanl. The emptying of the lake is
realized by evaporation since the lake is located in a closed reservoir.
Because it is located at the low slope, the big differences can be observed at the lake area
depending on the rains and the seasons. The lake area at the high level of water (in winter and
spring months) reaches 7800 hectares.
In summer season, the water level drops to 60-70 cm because of not getting enough rain,
using almost all of the water of the spring for irrigation purposes and excessive evaporation.
As a result of this, the water in approximately 4900 hectares of lands in drained totally and
vast muddy flats appear. The depth of the lake, which is very shallow, at the maximum water
level is 165 cms.
There are various islands at the temporary lake area where thousands of birds nest.
Most parts of the lake coasts are muddy and flat. There is a small delta formed at the place
where the water of the Seyfe springs, which are the most important sources for fresh water,
reaches to the lake. The water of the Seyfe springs are kept in a small reservoir and used for
watering in summer. There are poplar and willow groves around and on the west coasts of the
reservoir. There are vast steppe areas at north of the lake reaching up to the lake coast. Other
coasts are generally surrounded by fields.
At the north and south-east of the temporary lake area there are light salty and fresh water
marshy fields which are very important for the birds breeding and wintering at the lake.

2.1 Functions and Values

Wetlands regulate the water regime of their area by re-charging and de-charging
subterranean water, reducing the devastating effect of the floods and balancing the floor
water. Again they make positive effect on the local climatic elements, mainly on the rain and
temperature, by raising the humidity rate of their environment. In addition, they increase the
water quality by retaining the sediments, foodstuff, and poisoning materials. Meantime, wet
lands, together with the tropical forests, are the ecosystems of the world which perform the
most biological productivity.
For the season, wetlands make living possible for thousands of various living creatures
which have high ecological and commercial values.
Contrary to its poor benthic fauna, Lake Seyfe is quite rich in organic production. In some
years gathering of more than 100.000 birds is a most important indication of this rich
varieties.
Besides its richness in food stuff, the salty water lake, its wet lands varying from salty to
fresh water quality, vast steppe areas and safe conditions away from the predators, with secure
island at different ecological characters, has created to the thousands of different species of
birds ideal feeding, sheltering and breeding condition.
709

The artificial mounds around the lake, the naturalness of the vegetation, turquoise color
lake view formed by the bird groups have the potential which can attract many people to the
area with the help of the advertisement and the development of the infrastructure which shall
provide opportunity to the ornitho-tourism.

2.2 The Flora and Vegetation

The Middle Anatolia steppe climate is dominant in the Lake Seyfe reservoir. The areas
surrounding the lake are lack of trees and bushes. Concentrations of semi-bush forming
Lycium depressum are seen at the northern parts of the lake. Near the Lake Seyfe there are
orchards, poplar and willow groves.
No aquatic plants are seen because the water of the lake is salty. At the areas where the
salty water is effective and in the salt-water marshes the plants such as Halocnemum
strobilaceum, Salicornia prostrata, Salsola inermis, Panderia pisola, Petrosimonia brachiata,
Krascheninnikovia ceratoides, Cmaphorosma monspeliaca, Gypsophila perfoliata, Frenkenia
hirsuta, Limonium iconicum, Limonium globuliferum are dominant. At the off-shore areas,
especially north of the lake the beautiful species of the interior Anatolian vegetation can be
seen. Most common species seen in the are area; Alyssum desertorum var. desertorum, A.
Pateri subsp. pateri, spergularia media, Gypsophila parva, Noaea mucronata subsp.
Mucronata, Reaumaria alternatifolia, Astragalus kirshehiricus, A. lydius, A. karamasicus,
Peganum harmala, Bupleurum sulphureum, Scabiosa argentina, Anthemis wiedemanniana,
Achillea wilhelmsii, Artemisia santonicum, A. taurica, Gundelia tournefortii var. tournefortii,
Centaurea cordumeniaca, Marrubium parviflorum subsp. Oligodon, Tymus sipyleus subsp.
rosulands, Acantholimon acerosum var. acerosum, Secale montanum, Bromus japonicus
subpp. japonicus, B. tectorum B. tomentellus, Fectuca callieri, Stipa capillata.
Juncaceae, Pocaceae and Cyperaceae members are dominant at the pastures which are not
affected by salt-water.
In addition, at the places where the springs are mixed with the lake and in canals the fresh
water plants such as Phragmites australis, Sparganium erectum, Thypa angustifolia,
Potamogeton pectinatus, Iris pseudocorus.

2.3 The Fauna and Ornithological Importance

Most important living creatures at the lake are water-birds. No fish species are seen
because lake water is salty and mixed with sodium. Only two species of little fish such as
Aphanius chantrei and Sp. 5-6 cm in length live where the fresh water mixes with the lake.
Although this fish have no commercial value, they have important ecological value as food
for water-birds such as pelicans and herons.
Seyfe Lake Floor is very rich for Chironomidae larvae and shellfish such as Planorbis Sp.,
Lymnaea sp. and Physa sp. which are important for feeding of birds. Formation of big
gathering of birds at the place where the Chironomidae larvae are dense is a sign that the birds
are fed with this larva.

Ornithological Importance
187 different bird species are found at the lake and its surroundings after observations
made in this area. Bird existence at the lake reaches big number during the migration and in
water. Anser albifrons, Tadorna tadorna, Tadorna ferruginea, Anas crecca and Fulica atra
form big groups. In some years the number of the birds reaches up to 100 000.
Meantime the Lake Seyfe is one of the important breeding areas in Turkey for water-birds.
At the islands located at east of the lake, Flamingo (Phoenicopterus ruber), White Pelican
710

(Pelecanus onoctrotalus), Spoobill (Platelea leucorodia), Little Egret (Egretta garzette),
Glossy Ibis (Plegadis falcinellus), Red-crested Pochard (Netta rufina), Black-winged Stilt
(Himantopus), Avocet (Recurvirostra avosetta), Spur-winged, Lamwinh (Hoplopterus
spinous), Mediterranean Gull (Larus melanocephalus), Blackheaded Gull (Larus ridibundus),
Gull-billed Tern (Gelochelidon nitolica), Little Tern (Sterna albifrons) are the important
incubating bird species. Approximately 2000 pair of flamingos nests on the island which are
the most important breeding place after Lake Tuz, located east of the Lake Seyfe.
The steppes around the lake is the feeding and procreation area of the the Greast Bustard
(Otis tarda) which is one of the bird species threatened by extinction. The lake surroundings
also the gathering place of the storks before migration.

2.4 Historical and Cultural Values

Rich feeding conditions having different ecological characteristics such as shallow water
areas showing varieties from salty water to fresh water, marshy lands, wet grasslands and
steppe lands; existence of many island providing incubation facilities in a secure condition
away from the human and predators; being on the two migratory paths between Europe, Asia
and Africa which are connected on the Anatolian peninsula; being near to the Sultan Reed
and Salt Lake which are among the important wet lands of our country; has caused Lake
Seyfe having characteristics to be one of the important wet lands for birds of Turkey.
The findings obtained from the archaeological diggings have shown that the first
occupations occurred during Bronze Age (in the years 3500-2000 BC). There are 20 artificial
hill and tumulies of this age around the lake and its surroundings. Investigation has shown
that the people occupied the coast of the lake practiced throughout history of cultivating the
land. It is estimated that the people were fed on hunting the migrating birds while practicing
the agriculture. The antiques found in the tumulies are kept in the Krehir museum.

2.5 Human Activities

Agriculture
Main livelihood of the people living around the lake is field agriculture and livestock. 90%
of the people do this for a living. Dry agriculture is done at 91.7% of the reservoir and wet
agriculture at 8.3%.
Main agricultural products are wheat, sugar beet, barley, lentil, chickpea, bean, oat and
sunflower. Orchard and vineyards are among these though very few.
Since the pastures occupy vast area in the reservoir, pasturing has become important.
Excess grazing has been causing pasture to deteriorate and lost its functions. Conversion to
modern livestock husbandry shall reduce the pressure on the pastures and increase the
productivity.

Hunting
Hunting is totally prohibited after declaration of the area in 1990 as Nature Protecting
Area. But on effective control has not been obtained since a local institute has not been
formed for protection of the area. For this reason it is known that from time to time illegal
hunting occurs.

Tourism
Historical and cultural richness of the artificial mounds and tumulies in the area, rich bird
existence of the lake, beauty of the scenery formed by the lake and its surroundings, has
provided important potency to the lake and its surrounding in view of tourism.
711

The tourism can provide important contribution to the local economy by good
advertisement and if the infrastructure is developed.

2.6 Protection and Administration

By including the mentioned 10700 hectares of land to the Ramsar Agreement in the year of
1994, the protection of its natural formation and ecological characters at the international
level is no administrative mechanism to protect and observe the area yet.
For protection of the area and for securing the reasonable usage of it; preparing a
management plan for the area; in addition, strengthening the coordination and cooperation
between the related institutes, and for securing the observation, a local administration is
required to be established.

3- STATUS OF SEYFE LAKE AND ITS ENVIRONMENTS

DS (General Directorate of State Hydraulic Works) has been established in 1954
according to do law numbered 6200. Main duties of DS are preparing irrigation and drainage
projects, supplying of drinking water to the cities whose population are grater than 100.000,
constructing hydro electric power plants and realizing activities according to the
environmental law and regulations. The environmental sector is a new, but there are many
environmental projects prepared by DS. One of the environmental projects is Seyfe Lake
Ecological Preservation Project.
Laws have been enacted in Turkey to combat malaria and gain agricultural land with
regard to wetlands; these laws have introduced sanctions for the governmental establishments.
Various establishments have made studies in Seyfe Lake and Wetland and at the end of these
studies; the natural balance has begun deteriorating. In the recent years, the process of the
deterioration of the ecological balance has gained speed parallel to the drainage works of the
local people to acquire agricultural land in this wetland (with the assistance of governmental
credit) and the linkage of drainage canals to the lake. DS, which has conducted various
works in Seyfe Plain due to the requirements imposed by its statutory law since 1960, first
began works to dry up the lake entirely; there were plans to discharge the lake into the River
of Kzlrmak but this alternative had to be discarded in the light of the possibility of
Kzlrmak waters getting salty subsequently.
Late in the 1960s, due to the fact deep leakage did not occur because the ground was very
heavy in the flat areas located in the Northern region outside the existing lake in the middle of
the plain, the flood prevention canals linking such lakes to the main lakes for the purpose of
preventing small lakes forming in the years having much precipitation were opened by DS
in those plots of land where no surface flow realized as there were no slopes in the plain and
thus, it was possible to have agriculture in these areas. Agricultural benefits seen in this area
in the 1970s resulted in excavation of drainage canals by most farmers having links to those
flood prevention canals.
The issue again entered the agenda upon the pressure of the local people and agricultural
enterprises; it was considered to preserve the lake by building walls around it and to discharge
drainage and precipitation waters outside the basin; as this solution would take a long time to
be concluded, it prevented the economic feasibility. As outlined above, improvement works
sped up in the agricultural areas in the vicinity of the lake and as a result of this, extreme use
of fertilizers and herbicides as well as drainage works resulted in sharp deterioration of both
water quality of the lake and a change in its borders as large volumes of dirty water poured
into the lake. Current status plan for Mucur Seyfe Basin is shown below Figure 1.

712


Figure 1: Current Status Plan for Mucur - Seyfe Basin

As of the late 1970s, the society began having awareness about environment, growing an
interest in wild life. The richness and plentifulness of bird species and diversification of other
hunting animals help Seyfe become an attractive hunting area. Its closeness to Ankara and
existence of much less restrictions for members of the foreign missions in Ankara has further
increased the interest. Following this phenomenon, the wild life there has begun suffering
large scale destruction. This situation led to publicity about the area both in the country and
abroad, creating public opinion on the ecological preservation. Such publications and
emerging public opinion have changed the attitudes of the establishments towards the issue of
Seyfe enclosed basin as the agriculture dominated drying and improvement projects have
been replaced by ecology-oriented projects.
During the studies conducted in 1974 1975 in order to determine the hydro-geological
structure of Seyfe Basin, average reserves of 11 hm
3
/year of underground water were
determined and because the sources feeding the lake at present are used for irrigation, the
feeding of the lake is largely by direct precipitation and underground water.
Works involving excavation of a high number of drainage canals by farmers receiving
credit at the onset of the 1980s and lowering the bottom water which is located at a very high
level have gained momentum.



Non arable land: 20 674 ha
Area opened to agriculture by drainage: 2 733 ha
Area which is no longer suitable for cultivation: 3 750 ha
Area which will soon be no longer suitable for cultivation: 4 107 ha
Lake summer border: 1560 ha 100 cm
Lake winter border: 7 860 ha 140 cm
713


4- FIRST DRAINAGE AND RECLAMATION PROJECT BY DSI

In 1987, Mucur Seyfe Basin Reclamation Planning Report was prepared covering an area
of 23 226 ha.
Preservation of wild life gained currency again when the report began being taken up by
the governmental authorities. The decisions taken by the Council for Preservation of Cultural
and National Heritage and by Kayseri Council for Preservation of Cultural and National
Heritage in 1988 and 1989-1990, respectively, an area of 23 585 ha around the lake was
declared First Degree Natural SIT Zone as the rest of the drainage area of 1522 km
2
was
declared Second Degree Natural SIT Zone.
The fact that the basin has been declared a SIT zone displays the requirement that any
works to be carried out from now on must serve the preservation and improvement of the
ecological balance. The General Directorate of State Hydraulic Works has been authorized to
prepare a planning report in cooperation with the Environmental Under-secretary and other
establishments as a result of the studies and meetings conducted on the location with the
representatives of the relevant establishments and organizations under the coordination of the
Environmental Under-secretary following these developments. This meeting held on
20.07.1990 was attended by:
Under-secretary of Ministry of Environment
SPO Under-secretary
Ministry of Culture
Ministry of Tourism
General Directorate of State Hydraulic Works (DS)
Ankara University Agricultural Faculty
General Directorate of Agricultural Operations (TGEM)
As a result of the studies conducted under the chairmanship of the Environmental Under-
secretary, a decision of agreement consisting of 5 articles was given and signed.
According to this text, briefly:
1. The project will be an ecological preservation project,
2. The alternative canal, a large part of which remains outside First Degree SIT Zone, will be
examined.
3. For the purpose of determining the impact of this canal on :
a) agricultural areas
b) lake water balance sheet and water quality
c) wild life
d) water and soil quality in Kzlrmak and its basin following discharge, preparation of a
physical project by DS and submission of it to the SPO and Environmental Under-
secretary and it has been given the task of ensuring that an EIA (Environmental
Impact Assessment Report) be given by a council the members of which are to be
determined by Under-secretary of Ministry of Environment and DS by taking this
project as a basis. Following the acceptance of the feasibility report prepared by DS,
the project has been incorporated in investment program and implementation thereof
began in the same year.






714


5- MUCUR SEYFE BASIN ECOLOGICAL PRESERVATION PROJECT

5.1 Seyfe Lake Natural Ecosystem

Seyfe Plain and Lake are located on a closed basin of 1522 km
2
.

Of this area:
Lake water surface 1560 ha
Temporary marsh 6300 ha
Total 7860 ha (Lake is presently fed by two springs and Malya Drainage Canal)
The slope of the lands in the lake environs is very inclined.
Due to the fact that ground water level is high in seasons of precipitation, the saline content
of arable land in the environs has increased for many years and land has begun to be arid
and barren.

5.2 Deterioration of the Ecosystem

In the 1960s, MALYA State Agricultural Organization tried to lower the high flat area
water in the land which has become arid by digging drainage ditches. DS dug flood canals
to prevent floods in land.
After 1975, TOPRAKSU, was an institution for organization agricultural facilities in those
years, dug drainage canals to help villagers.
Both interventions led to the deterioration of the natural balance. The water of the closed
basin, of which some was lost as a result of capillary evaporation as some reached the lake
after a long time, reached the lake in a short time by means of the drainage canals,
expanding the lake area and increasing the lake water height.

In Conclusion
The saline content in the lake water surface has spread to a large area. The salinity in the
lake will decrease in the lake reservoir in the long term.
As the lake area has expanded, the land which people live nearby has remained under
water.
Because the evaporation process has increased, land has become non cultivable in the
months of cultivation because of high water level and topography. In addition, salinity has
increased.

5.3 Re-arrangement of the Deteriorated Ecosystem:

An area of about 23 000 ha is under the influence of salinity due to topography. Local
people cannot make cultivation or they obtain products at a low level.
The intervention by means of drainage has deteriorated the ecological balance to the
damage of local people.
The balance will also deteriorate to the damage of the bird population in the long term.
Agricultural pesticides and herbicides used in the drained areas will pollute the lake and
consequently, the ecosystem will be destructed entirely.
In case of failure to intervene, human beings would not hesitate to turn the ecosystem to
their own interests for the sake of protecting their own existence.
715

DS has developed a project which re-arranges the ecosystem in a manner protecting both
human existence and bird population. DS project is flexible; it is of a nature capable of
adjusting the optimum balance according to prevailing conditions over time.

5.4 Basis of DS Project

As seen in the map plot, the drainage canals performing fast discharges into the lake will
be cut off by a surrounding canal and collected water will be re-routed to the Kalayck branch
of Delice River which is an arm of Kzlrmak. If necessary, water will also be discharged into
the lake by means of the gates to be mounted on the canal.

a) Pollution of the lake will be prevented. (Pollution caused by agricultural pesticide herb
combat products, herbicides, etc.)
b) Decrease in the salinity of the lake reservoir because of the lake expansion in years of
heavy precipitation, will be prevented.
c) Lack of water that may arise in very arid years will be prevented by keeping the marshy
area humid all the time. (by providing water from the canal)
d) 15972 ha of land will again be suitable for cultivation and this will serve the prosperity of
people.


3 units of automatically controlled gates will be built in the Northern-Southwestern
direction under Seyfe Project to achieve control by means of the main discharge canal located
outside First Degree SIT zone to a large extent. While the canals were designed, the canal
bottom benchmark was kept at a level not draining the lake waters; during construction,
sophisticated works were avoided to the largest extent possible. 3 units of gates constitute the
most important element under the project.


716

These gates:
a) Gate 1: It is downstream according to the commencement point of Canal AB on the main
drainage canal of Malya State Agricultural Organization.
b) Gate 2: It is the starting point of Canal AB. By operating these two gates, drainage waters
going into the lake will be kept under full control. The drainage canal gate must be opened
in years of aridness; the lake water level will be kept at an optimum level as in normal
years by closing the main discharge canal gate.
c) Gate 3: (Basin Control gate) It is located at the basin exit of Canal AB. This gate is multi-
purpose.

1. By adjusting the gate level of open position and discharging water into the creek
which provides the discharge (Kalayck) under control, pollution of this creek and
downstream part will be prevented.
2. The main canal and stand-by canals can become a water storage facility by entirely
closing the gates when required; it can be used to regulate the lake water level and
water quality whenever necessary.
3. It is the control unit which will ensure adjustment of the DSI project according to
conditions.

In addition, reeds begin growing inside the canals of discharge over time; such areas of
reeds offer a very ideal environment for wild life. Strips each with an area of 40 km are never
so small to ignore.
Many alternatives have been studied to correct the ecological balance of Seyfe Basin and
to maintain it at the most suitable level; thus, it has been tried to reach the solution which
would inflict minimum harm on the nature while protecting the balance.
Under Seyfe Ecological Preservation Project, a canal with a length of 31 300 m has been
dug as part of Main Discharge Canal AB and 5 566 000 m
3
of excavation was carried out so
far. 13 300 m of stand-by drainage and 7 300 m of tertiary drainage have been dug. In
addition, a section of 14 000 m
2
between km: 0+000 4 + 000 of Canal AB has been coated
with stone pitching.
In 2003, 18 860 m of farm area drainage was dug as 124 000 m
3
of earthmoving was
carried out. 13 000 tons of iron cast gates manufacturing were carried out on the canal route.
In 2003, 10 000 m of farm drainage will be dug and installation of the electronic
measurement devices and control gates will be completed. In 2004, 21 000 m of farm
drainage will be dug and 200 000 m
3
of excavation will be realised.
The 10 km discharge gate numbered B1 which was incorporated by the project planning
was cancelled by DS.
Apart from the canal excavation in the basin, DS completes maps of the project area every
year to the maximum extent allowed by the relevant programs; it also carries out periodical
water observation at 5 points together with the additional points determined in 1990 and it
measures the lake level periodically. The results of these studies conducted are converted into
tables and graphs.
Seyfe Ecological Preservation Project prepared by DS is not a scheme solely created as a
result of DS views. During planning studies, views of all stakeholders and organizations have
been taken into consideration; joint studies were made in the field and a decision was then
taken on the formulation of the present planning.




717

6- CONCLUSION

The operation of remote controlled gates; which are sensitive to water level and
quality, constructed at the three points of AB, drenaj canal which is situated between fields on
the north and Seyfe Lake, is the most important characteristic of the project.

By using canal constructed with these three remote controlled gates, water can be
diverted to the Seyfe Lake also in drought seasons by closed completely two gates. In addition
water level and quality in the AB canal and lake are monitoring always and desired
concentration of salinity of lake can be obtained. Excess water, contains salt, pesticide and
fertilizer, comes from the drainage area can be diverted directly to Kzlrmak river in
desirable season. Pollution in branches of Kzlrmak River and Seyfe Lake can be prevented
by this way.
15972 ha of 23000 ha of saline land on the north of the Seyfe Lake can be rehabilited
and can be opened agricultural activities.

New breeding areas for birds are generated by anew design and alignment of AB
canal, situated 3-5 km far away from the maximum water level of Seyfe Lake. Moreover,
30900 m length drainage canal encased by reeds, present natural islands and manmade islands
in Seyfe Lake can be provide new breeding areas for birds and natural life.

Mucur-Seyfe Ecological Protection Project has contributed to economy of people live
in region and also ecology of Seyfe Lake and its surrounding.

718




GEOCHEMICAL MODELING OF GROUNDWATER
IN THE WESTERN NILE DELTA, NORTH EGYPT

Mohamed El Kashouty
*
and Shabaan Mahmoud
**

* Geology Department, Cairo University, Faculty of Science ** Research Consultant for Groundwater

ABSTRACT

Thirty-six groundwater samples, tapped mainly in the Pleistocene aquifer, are pumped to
cultivate the new projects in the western Nile Delta. Total dissolved solids (TDS) is strongly
correlated with major ions indicating leaching of agricultural wastewater. NO
3
is strongly
correlated with TDS, SO
4
, K, and Na, suggests the application of fertilizers and super-
phosphate fertilizers. TDS is strongly correlated with F, B, and NO
3
, shows the impact of
agricultural wastewater. Cu is strongly correlated with Pb and Cd, indicate the same
characteristics trend and origin, may be a combination of agricultural leaching and the
sorption-desorption processes in the geomedia especially in the aquifer area that intercalated
with high thickness of Pliocene clay. Fe is strongly correlated with Mn (r=0.9), delineate the
Fe-Mn oxyhydroxide phase. Seven rotated factors are recognized, each clarify
hydrogeochemical processes occurred in the aquifer system. The factors are, the main
agriculture fertilizer, toxic heavy metals, groundwater depth, dissolution of carbonate
minerals, co-precipitation phase, recharge influence, and leaching process. The cluster
analysis shows that cluster I is characterized by the lowest concentrations of TDS and NO
3
,
while cluster 2 contains the samples with higher TDS and nitrate concentrations than the
previous one. Cluster 3 is include samples of intermediate TDS that higher than the previous
one, and characterized by the increased NO
3
concentrations (10-40 mg/l). Clusters 4 and 5 are
the highest salinity and the NO
3
concentration ranges from 25 to 40 mg/l. The highest NO
3

concentrations and the TDS ranges from 2500-3500 mg/l dominate the last cluster. The last
three clusters are located beside the intensive irrigated areas (Nubaria and Bustan projects) to
be suffering groundwater degradation. The schematic organic matter (OM) and dissolved
oxygen (DO) scenario is distinguished into three major sub-areas, the southeastern part of <
10 mg/l NO
3
-N and expected to contain high OM and low DO contents. The second sub-area
of 10-40 mg/l NO
3
-N, is characterized by moderate OM and DO contents. The last sub-area in
the northwestern part of the study area of >40 mg/l NO
3
-N, is predicted to be of little OM and
much more DO content. The saturation indices for calcite and dolomite estimated by
WATAQp program, generally do not correlated with TDS and the calcite, dolomite, and
aragonite are dissolved in most of the studied locality, while minor sites are characterized by
precipitation. The dedolomitization process is play a minor role that restricted to local area,
while the dolomite dissolution model is more common in the study area. The results of the
Flownet program depict that the average time of groundwater to travel from the highest
potentiometric area (Nubaria mound in the northwestern part of the study area) to the
southeastern part, in Pleistocene aquifer, is 6500 yrs, with average velocity about 9 m/yr.

Key words: groundwater, statistical applications, saturation index, Flownet, western Nile Delta


FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


719
INTRODUCTION

The chemical composition of groundwater is controlled by precipitation, mineralogy of the
watersheds and the aquifer, climate, and the topography. In the study area (Fig. 1a), there is a
wide variety of climatic conditions, hydraulic regimes (aquitard, aquifers, faults, and joints),
land use application especially agricultural activity, and the sedimentary deposition. Thus the
samples from the area could potentially represent a variety of water types providing an
opportunity to test the performance of many of the available graphical, hydrogeochemical
programs, and statistical techniques. These methodologies used to classify water samples. The
use of the major ions as natural tracers (Back, 1966) has become a very common method to
delineate flow paths in aquifers. Generally, the approach is to divide the samples into
hydrochemical facies, that is groups of samples with similar chemical characteristics that can
then be correlated with location. The spatial variability observed in the composition of the
natural tracers can provide insight into aquifer heterogeneity and connectivity as well as the
physical and chemical processes controlling water chemistry. A variety of graphical,
statistical, and hydrogechemical techniques have been applied such as cluster analysis,
correlation coefficient, factor analysis, wataqF and Netpath programs, and the scenario of
organic matter and dissolved oxygen. This paper utilizes the data to review these techniques
and compare their ease of use and ability to sort water chemistry samples into groups.
Groundwater exploitation in the study area will continue to satisfy the demands of the
agricultural projects of the long term. It can result in lowering groundwater heads and increase
the salinity. The main groundwater resource is from the irrigation return flow, River Nile,
surrounding aquifers andprecipitation. Fig. 1b clarify the increase in abstraction rate with
time in the western delta region.

Cluster 1
Cluster 2
Cluster 3
Cluster 5
Cluster 4
Cluster 6
28
21
22
19 18
16
17
15
20
23
13
30
29
26
36
35
27
25
24
34
10
14
1
12
5
3
4
8
9
11
33
32
1
7
6
2
30 30
31 00 30 00
30 30
31 00
30 00
31
W
a
d
i E
l N
a
tru
n
T
a
n
t
a
0 13 km
Cairo-Alexandria
Desert road
Agricultural road
River Nile
Well location and
number
1

Fig. 1a Map view of HCA derived subgroup and group values for groundwater samples
in western delta area.





720

Fig.1b Abstraction rate within different periods
Hydrogeologic setting

The Nile reservoir is essentially a great
morphotectonic depression where the
Pleistocene sediments attain a thickness
exceeding 300-m and act as the main
groundwater aquifer that supplies the
surroundings. It is highly productive aquifer
(RIGW and IWACO, 1990), made up of
successive layers of sand and gravels with
clay intercalated. The aquifer is free water
table to semi-pervious and recharged from the
Nile Delta Pleistocene aquifer and the
seepage from the canals and drains. The local
development of gentle folds and faults (wadi
Natrun and wadi El Farigh anticline and Hamza syncline) produce the easier flow of
groundwater between the delta reservoir and the study area. Wadi Natrun is affected by
complex fold-fault systems, and represents negative areas to the west of the Nile Delta that
account for the pronounced groundwater pump. The up folding and possibly fault rising at
Khatatba-Wadi El Farigh brought the Miocene aquifer in the western Delta in juxtaposition
with Pleistocene aquifer in the Nile Delta. Westwards, from Wadi El Farigh to the Moghra
depression and the Qattara Depression, a series of a E-W and NW-SE faults may act as direct
conduits of groundwater. With respect to basaltic intrusion (base of Miocene aquifer), these
only impede the downward or upward flow of groundwater (Shabaan, 2002). Due south,
folded and block faulted area along the Cairo-Alexandria desert road is appeared, between
them a channel for instance wadi El Lulu and El Samya acting as water conduits. The
occurrence of major folding structure and unconformity in the nearby area of Abu Roash, lead
to potential feeding of upper most water-bearing sand and gravel from underlying Nubian
aquifer.


METHODS AND MATERIALS

The 36-groundwater samples used in this study were collected by the work of Shabaan (2002)
and analyzed in central lab of MWRI. EC, pH, and NO
3
, were measured in situ by portable
field kite, trace elements were estimated using ICP technique, Na and K were determined
using flamephotometer, and Ca, Mg, Cl, and HCO
3
were measured by titration, and SO
4

determination by spectrophotometer. Primarily the wells are used for public supply, domestic
and irrigation purposes. Samples were taken directly from each well, filtered, preserved, and
delivered to an analytical laboratory within prescribed maximum holding times.

RESULTS AND DISCUSSION

Correlation analysis

Total dissolved solids (TDS) is strongly correlated with K, Na, Mg, Ca, Cl and SO
4
,
indicating that salinity is most likely caused by the leaching of agricultural wastewater,
solutes concentrated by evapotranspiration, and diffusion from the geomedia. NO
3
is strongly
correlated with TDS (r=0.8), SO
4
(r=0.8), K (r=0.8), and Na (r=0.8), while moderately
correlated with Cl (r=0.6), Ca (r=0.5), and Mg (r=0.6), suggests the application of fertilizers



0
10
20
30
40
50
60
70
80
90
A
b
s
t
r
a
c
t
i
o
n

r
a
t
e
,

m
3
/y
r
Pre-1959 1959
1962
2000
*10
6
*10
6
*10
6
*10
7

721
and superphosphate fertilizers and the dolomitization and dedolomitization, which are
contributed partially to Ca and Mg contents in groundwater. The rock water interaction and
the mixing processes is established by the ion pair relationship (Collins, 1975). The best fitted
with the least square curves is figured (Fig. 2a) to determine the TDS against Ca, Mg, and Na.
The source of Ca content (Fig. 2a), most probably, is relics of the sedimentation water of the
carbonate minerals, suggests mixing processes between different water types. On the other
hand, the Mg-TDS best-fitted (Fig. 2b) explains that the Mg content is mainly derived from
the fertilizer application. The Na-TDS best-fitted (Fig. 2c) is obviously indicate the
agricultural waste source for Na and TDS. The main sources of these solutes are land use
application, followed by the diffusion from the geomedia. Na-Cl relationship (Fig. 2d) clarify
that half of the samples with low to intermediate salinity, are plotted close to the seawater
dilution line, while the other half with high salinity are plotted away. The excess Na over Cl is
caused by dissolution of Na-rich sediments. Since the Na/Cl ratio of local mixed fertilizer is
several orders of magnitude smaller than the Na/Cl ratio in seawater. This scattergram
suggests that fertilizers are contributed less in Na and Cl contents in groundwater. Cation
exchange reaction between aquitard clay layer and the solution phases, also can enhance the
contribution. It concludes from Fig. 2c and d that the Na concentration in groundwater is
attributed mainly to agricultural effluent and geomedia rather than fertilizers. The Ca vs. SO
4

(Fig. 2e) demonstrates that both ions are indeed correlated that characterizes the irrigated
water. The SO
4
vs Cl plot (Fig. 2f) indicate excess of SO
4
over Cl compared with their ratio
in seawater, reflects the application of mixed fertilizers with a high SO
4
/Cl ratio. Therefore,
the KCl fertilizer does not seem to be used frequently. Ca is strongly correlated with Mg
(r=0.8) reflect a common mineralogic sources, may be weathering of amounts of calcite and
dolomite (Stauffer and Wittchen 1991). Cl and SO
4
were correlated strongly with other
dissolved constituents supports the idea of land use application. TDS is strongly correlated
with F (r=0.9), B (r=0.9), and NO
3
(r=0.8), shows the impact of agricultural wastewater. Cu is
strongly correlated with Pb and Cd, indicate the same characteristics trend and origin, may be
a combination of agricultural leaching and the sorption-desorption processes in the geomedia
especially in the aquifer area intercalated with high thickness of Pliocene clay. Fe is strongly
correlated with Mn (r=0.9), delineate the Fe-Mn oxyhydroxide phase, favor the sorption-
desorption processes, depending on changes in physio-chemical conditions. B is strongly
correlated with F, shows the agricultural effluent.
722




































































































































































































































































































































































































































F
i
g
.

2

S
e
l
e
c
t
e
d

r
r
e
l
a
t
i
o
n
s
h
i
p
s

a
m
o
n
g

t
h
e

m
a
j
o
y

i
o
n
s

o
f

g
r
o
u
n
d
w
a
t
e
r

s
a
m
p
l
e
s






723

Rotated matrix of factor loading

The varimax rotation was used to achieve the rotated factor matrix after rotation with Kaiser
normalization (Davis, 1986). Seven rotated factors were extracted (Fig. 3), that explain 86.3
% of the total variance (Table 1). Therefore, these factors can be used to explain the
background hydrochemical processes. Factor 1 (35.8 % of the variance) has loading of K, Na,
Mg, Ca, Cl, SO
4
, NO
3
, B, F and TDS. It is attributed to agricultural effluent leaching and
fertilizer, called agricultural impact. Factor 2 (10.3 % of the variance) shows high loading of
Pb, Cu, and Cd, reflect the sorption-desorption process and the same origin, and called toxic
metal factor. Factor 3 (9.9 % of the variance) has loading of groundwater depth, residual
sodium carbonate, and PO
3
, indicate the influence of recharge rate. Factor 4 (8.9 % of the
variance) has loading of groundwater level, Ca, and Mg, clarify the dissolution of carbonate
minerals and called the dolomite dissolution or dedolomitization factor. Factor 5 (8.5 % of the
variance) has loading of Fe and Mn, characterize the Fe-Mn oxyhydroxide phase and called
co-precipitation factor. The lack of incorporation of the fore-mentioned toxic heavy metals in
the previous phase is caused by the presence of organic matter and clay minerals, which
enhance the sorption-desorption processes rather than this phase. Factor 6 (7.5 % of the
variance) has loading of K and Zn, can interpret that the land use application is contributed
little to medium concentrations of both elements in groundwater and mainly derived from the
geomedia. Factor 7 (5.7 % of the variance) is positively correlated with HCO
3
and negatively
with pH, indicate the role of surface meteoric water of agricultural impact.

Factor No. Principle component
Loading

Variance %

Variables contributed
1- Main agricultural impact 0.53-0.98 35.82 %
EC, SAR, K, Na, Mg, Ca, Cl,
SO
4
, TDS, NO
3
, B, and F
2- Toxic heavy metals 0.87-0.91 10.3 % Cu, Pb, and Cd
3- Groundwater depth 0.54-0.85 9.9 % GWD, RSC, and PO
3

4- Dissolution of carbonate miner 0.5-0.84 8.9 % Ca, Mg, and GWL
5- Co-precipitation phase 0.82-0.89 8.4 % Fe and Mn
6- Geomedia impact 0.47-0.88 7.4 % K, and Zn
7- Leaching process 0.82&-0.55 5.7 % HCO
3
and pH

Table 1 Varimax rotated principle components of the factors

The relationship among factors, determined by the program, is interpreted as follows: the
correlation between factor 2 and 3 is high (r=0.9), indicate the groundwater depth is greatly
enhance the sorption-desorption processes. It enhances by the pH changes through the
intensive agricultural waste and the open /or closed system conditions of the aquifer. The very
weak correlation between factor 4 and 6, shows that the Ca and Mg ions enters the
groundwater from both fertilizers and diffusion from the gemedia, and K and Zn are
contributed mainly by geomedia other than agricultural activity. The negative correlation
between factor 2 and 5, delineate the little contributes of the Fe-Mn oxyhydroide phase to the
sorption-desorption processes of these metals in the environment. Flocculation of humic
acids could lead to removal of metals from the solution (Sholkovitz, 1978), especially Cu,
because freshwater speciation is dominated by organic complexes (Hart and Davies, 1981;
Paulson et al., 1984; Kuwabara et al., 1989, Apte et al., 1990b). Fuller and Davis (1987)
showed that calcite dominated the sorption of Cd by aquifer sand. The main hosts of these
metals are the organic matter and clay minerals rather than Fe-Mn oxyhrdroxide phase. Factor
7 is correlated with 4, depict the dissolution of carbonate minerals is increased with leaching.
724




















































Fig. 3 Factor analysis of the major ions and trace elements
of the groundwater samples
725
Cluster analysis

Cluster analysis is a numerical technique for defining groups of related samples (Q-mode) based
on high similarity coefficient that computed between each pair of samples and these are then
clustered. Dendrogram, a two dimensional hierarchical diagram represents the end product of
cluster analysis. The result of the cluster analysis is shown by the dendrogram (Fig. 4), with
respect to the trace and major elements concentrations in all samples. Cluster I is characterized
by the lowest concentrations of TDS and NO
3
. It is much more diluted from the River Nile
(Rosetta branch) and the surrounding aquifers (depict from the flow lines). The Rosetta Nile
branch is considered a recharge zone to the aquifer, so it can improve the groundwater quality.
Cluster 2 contains the samples with higher TDS and nitrate concentrations than the previous one.
Both previous clusters are distinctive by the highest concentrations of Fe and Mn that co-
precipitated to form the Fe-Mn oxyhydroxide phase. The third cluster is include samples of
intermediate TDS that higher than the previous one, and characterized by the increased NO
3

concentrations (10-40 mg/l). Clusters 4 and 5 are the highest salinity and the NO
3
concentration
ranges from 25 to 40 mg/l. The highest NO
3
concentrations and the TDS ranges from 2500-3500
mg/l characterize the last cluster. Clusters 4, 5, and 6 are located beside the intensive irrigated
areas (Nubaria and Bustan projects) to be suffer groundwater degradation, lead to increased in
TDS and NO
3
concentrations. The regular trend of the clusters distribution is caused mainly by
distribution of grain size, lithology, organic matter, dissolved oxygen, the rate of agricultural
waste infiltration, sorption-desorption phenomena, topography, and the water levels. It is fair to
ask if these clusters of samples have any physical significance meaning, or are just a statistical
result. Plotting the subgroup value for each sample on a site map (see Fig. 1) can tested the
relationship of the statistically defined clusters of samples to geographic location.

5
12
32
33
8
9
2
7
4
6
11
3
1
14
31
34
17
26
13
15
10
29
27
16
30
36
23
24
19
22
21
25
20
18
35
28
1
2
3
4
5
6
Rescaled distance cluster combine
0 5 15 20 25 10
Cluster no.

Fig. 4 Cluster analysis of the major and trace elements of the groundwater samples in the
western delta

Table 2 shows the means for each of the parameters produced by the HCA. These values reveal
some trends between the major groups. Fig. 1a shows that there is a good correspondence
between spatial location and the statistical groups as determined by the HCA for instance cluster
1 is located besides the River Nile and is considered a recharge zone. Both clusters 1 and 2 are
726
located besides Rosetta Nile branch where a high thickness of Pliocene clay is represented and
decreases away from it. Both clusters are characterized by the highest Fe and Mn contents (Fe-
Mn oxyhydroxide phase) that coincide with the lithological trend. Apart from the River Nile, i.e.
towards the intensive cultivated area (Nubariya mound), the salinity and nitrate concentrations
are increased due to agricultural degradation. It appears that the technique can provide valuable
information to help define the hydrologic system. For instance, the high degree of spatial and
statistical coherence in this data set could be used to support a model of hydrochemical evolution
where the changes in water chemistry are a result of increasing rock-water interactions along
hydrological flow paths, besides the aforementioned anthropogenic degradation.

















Table 2 Mean water chemistry of the groundwater samples determined from HCA, pH (standard
units), mean concentration mg/l.

Scenario of dissolved oxygen and organic matter by nitrate trend

Nitrogen, phosphorous, and suspended sediment are contaminants that have widely contributed
to the deterioration in the water quality. These contaminants are derived from various sources
and are transported to surface water by surface (overland) runoff and groundwater discharge.
Surface agricultural and residential application of fertilizers, land application of animal waste,
discharge from septic tanks, and precipitation are major sources of nitrate (a common form of
nitrogen) that leaches groundwater. As groundwater flows through shallow aquifers and
discharges to surface water, it can contribute large amounts of nitrate to the surface water. The
factors affect groundwater flow and chemistry according Gary et al., (2001) are 1) the grain size
of soil and aquifer sediment 2) The amount of organic matter (OM), affecting water chemistry, is
minimum in coarse-grained deposits and tend to be more abundant in fine-grained. Organic
matter reacts with dissolved oxygen, removing it from the water environment, i.e. more organic
matter areas, dissolved oxygen (DO) can rapidly depleted. 3) Soil composition, topography, and
land use. Nitrogen is present in several forms in groundwater and it is transformed differently
among these forms depending on the availability of DO and OM. Where DO is abundant,
bacteria chemically oxidize (nitrify) ammonium to nitrate. Nitrates are of concern because it is
the one form of nitrogen that readily dissolves in, and flow with groundwater. Where DO is
limited, nitrate will not form and bacteria chemically reduce (denitrify) nitrate that is already
present to nitrogen gas or to ammonia if sufficient OM is present (Gary et al., 2001). The NO
3

content, water levels, and post map, can used to constructed the model (Fig. 5), it clarify the
Cluster
No. of
samples
TDS pH K Na Mg Ca Cl SO
4
HCO
3

1 12 658.1 7.98 4.8 139.1 13.6 44.3 116.3 135.7 205.2
2 4 1103.5 8.0 7.9 227.3 25.3 77.1 226.4 157.5 382.3
3 8 1694.8 7.8 18.2 447.6 28.3 56.0 365.4 442.4 334.9
4 2 3798.5 7.43 23.9 1112.3 80 117.4 1557.1 589.0 318.8
5 2 3582.5 7.8 23.4 1034.4 52.0 95.0 774.5 1453.4 149.2
6 8 2786.2 7.7 30.3 757.0 45.5 97.4 681.2 816.6 348.4
NO
3
Fe Cu B Mn Pb Cd PO
3
Zn F
1 3.095 0.571 0.039 0.135 0.237 0.055 0.034 0.154 0.034 0.536
2 5.83 1.025 0.011 0.308 0.583 0.006 0.001 0.184 0.019 0.820
3 20.08875 0.183 0.0113 1.101 0.097 0.021 0.001 0.265 0.258 2.150
4 26.685 0.311 0.002 1.44 0.065 0.005 0.001 0.108 0.0115 2.93
5 35.38 0.354 0.017 3.3 0.05 0.005 0.001 0.1475 0.0055 4.02
6 42.82 0.144 0.057 2.072 0.046 0.004 0.002 0.1652 0.336 4.0938
727
general trend of OM and DO. The schematic scenario (Fig. 5) is distinguished into three major
sub-areas, the southeastern part of <10 mg/l NO
3
-N and expected to contain high OM and low
DO contents. This favor the transport of the soil organic matter produced anoxic conditions, in
which the denitrification is predominant. The Holocene silty clay is concentrated in this area near
the Rosetta Nile branch, confirm the model output. The second sub-area of 10-40 mg/l NO
3
-N, is
characterized by moderate OM and DO contents. The last sub-area is in the western part of the
study area of >40 mg/l NO
3
-N, is predicted to be of little OM and much more DO content. The
oxic conditions are required for the persistence of NO
3
in groundwater. Starr and Gilham (1993)
showed that the depth to the water table is critical in maintaining oxic conditions and NO
3

stability. The low nitrate concentration in the southeastern part is likely resulted from a
combination of denitrification caused by abundant OM present in hypothetically fine-grained
aquifer and dilution resulted from recharge by precipitation and River Nile and good water
aquifers (reflected by flow lines). The magnitude and spatial distribution of NO
3
in groundwater
in agricultural watersheds is a function of numerous hydrogeologic and crop management factors
(Stephany et al., 1998). Crop type is linked directly to groundwater impacts because of differing
nitrogen-management practices. Agricultural drainage in areas of high water table (in the
northeast and northwest) (see Fig. 7d) can short-circuit transport to the water table. Subsurface
tile drains are installed at a depth of between 1 and 2 m, which imposes a greater threat to
surface water in drained watersheds because field tiles discharge to surface drains and ultimately
to streams.

0
Coarse-grained aquifercontaining
little organic matter
Medium-grained aquifercontaining
intermediate organic matter
Fine-grained aquifercontaining
abundant organic matter
Fine-grained confining unit
General groundwaterflow path
NO3 concentration, in mg/l
EXPLANATION
Pleistocene aquiferdue
north and Pliocene
due south
Pliocene clay
Wadi El Natrun
13 km


Fig. 5 Scenario of dissolved oxygen and organic matter trend in groundwater of the western delta
728
Thus much of the nitrogen would normally move slowly through the groundwater flow system
and is discharged directly to surface water bodies (Fenelon and Moore, 1998). These findings
have important implications for crop management practices, land use planning, and water
resources management. Where the high trend of OM in soil and sediment and the pathways of
groundwater flow enhance denitrification and otherwise limit the amount of nitrate in
groundwater that discharges to surface waters

Geochemical prediction of solid-solution phases interaction

The WATEQ family historically has addressed the problem of aqueous speciation and the
calculation of mineral saturation. Having determined the speciation, the calculations of ion
activity products (IAP) and saturation indices (SI) as log (IAP/K) are straightforward.
Calculation of a mineral saturation index (SI) is a convenient method of representing the
equilibrium condition of a solution with respect to a mineral. Saturation
index (SI) = log IAP/Kmineral
K is the solubility product defines as the solubility constant and calculated by the law of mass
action (Appello and Postma, 1993). By using the saturation index approach, it is possible to
predict the reactive mineralogy of the subsurface from the groundwater data without collecting
samples of the solid phase and analyzing the mineralogy. The purpose is to determine the
chemical potential of the groundwater at western Delta area, through the calculation of SI with
respect to minerals that could possibly be formed from these chemical compositions. The results
reflect its chemical potential for dissolution and precipitation with respect to a suite of minerals.
A range of saturation indices with respect to mineral between successive samples reflects
changes in the water chemistry due to stratification or mixing between two end members (El
Refaei, 2002). The saturation indices for calcite and dolomite generally do not correlated with
TDS (Fig. 6a). Nearly all water samples is in equilibrium or supersaturated with respect to both
minerals, an indication that carbonate minerals equilibria could be controlling concentrations of
calcium, magnesium, and bicarbonate in water in the aquifer system, in addition to the
agricultural waste water shares in the environment. Two components can be estimated from the
Fig. 6a.

1- The water sample with SI=1 is suspect because it is unlikely that groundwater would be this
far over-saturated with such a reactive mineral as calcite
2- Water samples with SI>1 for calcite, indicate that the analytical data and sample pH should be
checked for errors, as well as the input dataset for the SI calculation. In instances such as these,
the calculation of the SI identifies an error in sampling, analyzing or recording data.
Gypsum saturation indices for groundwater (Fig. 6b) increase with increasing sulfate
concentrations. Gypsum would tend to dissolve in water that is undersaturated with respect to
gypsum. Sulfate concentrations increases along flow path until saturation with gypsum occurs
and therefore serves as a Proxy for distance from the recharge area (Kehew, 2001). The general
oversaturation conditions of calcite through the flow system indicate that only a small amount of
anhydrite is necessary to produce oversaturation. The saturation index of dolomite is variable
(Fig. 6b). The saturation indices of selected minerals are illustrated in Fig. 6c, the groundwater is
all undersaturated with respect to halite and anhydrite, and saturate to supersaturated with respect
to aragonite. The saturation indices for calcite vs. dolomite (Fig. 6d), reflect the relatively flow
system length.




729


















































F
i
g
.

6

S
e
l
e
c
t
e
d

r
e
l
a
t
i
o
n
s
h
i
p
s

b
e
t
w
e
e
n

s
a
t
u
r
a
t
i
o
n

i
n
d
i
c
e
s

a
n
d

T
D
S

a
n
d

S
O
4

o
f

t
h
e

g
r
o
u
n
d
w
a
t
e
r

730
The saturation indices of selected minerals are plotted on a single diagram (Fig. 7a). Although
this type of plot does not as easily allow for identification of trends at a site, it does provide a
good method of displaying a considerable amount of data. Groundwater at this site appears to
be in equilibrium with all minerals except halite, which show the SIs separation with
anhydrite. It also indicate the saturation indices of evaporite minerals (anhydrite, gypsum, and
halite) decreased, while those for carbonate minerals are improved. Fig. 7a demonstrates the
groundwater represented by several chemical analyses has a qualitative resemblance in their
saturation levels that confirm its genetic similarity.

Fig. 8a, b, and c clarify the dissolution/precipitation conditions in the aquifer. Calcite,
dolomite, and aragonite are dissolved in most of the studied locality, while minor sites are
characterized by precipitation. For instance, Mg
2+
ions from the dolomite in the aquifer are
exchanging with Ca
2+
ions from the water in the void spaces. The subsaturation state of
calcite, dolomite, and aragonite prevails in the northwest and southeastern parts of the study
area. The left half side of the map (Fig. 8a, b, and c) reflects the entrance of intensive
irrigated water at Nubaria project taking the same directions of the groundwater flow,
therefore, the linking between the hydrogeology and hydrogeochemistry (Saturation index
values) is obvious. On the contrary, the right half side is fail to link between hydrogeology
and hydrogeochemistry. The hydraulic interconnection between Recent, Pleistocene,
Pliocene, Miocene, and Oligocene aquifers is facilitated by the free water table conditions.
The last side is recharged from the River Nile because of geographic position. Also the
vertical and horizontal changes in hydraulic condictivity favors the dispersion process of the
saturation indices through longitudinal and transverse directions, lead to decline in the
saturation index values with groundwater flow in the right half side. This fail is confirmed by
the very weak correlation between groundwater level and both TDS and ground level. These
saturation ranges for all mentioned minerals is supported by the presence of limestone,
dolomite, and evaporites in the surface exposure and lithostratigraphic sequence of the
aquifers. Table 3 presents the predicted sources of dissolution/precipitation as derived from
the WATEQp and the corresponding lithology.


Minerals/ S. I. Corresponding lithology
Aquifer Pliocene Pleistocene Miocene
Anhydrite -3.129 to 1.04
Grey clay of Gar El Muluk
Fm. or gypsous
carbonaceous Clay of Wadi
El Natrun Fm.
(Al Gamal et al. 2002)



Clay lenses of deltaic



Gypsous sandstone
Gypsum -2.86 to 0.77 deposits
Calcite -0.25 to 1.21

Reddish brown limestone


Reddish calcareous
loam


Limestone intercalating
sand clay of
Dolomite -0.9 to 2.21 of Alam El Khadim Fm. derived from Miocene Gebel El Khashab Fm.
Aragonite -0.39 to 0.66 rocks
Halite -6.81 to 4.5
Grey clay of Gar El Muluk Fm.
Or clayey beds in Alam
El Khadim Fm.
Clay lenses of deltaic
deposits
Grey clay of
Gebel El Khashab Fm.
Table 3 Sources of dissolution/precipitation with emphasized on the lithology




731













































Fig. 7 Saturation index of multiple minerals, dolomite dissolution trend
and the dedolomitization model of the groundwater
732





Fig. 8 Distribution of saturation index of selected menierals and water levels of the groundwater
and the topography

Dolomite dissolution and dedolomitization models

The reaction of calcite and dolomite with water and CO
2
can write as follows:
CaCO
3
(calcite) + H
2
O + CO
2
Ca
2+
+ 2HCO
3

CaMg (CO
3
)
2
(dolomite) +2H
2
O + 2CO
2
Ca
2+
+ Mg
2+
+4HCO
3

Typically CO
2
is derived from the oxidation of organic matter and root respiration in the soil
zone and is dissolved by recharge water. If dolomite dissolves according last equation, the
molar proportions of [Ca+Mg] to [HCO
3
] should be linear, with a slope of 0.5 (Sandra and
George, 2000). The majority of the water samples shows very close agreement (see Fig. 7b)
between simple dolomite dissolution model and the sample trend. On the other hand, few
water samples (samples 9, 23, 24, 28, and 30) shows a lack of agreement with simple
dolomite dissolution model, indicates that other processes control the chemistry of these
samples. Two predicted reactions may be producing the fore-mentioned water samples,
dedolomitization (dissolution reactions with carbonate minerals and gypsum) and sulfuric
acid neutralization (dissolution of carbonate minerals with sulfuric acid generated by the
oxidation of pyrite). On a plot of [Ca+Mg] as a function of [SO
4
+0.5HCO
3
], the
dedolomitization reaction yields a straight line with a slope of 1. The dedolomitization trend
was studied for the water samples (see Fig. 7c) and delineates samples 9, 28, and 30 depict a
good correlation with the dedolomitization model. The dedolomitization process is play a
minor role that restricted to local area, while the dolomite dissolution model is more common
733
in the study area. Samples 23 and 24 are disagreement with both models, suggest that these
samples neither suffer the dissolution of dolomite nor dedolomitization.

The FLOWNET model

The FLOWNET model was used to simulate the groundwater flow through the lithological
cross section from the northwest to the southeast of the study area in the western Delta. The
input data is hydraulic heads, boundary conditions, vertical and horizontal hydraulic
conductivity for Pleistocene aquifer and Holocene aquitard, effective porosity for both, length
of the profile, and thickness of the section (aquifers). The horizontal hydraulic conductivity of
the Pleistocene ranges from 12 to 80 m/d, while in the vertical direction ranges from 0.09 to 3
m/d. The horizontal hydraulic conductivity of the Holocene aquitard ranges from 0.38 to 3
m/d, while in the vertical direction ranges from 0.002 to 2 m/d. The estimated average time of
groundwater to travel from the highest potentiometric area (Nubaria mound, in the
northwestern part of the study area) to the southeastern part, in Pleistocene aquifer, is 6500
yrs, with average velocity about 9 m/yr. The conductivity of the aquifer and much lesser
thickness of the aquitard in this profile facilitate the high velocity.

Conclusions and recommendations

Grain size distribution and organic matter, recharge rates, flows, and discharge rates, are
primarily controls on the flow and chemistry of groundwater. The sorption-desorption
processes play an important role in the distribution of the contaminant plume, depending on
pH and the changes in the physio-chemical conditions of the environment. Statistical analyses
have helped to establish the cause and effect relationships of the groundwater quality. HCA
technique was judged more effcient than others, becaues it offers semi objective graphical
clustering procedure (dendrogram), which does not require pre-determining the final numbers
of clusters. Anthropogenic sources are responsible for almost all cases of high trace metal
concentrations in soil, deposits, and groundwater. The main source is metal containing
pesticides, fertilizers, and phosphate fertilizers, in addition to repeated application of
pesticides over many years have resulted in very high soil metal concentrations. In most cases
of trace metal contamination of agricultural soil, the impact on crops and on drainage water
quality is mitigated by the normal management and by the farmer. Practices such as periodic
liming and maintenance of high surface organic matter levels, together with the generally
better quality of agricultural soils-deeper, medium to fine textured, relatively high cation
exchange capacities, in most cases, to favor trace metal attenuation in the soil. In aquifers that
allow high pump rates like such circumstances elevated concentrations of organic carbon and
small amounts of clays or dissolved solids can enhance the movement of microorganisms
(Conboy & Goss, 2000) predicted from the farm practices. It is recommended to apply the
India model for abstraction rate that is estimated by scientific committee, otherwise the farmer
pay money or inter jail if he dont obey.

References

Al Gamal, S., Korany, E., Nada, A., and Al Sayed, S. (2002). Conceptual hydrochemocal
models using environmental isotopes data for Pliocene and Pleistocene aquifers in wadi
Al Natrun area, Egypt. Egyptian J . of geology, v. 46/1, pp. 301-316.
Appelo, C.A. and Postma, D. (1993). Geochemistry, groundwater and pollution.
A.A.Balkema, Rotterdam, Netherlands, 536p.
734
Apte, S.C., Gardner, M.J ., and Ravenscroft, J .E. (1990b). An investigation of copper
complexation in the Severe estuary using differential pulse cathodic stripping
voltammetry. Mar. Chem. 29: 63-75.
Back, W. (1966). Hydrochemical facies and groundwater flow patterns in the northern part of
the Atlantic coastal plain. US Geological Survey Professional Paper 498c, 16p.
Conboy, M.J . and Goss, M.J . (2000). Natural protection of groundwater against
bacteria of fecal origin. J . of Contaminant Hydrology 43: 1-24pp.
Collins, A. (1975). Geochemistry of oilfield waters. Elseviers Amsterdam, 475p.
Davis, J . (1986). Statistical and data analysis in geology, 2
nd
edition. NY, 646p.
El Refaei, A. (2002). Geochemical modeling of groundwater in the New Minia city, Egypt.
Egyptian J . Of Geology, v. 46/1, pp. 317-328.
Fuller, C. and Davis, (1987). Processes and kinetics of Cd
2+
sorption by calcareous aquifer
sand, Geochim, Cosmochim, Acta 51: 1491-1502.
Gary, S., Hamilton, P., and Woodside, M. (2001). Natural processes for managing nitrate in
groundwater discharge to Chesapeake bay and other surface waters more than Forest
buffers. U.S. Geological Survey, Water Supply paper-178, 13p.
Hart, B. T., and Davies, S. H. (1981). Trace metal speciation in the fresh water and estuarine
regions of the Yarra River, Victoria. Estuarine Coastal Shelf Sci. 12: 353-374.
Kehew, A. (2001). Applied chemical hydrogeology. Hall Inc., New J ersey, 368p.
Kuwabara, J ., Chang, C.Y., Closern, J .E., Fries, T.L., Davis, J . and Luoma, S.N. (1989). Trace
metal association in the water column of south san Francisco Bay , California, Estuarine
Coastal Shelf Sci. 28: 307-325.
Paulson, B.J . Feely, R.A., Curl, H.C. and Gendron, J .F. (1984). Behavior of Fe, Mn, Cu and
Cd in the Duwamish River esturay downstream of a sewage treatment plant. Water Res.
18: 633-641.
RIGW/IWACO, (1990). Vulnerability of groundwater to pollution in the Nile Valley and
Delta, TN 70.130-89-02, Cairo.
Sandra, M.B. and George, L.L. (2000). Regional groundwater flow and geochemistry in the
Midwestern basins and Arches aquifer system in parts of Indiana, Ohio,
Michigan,Illionois. US geological Survey, Professional paper, 1423-C, 100p.
Shabaan, M. (2002). Environmental impacts of development on the hydrogeology and
hydrochemistry, west of Delta, Egypt. Ph.D. dissertation, Fac. of Sci., Cairo Univ., 208p.
Sholkovitz, E.R. (1978). The flocculation of dissolved Fe, Mn, Al, Cu, Ni, and Co and Cd
during eastuarine mixing. Earth Planet Sci., Lett., 41:77-86.
Starr, R. and Gilham, R. (1993). Denitrification and organic carbon availability in two
aquifers. GroundWater 31, 934-947p.
Stauffer, R. and Wittchen, B. (1991). Effects of silicate weathering on water chemistry in
Forested, upland, felsic terrane of the USA: Geolchimica et Cosmochimica Acta, v. 55, p.
3253-3271.
Stephany, C., Kirby, M., and Kehew, A. (1998). Spatial and temporal variability distribution
of nitrate in an agriculturally impacted aquifer, Cass county, Michigan, Proceedings
Animal Production Systems and the Environment. Ames, IA: Iowa State Univ., 513-
518p.


735






APPLICATION OF DOUBLE CONTINUUM
POROSITY EQUIVALENT METHOD TO
INVESTIGATE THE KARST PROBLEM OF SALMAN
FARSI DAM IN IRAN

N. Talebbeydokhti, Professor of Civil Engineering
Shiraz University taleb@shirazu.ac.ir

A. Shamsai, Professor of Civil Engineering
Sharif University

Mohammad Hadi Fattahi, M.Sc., Graduate Student.
ABSTRACT:
Karst is a dry landscape situated on soluble carbonate rocks. Limestones
and dolomites with underground flow of water. Although karst terrains of the
world make a relatively minor percentage around 10% of the earth surface but
they present the most risk producing conditions for a fulfilling in project
objectives among the other geological formations.
In this research we are after to figure out the karst main problem,
seepage, in dams and reservoirs and study this phenomenon numerically. The
complexity of the karst system problems in Salman Farsi dam site is so serious
that a variety of remedial measures has been presented. Among those five main
suggestions are chosen to be modeled.
Choosing the best numerical method to analyze the karst problem needs a
complete study of all possible ways. After a full introduction on different
modeling methods we have a full investigation about Salman Farsi dam geology
and hydrogeology conditions. Complicated karst caves and caverns systems and
the dam right abutment fractured formation had encountered a lot of problems
for the dam water proofing. Finding the best water proofing method was an aim.
The numerical method chose for this research is Double continuum
Porosity Equivalent (DCPE). Our most important objective is to model the
seepage flow from Salman Farsi karstic formation. The visual MODFLOW
program has been used to grid the area and the dam characteristics inputting
data. The different suggested technical methods for water proofing are modeled
with DCPE and the results are compared, analyzed and tabled. The DCPE
method has three kinds of output which are the base of J udgment. "Constant

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


736

head OUT", "Other porosity" and "Total IN_OUT" are the program results
which the last one has been chosen as the criteria of choosing the best seepage
control method presented according to the minimum amount of seepage
downstream. The procedure, flowchart, Analysis, results and conclusion and
suggestions are presented in this paper.

INTRODUCTION:
Karstifed limestone terrains are very prolific sources of ground water. Their flow
characteristics have been studied in much detail at numerous locations. The complexity of the
flow system is primarily determined by the degree of certification.
The complexity of karst as a geological phenomenon is a difficult in studying as its
engineering karst problems. Karst with an estimated 10% over covering of the earth surface is
a vast case of research. Considering the day increasing need of man to water sources and the
optimal use of karst researches have promoted recently.
The water leakage from dams and reservoirs due to the karst effects and its features
such as fractures and holes and sinkholes and ponors occurs mostly after the priming. In some
dams this water loss exceeds to even 30m
3
/sec. This amount of water is like a question mark
in front of all the dams objectives. Its clear that finding, analyzing the problems and of
course the remedies of the case is so important.
Although the karst terrains consist only a small percentage of the earth surface, but
from the development and controlling the water sources and also running the hydro projects
are of the most complex and risky ones in the world and among all the other geological
formations. The importance of karst studying is considered from two points of view. First is
the better quality of these water sources and their less pollution and also their considerable
storage volume. Second is investigating the karst problems especially in regard to the hydro
projects. The karst subjects would be more important to us the time we know in a dry country
such as Iran and almost poor in rainfalls, there are a lot of karst formations full of water witch
can provide our needs.
The objective of this research is to study salman Farsi dam karst problems. This dam
located in Iran has encountered a lot of problems. The complexity of karst systems in site
area is of that range which has caused different theory about the remedial works. Studying
and analyze these theories and finding the best treatment is one objective in the case. Our
most important aim of this research is studying the various numerical models used to analyze
a karst media. As you know the complexity of karst area and its principle differences in flow,
storage and governing equation with a non karstic area and the troubles of the Darcy equation
in modeling have made the case so complicated. We have studied the different karst
numerical models and suggested the best one according to the advantages and disadvantages
of each. Among all the researched models, double continuum porosity equivalent (DCPE) is
the most modern and complete one. Using this model and the visual MODFlOW support, the
dam site is modeled and the seepage results are presented through program runs. All the
results are compared and presented at the end.

LITERATURE REVIEW:
Considering the karst complicacy, studying and research to solve its problems is a
great need nowadays. The greatest J ump in karst research for sure has been the numerical
movement from the last years of the twentieth century and up to this century. The first sparks
have been lightend by Sauter, Mohrlock, Teutsch and the others. Creating numerous models
to analyze the ground water flow in karst regions shows the importance of the case for the
scientists. Different researchers, scientists and karstmen throughout the world such as, Ford,
737

Breznik, Beck, white, sweed, Herake, Duinlan, J ennings, Wilson, Milanovic, Vlahovic,
Stringfield, parizan and many others have preformed a vast attempt to know karst, its
hydrology and hydrogeology and its different problems and of course the remedies in the
recent 30 years. But karst research is a nonfullfillment case of studying.


KARST NUMERICAL MODELS:
Two fundamental approaches to modeling karst water resource systems have been
considered. The black box model and distributed model.

BLACK BOX MODEL
In this model, the input data are given to a system and as a result of the system
function the output data are obtained. This system doesnt encompass a physical source and
just its results are considered. As a result of the scarcity of spatial data, the heterogeneity of
the aquifer parameters and because of its relative simplicity a black box approach has been
frequently used to study the simulation of karst aquifers. The used methods in these models
are various. Like the recession analysis (Atkinson, 1997), Transfer/kernel function (Dreiss,
1979), Mixing cell model (Simpson, 1988) and Simple Regression Analysis (Zaltsberg,1989)
but according to what described about these models, as dealing with a nonphysical nature and
condition, they can not be used to predict the aquifers internal formation, in other words, they
lack the predictive power.

DISTRIBUTED PARAMETER MODELS
The inadequacies of the black box models become apparent when one attempts to
model spatially variable output phenomena, such as characteristic water level fluctuations,
that have a definite physical basis. Frequently, geological information that could explain
observed differences, which in many cases is of a spatial nature has to be ignored in such
models. Moreover, they fail to consider the different processes that determine flow and
transport in a karst aquifer, i.e., the mechanism of ground water recharge, the influence of the
unsaturated zone and the phenomena in the aquifer itself. Furthermore, each of these factors
in turn has a different influence on fast and slow flow components.
A viable alternative is the distributed parameter modeling approach. Three major
methods have been used to describe the flow and transport through fractured porous media,
(1) the equivalent porous medium, (2) the discrete fracture and (3) the double porosity or
double continuum approach.

EQUIVALENT POROUS MEDIUM
When the fractures are narrow, evenly distributed, and if there is a high degree of
connectivity, an equivalent porous medium model (EPM) can be applied (Pankow et al.,
1986). A recent development for these kinds of model has been described by Newman
(1987). Sauter (1988) simulated the flow in the fractured porous rocks with an equivalent
porous medium model, integrating the permeability as a stochastic variable. Aquifers termed
"Deep Continuum Flow Carbonate Aquifers" by Thrailkill (1986) seem to fall into this
category. Most of the present karst aquifer models have been based on the assumptions of
the EPM.

DISCRETE FRACTURE MODELS
The second approach, the discrete fracture model usually implies that the effect of the
matrix is neglected. In its simplest form, flow is simulated by considering it as flow between
738

two parallel plates and it requires some detailed information of fracture apertures, fracture
length, density, orientation and connectivity (Snow, 1965; Irmay, 1964; Rom, 1966).
Especially as it is very difficult, if not often impossible to obtain the required
information at the relevant scale, the fracture networks are statistically simulated (Long et al.,
1985; Smiths et al., 1987). Kraemer and Haitjema (1989) theoretically studied the effects of
large fracture zones and statistically generated fracture networks for regional aquifer flow.
Although the theory for joint controlled flow is outlined in many textbooks and
publications (White, 1989, Ford and Williams, 1989, Cullen and LaFleur, 1984) and although
the laws and parameters governing turbulent open channel flow and pipe-flow were described
in detail, to the authors knowledge, these principles have not been applied to real aquifer
situations, except by Atkinson (1977) and Thrailkill (1991).

FRACTURE FLOW MODEL CONCEPTS:
A variety of modeling concepts have been proposed for the simulation of the
hydraulic behavior of fractured rocks. The concepts are based on either a discrete or
continuous representation of the fracture flow system, together with equation describing the
exchange between the fractures and the rock matrix.
In the discrete fracture approach, the location and the geometry of the fractures are
assumed to be either exactly known, or at least describable through a statistical model.
Several analytical and numerical studies have been conducted employing the discrete fracture
approach, including early laboratory and theoretical works by Wattle & Louis (1968),
Gringarten (1971), Grisak & Pickens (1980), and Smith & Schwartz (1984). Most of these
studies dealt with crystalline rocks for some other low-permeability system. It was therefore
in general assumed that groundwater flow occurs in the fracture system only, the adjective
flow within the porous matrix being negligible.
The major drawback of the discrete fracture approach is the necessity to describe the
geometry and hydraulic properties of all individual fractures within the flow domain. For
most regional aquifer systems this is not feasible in practice. So far, only a few very
intensively investigated radioactive waste disposal sites of very limited areal extent have been
characterized using the discrete fracture modeling approach. Bearing in mind that fracture
permeability is proportional to the square of the fracture width, no reliable regional rock
permeability assessment is to be expected using indirect underground investigation
techniques available today (e.g., geophysics). In the continuum approach, it is assumed that
the fractured medium can be described by one or more porosity equivalents. Often a double-
porosity approach is employed, one porosity representing the low-permeability porous or
fractured rock matrix, the other porosity representing the higher permeability fracture system.
The double-porosity model concept was introduced by Barenblatt et al. (1960). It was
extensively used and further developed in oil industry studies including work by warren &
Root (1963), Kazemi (1969), De Swaan (1976), Streltsova-Adams (1978), and van Golf-
Raacht (1982). At first, the magnitude of the cross-flow between the matrix and the fracture
system was assumed to be directly proportional to the lumped pressure difference between
the two systems.

DOUBLE POROSITY THEORY
The modeling of flow and transport in fractured rocks could benefit greatly from
studies carried out by petroleum engineers in the field of reservoir engineering and from the
research efforts in the search for safe repositories for radioactive wastes. The preferred
approach in these areas of research has been the double porosity approach (Barenblatt et al.,
1960; warren and Root; 1963; Duguid and Lee, 1977). The fractured medium and the porous
matrix blocks are modeled as to separate over lapping continua, each with its own flow
739

equation. The exchange of flow is controlled by the local difference in potentials. The
coupling of the two media is handled with a source/sink term in each equation. Details and
the development of the double porosity approach are described by strelsova (1988),
Sauveplare (1984) and Teutsch (1988), Narasimham (1988), developed this concept further
into multiple porosity systems and called the model MINC (multiple interacting continua). In
double porosity theory, the porous Hetrogen media is assumed as two separate systems that
are overlapped.
1. The fractured media with a low-permeability with its hydraulic characteristics are
regarded to the primary porosity
2. The system that includes the conduits with high permeability which is the secondary
porosity that its hydraulic characteristics are controlled through the thermal stresses and
tectonic events. Considering the conduits more openings, permeability and the flow
velocity in this system is much more than the fractured system. As in karst media a
double nature exists and for each of them a differential equation is available, Monech et
al. (1984) presented a REV (Representative Element Volume) for each of them. Applied
to a karst aquifer, the terminology of the double porosity or double continuum approach
requires some further clarification: The terms fractures and conduits are used
interchangeable for the fast transit system, and matrix and fissures are used to describe
the slowly draining part of the aquifers system. Although generally used in the literature,
it is preferred to use the term "fissured system" as opposed to "diffuse system", because
"fissured system" describes the actual permeable pore space of the slow system,
corresponding to the term "conduit system". The term "fractures" is used synonymously
for conduits in order to reflect the duality fracture/matrix of the double porosity
/continuum approach. The same terms apply to the unsaturated zone, in particular to the
epikarst. In double porosity theory, these are two hydraulic conductivities (k) and two
storage coefficients (s) assumed for each point. One "k" is referred to the matrix media
and the other one to the conduits. And there is the same situation about "s. The
relationship between these two can be expressed by this equation in which Q

is the
discharge between the systems.

Q

=- (h
a
-b
b
)

The above equation is taken under the pseudo-steady state and straight relation
between the two systems water head. In which, is the specific conversion factor (1/(T))
and (h
a
-b
b
) is the water head offset between two systems. There are differential equations
for each media in three dimensions as below:

( )
b a
a
a
a
a
zz
a
a
yy
a
a
xx
h h
t
h
S
z
h
K
z y
h
K
y x
h
K
x
+

0


( )
b a
b
b
b
b
zz
b
b
yy
b
b
xx
h h
t
h
S
z
h
K
z y
h
K
y x
h
K
x
+

0


h: water head
K
xx
, K
yy
, K
zz
: Hydraulic conductivity in 3 directions
S: storage coefficient
: Conversion factor between the two systems
S
s
: specific storage coefficient
"a" stands for conduit and "b" for matrix.
740

The double porosity model is prepared in different version. For this research, we have
tried the MOHRLOCK version which is the most complete and exact one.

GEOLOGICAL AND HYDRO GEOLOGICAL CONDITIONS

General Geological Conditions
The dam site is located in the northern flank of the Changal Anticline in the Asmari
limestone formations. The core of the Changal anticline consists of the Paleocene/Oligocene
Pabdeh Formation. Lithologically, this formation is a cherty, fossiliferous and conglomeratic
limestone, passing into shale, interbedded with thinly bedded marly limestone. This
formation is not prone to karstification. From the hydrogeological point of view, this
formation is watertight and can be declared as a regional hydrogeological barrier.
Overlaying Lower Asmari Formations (Oligocene to Early Miocene) consist of
regularly bedded limestone with and alternation of numbers of thin marly beds in the base.
The upper subunits of Lower Asmari are prone to karstification. Hydro geologically, the base
sequence (Subunit 24) can be declared as generally watertight.
The Middle Asmari (dam foundation area) is composed of marly limestone,
intensively karstified with vuggy porous crystalline nummulitic limestone. Large karst
features are developed in this sequence.
The structural and hydrogeological homogeneity of the northern flank of the Changal
Anticline is tectonically disturbed with a few rupture systems. Five of those systems were
detected in the field, but most important are the J 1 and J 2 systems. The intensity of crushing
along those systems and the aperture of openings varies from only a few centimeters to a few
meters in wide mylonite zones. The subvertical dip favors the ground water circulation by
gravity, provoking the karstification process. J -systems are frequently close to the gorge (dam
site) area. Because of the presence of shaly and marly interbeddings in the anticline core
those ruptures, when crossing the core, are compressed and impervious. Any water
movement penetrating through these formations is not possible.
The characteristics of the bedding planes are different. Some of them are very
compact without the possibility for water circulation by gravity. However, along many of the
bedding planes discontinuities are developed. In some cases, especially in middle and Upper
Asmari, interbeddings consist of thin clayey layers with a thickness of 1cm to 20cm.

GENERAL ASPECTS ABOUT KARST FEATURES
General characteristics of investigated karst channel at the Salman Farsi dam site are:
- Karst channels are developed along the dominating fractures (interbedding J -1 and J -
2). All channels and caverns are very poor with speleothems.
- The caverns and channels are very dry (permanent or temporary watercourse do not
exist above the river elevation).
- Hot microclimate (air temperature in natural conditions ranges between 30C - 35C)
On the base of investigated and observed karst features the following groups of
caverns are distinguished:
- Large caverns filled with piles of huge limestone blocks. In the lower sections of
those caverns, the blocks are surrounded by clayey matrix. The bottom of a solid rock
mass is mostly unknown.
- Caverns filled with well stratified, compressed and plastic clay are present at both
sides. Some of them are almost fossilized. The volume of those caverns varies from a
few cubic decimeters till a few hundreds of cubic meters. The clay color is different:
gray, reddish, dark brown and beige.
741

- Karstic channels, mostly subhorizontal are blanketed with dry clay deposits along the
bottom.
- Vertical or subvertical karst channels, circular or lenticular, are without any deposits
or speleothems.
The objective of the speleological investigations at the Salman Farsi (Ghir) dam site
was focused on morphological and local tectonically characteristics of the discovered
caverns. An analysis of the speleogenesis was not foreseen in this stage of investigations. A
comprehensive analysis of speleogenesis would need additional field investigations and
laboratory tests.

The karst problem in Salman Farsi dam
The karst features in the dam site consist the cave and caverns and complicated
fracture systems. Other karst features like Karrens and sinkholes and swallow holes are not
distinguished in the area. There are two main karst spring at the two flank of the dam site.
The one at the right is a cold water spring and at left there is a warm water hydrothermal karst
spring which is assumed to be the reason of karst features of this side, however the karst
features of the right side is said to be due to a fault zone on the right abutment of dam. As it is
said before the main karst features are internal ones. The most karstified part is the right
abutment with thousands of fractures and shafts and caverns and chimneys and caves. The
most spectaculars are and the most troublemaking karst feature in this side is the karstic
Golshan cave with a volume of about 150000m3. This space occupies a huge system of karst
caves and shafts which is regarded as the most intensive karst problem in the dam site.
(Figure 1).




Figure (1): lateral cross-section of Golshen cave in Salman Farsi Dam

The extent of karst found calls for a systematic decision making process about its
treatment. With the result of the speleological investigations available, such a process is now
possible. It was therefore found meaningful to screen reasonable options for treating this area
in a systematic manner, compares the different basic options selected from this screening.
More than is 15 treatment options are suggested which among them five options present
themselves in either sticking to the present alignment or bypassing the cave either upstream
or downstream and one choice continuing the alignment through the 15
th
Lithological unit.
The five selected options are as below:
1. Filling a part of the karsts and having a tall cut off wall while using the same grout
curtain.
742

2. Filling all the karsts and keeping the same grout curtain
3. upstream by pass
4. Down stream by pass
5. Continuing the grout curtain to 15
th
lithological unit.

Option 1 is hardly feasible because of the following reasons: Passing the cave needs a
concrete structure which has to resist the full hydrostatic pressure of up to 110 m water
pressure. Given the caves maximum height of approximately 80m, this would be an immense
gravity wall within the cave.
The foundation of this wall is hidden by an unknown depth of blocky material. To
excavate all this material is not realistic. However, a gravity wall of this size has to be
founded on rock (and not on the blocky material).
Filling the cave by self compacting concrete (SCC) is no solution, given the unknown
bottom of the cave below the blocky material. There is no control whatsoever in placing
(SCC) over such a height. Sandy layers might be among the material under the blocks. This
was typically found in other caves. Such layers would be critical leakage paths whiten the
"treated" part of the cave open.
Since option 1 is not conceivable, further investigations cover options 3 and 5 only.
The technical and economical comparison has been done about them and is presented in table
(1).
Our very first and main objective in this research is to run the double porosity
numerical model analyzing the water leakage which is always the worst problem in
hydroproject to study its amount. Among the different options of this research and finding out
the minimum leakage occurs in these five options. No need to say that the option of the
minimum leakage would be the best choice of the model.

Table (1): Table of Technical and Economical Comparison
NO. Alternatives Karst
Volume
Final Cost
(Billion
Rls.)
Technical Limitations
1 Filling all the karsts and
having the same grout curtain
150000m
3
108 Caves bed materials are not
known and filling all the karst is
so expensive
2 Filling a part of karsts and
having a tall out off wall
while using the same grout
curtain
50000 m
3
50 Risk of building a more than 80
m wall on unknown bed
material and the hydrostatic
pressure on the wall
3 Upstream bypass 20000 m
3
27 Grout curtain gets near to the
lake
4 Downstream bypass 0 26 The cave submerges and
Hydrostatic pressure on the
grout curtain while the
nonactive karst systems will be
activated
5 Continuing the grout Curtain
to lithological unit number 15
0 60 The risk of encountering a new
karst system

Salman Farsi dam modeling procedures:
Among all the numerical models described for karst regions in this paper, double
porosity continuum is identified as the best one to estimates the amount of leakage from a
karstic bulk. To analyze the water leakage in this project and more specific in the right
abutment and Golshans cave system, the model has been ran. As it said before, these are five
743

remedial methods. The model has to compute the water leakage in each choice. The modeling
procedures are presented here.
After preparing the geological maps, we should provide a mesh (noding) of the area in visual
MODFIOW program. The grid consists an area of 500m200m which is meshed in 10m
paces. Then the hydraulic specifications of the region such as hydraulic gradient and the rate
of storage and the characteristics of the stone layers are introduced to the program. The very
important factor that should be mentioned is the lugeon values of different sections of the
right abutment. Then, all the input data in visual MODFLOW are exported to the double
porosity model to be analyzed. MOHRLOCK program is the one that is used to work with in
double continuum porosity space.
Double porosity has three kinds of outputs. One is the "coustant head OUT" and the
other one "other porosity" and at last "Total IN-OUT". The base of decision and analysis is
the value of "IN-OUT" which shows the difference between the water given from the lake to
the abutment and the amount of water that has been kept in the area. The second amount of
water is called "other porosity" that indicater the water which is conversed between the lake
and the aboutment. But the leakage is assumed to be the water that has been added to the
river discharge downstream.
Here a flowchart that shows the procedures of modeling and running the DCPE
program is provided.






Figure (2): Salman Farsi modeling procedures flowchart

Start

Grid Golshans cave in MODFLOW


Lake constant head Input data "K" zones

Outlet stage of river Storage

Setting the primary and secondary
porosity according to max & Min
Lugeon



Running the conversion factor Exporting data from
MODFIOW to MOHR
LOCK


Running the program and getting the
results: Recharge, Other porosity &
Total IN-OUT



END


744

Conclusion:
To J udge about the leakage amount of each alternative expressed before, considering
two main element of "constant head" and "other porosity is necessary. Though the results of
other porosity in all the choices are so close to each other, it is possible to put the value of
"OUT constant head" as the base of J udgment.
According to the result in table (2), it is simply obvious that in the first choice of
filling the cave karst completely and continuing the same grout curtain, the minimum leakage
would be reached. And at the same rate, the fifth choice also shows the same amount of
leakage. The third choice also shows the same amount of leakage. The two other choices
show higher rates of leakage. According to what expressed before and mentioned in economy
a technical approach the first and the fifth choices arent suitable cases for this choice. So as
the leakage amount in third choice is as the same rate of the two others and besides its a
cheaper choice for running the upstream grout curtain, this choice is regarded as the best
chosen alternative for Golshans cave. The results are shown in table (2).


Table (2): The DCPE model results
Alternatives In Out
No
Constant head Recharge Constant head Other porosity
In-Out
1 Filling all the karsts and
having the same grout
curtain
0.1301*10
-1
m
3
/s 0.5099*10
-2
m
3
/s 0.7844*10
-2
m
3
/s 3648 m
3
/day 0.5615*10
-1
m
3
/s
2 Filling a part of karsts
and having a tall cut off
wall while using the
same grout curtain
0.1312*10
-1
m
3
/s 0.5306*10
-2
m
3
/s 0.8692*10
-2
m
3
/s 3640 m
3
/day 0.574*10
-1
m
3
/s
3 Upstreambypass 0.1311*10
-1
m
3
/s 0.501*10
-2
m
3
/s 0.7845*10
-2
m
3
/s 3624 m
3
/day 0.554*10
-1
m
3
/s
4 Downstreambypass 0.1315*10
-1
m
3
/s 0.5129*10
-2
m
3
/s 0.864*10
-2
m
3
/s 3659 m
3
/day 0.559*10
-1
m
3
/s
5 Continuing the grout
Curtain to lithological
unit number 15
0.1301*10
-1
m
3
/s 0.501*10
-2
m
3
/s 0.789*10
-2
m
3
/s 3624 m
3
/day 0.554*10
-1
m
3
/s

Suggestions:
1. Creating the lag time as in nature for the software. The model is designed to transfer
water through the Matrix cell into the conduit. Although the same thing happens in
nature, but it takes time not simultaneously. However the coefficient which is
designed to work out the same sake is not successful.
2. The sorting of matrix and conduit cells in this program is also one of the troublesome
making ones. Because in nature, the real system works differently and it causes two
troubles.
One: As we said the most part of flow in a karst area is transmitted through a small
percentage of it around 10%, but the program default doesnt provide it.
Two: The matrix and conduit cells are designed to be one by the other one. But in a
real karst system just the boundary cells of them are neighbors and the other matrix
cells are bulk and conduit cells form a conduit.
3. Maybe the assumption of a middle cell (not matrix/not Conduit) something a
metaphase cell (not 1/not 0) solves all the above troubles.
4. Using other models to analyze the karst water flow and comparing the results. For
example the CAVE hydrogeological model or the Discrete fractures Model.
5. Double porosity model is just used for regional works, developing the model to be
qualified for even global works is also a good suggestion for researchers.
6. Karst modeling is a new and a never ending branch of groundwater study and it needs
to be considered more.

745

References:
1. Altug S. 1976. Leakage Study of the West side of the oymapinar Reservoir, Turkey.
Bulletin of the International Association of Engineering Geology, No. 14.paris.
2. Bergado T.D. Areepitak C., Prinzl F. 1984. Foundation problems on karstic limestone
formation in western Thailand-A cases of Khoao Leam Dam. Florida, USA.
3. Bozovic A., Budanur H., Nonveiller E., Pavlin B. 1981. The keban dam foundation on
karstified limestone- A case History. Bulletin of the International Association of
Engineering Geology. No. 24.
4. Briznik M. 1982. Exploration, design and construction of Cut-offs in karst regions.
Commission International Des grandees brands barrages, swifter land.
5. Breznik M. 1985. Water and karst. Scientific conference.
6. Mostar, Bosnia and Herzegovina.
7. Beznik M. 1993. Exploration of Underground water reservoirs in karst. Int. symp. On
water Res. In karst, Shiraz, Iran.
8. Breznik M. 1998. Storage reservoirs and deep wells in karst regions. A. A. Balkema
publishers, Netherlands.
9. Cvijic J . 1918. Under ground Hydrology and Morphological evolution of karst. Society,
Vol35, Belgrad, Yogoslavia Djalay H. 1988. Remedial and water tightening of Lar dam.
Seizieme congress has Grandees Barrages, Sam Francisco. USA.
10. Ewert F.K. 1985. Rock Grouting with Emphasis on dam sites. Springer verlag.
11. Ewert F.K. 1992. The individual grouting of rocks. Water power & dam construction.
12. Ford D., Williams P. 1989. Karst geomorphology and hydrology. Unwin Hyman,
London, UK.
13. Herak M. 1976. Geological approach to classification of terrains for consolidation
treatments. Proceeding, I J USIK 76, Zagreb, Yugoslavia.
14. J ennings J .N. 1971. Karst. Australian notional university press, Canberra, Australia.
15. Milanovic P. 1977. Reservoirs in karst-common water tighening problems. Karst waters
& environmental impacts, balkema.
16. Milanovic p. 1981. Karst Hydrogeology. Water resources publications. Littleton,
Colorado, USA.
17. Milanovic P.l, Aghili B. 1990. Hydrological Characteristics and ground water
mismanagement of Kazerun Karstic aquiver, Zagros, Iran. IAHS publication No. 207.
18. Milanovic P. 2000. Geological Engineering in Karst: dams, reservoirs, grouting, ground
water protection, water tapping pummeling. Zebra publication, Belgrade, Yugoslavia.
19. Milanovic P., Keruzer H. 2002 Salman Farsi (Ghir) dam drilling and Grout curtain
investigation report mission report, Mahab Ghods Co. Publication Tehran, Iran.
20. Milanovic p., Kreuzer H. 2003. Salman Farsi (Ghir) dam serological investigations,
mission report, pancake klub, Beograd, Yugoslavia.
21. Nonveiller E. 1989. Grouting, theory and practice. Elsevier, Amsterdam, nether lands.
22. Saunter. Martin. Quantification and fore casting of regional ground water flow and
transport in a karst aquifer. 1992. Page 28-32.
23. Teutsch. George. An extended double-porosity concept as a practical modeling
approach for a karsit field terrain,. 1993.
24. White B. W. 1988. Applied Hydrology in karst. UAH, Huntsville, Alabama, USA.
746







QUALITY OF GROUNDWATER IN AN SEFRA AREA,
WESTERN SAHARA ATLAS, ALGERIA.

A. M. Tabet Helal
1
; E.F. MAMI
2
; N. Ghella
3
Lecturer and researcher
Hydraulic chemistry and Environment
University of Tlemcen Algeria
Earth Sciences and Agronomy Department
Research Laboratory 25
B.P. 119 Tlemcen Algeria
Phone number: 00213 43 2116 47 / 00213 71 27 35 74
Fax number: 00213 43 20 66 08
1- am_tabet@yahoo.fr 2- e_mami2003@yahoo.fr 3- na_ghellai@yahoo.fr

ABSTRACT

The great spaces of the South in Algeria remain dependant on
subsurface water for agriculture and breeding needs as well as drinking water for
people and cities.
About 400 km from Oran on the coast, An Sefra area is located south
west of Algeria, in the Western Sahara Atlas which is a mountain range of about
1000 km long dividing the country into two parts (Northern and southern
Algeria) and being a barrier for the generating rain marine influence.
In this area, the aquifer strata are rather poor and the wells give low
discharge.
Human activity has a direct impact on groundwater in this area.
We suggest to show the evolution of water quality of this aquifer, its
potability, its use for agriculture and animal life and to evaluate the water
amounts which will be necessary in this area by 2020.

Key words: Western Algeria Sahara An Sefra Water potability Irrigation water.






FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


747
General Geographical Framework of the Saharian Atlas

Located at approximately 400 km in the South of Oran, the Saharian Atlas constitutes a 1000
km long mountain separating Northern Algeria from the Sahara. It is limited to North by the
High Oranian Plains, in the South by the Sahara, the West by the High Moroccan East Atlas
and in the East by the Western edge of the Mounts of Mzab.

It corresponds to lengthened chains; directed coarsely South-West, North-East, following one
another a width from 100 to 200 km. From West to East we have : Mounts of Ksour (Western
Saharian Atlas) of djebel Amour (Central Saharian Atlas) and Mounts of Ouled Nal (Eastern
Saharian Atlas).

Geographical Situation

The studied area is localised in the central part of the Mounts of Ksour and it is limited:
In the South by the anticline of the djebel Assa.
In the North by djebel Morhad.
In the South by djebel Mekter and the town of An Sefra.


Orography

The studied area belongs to the Mounts of Ksour;some tops reach more than 2000 m of
altitude like djebel Assa (2236 m), djebel Mekter (2020 m) and djebel Mir (2109 m); which
form a topographic and climatic barrier between the High Plains in the North and the Saharan
platform in the south.

Climate

The climate of the area is of continental type with very cold winter of short duration. This
latter is limited to J anuary and December during which the temperature can reach as low as -
15 C.


Hydrography

The Mounts of Ksour are badly drained in the septentrional part: the surface waters are lost
rather quickly in Dayas and Sebkhas where evaporation is intense. In the southernmost part,
the more developed hydrographic network is many Wadis which water are absorbed by the
sands of the Great Western Erg.


Vegetation

The vegetable cover is characteristic of the semi-arid climate, it is primarily made up of
armoise "chih" and esparto. The latter disappears gradually from North towards the South
where it is replaced by halas-cylon.
In altitude grow pines of Alep isolated or grouped in groves.

748

Geographical Situation of Ain Sefra in Algeria






















Tirkount Depression in Ain Sefra (Algeria)

























749
Geological Map of Tirkount Depression

Geological Cross Section of Tirkount Depression





















Scale 1/100 000
750
Climate

The climatic study is made on the basis of data of the station of An Sefra.
(Period 1982-2002)


Station Altitude (m) Latitude Longitude
An Sefra 1075 3246 Nord 0036 West


Rain Fall Study
Distribution of annual average rainfall at the station of An Sefra

Year Rf (mm) Year Rf (mm)
1982-1983 54,85 1992-1993 214,64
1983-1984 65,8 1993-1994 123,6
1984-1985 77,63 1994-1995 118,5
1985-1986 130,01 1995-1996 167,96
1986-1987 333,4 1996-1997 85,26
1987-1988 246,6 1997-1998 63,83
1988-1989 193,36 1998-1999 132,53
1989-1990 183,97 1999-2000 52,9
1990-1991 451,47 2000-2001 79,05
1991-1992 600,52 2001-2002 98,6









751

Variations of Annual Rainfall




















Histogram of Monthly Rainfall












0
100
200
300
400
500
600
700
1
9
8
2
-
1
9
8
3
1
9
8
4
-
1
9
8
5
1
9
8
6
-
1
9
8
7
1
9
8
8
-
1
9
8
9
1
9
9
0
-
1
9
9
1
1
9
9
2
-
1
9
9
3
1
9
9
4
-
1
9
9
5
1
9
9
6
-
1
9
9
7
1
9
9
8
-
1
9
9
9
2
0
0
0
-
2
0
0
1
annes
P
r

c
i
p
i
t
a
t
i
o
n
s

(
m
m
)
Prcipitations (mm)
Moyenne

0
5
10
15
20
25
30
35
40
Pr cipit at ions
(mm)
S O N D J F M A M J J T A
mois

752
The Temperatures


Variation of the Monthly Average Temperatures






















Rainfall und Temperatures Graph























-10
0
10
20
30
40
50
60
S O N D J F M A M J J T A
moi s
T
e
m
p

r
a
t
u
r
e
s

(

C
)
min
max
moy


0
5
10
15
20
25
30
35
40
S O N D J F M A M J J T A
moi s
P
r

c
i
p
i
t
a
t
i
o
n
s

(
m
m
)
0
10
20
30
40
50
60
70
80
T
e
m
p

r
a
t
u
r
e
s

(

C
)
Tempratures
Prcipitations
753

Aquiferous Formation





































Hydrogeologic cross section









Scale horizontal 1/ 25 000
Scale vertical 1/ 25 000
754

Water Table Mape

184.00 186.00 188.00 190.00 192.00 194.00 196.00 198.00 200.00 202.00
252.00
254.00
256.00
258.00
260.00
262.00
264.00
266.00
268.00
270.00
0Km 2Km 4Km
P1
P2
P3
P4
P5
P6
P7
P8
F1
F2
S1
S2
184.00 186.00 188.00 190.00 192.00 194.00 196.00 198.00 200.00 202.00
252.00
254.00
256.00
258.00
260.00
262.00
264.00
266.00
268.00
270.00


P = Water place






755
Hydrochemical Study


Water place
Conductivity
(S/ cm)
Mineralisation
(mg/l)
Dry residue
(mg/l)
Turbidity
(NTU)
P2 2160 1836,93 1860 3.28
P3 2160 2160 1539,28 1.63
P4 1810 1810 1947,48 0.83
P5 2290 2290 1547,79 2.49
P6 1820 1820 639,45 17
P7 843 843 1505,26 3.91
P8 1770 1770 858,94 3.06
F1 1010 1010 398,77 1.37
F2 557 557 463,92 2.53
S1 648 648 450,03 3.77
S2 630 630 1199,11 0.99


Classification of Water for Irrigation
The following table gives the values of S.A.R and conductivity (S.A.R, Sodium Alcalinity
Ratio, U.S.Salinity LAB)
Water place
Conductivity (S/cm)
S.A.R (meq/l)
P2 1260 1.96
P3 1810 1.14
P4 2290 1.05
P5 1820 0.86
P6 843 0.64
P7 1770 1.20
P8 1010 0.49
F1 757 0.41
F2 648 0.33
S1 630 0.39
S2 1410 1.72

Good quality water for the irrigation in place F1, F2, S1.
Acceptable quality water for the irrigation in place P2, P3, P5, P6, P7, P8, S2
Water of bad quality for the irrigation in place P4

756
Potability of drinking water

Results of the chemical analyses

parameters

units

level
guides

maximum
permissible
concentration

Drinking water in
Water place

Non-
drinking
water

Algerian
Standard AS
Ca
+ +



mg/l
75


200


P2(178), P3(183)
P6(96), P7(178), P8(149),
F1(82), F2(87),
p4(260),p5(207),S1(87) ,S2(115).

AS
Mg
+ +



mg/l
-


150


P2(106), P3(62), P4(110),
P5(79), P6(41), P7(90), P8(33),
F1(20), F2(28), S1(25), S2(81).

AS
Cl

mg/l
200


500


P2(400), P3(147), P4(332),
P5(211), P6(77), P7(212),
P8(85), F1(43), F2(41), S1(39),
S2(171).

AS
SO

4



mg/l
200


400


P2(374), P3(379), P5(368),
P6(144), P7(337), P8(100),
F1(59), F2(43), S1(23), S2(207).
P4 (470).
AS
NO

3



mg/l

-


50


P2 (19), F1(5), F2(6), S1(5),
S2(49).
P3 (200),
P4(232),
P5(160),
P6(69),
P7(111),
P8(156).
AS
Dry residue

mg/l -


2000


P2(1860), P3(1460), P4(1860),
P5(1460), P6(660), P7(1260),
P8(860), F1(460), F2(660),
S1(560), S2(1160).

AS
PH


-
6.5
9.5


-

6.5 8.5
P2(7.5), P3(7.5), P4(7.4),
P5(7.5), P6(7.6), P7(7.7) ,
P8(7.7), F1(7.2), F2(7.7),
S1(7.9), S2(7.6).


Conductivity


S/cm


2880

P2(2160), P3(1810), P4(2290),
P5(1820),
P6(1843),
P7(1770), P8(1010), F1(557),
F2(648), S1(630),
S2(1410).
.

757
184.00 186.00 188.00 190.00 192.00 194.00 196.00 198.00 200.00 202.00
252.00
254.00
256.00
258.00
260.00
262.00
264.00
266.00
268.00
270.00
P2
P3
P4
P5
P6
P7
P8
F1
F2
S1
S2
184.00 186.00 188.00 190.00 192.00 194.00 196.00 198.00 200.00 202.00
252.00
254.00
256.00
258.00
260.00
262.00
264.00
266.00
268.00
270.00
0Km 2Km 4Km
184.00 186.00 188.00 190.00 192.00 194.00 196.00 198.00 200.00 202.00
252.00
254.00
256.00
258.00
260.00
262.00
264.00
266.00
268.00
270.00
P2
P3
P4
P5
P6
P7
P8
F1
F2
S1
S2
184.00 186.00 188.00 190.00 192.00 194.00 196.00 198.00 200.00 202.00
252.00
254.00
256.00
258.00
260.00
262.00
264.00
266.00
268.00
270.00
0Km 2Km 4Km
Isoconcentrations Mapes




















Mape of isoconcentration for ( Ca
+ +
)























Mape of isoconcentration for (Na
+
+ k
+
)
758
184.00 186.00 188.00 190.00 192.00 194.00 196.00 198.00 200.00 202.00
252.00
254.00
256.00
258.00
260.00
262.00
264.00
266.00
268.00
270.00
184.00 186.00 188.00 190.00 192.00 194.00 196.00 198.00 200.00 202.00
252.00
254.00
256.00
258.00
260.00
262.00
264.00
266.00
268.00
270.00
0Km 2Km 4Km
184.00 186.00 188.00 190.00 192.00 194.00 196.00 198.00 200.00 202.00
252.00
254.00
256.00
258.00
260.00
262.00
264.00
266.00
268.00
270.00
0Km 2Km 4Km
P2
P3
P4
P5
P6
P7
P8
F1
F2
S1
S2
184.00 186.00 188.00 190.00 192.00 194.00 196.00 198.00 200.00 202.00
252.00
254.00
256.00
258.00
260.00
262.00
264.00
266.00
268.00
270.00






















Mape of isoconcentration for ( Cl
-

























Mape of isoconcentration for Mineralisation


759



Water Need in Ain Sefra

1998 2005 2010 2020
Population increase

37.069 44.700 51.094 66.757
House water need
Mqm
2,0295 2,4473 2,7974 3,6549
Industry water need
Mqm
0,6216 0,7496 0,8568 1,1195


Water Need for Agriculture

Surface irrigable

Watered surface


996,50 He 226 He
31,3597 7,1272 Water need Mqm
1 litre water per second for one hectare


Water Need for Animals

Sheep

Goat

Cattle

Camel

Horse

numbers 68480 3210 2790 32 82
Water need
Qm /Jahr
74.985,6 3.514,95 15.275,25 175,2 448,95
(Water use: Sheep, goat: 2 to 5 l/day/head and cattle, camel, horse: 10 to 30 lday/head)


Entirely Water Need in Ain-Sefra until 2020

Population Population Agriculture Animals total

10,9291 3,3475 31,3597 1,2928 36 Mqm

760
MODELING OF LNAPL SPREADING AND
MIGRATION ON THE WATER TABLE BENEATH AN
OIL REFINERY

Y. Ansari, Graduate Student, Department of Civil Engineering, Shiraz
University, Shiraz, Iran
G. Habibagahi, Associate Professor, Department of Civil Engineering, Shiraz
University, Shiraz, Iran

ABSTRACT
A common problem associated with oil refineries is the subsurface
contamination resulted from leaking of light non-aqueous phase liquids
(LNAPLs) from storage tanks and underground facilities. In this paper, LNAPL
transport on the water table was modeled numerically to investigate the future
fate of the contamination and find possible suitable locations for their treatment.
The model assumes a sharp interface between the LNAPL and groundwater and
takes into account the groundwater velocity and pertinent soil parameters.
Spreading and migration of LNAPL mounds with time were predicted
using the Sharp interface model. Both presence and absence of the leaking
sources were considered in the analysis. Results show that migration is relatively
small. Consequently, sources of contamination are expected to be located in the
proximity of points with maximum LNAPL thickness. Clean-up process is vital
at the site due to soil volume that could be contaminated by existing LNAPL
mounds.

Key words: LNAPL; modeling; sharp interface; refinery; spreading; oil leak.

INTRODUCTION
Petroleum liquids are a basic building block of our modern lives. Uses include fuels,
lubricants & raw materials of manufactured products. Non-aqueous phase liquids (NAPLs)
are hydrocarbons that exist as a separate, immiscible phase when in contact with water and/or
air. Non-aqueous phase liquids are typically classified as either light non-aqueous phase
liquids (LNAPLs) which have densities less than that of water, or dense non-aqueous phase
liquids (DNAPLs) which have densities greater than that of water.

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


761
Contamination due to non-aqueous phase liquids (NAPL) has emerged lately as a major
environmental problem. The major concern caused by NAPLs is due to NAPLs persistence
and their ability to contaminate large volumes of soil and groundwater. An estimated 1.8
million underground storage tanks are in use in the United States. EPA estimates are that
280,000 tanks are leaking, of which more than 20% are discharging their contents directly
into the ground water [El-Kadi, 1992]. In other countries with numerous oil refineries
including various underground storage tanks like Iran the situation is not better.
The propagation of LNAPL through a porous medium constitutes multiphase flow. Two
different methods are considered to study and model the propagation and behavior of by-
products released beneath contaminated sites: concentration based and thickness based.
Concentration based is mostly adapted for soluble hydrocarbon products while thickness
based is applied in order to study separate and immiscible phases such as LNAPLs. When a
LNAPL leaks above an unconfined aquifer, the NAPL migrates through the unsaturated zone
as a separate phase under the dominant influence of gravity, leaving residual droplets in the
unsaturated zone. Once it reaches the water table, the LNAPL forms a free-product mound
floating on the water table. Then the LNAPL spreads laterally and moves in the direction of
decreasing hydraulic gradient, leaving residual LNAPL droplets.
Modeling the migration of non-aqueous phase liquids are considered in four general
groups: sharp interface models, immiscible phase flow models with capillarity, inter-phase
mass transfer models, and compositional models [Corapcioglu et al., 1993]. Each group
makes certain assumptions to generate a governing set of equations. Although sharp interface
models have some limitations, such as neglect of capillary forces in comparison to pressure
and gravity forces, these models can function as screening or site assessment tools for
practical purposes because of their relative simplicity (Corapcioglu et al., 1994). The
assumptions of sharp interface approach are (1) negligible capillary pressure gradient, (2)
sharp saturation changes across the oil-unsaturated soil and oil-groundwater interfaces, (3)
constant soil and fluid properties such as negligible compressibility, constant fluid densities
and constant porosity, (4) single phase (air in unsaturated soil, water in saturated soil)
displacement, (6) assumption of piston flow displacement , i.e., oil phase only, other phases
immobile at residual saturations (as noted by El-Kadi [1994] this assumption is valid for
coarse textured soils where gravity or fluid pressure predominates over capillary forces), (7)
negligible effect of water table on oil spreading and migration [Corapcioglu et al., 1996],and
(8) In the saturated zone, with only water and LNAPL present, water will be the wetting fluid.
This paper is based on actual field measurements, LNAPL thickness was measured over
groundwater at Shiraz Oil Refinery site, a sharp interface model was adopted to model
spreading and migration of LNAPL at the site subjected to two different scenarios. In the first
scenario it is assumed that the suspected sources of leakage are removed completely. In the
second scenario prediction is made assuming the sources continue to leak without any change.
Details of the investigation, modeling and results are presented in the subsequent sections.

SITE DESCRIPTIONS
Shiraz Oil Refinery with the annual capacity of 2 million tons of oil products was
designed and constructed in the early 1970s. The refinery started operating by the Iranian
engineers and experts in 1973. Main products include gas-oil, bitumen, motor gasoline, fuel
oil, kerosene, LPG (liquid petroleum gas) and sulfur. The refinery site was located
approximately 15 kilometers northeast of Shiraz and it has occupied an area of about 2.3
square kilometers. The geology at the site is composed of fine sand with varying amounts of
silt and gravel. Depth to groundwater at the refinery site ranges between 0.5-13 meters with
averages of 5 meters. Local groundwater flows to the north of the refinery site.
762
The primary investigations revealed that the industrial activities have led to subsurface
environmental contamination, particularly groundwater contamination. The studying area
included 30 pre-existing boreholes that were drilled in order to lower groundwater table.
Since the monitoring stations had been installed for a different task and some of them were
destroyed in the course of time, the existing network was augmented. By the beginning of this
research 35 monitoring stations were available and in use.

THEORETICAL FRAMEWORK
The mass balance equation of the NAPL phase in a porous medium is the starting point
to develop the governing equation in terms of oil thickness in order to simulate areal
spreading and migration of LNAPL mound with ambient groundwater flow. The mass
balance equation can be expressed as:


0
) (
.
0 0
0 0
=

+
t
n S
q

(1)

Where
0
is the density of the NAPL phase;
0
q is the specific discharge of the
NAPL phase; S
o
is the degree of saturation of NAPL phase; t is the time component and
n is the porosity of medium. The governing equation describing spreading and migration of
LNAPL mound with ambient groundwater flow in terms of NAPL thickness is developed by
averaging equation (1) along the vertical in the NAPL lens and rearranging as given in
Corapcioglu et al. (1996) :


) ( ) (
) ( ) ( ) (
) (
00
0
2
00
0 0

y x
Soun S n
Q
t
L
L
Soun S n K
K k q
L
Soun S n
L K k
C B
oo w
o or w
A
w
o w or
4 4 3 4 4 2 1 4 4 4 3 4 4 4 2 1 4 4 4 3 4 4 4 2 1
(2)

where
w
is the density of the water;
o w
K K , are the hydraulic conductivities of the
NAPL and water phases respectively; k
or
is relative permeability of the NAPL phase; S
oo
and
S
oun
are saturation degrees of mobile NAPL and residual NAPL, respectively; L is the floating
NAPL thickness on the water table defined as: ) , ( ) , , ( ) , , ( y x t y x h t y x L = under steady
state groundwater flow conditions (h is the elevation of the mound surface and is
the elevation of the groundwater from a specific datum); L
o
is a reference thickness of
the NAPL mound defined as the volume of oil per unit area of aquifer divided by porosity. q
w
is the specific discharge of water; Q
o
is the rate of leaking/pumping at the point ( , ); and
denotes the Dirac delta function used to represent point sources. A minus sign
before the second term on the right-hand side indicates a leak. Rearrangement of
equation (2) yields the following form in Cartesian coordinates:
) ( ) ( ) ( ) (
2
2
2
2
y x C
t
L
y
L
x
L
B
y
L
x
L
A

(3)

763
The term
y
L

in equation (3) would be equal to zero if x-direction coincides with the

direction of groundwater flow.
Two different states of displacement for NAPL phase is considered in the unsaturated
zone. The LNAPL migration on the water table shows two distinct surface areas between the
LNAPL phase and the air phase: some portion of the surface is in imbibition, and another
portion is in drainage (Kim & Corapcioglu , 2001). The LNAPL surface in drainage leaves
residual NAPL, because the NAPL phase is displaced by the air phase on the surface. The
LNAPL surface in imbibition does not leave residual NAPL because this portion displaces the
air phase in the pores. Therefore, with a continuous source, S
oun
in equation (4) becomes zero
because all of the surface areas are in imbibition. Absence of the continuous source causes
some of the surface areas in the drainage state, and the value of S
oun
becomes the value of
residual saturation (Kim & Corapcioglu, 2001).

NUMERICAL ANALYSIS & SOLUTION TECHNIQUES
The governing partial differential equation (equation (3)) is solved numerically using a
finite difference technique. 2-Dimensional governing equation can be solved using classical
Implicit, Alternate Direction Implicit (ADI), or explicit finite difference methods. Explicit
methods are conditionally stable; therefore for given values of horizontal and vertical space
intervals (x & y) stability condition places a limitation on the time step t. Therefore,
explicit methods are restricted to very small time steps and this limitation requires a large
amount of computational time to solve the transport equation. On the contrary, implicit
methods are unconditionally stable, however, in order to solve the resulting simultaneous
linear algebraic equations high-capacity computers are required.
In this research, Hopscotch Method presented by (Gourlay & McGuire, 1971) was
adopted. The method works by solving every other node explicitly, then solving the nodes in-
between using an implicit scheme. The method has been used mainly for two-dimensional
problems, which may be visualised as a chessboard (Figure 1). Initially all the white squares
(odd nodes) are solved explicitly:



) 4 (


Figure 1. Five-point stencil

SourceTerm
x
L L
B
y
L L L
x
L L L
A
t
L L
n
k j
n
k j
n
k j
n
k j
n
k j
n
k j
n
k j
n
k j
n
k j
n
k j
m

+
+

+
=

+ + +
+
) ( 2 ) (
2
) (
2
, 1 , 1
2
1 , , 1 ,
2
, 1 , , 1 ,
1
,
764
The black squares (even nodes) are then solved using the implicit equation:



) 5 (

Since in the implicit part all the neighboring points at time step (t+1) is determined in
the previous step, the algorithm is fully explicit but unconditionally stable and thus time steps
of any size may be used. This is the main advantage of Hopscotch method. Since only half the
nodes are solved in each 'step' in Hopscotch, the error propagation is relatively worse. This is
the only drawback of the method.

COMPUTER PROGRAMS
A computer program was thus developed to incorporate the above-mentioned algorithm
in order to simulate the migration and spreading of LNAPL mounds. Computer program was
developed in MATLAB environment. Two distinct computer programs were developed:

Program #1: Solves the governing differential equation assuming a continuous LNAPL
source with a prescribed discharge value to predict LNAPL spreading and migration in the
solution domain with time.

Program #2: Solves the governing differential equation in the absence of a continuous
LNAPL source. This program is capable of simulating migration and spreading of LNAPL
mound on an ambient groundwater surface by taking into account the residual mass loss
which is left behind in the pores.

Further steps required for numerical solution of the governing partial differential
equation are: (1) Meshing the solution domain; (2) Assigning boundary conditions; and (3)
Definition of initial condition.

Meshing the solution domain
Definition of the solution domain includes determination of location and dimensions of
the domain, shape of the domain, grid spacing, direction and location of the local axis of
coordinates. Location and dimensions of the domain are determined according to dimensions
and location of the significant LNAPL mound. Grid spacing (x,y) is limited to the
accuracy of calculation. Direction and location of local axis of coordinates of the solution
domain should satisfy following conditions:

Positive x-direction of local axis must coincide with the ambient
ground water flow direction.
Origin of the local coordinate system must coincide with the center of
LNAPL source.
A rectangular solution domain was chosen for this study.

Boundary Conditions
It is assumed that the meshing domain is large enough compared to the contaminated
zone and hence the following boundary conditions are applicable:

Normal to Y-direction :

SourceTerm
x
L L
B
y
L L L
x
L L L
A
t
L L
n
k j
n
k j
n
k j
n
k j
n
k j
n
k j
n
k j
n
k j
n
k j
n
k j
m

+
+

+
=

+
+
+

+ +
+
+

+ +
+
+
) ( 2 ) (
2
) (
2
1
, 1
1
, 1
2
1
1 ,
1
,
1
1 ,
2
1
, 1
1
,
1
, 1 ,
1
,
0 =

x
L
765

Normal to X-direction :


Initial Condition
Distribution of LNAPL thickness over groundwater is required as initial condition to
solve the problem. For modeling fresh leaks, the initial thickness of LNAPL is zero. However,
for case of modeling spreading and migration of an existing LNAPL, an estimation or direct
thickness measurement of the current LNAPL thickness is essential.

PARAMETER ESTIMATION
Parameters involved in the sharp interface model are divided into three categories: (1)
geotechnical parameters which describe subsurface characteristics of the site; (2) hydro-
geological parameters which contain detailed information on multi-phase flow of existing
fluids in the porous medium; and (3) physical parameters which introduce physical
characteristics of the problem.

Geotechnical Parameters
Soil porosity, soil water content, specific gravity of soil particles, and saturation degrees
of mobile NAPL (S
oo
) and residual NAPL (S
oun
) in subsurface are geotechnical parameters of
the model. Geotechnical parameters were estimated from tests on soil samples collected at the
level of groundwater table.

Hydro-geological Parameters
LNAPL spreading and migration above an unconfined aquifer contains multi-phase
flow of immiscible fluids including air, water, and LNAPL in the vadose zone. Hydro-
geological parameters present in the model describe the multi-phase flow of expected fluids.
These parameters are hydraulic conductivities of water and LNAPL in the soil matrix;
groundwater flow direction; groundwater flow velocity; relative permeability of NAPL phase;
and viscosity of NAPL phase.
Figure (2) shows groundwater flow direction and head contours at the site obtained
using geo-statistical interpolation of results from observation boreholes.

Physical Parameters
Determination of unknown physical parameters of the model is a difficult task. Physical
parameters participated in the model are: (1) actual LNAPL thickness in the formation if any;
(2) dimensions of the source if any; (3) rate of oil leak from the source; and (4) duration of
leak.

CASE STUDIED
Spreading and migration of LNAPL over groundwater beneath Shiraz Oil Refinery was
studied using the sharp interface model. LNAPL was encountered in some of the monitoring
boreholes drilled for this purpose. The apparent LNAPL thickness was measured using
electro-optical interface meter instrument which is capable of measuring LNAPL thickness
with 1mm accuracy. Geotechnical and geo-hydrological parameters were estimated based on
both in-situ tests such as Lefranc test (to measure hydraulic conductivity) and laboratory tests.
Table 1 presents model parameters estimated based on these investigations.

0 =

y
L
766

(a) Head contours (b) Velocity vectors
Figure 2. Head contours and velocity vectors of groundwater flow at the site

Site investigations and measurements of LNAPL thickness at Shiraz Oil Refinery site
concluded that 7 of 35 available monitoring stations had immiscible LNAPL phase. Two
distinct LNAPL mounds were identified. Details are presented in Table 1.

Table 1.Volume, area and thickness of LNAPL mounds
Mound
No.
Estimated volume of
LNAPL (m
3
)
Area of mound
(m
2
)
Maximum apparent thickness
(mm)
1 10311 187651 1130
2 403.34 110885 50

As clear from Table 1, mound no. 1 covered a larger area with a much bigger apparent
LNAPL thickness. Figure (3) shows the location & spreading of the two mounds.



(a) Mound no.1 (b) Mound no.2
Figure 3. Existing LNAPL mounds at the site


18
25
8
10
11
12
43
1
2
3
4
5
30
31
32
34
39
16
9
42
15
19
7
-811.00 1634.00 73.02
-535.50 1814.00 78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35 1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90 114.20
-209.10 877.50 89.61
74.60 949.20 93.62
F
G
76.70 1158.10 92.93
-336.00
466.75 93.72
-43.05 246.70 98.34
312.25 255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35 547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90 575.40
121.13
969.40 680.40 119.68
760.90 755.40 112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x= y= z=
z= y= x=
z=
x= y=
x= y= z=
z= y= x= x= y= z=
z= y= x=
z= y= x=
x= y= z=
z= y= x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
-811.00 1634.00 73.02
-535.50 1814.00 78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35 1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90 114.20
-209.10 877.50 89.61
74.60 949.20 93.62
F
G
76.70 1158.10 92.93
-336.00
466.75 93.72
-43.05 246.70 98.34
312.25 255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35 547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90 575.40
121.13
969.40 680.40 119.68
760.90 755.40 112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x= y= z=
z= y= x=
z=
x= y=
x= y= z=
z= y= x= x= y= z=
z= y= x=
z= y= x=
x= y= z=
z= y= x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
33
35
36
37
38
40
-1400 -1200 -1000 -800 -600 -400 -200 0 200 400 600 800
0
200
400
600
800
1000
1200
1400
1600
1800
Mound #2
18
25
8
10
11
12
43
1
2
3
4
5
30
31
32
34
39
16
9
42
15
19
7
-811.00 1634.00 73.02
-535.50
1814.00
78.24
-162.00 1983.75
87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35
1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90
114.20
-209.10 877.50 89.61
74.60 949.20
93.62
F
G
76.70 1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25
255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35
547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90
575.40
121.13
969.40 680.40 119.68
760.90 755.40
112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x=
y= z=
z= y=
x=
z=
x=
y=
x=
y= z=
z= y=
x= x=
y= z=
z= y= x=
z= y=
x=
x=
y= z=
z= y=
x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z=
y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
-811.00 1634.00 73.02
-535.50
1814.00
78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55
87.04
150.35
1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90
114.20
-209.10 877.50 89.61
74.60 949.20
93.62
F
G
76.70 1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25
255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35 547.15
99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90
575.40
121.13
969.40 680.40 119.68
760.90 755.40
112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x=
y= z=
z= y=
x=
z=
x=
y=
x=
y= z=
z= y=
x= x=
y= z=
z= y= x=
z= y=
x=
x=
y= z=
z= y=
x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
22
6
36
-800 -600 -400 -200 0 200 400 600 800 1000
600
800
1000
1200
1400
1600
1800
Mound # 1
18
25
8
10
11
12
43
1
2
3
4
5
30
31
32
34
39
16
9
42
15
19
7
-811.00 1634.00
73.02
-535.50
1814.00
78.24
-162.00
1983.75
87.74
-102.00
1695.75
87.42
-147.50
1440.55
87.04
150.35
1443.50
95.85
460.80
1444.35
105.58
640.55
1409.05 111.20
464.80
1373.00 106.58
125.60
1368.50
94.27
442.75 1179.45
105.54
C
468.70
1050.50
106.99
239.55
1073.50
99.26
1006.90
114.20
-209.10
877.50
89.61
74.60
949.20
93.62
F
G
76.70
1158.10
92.93
-336.00
466.75
93.72
-43.05
246.70 98.34
312.25
255.75
105.44
-86.15
641.65
91.67
115.30
590.40
98.82
281.35
547.15
99.95
D
A E
405.50
209.95 108.46
565.40
374.65
109.02
958.90
575.40
121.13
969.40
680.40
119.68
760.90
755.40
112.04
441.80
904.80
462.70
587.45
105.38
B
648.85
105.33
x=
y=
z=
z=
y=
x=
z=
x=
y=
x=
y=
z=
z=
y=
x= x=
y=
z=
z= y=
x=
z=
y=
x=
x=
y=
z=
z=
y=
x=
y=
z=
x=
y=
z=
x=
x=
z=
y=
x=
y=
z=
z=
y= x=
x=
y=
z=
x=
z=
y=
z=
y=
x=
x=
y=
z=
z=
y=
x=
x=
y=
z=
z=
y=
x=
z=
x=
y=
y=
x=
z=
y=
x=
z=
y=
z=
x=
x=
z=
y=
y=
z=
x=
x=
z=
y= x=
z=
y=
-811.00
1634.00
73.02
-535.50
1814.00
78.24
-162.00
1983.75
87.74
-102.00
1695.75
87.42
-147.50
1440.55
87.04
150.35
1443.50
95.85
460.80
1444.35
105.58
640.55
1409.05 111.20
464.80
1373.00 106.58
125.60
1368.50
94.27
442.75 1179.45
105.54
C
468.70
1050.50
106.99
239.55
1073.50
99.26
1006.90
114.20
-209.10
877.50
89.61
74.60
949.20
93.62
F
G
76.70
1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70
98.34
312.25
255.75
105.44
-86.15 641.65
91.67
115.30
590.40
98.82
281.35
547.15
99.95
D
A E
405.50
209.95 108.46
565.40
374.65
109.02
958.90
575.40
121.13
969.40
680.40
119.68
760.90
755.40
112.04
441.80
904.80
462.70
587.45
105.38
B
648.85
105.33
x=
y=
z=
z=
y=
x=
z=
x=
y=
x=
y=
z=
z=
y=
x= x=
y=
z=
z= y=
x=
z=
y=
x=
x=
y=
z=
z=
y=
x=
y=
z=
x=
y=
z=
x=
x=
z=
y=
x=
y=
z=
z=
y= x=
x=
y=
z=
x=
z=
y=
z=
y=
x=
x=
y=
z=
z=
y=
x=
x=
y=
z=
z=
y=
x=
z=
x=
y=
y=
x=
z=
y=
x=
z=
y=
z=
x=
x=
z=
y=
y=
z=
x=
x=
z=
y= x=
z=
y=
22
6
36
-800 -600 -400 -200 0 200 400 600 800 1000
400
600
800
1000
1200
1400
1600
1800
Water Level Contours
18
25
8
10
11
12
43
1
2
3
4
5
30
31
32
34
39
16
9
42
15
19
7
-811.00 1634.00
73.02
-535.50
1814.00
78.24
-162.00
1983.75
87.74
-102.00
1695.75
87.42
-147.50
1440.55
87.04
150.35
1443.50
95.85
460.80
1444.35
105.58
640.55
1409.05 111.20
464.80
1373.00 106.58
125.60
1368.50
94.27
442.75 1179.45 105.54
C
468.70
1050.50
106.99
239.55
1073.50
99.26
1006.90
114.20
-209.10
877.50
89.61
74.60
949.20
93.62
F
G
76.70
1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25
255.75
105.44
-86.15 641.65
91.67
115.30
590.40
98.82
281.35
547.15
99.95
D
A E
405.50
209.95 108.46
565.40
374.65 109.02
958.90
575.40
121.13
969.40
680.40
119.68
760.90
755.40
112.04
441.80
904.80
462.70
587.45
105.38
B
648.85
105.33
x=
y=
z=
z=
y=
x=
z=
x=
y=
x=
y=
z=
z=
y=
x= x=
y=
z=
z= y=
x=
z=
y=
x=
x=
y=
z=
z=
y=
x=
y=
z=
x=
y=
z=
x=
x=
z=
y=
x=
y=
z=
z=
y= x=
x=
y=
z=
x=
z=
y=
z=
y=
x=
x=
y=
z=
z=
y=
x=
x=
y=
z=
z=
y=
x=
z=
x=
y=
y=
x=
z=
y=
x=
z=
y=
z=
x=
x=
z=
y=
y=
z=
x=
x=
z=
y= x=
z=
y=
-811.00 1634.00
73.02
-535.50
1814.00
78.24
-162.00
1983.75
87.74
-102.00
1695.75
87.42
-147.50
1440.55
87.04
150.35
1443.50
95.85
460.80
1444.35
105.58
640.55
1409.05 111.20
464.80
1373.00 106.58
125.60
1368.50 94.27
442.75 1179.45
105.54
C
468.70
1050.50 106.99
239.55
1073.50
99.26
1006.90
114.20
-209.10
877.50
89.61
74.60
949.20
93.62
F
G
76.70
1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25
255.75
105.44
-86.15 641.65
91.67
115.30
590.40
98.82
281.35
547.15
99.95
D
A E
405.50
209.95 108.46
565.40 374.65 109.02
958.90
575.40
121.13
969.40
680.40
119.68
760.90
755.40
112.04
441.80
904.80
462.70
587.45
105.38
B
648.85
105.33
x=
y=
z=
z=
y=
x=
z=
x=
y=
x=
y=
z=
z=
y=
x= x=
y=
z=
z= y=
x=
z=
y=
x=
x=
y=
z=
z=
y=
x=
y=
z=
x=
y=
z=
x=
x=
z=
y=
x=
y=
z=
z=
y= x=
x=
y=
z=
x=
z=
y=
z=
y=
x=
x=
y=
z=
z=
y=
x=
x=
y=
z=
z=
y=
x=
z=
x=
y=
y=
x=
z=
y=
x=
z=
y=
z=
x=
x=
z=
y=
y=
z=
x=
x=
z=
y= x=
z=
y=
22
6
36
-800 -600 -400 -200 0 200 400 600 800 1000
400
600
800
1000
1200
1400
1600
1800
Water Flow Vectors
767
MODEL RESULTS
Results obtained using sharp interface model for the case studied are discussed in two
different parts:

A- Speculation of previous history and pattern of leaks
B- Prediction of future LNAPL plumes spreading and migration with time:

B-1: spreading and migration of current LNAPL mounds considering a
continuous leaking source.
B-2: spreading and migration of LNAPL mounds with no leaking source.

A- Speculation of previous history and pattern of leaks :
In this section, it was intended to simulate the formation of the current LNAPL mounds
in order to estimate the past history of the leak at the site. It was tried to reach the best fitness
between the model results and current situation of plumes at the site. The objective was to
find the radius of source, rate of LNAPL leak and leak duration that best matches the existing
LNAPL mounds. No optimization program was used but the variables were found using trial
and error knowing the physical constraints regarding each variable. Table 2 indicates the
parameters of source thus obtained. Figure (4) shows the simulated mounds based on the
estimated parameters.

Table 2. Estimated source parameters
Simulated Mound no.1 Simulated Mound no.2
Leaking duration (year) 20 14
Radius of source at
surface (m)
20 30
Rate of LNAPL discharge
(m
3
/day)
0.33 0.022


(a) Mound #1 (b) Mound #2
Figure 4. Simulated LNAPL mounds at the site

The discrepancy between simulated and existing conditions can be attributed to
assumptions and other simplifications inherent in the method. Comparison of LNAPL
characteristics between existing & simulated mounds are indicated in Table (3).



18
25
8
10
11
12
43
1
2
3
4
5
30
31
32
34
39
16
9
42
15
19
7
-811.00 1634.00 73.02
-535.50
1814.00
78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35 1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90 114.20
-209.10 877.50 89.61
74.60 949.20 93.62
F
G
76.70 1158.10 92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25 255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35 547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90 575.40
121.13
969.40 680.40 119.68
760.90 755.40
112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x=
y= z=
z= y=
x=
z=
x=
y=
x=
y= z=
z= y=
x= x=
y= z=
z= y= x=
z= y=
x=
x= y= z=
z= y= x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
-811.00 1634.00 73.02
-535.50 1814.00
78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35 1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90 114.20
-209.10 877.50 89.61
74.60 949.20 93.62
F
G
76.70 1158.10 92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25 255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35 547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90 575.40
121.13
969.40 680.40 119.68
760.90 755.40
112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x=
y= z=
z= y=
x=
z=
x=
y=
x=
y= z=
z= y= x= x= y= z=
z= y= x=
z= y=
x=
x=
y= z=
z= y=
x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
22
6
36
-800 -600 -400 -200 0 200 400 600 800 1000
600
800
1000
1200
1400
1600
1800
18
25
8
10
11
12
43
1
2
3
4
5
30
31
32
34
39
16
9
42
15
19
7
-811.00 1634.00 73.02
-535.50 1814.00 78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35 1443.50 95.85
460.80 1444.35 105.58 640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90 114.20
-209.10 877.50 89.61
74.60 949.20 93.62
F
G
76.70 1158.10 92.93
-336.00 466.75 93.72
-43.05 246.70 98.34
312.25 255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35 547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90 575.40 121.13
969.40 680.40 119.68
760.90 755.40 112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x= y= z=
z= y= x=
z=
x= y=
x= y= z=
z= y= x= x= y= z=
z= y= x=
z= y= x=
x= y= z=
z= y= x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
-811.00 1634.00 73.02
-535.50 1814.00 78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35 1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90 114.20
-209.10 877.50 89.61
74.60 949.20 93.62
F
G
76.70 1158.10 92.93
-336.00 466.75 93.72
-43.05 246.70 98.34
312.25 255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35 547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90 575.40 121.13
969.40 680.40 119.68
760.90 755.40 112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x= y= z=
z= y= x=
z=
x= y=
x= y= z=
z= y= x= x= y= z=
z= y= x=
z= y= x=
x= y= z=
z= y= x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
33
35
36
37
38
40
-1400 -1200 -1000 -800 -600 -400 -200 0 200 400 600 800
0
200
400
600
800
1000
1200
1400
1600
1800
768

B-1: Spreading and migration of mounds in the presence of a source:
In this part, future spreading and migration of current LNAPL mounds on a sloping
water table is studied. No infiltration is considered. The parameters used in the simulation are
listed in Table 4. The residual saturation is set to zero because of the continuous source.

Table 3. Comparison of LNAPL characteristics between existing & simulated mounds
Volume (m
3
)
Area Covered
(m
2
)
Max. Apparent Thickness
(mm)
Mound
No.
Exist. Simu. Exist. Simu. Existing Simulated
1

10311 11086 187651 132711 914 954
2

403.34 390.1 110885 67769 50 50.34

Figure (5) shows simulated spreading and migration of LNAPL mounds with a
continuous source after 20 years. There is considerable longitudinal spreading of LNAPL
plumes in the direction of local groundwater flow. Details on spreading and migration of
LNAPL mounds are given in Table 5.


Table 4. Parameters used in simulation
Parameter Symbol Value Unit
Porosity n 0.35
Darcy velocity q
w
0.015 m/d
Hydraulic conductivity of
water
K
w

0.864 m/d
Hydraulic conductivity of oil K
o
0.52 m/d
NAPL saturation in NAPL lens
S
oo

0.85
Residual water saturation
S
ow

0.15
Residual NAPL saturation
S
oun

0
water density
w
1000 kg/m
3

NAPL density
o
780 kg/m
3

NAPL relative permeability
k
or

0.8
NAPL source flow rate Q
o
0.33-0.022 m
3
/d
Radius of source
r
20-30 m



B-2: Spreading and migration of LNAPL mounds without leaking source:
Prediction of subsequent spreading and migration of current LNAPL mounds with
ambient groundwater flow in absence of the leaking sources was studied. Model parameters
were the same as in the previous simulation, except that the LNAPL leak rate is set to zero
and residual degree of saturation is equal to 0.15. The soil was subjected to contamination due
to residual droplets of LNAPL left in pores of porous medium through migration of LNAPL
mounds. Figure (6) illustrates spreading and migration of defined LNAPL mounds after 25
years. Details on spreading and migration of mounds are given in Table (5).



769

(a) Mound no.1 (b) Mound no.2
Figure (5). Simulated LNAPL migration and spreading after 20 years (with leaking source)


(a) Mound #1 (b) Mound #2
Figure (6) . Simulated LNAPL migration and spreading after 25 years (without leaking source)


Table 5. Details of LNAPL mounds after 20 years with source
Mound no.1 Mound no.2
Mound Volume (m
3
) 18676 911.18
Maximum mound
thickness (mm)
348.66 22.5


Table 6. Details of LNAPL mounds after 25 years without source
Mound no.1 Mound no.2
Volume loss (m
3
) 3812.57 277.63
Maximum mound
thickness (mm)
153.49 9.7

18
25
8
10
11
12
43
1
2
3
4
5
30
31
32
34
39
16
9
42
15
19
7
-811.00 1634.00 73.02
-535.50
1814.00
78.24
-162.00
1983.75
87.74
-102.00
1695.75
87.42
-147.50
1440.55
87.04
150.35
1443.50 95.85
460.80
1444.35
105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60
1368.50 94.27
442.75 1179.45 105.54
C
468.70
1050.50 106.99
239.55
1073.50 99.26
1006.90
114.20
-209.10
877.50 89.61
74.60 949.20
93.62
F
G
76.70
1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25
255.75
105.44
-86.15 641.65
91.67
115.30
590.40 98.82
281.35
547.15
99.95
D
A E
405.50
209.95 108.46
565.40 374.65 109.02
958.90
575.40
121.13
969.40
680.40
119.68
760.90
755.40
112.04
441.80
904.80
462.70 587.45
105.38
B
648.85
105.33
x=
y=
z=
z=
y=
x=
z=
x=
y=
x=
y=
z=
z=
y=
x= x=
y=
z=
z= y= x=
z=
y=
x=
x=
y=
z=
z=
y=
x=
y=
z=
x=
y=
z=
x=
x=
z=
y=
x=
y=
z=
z= y= x=
x=
y=
z=
x=
z=
y=
z=
y=
x=
x=
y=
z=
z=
y=
x=
x=
y=
z=
z=
y=
x=
z=
x=
y=
y=
x=
z=
y=
x=
z=
y=
z=
x=
x=
z=
y=
y=
z=
x=
x=
z=
y= x=
z=
y=
-811.00 1634.00 73.02
-535.50
1814.00
78.24
-162.00 1983.75
87.74
-102.00
1695.75 87.42
-147.50
1440.55
87.04
150.35
1443.50
95.85
460.80
1444.35
105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60
1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55
1073.50
99.26
1006.90
114.20
-209.10
877.50
89.61
74.60
949.20
93.62
F
G
76.70
1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25
255.75 105.44
-86.15 641.65
91.67
115.30 590.40 98.82
281.35
547.15
99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90
575.40
121.13
969.40
680.40 119.68
760.90 755.40
112.04
441.80
904.80
462.70 587.45
105.38
B
648.85
105.33
x=
y=
z=
z=
y=
x=
z=
x=
y=
x=
y=
z=
z=
y=
x= x=
y=
z=
z= y=
x=
z=
y=
x=
x=
y=
z=
z=
y=
x=
y=
z=
x=
y=
z=
x=
x=
z=
y=
x=
y=
z=
z= y= x=
x=
y=
z=
x=
z=
y=
z=
y=
x=
x=
y=
z=
z=
y=
x=
x=
y=
z=
z=
y=
x=
z=
x=
y=
y=
x=
z=
y=
x=
z=
y=
z=
x=
x=
z=
y=
y=
z=
x=
x=
z=
y= x=
z=
y=
22
6
36
-800 -600 -400 -200 0 200 400 600 800 1000
600
800
1000
1200
1400
1600
1800
18
25
8
10
11
12
43
1
2
3
4
5
30
31
32
34
39
16
9
42
15
19
7
-811.00 1634.00 73.02
-535.50 1814.00 78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35 1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90 114.20
-209.10 877.50 89.61
74.60 949.20
93.62
F
G
76.70 1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25 255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35 547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90 575.40
121.13
969.40 680.40 119.68
760.90 755.40
112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x= y= z=
z= y= x=
z=
x= y=
x= y= z=
z= y= x= x= y= z=
z= y= x=
z= y= x=
x= y= z=
z= y= x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
-811.00 1634.00 73.02
-535.50 1814.00 78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35 1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90 114.20
-209.10 877.50 89.61
74.60 949.20
93.62
F
G
76.70 1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25 255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35 547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90 575.40
121.13
969.40 680.40 119.68
760.90 755.40
112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x= y= z=
z= y= x=
z=
x= y=
x= y= z=
z= y=
x= x=
y= z=
z= y= x=
z= y= x=
x= y= z=
z= y= x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
33
35
36
37
38
40
-1400 -1200 -1000 -800 -600 -400 -200 0 200 400 600 800
0
200
400
600
800
1000
1200
1400
1600
1800
18
25
8
10
11
12
43
1
2
3
4
5
30
31
32
34
39
16
9
42
15
19
7
-811.00 1634.00 73.02
-535.50
1814.00
78.24
-162.00
1983.75
87.74
-102.00
1695.75
87.42
-147.50
1440.55
87.04
150.35
1443.50 95.85
460.80
1444.35
105.58
640.55 1409.05 111.20
464.80
1373.00 106.58
125.60
1368.50 94.27
442.75 1179.45 105.54
C
468.70
1050.50 106.99
239.55
1073.50
99.26
1006.90
114.20
-209.10
877.50 89.61
74.60
949.20
93.62
F
G
76.70
1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25
255.75
105.44
-86.15 641.65
91.67
115.30 590.40
98.82
281.35
547.15
99.95
D
A E
405.50
209.95 108.46
565.40 374.65 109.02
958.90
575.40
121.13
969.40
680.40
119.68
760.90
755.40
112.04
441.80
904.80
462.70 587.45
105.38
B
648.85
105.33
x=
y=
z=
z=
y=
x=
z=
x=
y=
x=
y=
z=
z=
y=
x= x=
y=
z=
z= y=
x=
z=
y=
x=
x=
y=
z=
z=
y=
x=
y=
z=
x=
y=
z=
x=
x=
z=
y=
x=
y=
z=
z=
y= x=
x=
y=
z=
x=
z=
y=
z=
y=
x=
x=
y=
z=
z=
y=
x=
x=
y=
z=
z=
y=
x=
z=
x=
y=
y=
x=
z=
y=
x=
z=
y=
z=
x=
x=
z=
y=
y=
z=
x=
x=
z=
y= x=
z=
y=
-811.00 1634.00
73.02
-535.50
1814.00
78.24
-162.00
1983.75
87.74
-102.00
1695.75
87.42
-147.50
1440.55
87.04
150.35
1443.50 95.85
460.80
1444.35
105.58
640.55 1409.05 111.20
464.80
1373.00 106.58
125.60
1368.50 94.27
442.75 1179.45 105.54
C
468.70
1050.50 106.99
239.55
1073.50
99.26
1006.90
114.20
-209.10
877.50 89.61
74.60 949.20
93.62
F
G
76.70
1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25
255.75
105.44
-86.15 641.65
91.67
115.30
590.40 98.82
281.35
547.15
99.95
D
A E
405.50
209.95 108.46
565.40 374.65 109.02
958.90
575.40
121.13
969.40
680.40
119.68
760.90
755.40
112.04
441.80
904.80
462.70 587.45
105.38
B
648.85
105.33
x=
y=
z=
z=
y=
x=
z=
x=
y=
x=
y=
z=
z=
y=
x= x=
y=
z=
z= y=
x=
z=
y=
x=
x=
y=
z=
z=
y=
x=
y=
z=
x=
y=
z=
x=
x=
z=
y=
x=
y=
z=
z=
y= x=
x=
y=
z=
x=
z=
y=
z=
y=
x=
x=
y=
z=
z=
y=
x=
x=
y=
z=
z=
y=
x=
z=
x=
y=
y=
x=
z=
y=
x=
z=
y=
z=
x=
x=
z=
y=
y=
z=
x=
x=
z=
y= x=
z=
y=
22
6
36
-800 -600 -400 -200 0 200 400 600 800 1000
600
800
1000
1200
1400
1600
1800
18
25
8
10
11
12
43
1
2
3
4
5
30
31
32
34
39
16
9
42
15
19
7
-811.00 1634.00 73.02
-535.50
1814.00
78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35 1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90 114.20
-209.10 877.50 89.61
74.60 949.20
93.62
F
G
76.70 1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25 255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35
547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90
575.40
121.13
969.40 680.40 119.68
760.90 755.40
112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x=
y= z=
z= y=
x=
z=
x=
y=
x=
y= z=
z= y=
x= x=
y= z=
z= y= x=
z= y=
x=
x=
y= z=
z= y=
x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
-811.00 1634.00 73.02
-535.50 1814.00
78.24
-162.00 1983.75 87.74
-102.00 1695.75 87.42
-147.50 1440.55 87.04
150.35 1443.50 95.85
460.80 1444.35 105.58
640.55 1409.05 111.20
464.80 1373.00 106.58
125.60 1368.50 94.27
442.75 1179.45 105.54
C
468.70 1050.50 106.99
239.55 1073.50 99.26
1006.90 114.20
-209.10 877.50 89.61
74.60 949.20
93.62
F
G
76.70
1158.10
92.93
-336.00
466.75
93.72
-43.05 246.70 98.34
312.25 255.75 105.44
-86.15 641.65 91.67
115.30 590.40 98.82
281.35 547.15 99.95
D
A E
405.50 209.95 108.46
565.40 374.65 109.02
958.90
575.40
121.13
969.40 680.40 119.68
760.90 755.40
112.04
441.80 904.80
462.70 587.45 105.38
B
648.85
105.33
x=
y= z=
z= y=
x=
z=
x=
y=
x= y= z=
z= y=
x= x=
y= z=
z= y= x=
z= y=
x=
x=
y= z=
z= y=
x=
y= z=
x=
y= z=
x=
x=
z= y=
x= y= z=
z= y= x=
x= y= z=
x=
z= y=
z= y= x=
x= y= z=
z= y= x=
x= y= z=
z= y= x=
z=
x= y=
y= x=
z=
y= x=
z=
y= z=
x=
x=
z= y=
y= z=
x=
x=
z= y= x=
z= y=
33
35
36
37
38
40
-1400 -1200 -1000 -800 -600 -400 -200 0 200 400 600 800
0
200
400
600
800
1000
1200
1400
1600
1800
770
CONCLUSION:
Sharp interface modeling was employed in order to screen current condition of leak and
to predict the consequences of spreading and migration of LNAPL mounds at a refinery site.
The governing differential equation was solved using Hopscotch finite difference method.
Geotechnical and hydro-geological parameters were estimated based on laboratory
experiments and empirical relationships. Physical parameters of the model (rate of leak,
radius of the source) were evaluated using finite element simulations. Two distinct problems
were considered in this study: (1) simulation of history and pattern of leaks; and (2) prediction
of further spreading and migration of LNAPL mounds both in presence and absence of
continuous LNAPL source.
Two distinct LNAPL mounds were distinguished at the refinery site. LNAPL mound #1
with oil volume 25 times of the mound no. 2 was the dominant LNAPL plume of the site.
Both LNAPL mounds were simulated and duration and pattern of leaks leading to the existing
condition were estimated. Results of further spreading and migration of mounds with a
continuous LNAPL source after 20 years are obtained using the sharp interface model.
Migration of mounds with ambient groundwater flow after 25 years was evaluated and
decrease in mound volumes due to residual saturation in absence of leaking source was
determined.
Furthermore, volume of soil that could be contaminated by each mound was calculated
and it was concluded that clean-up process is unavoidable. According to model results further
migration of LNAPL mounds is small. Consequently, sources of LNAPL contamination are
located in the proximity of maximum LNAPL thicknesses for each mound. This fact helped
locating possible source of leakage at the site. The insignificant migration of mounds is due to
low groundwater velocity and low hydraulic conductivity of porous medium.

REFERENCES:

Gourlay, A.R., Mcguire, G.R., 1971. General hopscotch algorithm for the numerical solution of partial
differential equations, J . Inst. Math. Appl., 7, 216-227.


Abdul, A.S., Kia, S.F., and Gibson, T.L., 1989. Limitations of monitoring wells for the detection and
quantification of petroleum products in soils and aquifers, Ground Water Monit. Rev., 9(2), 90-99.


Corapcioglu, M. Y., Lingam, R., Kambham, K. K. R., and Pandy, S., 1993. Multiphase contaminants
in natural permeable media: Various modeling approaches, in Migration and Fate of Pollutants in Soils
and Subsoils, edited by D. Petrozelli and F. G. Helfferich, Speringer-Veriag, NewYork, , pp. 191-220.


Corapcioglu, M.Y., Tuncay, K., Ceylan, B.K., 1996. Oil mound spreading and migration with ambient
groundwater flow in coarse porous media, Water Resour. Res. 32 (5), 1290-1308.


Corapcioglu, M.Y., Tuncay, K., Lingam, R., Kambham, K.K.R., 1994. Analytical expressions to
estimate the free product recovery in oil-contaminated aquifers. Water Resour. Res., 30 (12), 3301
3311.


De Pastrovich, T.L., Baradat, Y., Barthel, R., Chiarelli, A., and Fussell, D.R., 1979. Protection of
ground water from oil pollution, CONCAWE, The Hague, 61 pp.


771
El-Kadi, A.I., 1992. Applicability of sharp-interface models for NAPL transport: 1.
infiltration, Ground Water, 30(6), 849-856.


El-Kadi, A.I., 1994. Applicability of sharp-interface models for NAPL transport: 2.
spreading of an LNAPL, Ground Water, 32(5), 784-795.


Hampton, D.R., and Miller, P.D.G., 1988. "Laboratory investigation of the relationship between actual
and apparent product thickness in sands, in Proc. Conf. on Petroleum
Hydrocarbons and Organic Chemicals in Ground Water: Prevention, Detection, and Restoration, Natl.
Ground Water Assoc., Dublin, OH, 157-181.


HOCHMUTH, D.P., and SUNADA, D.K., 1985. Groundwater model of two-phase immiscible flow in
coarse material, Ground Water, 23, 617626.


Kemblowski, C M.W., Chiang, C.Y., 1990. Hydrocarbon thickness fluctuations in monitoring wells,
Ground Water, 28, 244252.


Kim, J ., Corapcioglu, M.Y., 2001. Sharp interface modeling of LNAPL spreading and migration on
the water table, Environ. Eng. Sci., 18 (6), 359 367.
Lenhard, R.J ., and Parker, J .C., 1990. Estimation of free hydrocarbon volume from fluid levels in
monitoring wells, Ground Water., 28(1), 57-67.


Mercer, J .W., and Cohen, R.M., 1990. A review of immiscible fluids in the subsurface: Properties,
models, characterization, and remediation, J . Contam. Hydrol., 6, 107-163.


Noye, J ., 1982. Numerical Solutions of Partial Differential Equations, Part one, North-Holland
Publishing Company, Proceedings of the 1981 conference on the numerical solutions for partial
differential equations, Melbourne University, Australia.

U.S.EPA, 1994. Light Nonaqueous Phase Liquids, EPA/540/S-95/500, Newell, C.J ., Acree,S.D.,
Ross, R.R., and Huling, S.G., Washington,DC.

772
COASTAL HABITATS DEGRADATION DUE TO
CHRONIC AND RECENT LANDFILLING ALONG
THE RED SEA

Mahmoud A. Radi Dar
National Institute of Oceanography and Fisheries, Red Sea, Hurghada, Egypt
E-mail: Mahmoud_rady@yahoo.com
ABSTRACT
In the period between 1980 and 1994, before the environmental law
4/1994 proceeding, the total recorded infringements to the land inside tidal flat
of the Red Sea in Hurghada coastal area are 175 cases. The filled tidal flat areas
as a direct result of these infringements are about 2,198,643 sq. meters with
water depth variation between one meter and 26 meters. The total amount of the
filling sediments inside the tidal zone is approximately about 30,000,000 cubic
meters along the coast.
The main hazardous effect of the artificially transported sediments is
concentrated in the finest fractions that agitating, wining and transporting with
the waves and currents and re-deposited in the calm zones over the benthic
communities. These huge amounts of the fine loading sediments are altering the
marine ecosystem in Hurghada area. Three main alteration categories are
associated with the landfilling; geo-facies, bio-facies and the seawater.
These later landfills are added to the chronic problem which is
accompanied raw materials shipping harbors in Safaga, El-Hamraween, Qusier
and Abu-Ghsoun. Longtime ago, these harbours were used to export the
grinding raw materials of phosphates, feldspars and ilmenites. They were used a
very rudimental technology in the raw materials transfer and shipping. The
hazardous effects of smothers and the fine particle sediments on the coastal
habitats were extending for many kilometers to the south along the coast.
The continuous landfill loads are destroying the different marine
resources. Consequently, this situation requires more attention from the
scientific authorities in complementation with the decision makers to dilute
these problems inside national pretension plans that provide the practical and
technical supports to preserve and develop these resources.

Keywords: Landfill, Red Sea, Geo-facies, Bio-facies, seawater, Harbours and Technology.

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


773
INTRODUCTION
Tidal flat ecosystem differs from the aquatic ecosystem in the sense that the former is
exposed to sunlight directly at ebb tide and is submerged at flood tide. In tidal flat the primary
producers are not phytoplanktons but phyto-benthos (Hard corals and many species of the soft
corals). Tidal flat develops under turbidity currents in the semi-enclosed coastal area whereas
the finer particles are trapped in the flat and sediments show strong reducing conditions
sometimes to sediment surface. Wave and current energies are important factors in
determining the physicochemical and biological conditions in coastal ecosystem (Havens et
al. 1995).
The artificially transported sediments and the coastal landfilling are the most
significant environmental problems that related to the present coastal uses and activities. It
irreversibly destroys the intertidal habitats of the coast. Landfilling is usually carried out in
the shallow waters, and results in burying and obliteration the biologically productive
intertidal and subtidal habitats including; the salt marshes halophytes, mangroves, intertidal
mud flats, lagoon habitats, sand/mud biotopes and coral reefs. The indirect effect of the
landfilling is the interruption of water movement by the barrier creations as causeways,
concrete marinas and artificial spits (Nawar et al. 2001). Tourist developments along the Red
Sea coast and the gulfs show significant dredging and filling operations, to develop huge
marine structures, with artificial lagoons and sandy beaches (UNEP, 1997).
Dredging and the dredged disposal materials are also considered an important source
of the filling materials in the tidal flat areas (Barbosa and Almeida, 2001). Landfill speciation
is referring to physical fractionation including; particulate, colloidal and dissolved forms,
while the chemical fractionation including; organic complexes, inorganic complexes and free
metal ions (Larocque and Rasmussen, 1998; Hoz et al. 2000). The suspended solids generated
during these operations cause significant damage to the benthic communities, as the
suspended solids can be carried by the prevailing water current over long distances and thus
cause widespread damage to the marine life. Landfills might contain among many other
constituents, heavy metals in considerable concentrations that may constitute an
environmental problems if they migrate into surface water, the surrounding groundwater
(Baun and Christensen, 2004) or bioaccumulate in the faunal tissues (Cubadda et al. 2001;
Hanna and Muir, 1990; Scott 1990).
As with landfilling, dredging causes immediate destruction for the surrounding
environment. Dredging and disposal dredged materials destroy the benthic communities by
eliminating their habitats (Nawar et al, 2001). The resulting suspended fractions may alter the
water quality and circulation, water mass exchange, surface waves, and long-shore current
patterns, which adversely affect coastal and marine resources.
Many of the tidal flat areas are sustained due to the continual supply sediments which
may be transported along the coast by a variety of processes upon being dispersed into the
marine setting. With greater pressure put on the coastal zone it remains essential to establish
linkages between the sediment supplying and the environments they impact (Nittouer and
Wright, 1994). Thus the understanding of the significance specific discharge classes in
sediment transporting in shallow marine area and the tidal flats may provide delivery insight
into the terrigenous clastic sediments supplied to the coastal zones (Hudson and Mossa,
1997).
The aims of the present study are delineating the causes and sources of the landfilling
and dredging processes along the Red Sea coast, their short and long term effects on the
marine environment, the direct expectable alterations and worse environmental defections.
Also, the study targets to demonstrate the roles of the scientific authorities and decision
makers to dilute the problems and planning for the futurity resources preserving and
developing technically and practically.
774
Al -Gona
Marine
Biol ogi cal
St ati on
Al -Fanadi r
Magawi sh
27 22
27 24
27 20
27 18
27 16
27 14
27 12
27 10
-
-
-
-
-
-
-
-
---
-
-
-
33 40 33 42
33 44
33 46
33 50
33 52
0 0 . 2 0 . 4
0
5
10 km
Sheraton Hot el
----------------------------------
-------------------------
Northern l ocati on
Nort hern Mi dpoi nt
Sout hern Mi dpoi nt
Sout hern Locati on
Sampl ing l ocat i ons
B
Ras Gharieb
Hurghada
Saf aga
Al-Qusier
0 150 300 Km
-
-
-
-
---
32 34 36
24
26
28
30
Mers a Alam
Hamata
Abu Ghsoun
Sinai
Suez
Al-Hamraween
A
34.175 34.18 34.185 34.19 34.195 34.2
26.25
26.255
26.26
26.265
26.27
26.275
Al-Hamar aween Harbour
Phosphate Factory
A
l-
H
a
m
r
a
w
e
e
n
D
w
e
llin
g
s
R
E
D
















S
E
A
W
a
d
i A
b
u
-H
a
m
ra
Wadi Al-Hamr aween
Phosphate Mi nes
Sc ale 1 : 2000
C
B

Fig. (1). Location map shows the studied sites.

775
MATERIALS AND METHODS
The sediment samples were collected to represent the worse and the current situation
in Hurghada city and Al-Hamraween shipping port definitively (Fig., 1). Hurghada marine
area was divided into five sampling sites, infilling sampling zone (7 samples), northern
Hurghada (7 samples), northern midpoint (6 samples), southern midpoint (9 samples) and
southern Hurghada.
In order to evaluate the mount of the filled materials in the navigation basin; the
plunging zone, the embayment and the area under the wharf, Al-Hamraween harbour was
surveyed during the period from June, 30, 2003 to July, 5, 2003 by the Egyptian Navy. The
Navy surveyors divided the navigation basin of the harbour to 27 sections with about constant
intervals (10m) between each two sections. The depth interval was measured each one meter
in order to estimate any probable changing in the depth and to ensure that the change variation
is representing the actual situation. With respect to the actual depths of the basin, the new
depth recorded in the profiles was correlated with the essential depth of the basin during
construction. The grain size fractions of the infilling materials in the basin were studied by
dividing the navigation basin into three sections normal to the beach and 9 samples were
collected (3 samples in each section).
Twenty seawater samples were collected using water sampler (PVC tube 3L).
Seawater characteristics as; salinity (), total dissolved salts (TDS), hydrogen ion
concentration (pH) and dissolved oxygen (DO) were measured using Hydrolab. Instrument
(Surveyor 4, USA).
The monthly maximum, average and minimum wind speeds and directions in
Hurghada locality throughout the last year 2003 were estimated from the weather station of
the National Institute of oceanography and Fisheries, Hurghada branch.
In the laboratory, the collected samples were washed, air dried, dis-aggregated with
fingers and sieved every one-phi () interval relative to Wentworth scale (Folk 1974). There
are seven different fractions were obtained including; gravel (
-1
), very coarse sand (
0
),
coarse sand (
1
), medium sand (
2
), fine sand (
3
), very fine sand (
4
) and mud (
5
).
Statistically, the sediment fractions were categorized into three groups, each of them
represent mode of transportation according to Visher (1969); surface creeping or the bedload
transportation (
-1
+
0
), saltation mode (
1
+
2
), and the true suspension mode (
3
+
4
+
5
).
the percentage of each mode relative to the other modes were calculated in each zone of
study; the infilling locations, near the beach and in the zones of fine sediments accumulations,
these percentages illustrate the effective mode of sediments transportation in each location.

RESULTS AND DISCUSSION
Sediments are often highly heterogeneous; their nature, composition, and contaminant
characteristics can vary greatly within small areas. The main sediment constituents are mainly
controlled by the feeding sources and the local conditions of the zones of accumulation zone
(winds, waves and currents) in addition to the nature of the cumulated sediments, therefore
the sediments can be composed of widely variable proportions of constituents from both
natural and anthropogenic sources, such as organic and inorganic erosional materials, clays,
detrital materials, carbonate precipitates, calcareous organism and plant remains.
Sedimentary cycles for the tidal flats in the semi-enclosed tidal zones appear to be
complicated. The most important factors to induce different sedimentary cycles has been
considered to be the inner coastline geometry of the area immediately seaward of the surf
zone where waves frequently agitate the bed in the inner shelf (Wright, 1995) and hence the
degree of effects of these prevailing wind-generated waves and currents on different places
relative to the direction of winter monsoonal winds and the sediment source (Crockett and
Nittrouer, 2004; Ryu et al. 2004).
776
The coral reef system fronting most of the coastline of the Red Sea provides natural
protection to the aquatic system. Its pronounced morphologic features protect coastal
recreation facilities located behind the reef (Frihy et al. 2004). They added, the rapid and
uncontrolled development of the Red Sea coast has resulted in a series of disturbances that are
not friendly with the surrounding environment. Theses developments involve excavations for
creating artificial lagoons and private wharfs for the safari and diving trip boats. They also
documented, large amount of fillings on the reef flat are dumped to create space and solid
embankments are built as jetties and mooring areas.

Transportation mechanism
In the coastal areas, the effect of wave winnowing and near shore currents on the
different sediment sizes result in three modes of transportation; surface creep or bedload,
saltation and true suspension transports. Surface creep or the bedload transport involves the
coarse-grained fractions that approximatly more than 1.00mm diameter with a different
degree of sorting more than the other two modes. Saltation mode involves the grain size
distribution of the moving particles approximatly between 0.75mm to 1.00mm. The sediment
grades of this mode are relatively the same and have high degree of sorting. True suspension
caused by turbulence where there is no vertical change in the grains occurs in the very fine
sand range, typically less than 0.1mm. These values depends on the actual velocities on site
resulting from long shore currents bed resulting from waves.The sediment particles that may
be held in suspension depend upon turbulence; consequently; the break or truncation point
between suspension and bedload transport may be highly variable and reflects physical
conditions at the time of deposition (Visher, 1969). The bedload transport is the mode in
which the grains are displaces from their niche positions shoreward with the onshore flow and
settle back to the stationary position on the bed before entrained by the returning offshore
flow (Hardisty, 1994). He added, the intensity of the flow conditions can entrain the particles
into the flow body and then undergo suspended load transport whereas, the sediment
transported in the beach system is moved in suspension.
The break between saltation and suspension was in the very fine sand near
3
to
5
.
The fine saltation suggests winnowing by wave action and the poorly sorted sediments
suggest the dumping from highly turbulence graded suspension-traction carpet. The
mechanism of the graded suspension is controlled by the current velocity. The turbidity
current deposits are characterized by large, poorly sorted suspension population, which
includes grain sizes from clay and silt to about 1mm (Visher, 1969).
Winds in the Red Sea mainly blow from the NW and N during most of the year
(Morcos, 1970). Currents are induced by tidal current, wind driven current, currents generated
by the wave breaking close the shoreline (Frihy et al. 2004). Tides are semidiurnal with an
average spring range of 0.5m and with a mean sea level up to 1m higher in winter and a six
hour time difference exists between tide times (UNEP, 1997). The fundamental movements of
surface water in the Red Sea follow the winds, so that the northerly wind of summer drives
surface water south for about four months at a velocity of 12-50 cm/sec, while in winter, the
follow is reversed (Sheppard et al. 1992).
With respect to the wind speed and direction (Fig., 2) recorded measurements in
Hurghada area (Table, 1), throughout the period between January and August, the wind blow
NW and N directions. The maximum recorded speed 9.05 m/sec while the mean average
through the same period is 2.62 m/sec. During September, the wind direction is inclined
Between W and NW mainly with average speed of 2.70 m/sec. During the period between
October and December, the wind direction is completely changing with reverse direction from
NE to E with the mean average speed 8.33 m/sec. The recorded data indicates that the areas of
investigation are considered windy but not stormy almost the year and the winds are
responsible about the infilling sediment motion from the original location to the re-
777
depositional basins. Subsequently, the effectiveness of each transportation mode is controlled
by the local conditions at the source sediments (impact source). The wind induced waves and
currents strength along the shoreline are the essential forces for the transportation.
The recorded wind speeds indicate that the modes of transportation have differential
movement relative to the effective waves and currents which means, the sediments that
belong to the surface creep or the bedload mode are formed in situ or artificially transported
to the present locations, while the saltation mode sediments are slightly move from the
original location with the high wave effect along the shore line and redeposit in the nearest
tidal zones, also the sediments belongs this mode are mainly have generated by the natural
erosion for the underlying and surrounding terraces, hard skeletons of the benthos and the
foraminiferal tests. In saltation mode, the particle movements may be prevented by the
construction of jetties, breakwaters and groynes. Such structures act as dams to the shore-
parallel transport, causing a buildup of the beach on the updrift side and corresponding
erosion in the downdrift direction. This shore-parallel transport has been termed longshore
transport, littoral drift or littoral transport which is induced by the tidal currents (Hardisty,
1994). He added that waves induce longshore currents. They are the main cause of sediment
movement while the other currents are only effective under exceptional circumstances. These
investigations are completely coincided with the present situation in Hurghada.
The true suspension mode is the essential mode of transportation along the coastal
zone of the Red Sea. This mode includes; the fine, very fine, suspension, particulate and
colloidal materials in addition to the dissolved portions of the infilling constituents. The
harmful effects on the marine environment are accompanied with this transportation mode.

January
February
March
April
May
June
July
August
September
October
November
December
Frequency (%)
I I I
0 10 20

Fig., (2) Rose Diagram for the main wind directions through the year, 2003.

Table (1). Maximum, average and minimum wind speeds with the direction through a year:
Max. (m/sec.) Average (m/sec.) Min. (m/sec.) Main Direction
Jan. 2003 5.28 2.58 0.03 NW - N
Feb. 2003 5.28 2.55 0.001 NW - N
Mar. 2003 7.87 2.61 0.001 NW - N
Ap. 2003 5.12 2.64 0.001 NW - N
May 2003 5.22 2.6 0.001 NW - N
June 2003 7.77 2.64 0.001 NW - N
July 2003 9.05 2.66 0.001 NW - N
Aug. 2003 7.83 2.67 0.001 NW - N
Sep. 2003 7.94 2.69 0.001 W - NW
Oct. 2003 7.94 2.08 0.003 NE - E
Nov. 2003 2.51 2.39 2.26 NE - E
Dec. 2003 2.61 2.55 2.47 NE - E
778
Hurghada case (Recent landfill)
The wide spread impact of the sediments has been observed in Hurghada, where such
sediments have spread to extensive fringing reefs down the coastline and to the adjacent
islands and offshore reefs, forming a thick veneer of fine sediments on the surface of the
corals (UNEP, 1997).
Frihy et al. (2004) reported that the coastline of Hurghada is naturally sheltered
against waves dominantly blown from the north and northwest. Coarse sand partially covers
the beach and the inner shallow part of the tidal flat, about half of its width. They added, the
outer part of the tidal flat and its contiguous tidal slope including tidal crest are mainly
composed of reefal rocks. Sand also covers the bottom of the natural lagoons pools which
exist on the outer part of the tidal flat. The shallow marine embayments act as sediment traps
and can preserve the signature of the source (Preda and Cox, 2005).
Geomorphologically, the marine area off Hurghada can be divided into three main
zones; the northern zone extends from El-Gona to northern El-Fanadir Rocky Islands, middle
zone extends between the southern limit of El-Fanadir to the end of Umm-Gawish Island and
the southern zone that extending to the south of Umm-Gawish Island (Fig., 1b). The northern
zone is characterized by the very wide tidal zone extension with extending fringing corals
face directly the N-NW wind blowing. This situation provides high protection for the coastal
area from the intensive surge waves and currents. Also this zone characterized by the presence
of many channels, lagoons and embayments that provide natural sedimentation basins for the
fine and suspended loads of the water column.
The second zone is distinguished by the presence of the islands series including; El-
Fanadir, Abu minqar, the Great Gifftun, the Small Gifftun and Umm-gawish islands that
cause a protection for the coastal areas and provide some sheltered places suitable for the
sediment load redeposition but with lesser degree than the northern zone. Dar (2002) reported
that the area between El-Fanadir and the northern edge of Abu Minqar Island tidal flat
restricts a wide path for the water masses displace with the main direction of winds.
The third zone south of Umm-Gawish misses the protection that present in the other
zones, whereas the coastal areas are directly facing the surge waves and currents.
In the infilling zone sediments, the coarse fraction are dominant averaging of (49-
40%) followed by the saltation mode sediments (33-46%) and the rest percentage (17-13%)
represents the suspension mode sediments (Fig. 3A). It is clear that if these sediment mixtures
are exposed to the weakest wave agitation, vibration and winnowing, more than 15% of them
will load with the water column and move away from the original location. This type of the
infilling sediments is widely distributed in Hurghada to fill the tidal areas for many purposes;
land spites, jetties, beach enhancement and to provide sandy beaches in the rocky tidal flats.
In addition to this sediment type, some investors tend to reduce the transpiration cost of this
sand type by use excavate the coral terraces or bringing dust or shale rocks from the nearest
quarries. These materials are mainly composed from the finest sand grades and subsequently
the suspension load will geminate. Also, the construction rubbishes were widely thrown in the
tidal zones in order to reduce and hide the illegal infringements.
We expect that, the situation in the four selected sectors in marine area off Hurghada
represents the net summation of all these activities along the coast.
In northern Hurghada (Fig. 3B), true suspension sediment mode represents more than
half (av. 55.31%) of the total sediment fractions followed by saltation mode sediments (av.
34.90%), while in the northern midpoint this sediment mode is the dominant (av. 76.29%)
with relatively lower representation for the other modes (Fig. 3C). In the southern midpoint,
the situation appears to be more gracious than the northern midpoint and more likely to
northern Hurghada whereas, the suspension mode represent a half of the total sediment
779
percentage followed by the saltation mode (Fig. 3D). In southern sector, the saltation mode is
the dominant (54.32%) and the other two modes are of subequal percentages (Fig. 3E).
In northern Hurghada and northern midpoint, the fringing reef protection allows the
fine particles of the sediments settle down on the back reefs, sheltered zones, natural lagoons
and the embayments densely. The intensive deposition in the mid point is due to the fine
particles that come from the coastal areas and the water loads that driven by wind and coming
from the north. In the southern midpoint, the water motion between El-Fanadir and Abu
Minqar islands help to minimize the filling impact on the surrounding environment despite
the huge fillings in this zone from both landfill and dredging operations. The highly impacted
areas are present in the sheltered sides of the islands. The field observations illustrated that the
benthos communities of the western side of the Great Gifftun Island are completely destroyed
Fig., 3F).

A
Bedload
50%
Saltation
33%
Suspension
17%
B
Saltati on
35%
Suspensi on
55%
Bedl oad
10%

C
Sal tati on
16%
Suspension
76%
Bedl oad
8%
D
Suspensi on
50%
Sal tati on
38%
Bedl oad
12%

E Bedl oad
22%
Suspensi on
23%
Sal tati on
55%
F
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
I
n
fi
l
l
i
n
g
z
o
n
e
N
. H
u
r
g
h
a
d
a
N
. M
i
d
p
o
i
n
t
S
. M
i
d
p
o
i
n
t
S
. H
u
r
g
h
a
d
a

Fig. (3) The sediment grain sizes of the different transportation modes in Hurghada;
A- Infilling zone, B-Northern location, C- Northern midpoint, D- southern midpoint,
E- Southern location and, F- The correlation between the different zones.

780
There are 175 infringement cases on the tidal zones of the Red Sea were recorded in
Hurghada city in the period between 1980 and 1994. The total filled areas reach to about
2,198,643 sq. meters. The filled tidal areas are varying between 140 sq. meters and 213,000
sq. meters with water depth variation between 1m and about 26 m. Consequently, the infilling
sediment volume is more or less 30,000,000 cubic meters. In addition to many cases of
dredging and terrace excavation by very rudimental technology and dynamite were recorded
especially in northern and far south of Hurghada for artificial lagoons, swimming pools,
private marinas and jetties.
According to the sediment nature of the infilling materials, excavation and dredging
operations, the suspended materials that move from the infilling location and impact the
marine ecosystem may reach to about 20% from the total filled and dredged sediments. This
suggestion in accompanied with the present situation in the different impacted and sheltered
areas.

Al-Hamraween case (Chronic landfill)
Al-Hamraween is located about 60 km southern of Safaga, 20 km northern of Qusier
City and about 120 from the Capital City of the Red Sea governorate (Hurghada). It is
dominated between lat. 26 15N & 26 17N and long. 34 10E & 34 13E. The harbour
occupies the southern part of the settlement lying directly on small embayment. The harbour
includes one wharf extending inside the navigation basin to the suitable depth for shipping
and contains all the loading/unloading facilities (Fig., 1C).
Al-Hamraween locality contains the biggest factory for grinding and sieving the raw
boulders and the big stones till reach the allowable sizes to the shipping transfer. The factory
is located directly near the port zone. The grinded raw materials contain a mixture of all size
grains beginning from cobbles and pebbles to the dust size. Throughout the transferring cycle,
this mixture is exposed to the wind effect which leaches directly the finest particles especially
during the windy days relative to the wind strength. The winds direction is perpendicular to
the Wharf that can emanate large amounts of the fine size materials including the most sand
grades, dust and smothers. This process is continuously occurring each loading cycle in every
day. In the reverse wind days, the amount of volatized materials geminates to more than four
or five folders due to the eddy winds that generate due to the reverse wind cycle. Most of the
volatilized materials were settled down in the navigation basin especially below the shipping
wharf.
These settled sediments are distributed under and around the wharf occupying all the
marine area around the harbour with increase the bed thickness southward. The infilling
amounts are increasing day by day with respect to the diurnal loading and with time whereas
they are threaten the maritime activities inside the harbour by reducing the appropriate
plunging depth to less than the required depth for navigation. The amount of the filling
sediments in the navigation basin were estimated by about 12,500 cubic meters distributed in
the plunging basin and increasing southward.
The bedload sediments are the major components of the sections I and II (av. 45.76%
and 49.13% respectively) as shown in the figures (4A & B), followed by saltation mode
sediment (40.77% and 42.56% respectively), Fig. 4C, while in section III, the saltation mode
sediment is the dominant category (av. 54.09%) with slightly convergent values for the other
modes. These data illustrate that, the bedload mode is dominant directly under the shipping
wharf and decreases southward, while both the saltation mode increases south ward to the end
of the navigation basin (Fig., 4D). The true suspension mode is progressively increasing
southward for very long distance whereas its impacts were observed in many kilometers to the
south belong the main wind direction.
The dredging involves slurry the sediments with water in one part sediment to four
parts water mixture (Jones and Lee, 1978), where this mixture is typically then pumped as
781
slurry to either open water or to confined upland disposal or to a hopper of a hydraulic dredge.
The sediment slurries result in mixing of the sediments with water which tends to promote the
release of constituents from the sediments and in the interstitial water into the seawater (Lee
et al. 1978). The excess water is allowed to drain off at the dredging site and consequently
mixes with the sea water and transport with the currents to the neighbor sheltered location
whereas the slurry settles down over the present living communities.
This is the present situation of the harbour and it is also present in many harbours
along the Red Sea. The harbour rehabilitation by removing the sediment deterrents in the
navigation basin is repeated every two or three years. The severity is branched and divided
between the suspended loads, dust smothers and agitation the settled particles again through
the dredging operations. All these materials are moving to the endpoint of settlement over the
coastal habitats southern of the harbour and can easily recognize.
In addition to the complete degradation in the coastal habitats, the dredging activities
are also increasing the water column turbidity for long time depending on the local conditions
in the Red Sea. This turbid status reaches to the point where the amount of photosynthesis that
can occur in the water is curtailed with the result that there is an overall impairment of the
function of the ecosystem algal growth.
It is clear that, the applying technologies in the shipping operations in all Egyptian
harbours require new lock practically and financially to change the rudimental manners and
displace the old machineries with the advancing and appropriate techniques.

A
Bedl oad
46%
Sal tati on
41%
Suspension
13%
B
Bedload
49%
Sal tation
43%
Suspension
8%

C
Bedload
28%
Saltation
54%
Suspension
18%
D
0%
20%
40%
60%
80%
100%
Northern S. Middle S. Southern S.

Fig., (4) The sediment grain sizes of the different transportation modes in Al-Hamraween
Hrbour; A- Northern section, B- Middle section, C- Southern section and D- The
correlation between the different sections.
GEO-FACIES ALTERATION
Generally the marine sediments are mainly carbonates that have biogenic nature
consisting mainly from; terrace erosional clasts, coral reefs debris, foraminifera, coralline
algae, shells and shell fragments, bryozoans, echinoderms, brachiopods and sponges, in
addition to many other types of the minor constituents (El-Sayed, 1984). These sediments
782
contain considerable portions of the terrigenous sediments that may transfer to the marine
ecosystem with winds or through the floods times. The natural terrigenous constituents in the
marine environment are gradually decreasing seaward.
These biogenic carbonate sediments are mainly of coarse grain sediments with very
low percentage of silt grains. This nature allows the oxygen to penetrate the sediment layer
and providing oxidizing conditions over/in and around the tidal flat sediments. The oxic
conditions are permitting the different types benthos (Fauna and flora) growing up healthy
with observable biodiversity.
Dar (2004) recorded that the high landfilling changes the bottom sediment
characteristically. He added, the areas that covered with fine-grained sediments are
characterized by high reducing nature (low Eh), which decrease the chances of living biota
on and through them. On the other hand, the areas covered with the coarser sediments
(mixture of gravel, sand and mud) are more oxidize (high Eh).
The availability of a sediment associated contaminant to aquatic life is controlled by
the bonding affinity of the contaminant for the various solid phases present in the sediments
(Wright, 1992). In general, the reason that contaminants are present in sediments is that they
are particulate or tend to bind to some component of the sediment that has a density greater
than water as; clay minerals and clay-sized particles, hydrous oxides of iron, manganese,
natural terrestrial and aquatic detrital organics and organic carbon introduced from the
activities of man causing it to settle under quiescent conditions (Wright et al. 1992).

SEAWATER ALTERATION
It is further recognized that dredging and dredged sediment disposal practices as a
result of landfill or as part harbor navigation depth maintenance will have hazardous impacts
on the water quality and consequently the living communities inside and around the zone of
the impacts. The significant problem in determine the water quality to filling and dredged
sediment disposals is that the particulate matter constituents. This mixing between the fine
particle sediments and the seawater causes a change in liquid-solid ratio which would tend to
promote release of dissolved salts from a sediment surface to the seawater (Long and Morgan,
1990).
The suspended sediments in the water column consist of two active regions; the
uppermost region is the water column in which there is typically a small amount of suspended
particles, the next region is just above the sediment water interface where there is typically a
significant increase in the number of suspended sediment particles which can form slurry
suspended sediments (Long et al. 1995). This is the area where constituents released from the
bedded sediments are mixed into the overlying waters. This is also the area where those
organisms that are present in the sediment that depend on oxygen as well as exposure to
constituents present either in the water column or released from the sediments.
Orpin et al. (2004) reported that the water turbidity in coastal areas is primarily a
function of eave-generated bottom shear stress whereas, when the stress is below a critical
value, no sediment is suspended from the bed. Fahmy (2003) reported that, the total
suspended matter in the coastal water of the Red Sea ranged from 4.47 to 7.58 mg/l with an
annual mean 5.85mg/l and the Secchi disk transparency ranged from 3.75 to 14.63m with an
annual average of 7.71m.
Turbidity in the shallow and tidal areas of the Red Sea is not predominant but it is
controlled by the local conditions and the suspended sediment availability. Generally, in the
windy days, the wave winnowing agitates the coastal sediments and the turbidity rates
increase gradually. The other source of turbidity is the deposited fine sediments in the zones
of accumulation, in the period of wind reversible direction (October to December), the
direction change causes suspended matter vibration and subsequently the turbidity increases.
783
Reef corals exist in seawater salinities ranging from 25 to 42. At the lower end of
the salinity tolerance range, many examples exist of lethal and sublethal effects of lowered
salinities following storm and flood events. Salinity impacts to corals are confounded by other
flood related stresses such as sedimentation, turbidity and increased ultraviolet radiation
exposure (David, 2001). Hanna et al. (1988) recorded that the seasonal variation of the
average salinity in Hurghada ranged from 40.12 and 40.65. Madkour (2004) recorded
salinity variation between 41.70 and 42.30. The recent salinity measurements in the
present study recorded variation between 41.00 and 43.37. The gradual salinity
increasing indicates to the water quality degradation with increasing suspended load and the
dissolved slats from the coastal activities. These renewing conditions comprise intensive
stress on the benthic communities and can easily change the interrelation between the
different components of the organisms.
BIO-FACIES ALTERATION
Jones and Lee (1978) assert that increased suspended sediment concentrations are
significantly adverse to aquatic life. The suspended sediments typically have greater impact
on fringing reefs than on reefs distant from shore. The sediment reaching a coral reef reducing
sunlight availability (turbidity), which is required by the zooxanthellae for photosynthesis,
thereby affecting the ecology and composition of reef communities (Thomas et al. 2003) also
affect the growth of the coral (Nugues and Roberts, 2003). In severe cases, sedimentation can
kill corals outright through smothering (Anthony, 1999). Furthermore, sediments supply to
reefs is responsible for the nutrients, heavy metals as well as other contaminant enrichments.
The light penetration reduction with the high turbidity rate in the high filling places
are the important reasons for changing the benthic community structures. The increasing in
dissolved nutrients leads to the proliferation of algae including phytoplankton, benthic
macroalgae and algal epiphytes (David, 2001) rather than the other communities as the coral
reefs. The severe turbidity increasing in the water column is responsible about tidal habitat
destruction and the swimming fauna migration, subsequently the new conditions are favorable
for the undesired benthos as some sea urchin species (Diadema setosum).
CONCLUSION
The landfill along the Red Sea coast is the essential problem that threatens the marine
environment. This problem is essentially engaged with the mining and shipping activities of
the harbours, whereas it is considered chronic and renewable case. A major engineering effort is
often required to remove the filling sediment from the navigation basins entrance by dredging these
basins and placing the dredged material at feasible and physically appropriate locations (Seabergh
and Kraus, 2003).
The recent landfill in Hurghada that engaged with the tourist activities is relatively
easier than the chronic case. The political disciplines and legal punishment for the
infringement cases will surely save the marine resources.
The management of harbor development, shore protection, and system conservation
will require a more focused attention in the future and creative solutions must be sought for
the multifaceted problems of the coastal zone (Cuadrado et al. 2005). Monitoring programs
are necessity in the harbour developments, coastal and tidal habitats rehabilitation in order to
preserve the marine resources inside ambitious national plans in assortment between the
Egyptian Scientists and the decision makers.
REFERENCES
Anthony, K.R.N. (1999). Coral suspension feeding on fine particulate matter. J. Exp. Mar.
Biol. Ecol., 232: 85-106.
Barbosa, M.C. and Almeida, M.S. (2001). Dredging and disposal of fine sediments in the
state of Rio de Janeiro, Brazil. Journal of Hazardous Materials, 85: 15-38.
784
Baun, D.L. and Christensen, T.H. (2004). Speciation of heavy metals in landfill leachate: a
review. Waste Management & Research, 22: 3 23.
Crockett, J.S. and Nittrouer, C.A. (2004). The sandy inner shelf as a repository for muddy
sediment: an example from Northern California. Continental Shelf Research, 24: 5573.
Cuadrado, D.G.; Gomez, E.A. and Ginsberg, S.S. (2005). Tidal and longshore sediment
transport associated to a coastal structure. Estuarine, Coastal and Shelf Science, 62: 291-
300.
Cubadda, F.; Conti, M.E. and Campanella, L. (2001). Size-dependant concentrations of trace
metals in four Mediterranean gastropods. Chemosphere, 45: 561-569.
Dar, M. A. (2004). Holothurians role in the marine sediments reworking processes.
Sedimentology of Egypt, 12: 173-183.
Dar, M.A. (2002). Geological bases to study the environmental defect in the marine
ecosystem as a result of touristic activities in Hurghada area and surroundings, Red Sea,
Egypt. Ph. D. Thesis, Sci. Faculty, Suez Canal Univ., 218p.
David, H. (2001): Great Barrier Reef Water Quality: Current Issues. Great Barrier Reef
Marine Park Authority. Townsville, 97p.
El-Sayed, M. K. (1984). Reefal sediments of Al-Ghardaqa, Northern Red Sea, Egypt. Marine
Geology, 56: 259-271.
Fahmy, M.A. (2003). Water quality in the Red Sea coastal waters (Egypt): Analysis of spetial
and temporal variability. Chemistry and Ecology, 19(1): 67-77.
Folk, R.L. (1974): Petrology of sedimentary rocks. University of Texas, Hemphill Pub. Co.,
182p.
Frihy, O.E.; El Ganaini, M.A.; El Sayed, W.R. and Iskander M.M. (2004). The role of
fringing coral reef in beach protection of Hurghada, Gulf of Suez, Red Sea of Egypt.
Ecological Engineering, 22: 17-25.
Hanna, R. G. and Muir, G. L. (1990). Red Sea Corals as biomonitors of trace metal pollution.
Environ. Mon. Assess., 14, 211-222.
Hanna, R.G.; Saad, M.A. and Kandeel, M.M. (1988). Hyrographical studies on the Red Sea
water in front of Hurghada. Marine Mesopotamica, 3(2): 139-156.
Hardisty, J. (1994). Beach and nearshore sediment transport. In: Sediment transport and
depositional processes. Pye K., Blackwell Sci. Pub., 397p.
Havens, K.J.; Varnell, L.M. and Bradshaw, J.G. (1995). An assessment of ecological
conditions in a constructed tidal marsh and two natural reference tidal marshes in coastal
Virginia. Ecol. Eng., 4: 117-141.
Hoz, L.R.; Edwards, A.C. and Hernandez, M.L. (2000). Heavy metals in sediments of a large
turbid tropical lake affected by anthropogenic discharges. Envrion. Geo., 39 (3-4): 378-383.
Hudson, P.F. and Mossa, J. (1997). Suspended sediment transport effectiveness of three large
impounded rivers, U.S. Gulf Coastal Plain. Environmental Geology, 32 (4): 263-273.
Jones, R. A. and Lee, G. F., (1978). Evaluation of the elutriate test as a method of predicting
contaminant release during open water disposal of dredged sediment and environmental
impact of open water dredged material disposal. Technical Report, Vol. I- Discussion,
Army Engineer Waterways Experiment Station, D: 45-78.
Larocque, A.C.L. and Rasmussen, P.E. (1998). An overview of trace metals in the
environment, from mobilization to remediation. Environ. Geol., 33: 85-90.
Lee, G. F., Jones, R. A., Saleh, F. Y., Mariani, G. M., Homer, D. H., Butler, J. S., and
Bandyopadhyay, P. (1978). Evaluation of the elutriate test as a method of predicting
contaminant release during open water disposal of dredged sediment and environmental
impact of open water dredged material disposal. Technical Report, Vol. II- Data Report,
Army Engineer Waterways Experiment Station, D: 78-125.
785
Long, E. R. and Morgan L. (1990). The Potential for Biological Effects of Sediment-sorbed
Contaminants Tested in the National Status and Trends Program. NOAA, Technical
Memorandum NOS, OMA, 52p.
Long, E. R.; MacDonald, D. D.; Smith, S. L. and Calder, F. D., (1995). Incidence of Adverse
Biological Effects Within Ranges of Chemical Concentrations in Marine and Estuarine
Sediments. Environmental Management, 19: 81-97.
Madkour, H.A. (2004). Geochemical and environmental studies of recent marine sediments
and some invertebrates of the Red Sea, Egypt. Ph. D. Thesis, Sci. Faculty, South Valley
Univ., 319p.
Morcos, S.A. (1970). Physical and chemical oceanography of the Red Sea. Oceanographic
and Marine Biology Annual Reviews, 8: 73-202.
Nawar, A.H.; Soliman, F.A. and Dar, M.A. (2001). Landfill effect on the bottom sediments of
the urban zone of Hurghada marine locality, Red Sea, Egypt. Journal of Environmental
Research, Zagazig Univ, Egypt, 3: 70-92.
Nittrouer, C.A., Wright, L.D. (1994). Transport of particles across continental shelves.
Review of Geophysics, 32: 85113.
Nugues, M.M. and Roberts, C.M. (2003). Partial mortality in massive corals as an indicator of
sediment stress on coral reefs. Marine Pollution Bull., 46: 314-323.
Orpin, A.R.; Ridd, P.V.; Thomas, S.; Anthony, K.R.; Marshal, P. and Oliver, J. (2004).
Natural turbidity variability of anthropogenic sediment discharge near sensitive
environments. Marine Pollution Bull., 49(7-8): 602-612.
Preda, M. and Cox, M.E. (2005). Chemical and mineralogical composition of marine
sediments, and relation to their source and transport, Gulf of Carpentaria, Northern
Australia. Journal of Marine Systems, 53: 169-186.
Ryu, S.O; Lee, H.J. and Chang, J.H. (2004). Seasonal cycle of sedimentary process on
mesotidal flats in the semienclosed Muan Bay, southern west coast of Korea: culminating
summertime erosion. Continental Shelf Research, 24: 137147.
Scott, J.B. (1990). Chronic pollution recorded in coral skeleton in Hong Kong. J. Exp. Biol.
Ecol., 139: 51-64.
Seabergh, W.C. and C. Kraus, N.C. (2003). Progress in management of sediment bypassing at
coastal inlets: Natural bypassing, weir jetties, jetty spurs, and engineering aids in design.
Coastal Engineering Journal, 45(4): 533-563.
Sheppard, C.; Price, A. and Roberts, C. (1992). Marine ecology of the Arabian Region.
Academic Press, San Diego, California, 358p.
Thomas, S.; Ridd, P.V. and Day, G. (2003). Turbidity regimes over fringing coral reefs near a
mining site at Lihir Island, Papua New Guinea. Marine Pollution Bull., 46: 1006-1014.
UNEP (1997). Assessment of land-based sources and activities affecting the marine
environment in the Red Sea and Gulf of Aden. Untied Nation Environmental Program
(UNEP), Regional seas reports and studies, No. 166, 62p.
Visher, G. S. (1969): Grain size distribution and depositional processes. J. Sed. Petrol., 39(3):
1074-1106.
Wright, L.D. (1995). Morphodynamics of Inner Continental Shelves. CRC press, Boca Raton,
FL, 241p.
Wright, T. (1992). Evaluation of dredged material for open-water disposal: Numerical criteria
or effects-based. In: Handbook of Dredging Engineering, McGraw-Hill, NY, 959-967.
Wright, T., Engler, R., and Miller, J. (1995). Effects-based testing and sediment quality
criteria for dredged material. IN: Water Quality Standards for the 21st Century EPA, 207-
216.
786
GROUNDWATER CONTAMINATION AND
PROTECTION IN EXTRAORDINARY CONDITIONS

Prof. Eng. Jozef Hulla, DrSc.
Slovak University of Technology, Civil Engineering Faculty, Bratislava, Slovak
Republic
Assoc. Prof. Dr. Eng. Mohamed Taha Abdel-Rahman
Ain Shams University, Civil Engineering Faculty, Cairo, Egypt
Eng. Julius Plsko
EKOSUR Piestany, Slovak Republic
ABSTRACT
Several serious problems concerning groundwater contamination were to
be solved in the Slovak republic conditions in the past. Transport means
accidents often took place on the roads, railway and water flows. Contaminant
substances often leaked from chemical and petrochemical plants and from waste
deposits. Special attention was paid to groundwater contamination near nuclear
power stations. Groundwater quality may have been endangered even at huge
construction works which took place in deep excavations with artificial sealing
elements. Flood situations are also dangerous for ground water resources. One-
or two-dimensional computation models are most often used to study
contaminant substances transmission in porous or fissure media. The models are
input by groundwater flow characteristics (filtration velocities, flow velocities,
effective porosities) and substances transmission characteristics (diffusion
coefficients, dispersion coefficients, dispersivities, retardation factors, half-life
decay). If the most realistic contaminant substances transmission prognosis are
to be created and the most effective groundwater sources protection systems are
to be suggested the characteristics mentioned are preferably to be gained in field
conditions.

Key words: groundwater flow, contaminant transmission, protection measures

1. INTRODUCTION
Significant part of our territory is formed by quartenary gravel and sand sediments
with a relatively high groundwater level. Surface silt layers are of various thicknesses and on
several locations these do not take place. Thicker silt surface layers play a significant
protection role at transport accidents. Locations with small silt surface layers thicknesses or
without them are the most sensitive to accidental groundwater contamination.

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


787
Groundwater contaminant substances transmission prognosis enable us to judge the
resource water quality impairment danger. No protective measures are necessary to be taken
in many cases because the processes connected with substances transmission in the medium
reduce harmful substances concentration to an acceptable level. If the processes mentioned
were not efficient enough it would be necessary to provide suitable protective measures. For
example contamined water pumping from bored wells with a following clean-up and dillution
can be taken into consideration. Substances transmission trace can be made longer with the
aid of diaphragm wall or the contamined water medium can be completely closed and
stopped. However, the way out for such decisions is the trustful transmission prognosis
significantly depending on input data impartialty. Ecological disasters usually do not provide
enough time for thoroughgoing research, thus well estimated characteristics making an
appropriate picture of the conditions have to be trusted.
Basically, we pay attention only to impulse contamination which is typical for
accidental or break-down contaminations. Long time or continual contamination processes
are very dangerous for the groundwater quality and all the measures to minimise them must
be taken as soon as possible.
Hydrodynamic dispersion is the most significant process at higher groundwater
flow velocities in gravel and sand soils. Diffusion processes are important especially at
contaminant sealing walls penetration.

2. COMPUTATION MODELS
Differential equation (Bachmant & Bear, 1964) is usually applied for one dimensional
transmission of those conservative substances which are stable and do not adsorb to the solid
particles surface
x
c
v
x
c
D
t
c
L

2
2
(1)
The solution of the equation is given by Lenda & Zuber (1970) as

=
t D
vt x
t D An
V c
t x c
L
L ef
o o
4
) (
exp
2
) , (
2

(2)
In the equations (1) and (2): c is the concentration, t - time, D
L
- longitudinal
dispersion coefficient, x - distance, v - flow velocity, c
o
- initial concentration ( for x = 0), A -
flow area, n
ef
- effective porosity.
If dispersivity (
L
= D
L
/ v, where v = x / t
o
and t
o
is the top time) is used instead of
the dispersion coefficient (D
L
) the equation (2) for conservative substances gets to

=
o
L
o
o
L ef
o o
t
t
x
t
t
x x
t
t
x An
V c
t x c

4
) (
exp
2
) , (
2
(3)
For unstable substances liable to decay (radioactive), degradation (organic) which can
optionally be adsorbed to solid soil particles the equation (3) gets to

=
T
t
t
t
t
t
xR
t
t
xR An
V c
t x c
o
L
o
o
L ef
o o
693 , 0
4
) 1 (
exp
2
) , (
2

(4)
788
where apart from the known symbols R is the retardation factor and T the half-life
decay.
For practical purposes the maximum substance concentration (c
max
) at a given
distance from the place of application is useful to be known. In such case t = t
o
and the
equation (4) simplifies to
) 693 , 0 exp(
2
max
vT
xR
xR An
V c
c
L ef
o o
=

(5)
One-dimensional substance transmission usually takes place in laboratory conditions
during experiments in cylindrical tanks filled with soil. Even the substances transmission
taking place in two-dimensional field conditions can be replaced by one-dimensional
transmission with some simplifications (Hulla et.al. 1998).
At one-dimensional groudwater flow and two-dimensional dispersion conservative
substances transmission the conditions for the following equation (Bachmant and Bear, 1964)
are fulfilled

c
t
D
c
x
D
c
y
v
c
x
L T
= +
2
2
2
2
(6)
with a solution (Lenda and Zuber, 1970)
c x y t
c V
hn D t
x vt
D t
D t
y
D t
o o
ef L
L
T
T
( , , ) exp
( )
exp( ) =

2
4
1
2
4
2 2

(7)
or
c x y t
c V
hn vt
x vt
vt
vt
y
vt
o o
ef L
L
T
T
( , , ) exp
( )
exp( ) =

2
4
1
2
4
2 2

(8)
Apart from the known symbols in equations (6) to (8) D
T
is the transverzal dispersion
coefficient,
T
- transverzal dispersivity, h - thickness of the watered layer.
The equations (6) to (8) fulfill the substances transmission assumptions from
permeable boreholes. However, they can also be used when contaminant substances
penetrated from the surface through an insaturated zone to the groundwater table. In the
beginning three-dimensional substances transmission takes place, having covered some
distance further transmission becomes two-dimensional. For smaller thicknesses of the
permeable layers the area with three-dimensional transmission gets negligibly small.

3. GROUDWATER FLOW CHARACTERISTICS
The basic groundwater flow characteristics is the filtration velocity. Knowing the
flow field and direct water amount flowing away from the drain or a cylinder filled with soil
measurement the filtration velocity can be simply calculated by the continuity equation.
One borehole tracer methods (Halevy et.al., 1967) are advantageous in field
conditions. Electrolytes, dyes or radioactive substances are commonly used as tracers. The
tracer dillution process or its vertical motion can be monitored in perforated tubes with
filtration fill in boreholes.
Based on the tracer dillution by flowing water proces monitoring approximately in the
horizontal direction through a vertical perforated tube the filtration velocity can be calculated
from the formula
n
n o
f
c c
c c
t
d
v

= ln
4

(9)
789
where d is the perforated tube inner diameter, - borehole drainage influence for
horizontal flow, c - tracer concentration in time t, c
n
- natural concentration before artificial
tracer introduction.
Because of the interconnection of various pressure horizons vertical water motion
emerges in the borehole perforated tube. With the aid of tracers or screws vertical discharges
(q
v
) can be measured and their depth dependency gained. The filtration velocity for
approximately horizontal groundwater motion in the vertical borehole nearness is given by
the equation
h d
q
v
v
f

(10)
where q
v
is the vertical discharge increase (where the water flows into the borehole)
or the decrease (where the water flows out of the borehole) in the depth of h, - borehole
drainage influence for vertical flow, d - perforated tube inner diameter.
From the formulas (9) and (10) filtration velocities for given depths are gained, the
average value for all the borehole, i.e. for all the watered perforated part is calculated as

=
h
h v
v
f
f (11)
At one location, the permeability coefficient was estimated from Darcys law by
means of the tracer method (k
T
) for each borehole. Apart from these results, permeability
coefficients from pumping test (k
P
) and from grain-size analysis from Beyer-Schweiger (k
B
)
and Carman-Kozeny (k
C
) formulas were estimated.
Statistical analysis results for the area tested are shown in Fig. 1. Note that the
permeability coefficients k
C
are generally lower than k
B
; the relation can be expressed by the
formula
8035 , 0
1367 , 0
B C
k k = (12)
(the reliability of this formula is very high, R
2
= 0,93). The pumping tests, vertical
flow tracer measurement median and Beyer-Schweiger median give in gravel soils similar
results.
Groundwater flow velocities are the most reliably found by multi-borehole tracer
methods. Tracer is introduced into one borehole and its motion is monitored in other
boreholes which should be built in the groundwater flow direction at certain distances (x). In
these the
790

Fig. 1. Correlation dependence between the permeability coefficient in the tested area
(k
B
Beyer-Schweiger, k
C
Carman-Kozeny) from grain-size analysis (o average, + -
median), vertical flow tracer measurements ( average, median), and from the pumping
tests ().

concentration dependencies are obtained fromwhich the times corresponding
maximum concentrations (to) can be reliably got. Flow velocities are then calculated from the
formula
o
t
x
v = (13)
An important matter for multi-borehole experiments in gravel soils is the vertical
water motion in the boreholes requiring suitable measures to be provided in order to get
successful and apropriately exact results.
Fig. 2 shows one concentration dependence from the measurements in Jaslovsk
Bohunice for the distance x = 10,3 m, top time

t
o
= 1,65 h = 5940 s and flow velocity v =
10,3/5940 = 0,0017 m/s.
Effective porosity can be found from the formula
v
v
n
f
ef
= (14)
where v
f
is the filtration velocity (e.g. from the one-borehole methods), v - flow
velocity. Imaginary unreal small values of effective porosities in gravel soils are usually
caused by high
values of real flow velocities in thinner degraded layers where sand particles are not
present.
791

Fig. 2 Concentration curve from the measurements in the borehole x = 10,3 m far
from the tracer (NaCl) introduction with symbols for flow velocity and transmission
characteristics calculation in Jaslovsk Bohunice.


4. SUBSTANCES TRANSMISSION CHARACTERISTICS
The most important characteristic for substances transmission analysis is the
longitudinal dispersion coefficient (D
L
). For laboratory experiments in cylinders with sand
soils fulfilling the conditions of one-dimensional substances transmission with short flow
distances (about 0,5 m) a dependence was got
b
L
v a D = (15)
with values of a = 0,09 to 0,4 m, b =1,04 to 1,12 (Rahman 1996).
The concentration curve is usually the start-point for field experiments and the
longitudinal dispersion coefficient is calculated as
o t L
t v D 2 / ) (
2
= (16)
Apart from the known symbols
t
is the deviation; for the results from Fig. 2
t
=
2700 s and D
L
= 0,0018 m
2
.s
-1
.
The transversal dispersivity coefficient can be calculated from an approximate
formula D
T
= D
L
/10.
Another important characteristic is the longitudinal dispersivity
v
D
L
L
= (17)
which for example shall be
L
= 0,0018 / 0,0017 = 1,06 m.
Number of laboratory and field experiments held in gravel and sand soils as well
as in fissured rocks in our country and abroad wittness about a significant influence of the
distance, i.e. the transmission trace (x) upon the longitudinal dispersivity (
L
). The relation
plotted with a full line in Fig. 3 can be defined by the formula
2 2
) 515 , 12 (ln 515 , 12 77 , 4 exp( + = x
L
(18)
and can be used for accident situation contaminat substances transmission orientation
analysis where further information about geological and hydrogeological conditions is not
792
available. For the location significant from the groundwater quality point of view the
dispersivity-distance relation is reasonable to be found or made more exact by an experiment.
The dispersivity increase is obviously interconnected with the penetration of
conservative substances into dead pores in which groundwater does not move. The
substances being adsorbed to soil particles behave in a similar way. As a result of adsorbtion
their motion is slower and can be expressed by a retardation factor
v
v
R
a
= (19)
In the relation (19) v
a
is the adsorbed substance real flow velocity, v - real clean
water flow velocity. The retardation factors for harmful substances are found in laboratory
conditions.
In the formulas (4) and (5) there are half-life periods (T) of some organic or
radioactive substances. For the radioactive hydrogen - tritium - the T = 12,26 years, for
radiocarbon
14
C T


Fig. 3 Dependence of dispersivity (
L
) on the distance (x) according to Klotz and
Moser (1974), Seiler (1985), Gelhar (1986), Luckner and estakov (1986), Drost et.al.
(1991), Plko and Hulla (1990-1996) and others.

= 5730 years, for cesium
137
Cs T = 30 years, for radioactive iodium
131
I T = 8,14 days,
but for
129
I T = 1,7.10
7
years, for radioactive bromium the half-life decay is only 36 hours.

5. GEOTECHNICAL PROTECTIVE MEASURES
According to the pollution prediction, if the maximum pollutant concentration in local
resources (equation 5) is considerably higher than the permissible values, it would be
necessary to take into account supplementary pumping, sealing or reactive protective
measures.


793
5.1 Pumping Systems
If a drilled well is constructed to remove pollutants from the groundwater, or if
suitable well creating part of the exploited water resources is used, the flow field in its
surroundings will undergo changes due to water pumping (Fig. 4).

Fig. 4 Flow field in well surroundings.

From the special area having a width b, the pollutants may be transmitted into the
well. The capture zone width from which the water is drained into the well depends mainly
on the pumping amount Q. Results of our parametric study (Hulla et all, 1996) are presented
in Fig. 5. They are based on the simple Dupuit relationship for a hydraulically perfect well
into which water with unconfined groundwater level is flowing (water depth = 10,0 m, well
diameter = 0,3 m).

Fig. 5 Dependence between the relative pumped amount Q/Q
o
and the relative width
of capture zone b/b
o
.

Under certain conditions, actual basic values of b
o
and Q
o
are specified over the
pumping test, it is feasible by means of graf in Fig. 5 to determine by adequate accuracy also
the values of b and Q for other conditions.
794
A pumping system creating a hydraulic screen (21 wells) has been providing for the
protection of high quality drinking water resources of the Danube Island downstream of the
petrochemical refinery Slovnaft, situated close to Bratislava. In 1966 practically all wells
were contaminated and the first traces of petroleum substances were found in the important
drinking water resources supplying a population of 400000 in Bratislava, having a capacity of
nearly 1 m
3
/s. Contaminated water from the hydraulic screen and also wastewater from the
refinery are treated in an advanced wastewater treatment plant delivered by the Japanese
company Kurita.
At the recycling system the contaminated water from the pumping well is conveyed
through a pipe into recharging well in the counter current direction. The contamined water is
closed and circulates between two wells. It is also possible to develop recycling systems by
means of several pumping and recharging wells.

5.2 Sealing Systems
Protection against accidental emergency pollutant losses is possible, under certain
conditions, using sealing protection systems. It is evident that there would be no time to
develop them if the spillage already has occurred. However, they may be constructed as
preventive measures for the intensification of self purification processes, especially in
locations where traffic roads are close to important water resources.
Sealing systems are constructed as diaphragm walls of various thickness (from 0,1 to
0,6 m) and made using various technologies. Sealing walls are suitable in locations where
they could be embedded into a relative impermeable layer. Anisotropy is a favorable factor in
gravel soils reaching down to deep layers.
Application of sealing diaphragm walls can provide a more intensive self purification
capacity or or completely confining the space containing the contaminated groundwater. The
principle of the sealing wall effect on self purification processes is illustrated in Fig. 6.
Arelatively non-permeable wall forces the polluted water to bypass it. The flow lines get
longer, and thus intensify self purification processes (especially dispersion and adsorption).

Fig. 6 Flow lines bypassing the sealing wall.


When the sealing walls have fully confined the space with contamined water, it is
necessary to pay attention to the fact that the water level in the confined space should always
be lower than the clean water level in the surrounding area. However, the amount of pumped
water is depend to agreat extent upon the wall permeability (k
w
10
-8
m/s).
795

5.3 Permeable Reactive Walls
The principle of groundwater protection by permeable reactive walls is based on the
deepening of vertical slot, filled by permeable reactive medium (Jirasko 2004). Contaminated
groundwater flow perpendicular through the wall, (Fig. 7 a). In the ractive medium take
place physical, chemical or biological processe, adsorption, reduction, oxidation,
degradation. Contaminant is stoped or transformed to less dangerous matter and from the
wall outlet clean water. In some cases is advantageous combination of reactive and classical
sealing walls (Fig. 7 b).

Fig. 7 Protection of groundwater by permeable reactive walls.


Permeable reactive walls can be made as permanent, temporary or exchangeable. For
adsorption process are used zeolithen, active coal, peat, for degradation elementary iron. It is
main job for chemical people to find suitable substances for necessary reactions with
contaminants.
From geotechnical point of view is very important design and construction of
permeable reactive wall in given conditions and judgement of hydrogeological regime
changes. Interested problem is connected by reactive fill exchange. For dangerous
contaminats is necessary systematic quality of water flowing from reactive wall.

6. CONCLUSIONS
Based on the grounwater motion characteristics in the monitored location
hydroisohypses are calculated and their development is compared to hydroisohypses gained
from water table measurements. Having reched an appropriate correspondence the
groundwater flow model is debugged.
In the following stage of the calculation substances transmission for one-shot
contamination is investigated. The result can be the concentration, volume activity or other
characteristic isolines which are compared to the samples taken analysis. In order to rech an
appropriate correspondence the calculations can be repeated with other input data.
If the maximum pollutant concentration in local groundwater resources is
considerably higher than the permissible values, it is necessary to take into account
supplementary pumping, sealing or reactive protective measures.

REFERENCES
Bachmant, Y. and Bear, J., 1964: The general equations of hydrodynamic dispersion in
homogeneous, isotropic porous mediums. J. Geophys. Res. 69, 2561-2571.
Drost, W., et al. 1991: Estimates of dispersivities by interpretation of statistically anisotropic
filtration velocity vectors in fluvioglacial aquifers. In: Extended synopses IAEA-SM-319, Vienna,
168.
796
Gelhar, L., W., 1986: Stochastic subsurface hydrology from theory to applications. Wat. Res.
Research, 22, 9, 135.
Halevy, E. et al., 1967: Borehole dilution techniques: a critical review. In: Isotopes in Hydrology,
IAEA Vienna, 531-564.
Hulla, J., Gramblikov, D. and Abdelrahman, M.T., 1996: Pollutant transport and groundwater
protection at the accident pollution. Podzemn voda, 2, 1, 32-41.
Hulla, J., Bednrov, E., Plko, J. and Kostolansk, M. (1998): Tracer methods for groundwater
flow and pollution transport characterization. In: Geotechnical Site Characterization, Balkema
Rotterdam, 693-698.
Jirasko, D.: Application of permeable reactive barriers for contaminant groundwater remediation
(in czech). In: Foundations, Czech geotechnical society, Brno 2004, 129-134.
Klotz, D. and Moser, H. 1974: Hydrodynamic dispersion as aquifer characteristic. In: Isotopes
techniques in groundwater hydrology, IAEA Vienna, 341-355.
Lenda, A. and Zuber, A., 1970: Tracer dispersion in groundwater experiments. In: Isotope
Hydrology, IAEA Vienna, 619-641.
Luckner, L. and Schestakov, 1986: Migrationprozesse im Boden und Grundwasserbereich. DVG
Leipzig.
Plko, J. et al. 1996: Radioactive pollution and protection of groundwater. In: Environmental
Geotechnics, Balkema Rotterdam, 1173-1178.
Rahman, M. T.: Hydrodynamic dispersion in cohesionless soil and groundwater protection. A
thesis for the degree of Doctor of Philosophy. Slovak university of technology, Bratislava 1995,
164.
Rahman, M., T., and Hulla, J. 1996: Dispersivity and adsorption of conservative pollutants in
cohessionless soils. J. Hydrol. Hydromech., 44, 2-3, 118-131.
Seiler, K., P., 1985: Results of field experiments on hydrodynamic dispersion in Quaternary
gravels of southern Germany. In: Scientific Basis for Resources Management, IAHS, Jerusalem,
351-360.


The paper emerged solving the grant project No. 1/2148/05





















797

BASIN SHAPE CHARACTERISTICS OF MAJOR
EASTERN
DESERT HYDROGRAPHIC BASINS (EGYPT)
BY
ATEF K.F.SAAD
Central Laboratory for Environmental Quality Monitoring, National Water Research Center, El-
Kanatir , Qualubia. (a.saad@cleqm.org)


ABSTRACT:
The Eastern Desert (E.D) which covers an area of about 222,000 km
2
omprising
54 hydrographic basins (wadis), ranging from large to very small basins in area.
The occasional runoff emitting from these hydrographic basins depends specifically
on two main factors which are; climatic factors and physiographic factors. The
climatic factors are composed generally of rainfall, interception, evaporation and
evaptranpiration. The physiographic factors are composed of both the drainage
basin characteristics as represented by its geomorphology and physical setup, and
the water channel characteristics which are represented by its carrying and storage
capacity.
The present paper deals mainly with Basin shape characteristics specifically
basin Length (L
b
), Rotundity factor (R), Form Factor (R
f
), Circularity ratio (R
c
),
Elongation ratio (R
e
) and Channel maintenance (C). These factors which are
quantitative expressions of the drainage basins can best represent the behavior of
anticipated runoff risk. In respect to the geomorphologic studies, the focus is made
specifically on the determination of geomorphologic characteristics of the
hydrographic basins, which are mainly bifurcation ratio (R
b
), density (D) and
frequency (F). Knowledge of both the geomorphologic factors and the basin shape
characteristics are used to identify the risk value for each basin, its respective
ranking with respect to flooding probability and the importance of each of these
parameters. Only the Nine major basins in the E.D. are were taken into account
namely W. Sannur, W. Tarfa, W. Assiuty, W. Qena, W. Laquitta, W. abbad, W.
Shait, W. Kharit and W. Allaqui.


FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


798


INTRODUCTION:

The Eastern Desert (E.D) occupies an area extending between latitude 31
0
00 to the north and
latitude 22
0
00

at the Egyptian-Sudanese border to the south, and the strip of land laying between
the Red Sea and gulf of Suez at the east, and the Nile Valley to the west. The Eastern Desert
covers an area of about 222,000 km
2
within the Egyptian territories only. This area comprises 54
hydrographic basins (wadis), ranging from large to very small basins in area. Although rain falls
infrequently in arid and semi-arid lands, e.g. lower Egypt and the Eastern Desert, it occasionally
comprises considerable amount of water that may form a sheet of runoff flow, particularly in
basins of large catchments areas. This runoff may turn to be destructive to the adjacent
environment.

The occasional runoff emitting from hydrographic basins depends specifically on two main
factors which are; climatic factors and physiographic. The climatic factors are composed
generally of rainfall, interception, evaporation and evaptranpiration. The physiographic factors
are composed of both the drainage basin characteristics as represented by its geomorphology and
physical setup, and the water channel characteristics which are represented by its carrying and
storage capacity.

The majority of these basins has been studied from various aspects i.e. geologic and
geomorphologic, by several scientists, e.g. Said (1990), El-Shamy, (1988), Shata.A.,(1992),
BRGM (1980), El-Gindy (1982), Tahlawy (1972), and Shahin (1985), as well as by oil
companies, the geological survey, Ministry of Public Works and Water Resources (RIGW) and
Academy of Scientific Researches and Technology.

The present paper deals mainly with Basin shape characteristics specifically basin Length (L
b
),
Rotundity factor (R), Form Factor (R
f
), Circularity ratio (R
c
), Elongation ratio (R
e
) and Channel
maintenance (C). These factors which are quantitative expressions of the drainage basins can best
represent the behavior of anticipated runoff risk. In respect to the geomorphologic studies, the
focus is made specifically on the determination of geomorphologic characteristics of the
hydrographic basins, which are mainly bifurcation ratio (R
b
), density (D) and frequency (F).
Knowledge of both the geomorphologic factors and the basin shape characteristics are used to
identify the risk value for each basin, its respective ranking with respect to flooding probability
and the importance of each of these parameters. Only the Nine major basins in the E.D. are were
taken into account namely W. Sannur, W. Tarfa, W. Assiuty, W. Qena, W. Laquitta, W. abbad,
W. Shait, W. Kharit and W. Allaqui as shown in (figure 1).

799






Figure 1: Major basin studied in the eastern desert
Study Area
800

GEOMORPHOLOGY:

The landscape of the Eastern Desert (E.D) is characterized by high relief which is responsible
for the relatively wetter climate that produced the great water courses of this area. Most
authorities agree that the mountainous areas of the E.D. seem to have received lot of moisture .
The E.D. is intensely dissected by valleys, wadis and ravines of various magnitudes of
catchments areas forming thus several hydrographic basins. These basins are subdivided into two
main drainage patterns having a general north-south divide located at about 50 km west of the red
sea. The eastern drainage pattern is composed of numerous individual small basins draining
directly to the Red sea and the Gulf of Suez. The western pattern is composed of various
catchment areas draining directly to the Nile River. The western hydrographic basin are about 54
basins with catchment areas ranging from as large as 70,000 km
2
(Al-Allaqui "Egypt and Sudan"),
to as small as 190 km
2
. In general the E.D. can be characterized by the following geomorphologic
units; (Said, 1962).

The structure plateau and ridges underlain both by rugged crystalline rocks and by almost
flat carbonate rocks. These occupy the major part of the area of the E.D.
The structural plains; mostly sandstone and are particularly represented in the southeast
portion of the area.
The coastal plains mostly beach sand and lagoonal mud, in the north, and by stony and
reefal raised beaches in the south.
The fluviatile plain.

GEOLOGICAL ASPECTS:

The E.D. is occupied by sedimentary and crystalline rocks. Structurally, the E.D. is composed
of a number of segments separated by major N.NW-S.SE faults. The geologic history of the E.D.
can be briefed, from Said, 1990 and others, as follows:-

Paleozoic: scattered in the area, occurs also in the subsurface at the Northern part of the
area. The Paleozoic is dominated by sand and shale facies of epicontinental origin. a
marine regression occurs at the end of the Paleozoic and continue to The Triassic_period
(Early Mesozoic).
Mesozoic: The area becomes a continental plain, where some salty lagoons were formed
in the northern part of the E.D. A new transgression occurs in the Early J urassic period in
the north part to the Gulf of Suez. During the Cretaceous the sea retreated, followed by a
new transgression during Middle Cretaceous and formed the Nubian Sandstone complex
and marine limestone facies in the Upper Cretaceous.
Cenozoic: Eocene Paleocene Period is characterized by high structure events and, thin
clastic sediments in the north, followed by thick calcareous deposits (Upper Carbonate
layer)
Paleocene: Early Eocene: composed of chalky shale, evaporitic at the top followed by
calcareous shale (Esna Shale), this later layer separate the Upper Cretaceous from the
Eocene.
The Oligocene is characterized by a general uplifting in Egypt.
Miocene started by a marine transgression in the Gulf of Suez. The lower Miocene is up
801

to 1000 m thick and composed of Fossilferrous Marl, Limestone and Sandstone beds.
During Middle Miocene a lagoonal period prevails in the Gulf of Suez and attains a
thickness of 3000 m of Fossilferrous shale, limestone, salts and gypsum.
During the Pliocene an uplifting occurs and the area remains more or less at its present
situation.
The Pleistocene is characterized by a fluviatile phase, the sea retreated and intense erosion
took place with the formation of thick gravel terrace along the Red Sea.
The arid climate prevails in the Holocene times with the formation of sabkha flats and
sand dunes.

SCOPE OF WORK:

The scope of the present paper is to identify the basin shape characteristics as well as the
geomorphologic factors of nine drainage basins in the eastern desert which are W. Sannur, W.
Tarfa, W. Assist, W. Qena, W. Laquitta, W. abbad, W. Shait, W. Kharit and W. Allaqui. Also, a
study is performed in order to identify the relationship between the basin shape characteristics and
the geomorphologic factors. The integration of the parameters and factors characterizing both the
basin shape and the basin geomorphology leads to identify a risk assessment in respect to flooding
probabilities.

Calculation Of The Drainage Basins Properties:

Basin Shape Characteristics: Quantitative expressions of a drainage basin have been
given in several forms as follows;
o Form factor R
f
:a dimensionless ratio of a basin area (A
u
)to the square of basin
length (L
b
2
) (Horton,48 in chow, 64), small values of (R
f)
indicates basins
tending to elongation in shape.
R
f
= A
u
/L
b
2

o Circularity ratio R
c
: dimensionless ratio of a basin area (A
u
) to the area of a
circle having the same perimeter as the basin(P
b
). (Miller;56 in Chow, 64),
values of (R
c
) closer to 1 tend to circularity in shape.
R
c
= A
u
/A
c
Rc= 4A
u
/P
b
2

o Elongation ratio R
e
: is the ratio of diameter of a circle(D
a
) of the same area of
the basin (A
u
) to the maximum basin length (L
b
) (Schumm,56 in chow, 64),
smaller values of (R
e
) indicates basins tending to elongation in shape.
R
e
= D
a
/L
b
R
e
=2(A
u
/)/L
b

o Rotundity Factor R : is the ratio of the area of a circle having diameter (L
b
) to
the area of the basin (Chorley et al, 57 in chow, 64), smaller values of (R)
indicates basins tending to circularity in shape
R = L
b
2
/4A
u

o Constant of Channel Maintenance C: it is the inverse of the Drainage
Density (D), (Schumm, 56), where D is the ratio of the total length of channel
segments for all orders to the basin area.
C = 1/D
o Length of overland flow L
g
: one of the most important variables that affect
the hydrologic and physiographic development of a drainage basin, it is the
802

length of flow path, which is approximately half the average distance between
stream channels and is expressed by Horton,48 as in Chow, 64), where:
L
g
= 1/2D = C/2

Table 1 shows the measured parameters (L
b
and P
b
) of the 9 major basins in the E.D., using
the chartometer and the calculated values of their respective basin shape ratios and factors i.e. Rf,
Rc, Re, R, C and Lg..
The values of the shape form parameters, thus calculated show in general a uniform trends
of the basins that tend to be elongated as represented by W. Abbad and W. Tarfa, and those
tending to be circular in shape as Allaqui and Assiuty. The other basins are also in harmony with
the general trend ranging between elongation and circularity.

Geomorphologic Factors: the most significant properties are the bifurcation ratio
(R
b
), the stream frequency (F), and the drainage density (D).
o Bifurcation ration Rb: characterises the order of the stream, which reflects
meanwhile the shape of the drainage basin (Horton, 45 as in Chow, 64).
Whereas high values of (R
b
) which are normally associated with elongated
basins would yield a low but extended peak flow, low values of (R
b
) that are
associated with circular shape basins would produce a sharp flooding peak.
Therefore, hydrographic basins tending to have a circular-like shape are
expected to be more dangerous to the environment particularly at their outlet
localities.
o Stream Frequency (F): is defined as the ratio between the total number of all
segments of all orders to the basin area or
F= N
u
/A
u

o Drainage Density (D), characterise in general the closeness of spacing of
channels within the hydrographic basin. It is defined as :
D=L
u
/A
u

Where L
u
is the total length of all segments of all orders, A
u
is the area of the basin. The
stream frequency (F) and the drainage density (D) are mathematically related as revealed from
extensive field observations (Melton, 1958), as follows
F=0.694D
2


Table 2 shows the geomorphologic characteristics of eight selected basin in the Eastern Desert.
These parameters are compiled from (El-Shamy 1992), which indicates in general, comparatively
low values of both Rb and D associated normally with circular basin trends and steep slope and
hard rock surface respectively.

RELATIONSHIP OF BASIN SHAPE AND GEOMORPHOLOGIC CHARACTERISTICS:

1. Relationship between basin shape characteristics.

Figure (2) represents the relationship between the various basin shape characteristics. It is
clear that Channel Maintenance (C) and the overland flow length (L
g
) has exactly the same trend.
Also, the plot shows that the Rotundity Factor (R), the Circularity Ratio (R
C
) and the Elongation
Factor (R
e
) have the same behavior. However, the Form Factor (R
f
) has a unique smooth trend.
803




2. Relationship between Geomorphologic characteristics.

Figure (3) represents the relationship between geomorphologic characteristics of the eight
basins, the plot shows that stream frequency (F), Bifurcation Ratio (R
b
) and Drainage Density (D)
have almost the same trend.

3. Relationship between geomorphologic characteristics and basin shape
characteristics.

Figures 4 through 7 represent the relationship between selected geomorphologic
characteristics and some selected basin shape factors. In figure 4 the relation between stream
Frequency (F) and overland flow length (L
g
) the plot show the identical trend of the two
parameters for the nine studied basins. Also, figure 5 shows the identical trend between area (A)
and form factor (R
f)
. Figure 6 shows the same trend between total length of a segment (L
u)
the
longest path of channel flow (L
b)
. Also figure 7 shows almost the same trend between bifurcation
ratio (R
b)
and the Form Factor (R
f
).

Risk Analysis:

A risk analysis study is performed to identify the most risky Wadi, in respect to flooding
probabilities among the nine wadis studied in the Eastern Desert form the basin shape
characteristics point of view as well as from the geomorphologic factors point of view. Also, the
study includes the identification of the most important geomorphologic factor relative to the other
factors and the most important basin shape parameter relative to the other parameters.

Risk Value:
In order to perform risk analysis study the following linear equation (Wagdi, 2004) is used to
estimate the relative risk value of a Wadi relative to the highest and lowest values.
Risk Value= 4* (X-X
min
) / (X
max
-X
min
) + 1
Where X, Xmin, Xmax are the values of the parameters under consideration.

Factor of Important Coefficient:
The Factor of Important Coefficient (FIC) is defined as the factor that has the most critical
impact in an evaluation process relative to other factors. It is also considered as the key step
involving judgment in selecting among different parameters (Canter, 1977). The FIC is developed
for all basin shape characteristics as well as for geomorphologic factors. It is important to note
that the sum of all FICs must be equal to one.

Results and Discussion:

The risk values for each parameter for both the basin shape characteristics and geomorphologic
factors, for each basin, are calculated as shown in tables (3) and (4) respectively. These values
shows, meanwhile the relative importance in view of flooding risk probability for each basin in
804

respect to the others.
From table 3 raw 41 shows the FICs calculated according to Basin Shape characteristics and
raw 42 (Rank a) the ranking of different Basin shape parameters relative to each other. It is clear
that the Elongation ratio (Re) and the Form Factor (Rf) are the most important parameters to be
considered when dealing with the evaluation of a Wadi according to the basin shape
characteristics.
In table 4 column H (Rank b), shows that Wadi Qena is the most risky Wadi followed by Wadi
Assiuty. Table 4 raw 58 shows the FICs calculated for all parameters, the ranking of these
parameters in shown in raw 59(Rank d) of table 4. It is clear that the Drainage Density (D) and
Bifurcation Ratio (Rb) are the most important factors when dealing with geomorphologic factors.

SELECTED REFERENCES:

1. Shata,A,A. (1992)" Climatic Changes and Tectonic Activities and the Development of the
Nile Drainage Basin In Egypt", Butter Worth, Heinemann Publisher, Linacre House, J ordan
Hill Oxford OX2 8DP, First Edition 1992.
2. Shahin, M. (1985)" Hydrology of the Nile Basin" ELSEVIER, Amsterdam 1985.
3. Said, R. (1962)" The Geology of Egypt" ELSEVIER, Amsterdam 1962.
4. Said, R. (1990)" The Geology of Egypt" ELSEVIER, Amsterdam 1990.
5. RIGW & IWACO (1987)" Hydro geological Map of the North Eastern Desert (1:100000)"
Cairo, Egypt 1987.
6. RIGW & IWACO (1988)" Hydrogeological Map of Egypt (1:2000000)" Cairo, Egypt 1988.
7. El-Shamy, I, Z. (1992)" Recent Recharge and Flash Flooding Opportunities in the Eastern
Desert, Egypt Annals Geolo. Survey. Egypt. V.XVIII, p. 323-334, 1992.
8. El-Shamy, I, Z. (1988)" Quantitative Geomorphology and Surface Runoff Control For Wadi
Qena, Central Eastern Desert" E.G.S. Proc of the 6th Ann. p 13-26. 1988.
9. Canter, L.W., Environmental Impact Assessment McGraw-Hill, Inc. 1977.
10. Wagdy, A., Flood Risk Assessment for Non-Gauged watersheds Second regional
Conference on Arab Water, 2004.
11. Chow, Van Te., 1964:" Handbook pf Applied Hydrology" Mc Graw Hill, 1964.






805

Table 1 : Basin Shape Charateristics Calculations
B
A
S
I
N

N
A
M
E
Basin Shape Charateristics (Calculated and/or Measured)
P A L
b
D
a
R
f R R
c
R
e C L
g
W.Sannur 360 6500 140 114.650 0.332 2.367 0.630 0.650 4.082 0.1225
W.Tarfa 380 11600 200 121.019 0.290 2.707 1.009 0.608 2.985 0.1675
W.Assuti 400 6200 140 127.389 0.316 2.482 0.487 0.635 2.488 0.201
W.Qena 600 18000 220 191.083 0.372 2.111 0.628 0.688 2.778 0.18
W.Laquitta 360 7700 180 114.650 0.238 3.303 0.747 0.550 3.135 0.1595
W.Abbad 340 6900 180 108.280 0.213 3.686 0.750 0.521 3.534 0.1415
W.Shait 400 7700 160 127.389 0.301 2.610 0.605 0.619 3.817 0.131
W.Kharit 680 19100 220 216.561 0.395 1.989 0.519 0.709 4.878 0.1025
W.Allaqui 1300 67500 400 414.013 0.422 1.861 0.502 0.733
Table 2: Geomorphologic Characteristics (Compiled from El Shamy, 1992)
B
A
S
I
N

N
A
M
E
Basin Shape Charateristics (Calculated and/or Measured)
B
A
S
I
N

N
A
M
E Geomorphologic Characteristics
N
u A L
u F D
R
b
W.Sannur 271 6500 1593 0.042 0.245 3.72
W.Tarfa 905 11600 3886 0.078 0.335 3.11
W.Assuti 696 6200 2492 0.112 0.402 3.36
W.Qena 1534 18000 6450 0.085 0.36 4.22
W.Laquitta 593 7700 2685 0.077 0.319 3.16
W.Abbad 360 6900 1960 0.052 0.283 2.82
W.Shait 301 7700 1600 0.049 0.262 3.75
W.Kharit 575 19100 3915 0.03 0.205 3.81
Table 3: Ranking & FIC according to basin Shape Charateristics.
B
A
S
I
N


N
A
M
E
B
A
S
I
N

N
A
M
E Geomorphologic Characteristics
Calculated Basin Shape Charateristics
P L
b
D
b
R
f R R
c
R
e C L
g
W.Sannur 1.24 1.00 1.23 3.61 1.89 2.10 3.74 3.67 3.67
W.Tarfa 1.47 4.00 1.47 2.70 2.69 5.00 2.85 1.83 1.83
W.Assuti 1.71 1.00 1.71 3.28 2.16 1.00 3.42 1.00 1.00
W.Qena 4.06 5.00 4.06 4.50 1.29 2.08 4.56 1.49 1.49
W.Laquitta 1.24 3.00 1.24 1.54 4.10 2.99 1.62 2.08 2.08
W.Abbad 1.00 3.00 1.00 1.00 5.00 3.01 1.00 2.75 2.75
W.Shait 1.71 2.00 1.71 2.93 2.46 1.90 3.09 3.22 3.22
W.Kharit 5.00 5.00 5.00 5.00 1.00 1.25 5.00 5.00 5.00
W.Allaqui
FIC 0.09 0.13 0.09 0.13 0.11 0.10 0.13 0.11 0.11
RANK a 8 3 9 2 6 7 1 4 4 1.00
Table 4: Ranking & FIC according the geomorphologic Factors.
B
A
S
I
N

N
A
M
E
B
A
S
I
N


N
A
M
E
Calculated Geomorphologic Factors
N
u A L
u F D
R
b
W.Sannur 1.00 1.09 1.00 1.59 1.81 3.57
W.Tarfa 3.01 2.67 2.89 3.34 3.64 1.83
W.Assuti 2.35 1.00 1.74 5.00 5.00 2.54
W.Qena 5.00 4.66 5.00 3.68 4.15 5.00
W.Laquitta 2.02 1.47 1.90 3.29 3.31 1.97
W.Abbad 1.28 1.22 1.30 2.07 2.58 1.00
W.Shait 1.10 1.47 1.01 1.93 2.16 3.66
W.Kharit 1.96 5.00 2.91 1.00 1.00 3.83
W.Allaqui
FIC 0.14 0.15 0.14 0.18 0.19 0.19 1.00
RANK
b 6 4 5 3 1 2
B
A
S
I
N

N
A
M
E Calculated Geomorphologic Factors


806


Relationship between basin shape
0
1
2
3
4
5
6
7
8
9
10
W
.
S
a
n
n
u
r
W
.
T
a
r
f
a
W
.
A
s
s
u
t
i
W
.
Q
e
n
a
W
.
L
a
q
u
i
t
t
a
W
.
A
b
b
a
d
W
.
S
h
a
i
t
W
.
K
h
a
r
i
t
W
.
A
l
la
q
u
i
Rf R Rc
Re C Lg

Relat ionship bet ween geomor phologic
char act er istics of t he study basin in t he ED.
0.01
0.10
1.00
10.00
W
.
S
a
n
n
u
r
W
.
T
a
r
f
a
W
.
A
s
s
u
t
i
W
.
Q
e
n
a
W
.
L
a
q
u
i
t
t
a
W
.
A
b
b
a
d
W
.
S
h
a
i
t
W
.
K
h
a
r
i
t
W
.
A
l
l
a
q
u
i
F D Rb

Figure 2 : Relationship between basin Figure 3: Relationship between geomorphologic
shape characteristics of the studied basins. Factors

0
1000
2000
3000
4000
5000
6000
7000
W
.
S
a
n
n
u
r
W
.
T
a
r
f
a
W
.
A
s
s
u
t
i
W
.
Q
e
n
a
W
.
L
a
q
u
i
t
t
a
W
.
A
b
b
a
d
W
.
S
h
a
i
t
W
.
K
h
a
r
i
t
W
.
A
l
l
a
q
u
i
L
u
0
100
200
300
400
500
600
700
L
b
Lu Lb


0
0.05
0.1
0.15
0.2
0.25
W
.
S
a
n
n
u
r
W
.
T
a
r
f
a
W
.
A
s
s
u
t
i
W
.
Q
e
n
a
W
.
L
a
q
u
i
t
t
a
W
.
A
b
b
a
d
W
.
S
h
a
i
t
W
.
K
h
a
r
i
t
W
.
A
l
l
a
q
u
i
L
g
-0.01
0.01
0.03
0.05
0.07
0.09
0.11
0.13
0.15
F
Lg F
Figure 4: relation between F & Lg Figure 5: relation between A & Rf


0
10000
20000
30000
40000
50000
60000
70000
80000
W
.
S
a
n
n
u
r
W
.
T
a
r
f
a
W
.
A
s
s
u
t
i
W
.
Q
e
n
a
W
.
L
a
q
u
i
t
t
a
W
.
A
b
b
a
d
W
.
S
h
a
i
t
W
.
K
h
a
r
i
t
W
.
A
l
l
a
q
u
i
A
0.000
0.100
0.200
0.300
0.400
0.500
R
f
A Rf

0
1
2
3
4
5
W
.
S
a
n
n
u
r
W
.
T
a
r
f
a
W
.
A
s
s
u
t
i
W
.
Q
e
n
a
W
.
L
a
q
u
i
t
t
a
W
.
A
b
b
a
d
W
.
S
h
a
i
t
W
.
K
h
a
r
i
t
W
.
A
l
l
a
q
u
i
R
b
0.000
0.100
0.200
0.300
0.400
0.500
R
f
Rb Rf

Figure 6: relation between Lu & Lb Figure 7: Relation between Rb & Rf

(The basins are arranged according to their locations, from North to South)
807

808

ESTIMATION OF SOIL SOLUTE ELECTRICAL
CONDUCTIVITY USING WET-SENSOR

Ghada Samy
1
, Magnus Person
2
, Samia Abu El-Fotouh
1
, Soheir Kamel
1
, and Ronny
Berndtsson
2

1
Irrigation and Hydraulics Depart, Faculty of Engineering, Ain Shams University,
Cairo, Egypt.
2
Water Resources Engineering Depart, Lund University, Lund, Sweden.


ABSTRACT

Studies of solute transport through soil and attendant
environmental impacts are hampered by the lack of methods for
continuous monitoring of solute concentration. This paper presents the
introductory part of an integrated research project to investigate the
salinization effects due to large-scale use of saline and semi-saline water
for irrigation in El-Salam canal project. The proposed project mainly
targets the prediction of future soil and ground water environmental state
associated with such large scale saline water-based irrigation schemes.
However, this paper concentrates on the experimental calibration session
of a proposed rapid, nondestructive sensor to investigate the salinization
process by measuring the dielectric properties of the soil to estimate both
the soil water content () and (
p
) for different common soil types of the
region under consideration. The proposed sensor depends on frequency
domain reflectometry (FDR) technique and it is called WET sensor. It
measures the dielectric constant (k
a
) and bulk electrical conductivity (
a
)
of soil. Then, it utilizes both of them to estimate () and (
p
). The study
involves experimental measurements in the laboratory using four
different soil types with () in the range 0.10 m
3
m
-3
up to saturation. In
each soil type, three different electrical conductivity solutions (
w
) were
used (0.70-1.46-1.88) dSm
-1
. The results revealed that the calibration
coefficients of water content and the soil parameter are significantly
dependent on the soil type, while, slightly affected by electrical
conductivity of the moistening solution.




FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


809
INTRODUCTION

Soil water is essential for plant growth and is the vehicle for solute transport,
including nutrients and soil contaminants. Accurate measurement of water content
and solute concentrations are crucial for the better management of irrigation water and
rainfall capture. The technique of measurement should be rapid, reliable, simple, and
non-destructive.
The common direct method for water content and salinity measurement is soil
sampling. The water content can be determined by oven drying while the soil extract
method is used for salinity measurements. This method is not practical when many
samples are needed. Furthermore, it is destructive as only a single measurement can
be made for a sample soil volume.
Great effort has been devoted in the last decades to the development of new
and more accurate methods of measuring water content and salinity. TDR is an
indirect method of determining soil water content. The method involves measuring
the propagation velocity of an electromagnetic pulse traveling along a parallel
metallic probes embedded in the soil. This measurement later converted to the
volumetric moisture content of soil () by various models (Topp et al., 1980, Dalton
et al., 1984). There are several advantages associated with the TDR technique: it is an
accurate instrument and it can easily be automated to take scheduled readings. Some
important disadvantages are: water content measurements can not be made in highly
saline soils and the initial costs are relatively high compared with other methods.
FDR sensor was also developed for continuous measurement of soil water
content. This system uses the dielectric properties of water in a different approach
than TDR (Bilskie, 1997). The FDR sensor sends an electromagnetic wave along its
probes and measures the frequency of the reflected wave, which varies with water
content. Hilhorst and Dirksen (1995) reported that ionic conductivity can be measured
more easily and more accurately with the FDR sensor than with existing time domain
sensors. FDR has several advantages over TDR: interpretation of data is direct; it has
low power consumption, it is inexpensive for multiple site measurements. Because of
these advantages, FDR is being applied on irrigation scheduling. Laboski et al. (2001)
and Kukangu et al. (1999) successfully used this technique for irrigation practices.
However, both TRD and FDR are indirect techniques in measuring the soil
water electrical conductivity (
p
). Each of them measures the bulk soil electrical
conductivity (
a
) which depends on both (
p
) and . Thus, the (
p
) can only be
predicted if is constant, or if the relationship between
p
,

a
, and

is determined.
Several different models of
p
-
a
- relationship have been developed (e.g. Rhoades et
al., 1976; Mualem and Friedman, 1991). However, these models have several
drawbacks such as the high dependency on soil type and requiring details soil specific
calibration. Recently, Hilhorst (2000) presented a theoretical model describing a
linear relationship between bulk electrical conductivity,
a,
and dielectric constant, k
a
in moist soil. He found that using these linear relationship measurements of the pore
water electrical conductivity,
p
, can be made in a wide range of soil type without
soil-specific calibration. Persson (2002) presented an evaluation of the linear model
using detailed TDR measurements in three sandy soils. He showed that the linear
model was as good as other commonly used models for
p
predictions with significant
dependency of the linear model on soil type. Hamed et al. (2003) presented a further
evaluation of the linear model in several different soil types using TDR
measurements. He showed that the linear ka-
a
was reasonably well predicted by
Hilhorst model and the results could be improved using a soil specific calibration.


810


RESEARCH OBJECTIVES

This paper presents the first part of an integrated research project to
investigate the salinization effects due to large-scale use of saline and semi-saline
water for irrigation in El-Salam canal project. Studies of the salinization process will
be made by use of developed FDR sensor (WET sensor) and targets the prediction of
future soil and ground water environmental state associated with such large scale
saline water-based irrigation schemes.. This research part concentrates specifically on
the calibration of the senor which measures the dielectric properties of the soil to
estimate the water content and the pore water electrical conductivity. This paper
studies also the effect of the electrical conductivity of the moistening solution on the
calibration parameters. For this purpose, a series of laboratory experiments was
conducted in four different soil types using three different
w
levels over a wide range
of .


THEORY

Dielectric Properties

When an electrical field passes through a material (such as soil) some of the
energy in the field is transmitted, some is reflected, some is stored and finally some is
absorbed and converted into heat. The extent to which each of these occurs within the
soil is determined by its dielectric properties. These are quantified by a parameter
called the relative electric permittivity of a material, which characterizes its response
to the polarizing effect of an applied electric field. It is usually represented as a
complex number. The real part of the permittivity represents the energy stored, and
the imaginary component represents the total energy absorption or loss. For a static
real field the real part of the permittivity is often referred to as the dielectric constant
k
a
.

Soil Moisture

Whalley (1993), White et al. (1994) have shown that there is a simple
relationship between the measured permittivity of the soil, k
a
, and volumetric water
content, , of the form:

k
a
= b

+ b
1
(1)

This equation appears to work very well for most non-magnetic soils over a
range of frequencies between 1 MHz and 10 GHz. As the sensor measures (k
a
)
directly, an experimental calibration to determine the pairs of coefficient, b
o
and b
1
is
used.





811
Pore Water Conductivity

The electrical conductivity of the bulk soil,
a
, is a function of both soil water
content, , and the pore water conductivity,
p
. Several different models have been
developed for the
a
-
p
- relationship.

Malicki et al. (1994) found a high degree of linear correlation between values
of
a
and k
a
for a broad range of soil types. An empirical
a
-
p
- model was also
represented;


p
= (
a
0.08) / (k
a
6.2) (0.0057 + 0.000071S) (2)

where S is the sand content in % by weight. Inspired by this work, Hilhorst (2000)
recently presented a theoretically based linear
p
-
a
-k
a
relationship:


p
= k
p

a
/ (k
a
k
o
) (3)

By arranging the equauion;

k
a
= (k
p
/
p
)
a
+ k
o
(4)


where k
p
is the dielectric constant of the pore water and k
o
is the k
a
value when
a
= 0.
However, the parameter k
o
is not the k
a
values of dry soil, but appears as an offset of
the linear relationship between k
a
and
a
. As the sensor measures (k
a
) and
a
directly,
the equation is only contains one fitting parameter k
o
which depends on the soil type
and is called soil parameter.

This highlights the significant need for an integrated experimental calibration
session of the proposed sensor to determine the coefficients of water content and the
soil parameter of the soil under investigation.


THE SENSOR

The WET sensor is a new dielectric sensor. It consists of 0.068m rod, 0.003m
in diam. The probe is built around an Application Specific Integrated Circuit (ASIC)
which measures real and imaginary part of the complex dielectric permittivity
simultaneously by the sensor rods and the soil at the single frequency of 20 MHz. The
ASIC increases the accuracy of the measurements and eliminates influences of
lengths of cables, quality of cables, connectors, and switches, making multiplexing
easier and cheaper. This operates as a vector voltmeter to make precision
measurments of
a
and k
a
. The schematic sketch of the ASIC is presented in figure
(1.a).


812


Figure (1.a): Schematic representation of the ASIC circuitry
for measuring dielectric properties


The sensor is connected to HH2 Moisture Meter [see figure (1.b) and (1.c)]
which applies power to the sensor and measures the output signal voltage returned.
The measuring freuency is 20Mhz. The sensor probe detects the changes to the
20MHz signal and sends this information to the HH2 which measures the capacitance
(C) and conductance (G) of the material between the rods (soil). Then, it infers the
dielectric properties using the sensor calibration file, which contains sets of
capacitance and conductance readings obtained when the sensor was calibrated in
various reference fluids with known electric properties. Finally, it calculates soil
moisture, , and the pore water conductivity,
p
using the calibration coefficients.
[See figure (2)].




Figure (1.b): Sensor probe. Figure (1.c): HH2 Moisture meter



813
WET Sensor Output
Dielectric properties conversion
& Temperature offset adjustment
Soil Properties Conversion
Temperature
compensation
Display
Sensor calibration
Soil Calibration
Soil parameter
HH2 compensation
settings
C G
T
k
a

a T

p T


Figure (2): The sensor measurement algorithm



THE EXPERIMENTAL WORK

Soil samples from four different locations in north Sinia were used in the
study. A summary of some selected soil properties is presented in Table1. Solutions
with different Nacl concentration were prepared. A laboratory conductive meter was
used to measure their electric conductivities (
w
) (Their respective
w,
were 0.70, 1.46
and 1.88).

Table 1: Some selected soil properties


The experiments had been carried out at constant temperature to eliminate the
temperature effect. The experiment procedure can be summarized in the following
sequential steps.

Location Sand% Silt% Clay% Soil type
Bulk density
g cm
-3

Soil (1)

Soil (2)

Soil (3)

Soil (4)
53.39

62.15

99.74

83.48
35.81

27.44

______

11.55
9.39

7.47

______

4.97
Sandy loam

Sandy loam

Medium sand

Loamy sand
1.42

1.38

1.41

1.54
814
The soil samples were dried and passed through 2mm sieve then leached with
distilled water and the excessive water was drained.
The WET sensor reading was taken several times until the measured
a
reached a constant value then the soil was redried.
The soil sample was then moistened with Nacl solution in seven increments to
cover the range of moisture from water content = 0.1 up to saturation. First,
the soil was physically mixed with a small amount of solution to bring the
target water content. Then, the sample was packed in small layers into a
cylinder (0.075 m diameter, and 0.103 m height). Each layer was compacted
using a rubber cork as uniformly as possible. The cylinder was filled this way
so as to keep the normal density of the soil.
The Wet Sensor readings were taken for k
a,

a
three times to find the average
value.
This procedure was repeated seven times until saturation.
Finally, an upward infiltration process was performed -with the same solution-
to drain sufficient water on the top of the saturated sample. The drained water
was collected and the
p
was measured using the conductive meter.
Procedure was repeated for each solution and sample.


RESULTS AND DISCUSSION

Water content
For each soil type, the relationship between k
a
and was plotted. The offset
and the slope of the k
a
- relationship were determined by linear regression which
represented the calibration coefficients b
o
, b
1
respectively.

Figure (3.a) presents an example of the water content calibration results which
shows a linear relationship between k
a
and . It is noticed that the conductivities of
Nacl solutions do not affect the relationship. The data of each soil type -regardless
w
-
were, therefore, used for fitting global values of b

and b
1
as presented in table 2.


Table 2: Results of the linear regression of k
a
data obtained in calibration
experiment

Soil type
Range in
(m
3
.m
-3
)
b

b
1
r
2
Soil (1) 0.10 0.40 2.12 9.91 0.925
Soil (2) 0.10 0.35 2.10 10.23 0.929
Soil (3) 0.10 0.30 1.92 8.63 0.976
Soil (4) 0.10 0.35 2.29 9.56 0.950






815


y = 9.91x + 2.12
R
2
= 0.9253
0
2
4
6
8
0 0.1 0.2 0.3 0.4 0.5
(m
3
m
-3
)

k
a
0.7 ds/m
1.46 ds/m
1.88 ds/m


Figure (3.a) The k
a
against water content () during the calibration
experiments in soil sample (1). The solid line is the linear regression of ka- data


The measured k
a
was used with the global b

, b
1
values for the calculation of
water content; . Figure (3.b) shows an example of the results which reveals a high
correlation.


0.00
0.20
0.40
0.60
0 0.2 0.4 0.6
True [m
3
.m-
3
]
C
a
l
c
u
l
a
t
e
d

[
m
3
.
m
-
3
]
0.7 ds/m
1.46 ds/m
1.88 ds/m
1:1 line


Figure (3.b) The water content () calculated using the equation (1) vs. the
true () in soil sample (1) for the three different NaCl solutions.


816
Pore water conductivity

For each value of
w
, the relationship between k
a
and
a
was plotted. An
example of the results is shown in figure (4.a). The relation k
a
-
a
is linear. Also, the
figure shows that the slope is remarkably affected by electrical conductivity of the
moistening solution
w
. It is noticed that the slope decreases with the increase of
w
.
Furthermore, the offset (k
o
) is almost the same for all
w
values. The results are
agreed with the original study by Hilhorst (2000), because he only presented one
value of k
o
for each soil type. The offset (k
o
) was determined and averaged for each
soil type and presented in table (3).

0
5
10
15
20
25
0 0.1 0.2 0.3 0.4 0.5
a (dS.m
-1
)
k
a
0.70(ds/m)
1.46(ds/m)
1.88(ds/m)


Figure (4.a) Measured values of dielectric constant (k
a
) and bulk electric conductivity
(
a
) in the soil sample (1) The solid lines are the trend lines to the data points.

Table 3: Average values of the offsets (k
o
)

Soil type
Range in
(m
3
.m
-3
)

k

Soil (1)
Soil (2)
Soil (3)
Soil (4)
0.10 0.40
0.10 0.35
0.10 0.30
0.10 0.35
7.84
8.78
4.95
8.32

The averaged values of k

was used with the measured k

a
and
a
in equation
(3) for the calculation of the pore water conductivity
p
. Figure (4.b) shows an
example of the results which reveals a good correlation between the measured and
calculated
p
.
817

0
1
2
3
0 0.5 1 1.5 2 2.5 3
Measured
p
[dSm
-1
]
C
a
l
c
u
l
a
t
e
d

p

[
d
S
m
-
1
]
0.7 ds/m
1.46 ds/m
1.88 ds/m
line 1:1

Figure (4.b) The pore water electrical conductivity (
p
) calculated using the equation
(3) vs. the measured (
p
) in soil sample (3).



SUMMARY AND CONCLUSION

This paper presents the introductory part of an integrated research project to
investigate the salinization effects due to large-scale use of saline and semi-saline
water for irrigation in El-Salam canal project. It concentrates on the experimental
calibration session of a proposed rapid, nondestructive FDR-based technique to
investigate the salinization process by measuring the dielectric properties of the soil.
The coefficients of water content and the soil parameter values of the different
common soil types in the region under consideration have been determined and stored
in the HH2 Moisture Meter. Furthermore, the paper studies also the effect of the
conductivity of the moistening solution (
w
) on the calibration parameters. The results
revealed that the calibration coefficients of water content and the soil parameter are
significantly dependent on the soil type, while, slightly affected by electrical
conductivity of the moistening solution.

However, this introductory calibration session has made the FDR instrument
ready for carrying out the proposed research which, specifically, targets the prediction
of future soil and ground water environmental state associated with such large scale
saline water-based irrigation schemes. Hopefully, by the end of the proposed project,
the lessons learned are expected to be used as pedgological example not only in Egypt
but potentially also for other Arab and regional countries with similar environmental
circumstances.


REFRENCES

- Bilskie, J. (1997). Using dielectric properties to measure soil water content. Sensor
Magazine, 14: 26-32

818
- Dalton F. N., Herkelrath W. N., Rawlins D. S. and Rhoades J. D. (1984). Time
domain reflectometry: Simultaneous measurement of soil water content and electrical
conductivity with a single probe. Science (Washington DC) 224:989-990.
- Hamed, Y., Persson, M., and Berndtsson, R. (2003). Soil solution electrical
conductivity measurements using different dielectric techniques. Soil Sci. Soc. Am.J.
67,1071-1078.
- Hilhorst, M. A. (2000). A pore water conductivity sensor. Soil Sci. Soc. Am.J.
64,1922-1925.
- Hilhorst, M.A and Dirksen C. (1995). Dielectric water content sensors: Time domain
reflectometry versus frequency domain. In College on Soil Physics; Dielectric soil
water content and electrical conductivity Measurements. Int. Center for Theoretical
Physics, Trieste, Italy.
- Kukangu, G., M.J. Savage, and M.A. Johnston. (1999). Use of sub-hourly soil water
content measured with a frequency-domain reflectometer to schedule irrigation of
cabbages. Irrigation Science, 19: 7-13.
- Laboski, C. A. M., Lamb, J. A., Dowdy, R. H., Baker, J. M. and Wright, J. (2001).
Irrigation scheduling for a sandy soil using mobile frequency domain reflectometry
with a checkbook method. Journal of Soil and Water Conservation, 56 (2): 97-100.
- Malicki, M. A., Walezak R.T, Kock, S. and Fluhler, H. (1994) Determining soil
salinity from simultaneous readings of its electrical conductivity and permittivity
using TDR. In: Processing of the Symposium on Time Domain Reflectometry in
Environmental, Infrastructure, and Mining Applications (Evanston, Illinois, 7-9
September, 1994).
- Mualem Y. and Friedman (1991) Theoretical prediction of electrical conductivity in
saturated and unsaturated soil. Water Resources Research, Vol. 27, No.10. , 2771-
2777.
- Persson, M. (2002). Evaluating the linear dielectric constant-electrical conductivity
model using time domain reflectometry. Hydrol. Sci. J.47(2),269-277.
- Rhoades, J.D., Ratts, P. A., and Prather, R. J. (1976). Effects of liquid-phase
electrical conductivity, water content, and surface conductivity on bulk soil electrical
conductivity. Soil Sci. Soc. Am.J. 40:651-655.
- Topp, G. C., Davis, J. L. and Annan, A. P. (1980). Electromagnetic determination
of soil water content: Measurments in coaxial transmission lines. Wat. Res. 16, 574-
582.
- Whalley, W. R., (1993). Considerations on the use of time-domain reflectometry
(TDR) for measuring soil moisture content. Journal of Soil Sci. 44, 1-9.
- White, I., Knight, J.H., Zegelin, S. J., and Topp, G.C. (1994). Comments on
considerations on the use of time-domain reflectometry (TDR) for measuring soil
water content by W. R. Walley. Journal of Soil Sci. 45,503-508.



819

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005









DRIP IRRIGATION EXPERIMENT IN CLAYEY AND
SANDY SOILS USING MULTIPLE TRACERS

Yasser Hamed
a1
, Fethi Bouksila
b
, Fairouz Slama
b
, Ronny Berndtsson
c
, Akissa Bahri
b
,

a
Civil Engineering Depart., Faculty of Engineering, Suez Canal university, Egypt
b
INRGREF, Tunisia.
c
Water Resources Engineering Depart., Lund University, Sweden

ABSTRACT
Drip irrigation is frequently used as a mean to improve irrigation efficiency. Recent
research, however, has shown that unsaturated solute transport may be strongly affected by
preferential flow, especially due to hydrophobic conditions in dry soil. The degree of preferential
flow using typical drip irrigation with moderately saline water for two typical soil types was
investigated using multiple tracers (dye and bromide). The experiments were carried out in two
different types of dry soil, clayey and sandy soil. Three plots at each site were chosen. An amount
of solution containing dye and bromide, equal to a typical daily irrigation volume was discharged
through a single irrigation dripper at each plot. On the following day, horizontal 5-cm trenches
were dug and dye pattern and bromide concentrations were recorded. The effects of
hydrophobicity and preferential flow appeared not to be significant for the soil down to tillage
depth. For the clayey soil preferential flow was not significant in spite of clay aggregates. Cracks
below tillage depth were the main cause for preferential flow. For the dry sandy soil preferential
flow effects were also not significant. The infiltrated solute wetted twice the soil volume for the
sandy soil as compared to the clayey soil.

Keywords: Drip irrigation; bromide; dye; sigma probe; preferential flow.

1. INTRODUCTION

Drip irrigation in general is a way to improve irrigation efficiency. An advantage using drip
irrigation is that small amounts of water can be used even for saline water. Drip irrigation
minimizes salinity and matric stresses in the root zone, though salts accumulate in the periphery
of the wetted area. Higher levels of salinity in the irrigation water can be tolerated with drip as
compared with other methods of irrigation (Rhoades et al., 1992). The distribution of soil water


820
content is reverse, decreasing away from the point source. This results in a root distribution
pattern in which most of the roots are typically found in the highly leached zone beneath the
drippers (Shalhevet et al., 1983). The drip irrigation provides good conditions of total soil water
potential for a given quality of irrigation water. Hence, drip irrigation is the best method for
applying saline water to crops, avoiding leaf injury and at the same time providing optimum soil
water conditions. There are, however, needs to remove salts that accumulate at the wetting front
(Shalhevet 1994).
It is generally accepted that water may flow through the soil via preferential paths, bypassing
large parts of the soil matrix (e.g., Gee et al., 1991). This reduces the availability of water and
nutrients to plants, and causes accelerated transport of pollutants. A main cause of preferential
flow in the unsaturated zone may be soil water repellency (hydrophobicity). This has a great
effect on the affinity of soil to water. Soil water repellency may be the reason for soil resisting
wetting for periods ranging from a few seconds to hours, days or weeks (e.g., King, 1981). A
hydrophilic surface allows water to spread over it in a continuous film whereas water on a
hydrophobic surface water balls up into individual droplets (Adams, 1963). It was thought in the
past that soil water repellency was generally associated with coarse-textured, sandy soils (e.g.,
Roberts and Carbon, 1971; Wilkinson and Miller, 1978; McGhie and Posner, 1980; Debano,
1991). Coarser particles are more capable to developing water repellency because of their smaller
surface area per unit volume compared with soils with soils finer texture in which hydrophobic
substances can easily coat soil particles (Giovannini and Lucchesi, 1983; Blackwell, 1993).
Crockford et al. (1991) found an increase in water repellency with particle size within a soil
sample. Debano (1991) concluded that water repellency is most likely to develop in soils with
less than 10% clay content. However, recently it was found that even severe water repellency is
possible in soils with considerable clay content. Soils with 25-40% clay have been found to
exhibit extreme water repellency (Crockford et al., 1991; Chan, 1992; Dekker and Ritsman,
1996b). This can happen only as long as the clay forms aggregates, thus reducing the surface area
to be covered with hydrophobic skin (Wallis et al., 1991; Bisdom et al., 1993). An important
factor in soil water repellency is soil moisture. Hydrophobicity becomes more severe in dry soil
and decreases as soil moisture increases until critical moisture content is reached, above which a
soil becomes hydrophilic (e.g., DeBano, 1971; Witter et al., 1991; Carter et al., 1994). There are,
however, still few studies on how hydrophobic soils may affect drip irrigation.
Since preferential flow is a three-dimensional process occurring at the scale of individual soil
pores it is difficult to map this process in the field. One way to reveal spatial flow patterns though
may be by using dye and/or tracers. Using dye, flow patterns can be studied in a rather large
undisturbed soil volume, and the spatial flow patterns are revealed with a high resolution. The
results are, however, instantaneous, and the experiments can only be done once at the same site.
Some recent dye investigations using dye (Brilliant Blue FCF) can be found in Flury and Flhler
(1994), Kung (1990), Lin and Mcinnes (1995), Yasuda et al. (2001), and hrstrm et al. (2001).
Adsorption of the dye particles varies, however, between soil types; soils with high clay content
and low content of organic carbon tend to adsorb more dye than others (Ketelsen and Meyer-
Windel, 1999). Other factors that affect the adsorption are, for instance, pH and calciumcontent
(Flury and Flhler, 1995). By combining dye with tracers, e.g., bromide (Br
-
), the retardation of
dye can be quantified. Zehe and Flhler (2001) combined Brilliant Blue and bromide and found
that the retardation factor ranges between 0.86 and 2.16, depending on location. In general,
however, there are only few studies on how the retardation of dye varies with soil type and soil
water conditions.
821
In view of the above, there are needs to further investigate infiltration patterns and possible
hydrophobicity for dry soils. By necessity this has to be done using tracers, possibly several joint
tracers. Consequently, the objective of this study was to investigate the degree of preferential
flow and possible hydrophobicity when using typical drip irrigation flows for two representative
initially dry soil types. To get a detailed description of spatial flow patterns and salt distribution
under the dripper two different tracers (dye and bromide) were used in the experiments. In this
way, the retardation factor for dye could be determined horizontally as well as vertically for the
two soil types. The paper is closed with a discussion on practical implications of the results.

2. MATERIALS AND METHODS

2.1 Area description

The experiments were carried out in early October 2001 at the end of the dry season at two
sites in northern Tunisia. One experimental site was selected at Kaalet El-Andalous (Fig. 1a),
which is situated approximately 50 km northwest of Tunis. The soil here is silty clay (see Table
1). The USDA classification of the soil is Vertic Xerofluvent (Soil Survey Staff, 1975). The
water table is located at about 1.5 m depth and kept constant by tile drains, with spacing 40 m.
The field was tilled and lay bare before the experiments. Three plots (S1, S2, and S3) were
chosen with an inter-plot distance of 2.5 m between two drains (Fig. 1a). The soil was dry before
the experiments (12-13 m
3
m
-3
) (Table 1). Cultivation depth at the site is approximately 20-30
cm. Below this depth the soil is little evolved and characterized by prismatic clay structure with
many cracks. These cracks are formed by shrinkage resulting from desiccation of the clay. The
cracks may partly be invisible but sometimes several mm wide with a length ranging from a few
mm up to more than one meter.
The second site was selected at Nabeul, which is located approximately 70 km southeast of
Tunis. Soils here consist of loamy sand (Table 1). The experimental plot was located at the first
third of a 40 x 40 m experimental field area. The water table is located at about 1.5 m depth. The
field was tilled and drip irrigation was used one year before the experiments to irrigate potatoes.
The soil is rather homogeneous with depth regarding texture. Apparent structural soil horizons
exist however. One such soil horizon is the tillage depth corresponding to 30-40 cm. Three plots
(S4, S5, and S6) were chosen with an inter-plot distance of 2.5 m (Fig 1b). Also, here the soil was
initially dry before the experiments (7-8 m
3
m
-3
).
The climate at both sites is Mediterranean, characterized by mild winters receiving the
major part of the annual precipitation (450 mm on average), and hot and dry summers. Total
rainfall and distribution are highly variable from year to year. Average annual potential
evapotranspiration is 1370 mm.

2.2 Field Experiments

Local irrigation water was used for the experiments. The irrigation water had an average
electrical conductivity (
w
) of 3.87 and 3.95 dS/m for the clayey and sandy soil site, respectively.
The irrigation water was mixed with dye (6 g/l) and potassium bromide (4 g/l), resulting in a total
electrical conductivity (
w,tot
) of about 10.5 dS/m. The solute was applied through a single
dripper with a constant average flux of 2.5 l/h. This flux is typically used in the area when
irrigating vegetables, e.g., tomatoes or cucumbers. Approximately 8 liter was discharged from a
small tank through the single dripper and a constant pressure was maintained using a small
822
battery-driven pump. The amount of 8-10 liters is typically used in the area as a daily irrigation
volume through a single dripper. At both sites drip irrigation is common to irrigate vegetables
and other crops.
The dye tracer used was the food-grade dye pigment Vitasyn-Blau AE 85 (Swedish Hoechst
Ltd.). This dye has the same chemical composition as the dye Brilliant Blue FCF, which has been
used frequently in several field experiments (e.g., Flury and Flhler, 1994; Aeby et al., 1997).
After infiltration, the plots were covered with plastic to avoid evaporation and to protect from
rain. About 15 hours after the infiltration, horizontal soil surface sections were dug with 5 cm
intervals at each plot. A scale within a 0.5 by 0.5 m wooden frame with origin coinciding with
the position of the dripper was put on the soil surface before taking photos. The position of the
frame was determined using two fixed points adjacent to each plot. Photos were taken of the
horizontal dye stained soil sections with a both a digital and a manual camera. The Sigma Probe
(EC1 Sigma Probe, Delta-T Devices Ltd., Cambridge, UK; Hilhorst, 2000) was used to measure

w
at 5 cm intervals in a spatial grid within the 0.5 by 0.5 m scale. The Sigma Probe measures
pore water conductivity with high degree of independence from both soil moisture content and
the degree of contact between the probe and soil. The complex permittivity of the bulk soil
(dielectric constant) K
a
, has been found to be a function of both soil water content , and the
permittivity of the pore water K
w
(e.g., Nyfors and Vainikainen, 1989). In similar way, the
electrical conductivity of the bulk soil,
a
, is a function of both and the electrical conductivity
of the pore water
w
. The
w
measurements were converted to relative electrical conductivity
according to
rel
=(
w,tot
-
in
)/(
p
-
in
)], where
in
is the average natural soil conductivity and

p
is the electrical conductivity of the applied pulse. This quantity corresponds to relative resident
concentration since the electrical conductivity is linearly related to solute concentration (Marion
and Babcock, 1976).
Soil samples were collected at each site between the plots and at the dripper position to
investigate soil and soil water properties (Fig. 1). These were texture, soil water content, bulk
density, particle density, porosity, structural stability, and content of organic matter. The samples
were collected with an auger at depths 0-10, 10-20, 20-30, 30-40, 40-50, and 50-60 cm. Figure
(1) shows the sample positions for both sites.

2.3 Image analysis

Horizontal soil sections were photographed with a digital camera at about 1.5 m height. As a
back-up soil sections were also photographed in a similar way by a standard color slide camera.
The digitized images were analyzed using Adobe Photoshop (version 5.5, Adobe Systems,
Inc.) and UNIX Matlab
R
(version 6.0, The Mathworks Inc.) software. The dyed area was
identified using Photoshop toolbox. Also, the wet area of the clay soil was identified using the
same technique. The wet area of the sand could not easily be identified and this analysis for the
sandy soil was therefore abandoned. The area of the dye and wet soil (clayey soil) was calculated
using Matlab
R
image processing toolbox. In general, soil sections were excavated until no dye
traces were seen. This meant in most cases down to a depth of 50 cm and an average of eleven
pictures at each plot. These pictures were used to calculate dye (and wet area for the clay soil)
and bromide horizontal coverage at every section.




823
3. Results and Discussion

3.1 Dye and tracer analysis

Figure 2 shows the mean soil water content before and after infiltration for the clayey and
sandy soil, respectively. The effect of infiltration is clearly distinguishable especially for the
clayey soil. Soil water content increased especially for the clayey soil from 12-13 to 30-35 m
3
m
-3
.
The effect in the sandy soil was less pronounced. The initial soil water content for the sandy soil
was almost uniform with depth. This is probably due to relatively homogeneous soil texture. In
the clayey soil, on the other hand, soil water content was approximately uniform down to a depth
of 30 cm. At larger depths, the water content increased. This is probably an effect of the
cultivation depth and the difference in soil structure. After infiltration, the soil water content was
higher at the soil surface. This indicates that the infiltration process is still continuing after about
15 hrs.
The experiments were done at relatively dry soil conditions for both the sandy and the
clayey soil. This made both the sandy and the clayey initial soil color relatively light and dyed
and wet areas easily distinguishable (except wetted areas in the sand) from non-affected areas.
Figure 3 shows the dye-covered area with depth for both the clayey and sandy soil. In general, the
dye-covered area was larger for the sandy soil as compared to the clayey soil at all depths. The
dye cover in the clayey soil increased gradually from soil surface to 15-20 cm depth. Further
down, the dyed area decreased gradually. This pattern can be explained by the general
appearance of the soil structure. Cultivation depth is approximately 20-30 cm. Below this depth
soil remains little developed. This means that soil structure is relatively homogeneous down to
20-30 cm depth. Consequently, similar advective and dispersive forces act on the solute in this
layer. Below this depth the soil structure is prismatic which means that flow mainly occurs in
cracks between the clay prisms. Consequently, onset of most of the preferential flow is at
approximately 20-30 cm depth. Below about 50 cm only little dye remained. Traces of dye,
however, were found down to 75,100 and 67 cm for plot 1, 2, and 3, respectively. Both the
amount of dye penetration to the lower layers and penetration depth are relative to the amount of
cracks and their lengths. Plot 2 has the largest amount of cracks while plot 3 has the lowest
amount.
For the sandy soil the dye cover was in general at least the double. The dye cover with
depth at each of the three sites had approximately the same shape. There are two peaks, one upper
at the soil surface layer and a second at about 30-40 cm depth. The first upper peak is probably
due to hydrophobic properties of upper soil layers. This caused solutes to spread horizontally
until gravitational forces of infiltrated solute overcome the hydrophobic strength. The second
deeper peak probably demarks the tillage depth and a soil layer with different bulk density.
Maximum depth of the penetrating dye was found at 48, 47, and 60 cm for plot 4, 5, and 6,
respectively. The larger dye penetration depth at plot 6 was probably causes by more coarse sand
at deeper layers (40-60 cm =70%; see Table 1).
Figure 4 shows a comparison between the area covered by dye and extent of wet area with
depth for the clayey soil. The overall impression according to the above is further confirmed. The
wet area principally shows the areal extent of bromide. This was confirmed by Sigma Probe
readings (see below). Consequently, it is clearly seen how dye is retarded in relation to bromide
for the actual clay soil.
Figure 5 shows a comparison between dye pattern with depth for the clayey and sandy soil,
respectively. The clayey soil is represented by plot 2 and the sandy soil by plot 5. The onset of
824
preferential flow is seen at approximately 30 cm depth for the clayey soil. Although, the clayey
soil structure formed by tillage is very coarse, water repellency does not produce preferential
flow. These results to some extent contradict the conclusions of Crockford et al., (1991); Chan,
(1992); Dekker and Ritsman, (1996b); Wallis et al., (1991); Bisdom et al., (1993) who concluded
that severe water repellency may take place in soils with considerable clay content (25-40%)
under the condition the clay forms aggregates. For the sandy soil represented by plot 5
preferential flow does not seem to be important. Several studies before have, however, found
considerable preferential flow in similar sandy soils (e.g., McGhie and Posner, 1980; Debano,
1991).
This is probably an advantage for using drip irrigation. A general reason for larger dye cover
areas in the sandy soil is the greater adsorption of dye in the clayey soil as compared to that of
sand.
Figure 6 shows a comparison between wet areas (gray color), dye cover (black color), and
isolines for relative bromide concentration (contour maps) in the clayey soil. The isolines of
bromide represent detected bromide concentrations by the sigma probe. Outside the outer isoline
the bromide content was too low or soil was too dry to allow bromide to be detected by the sigma
probe. In general, there is a good match between wet area and detected bromide concentrations.
In the sandy soil, the extent of the wet area could not be clearly detected due to only minor
changes in soil color before and after infiltration.

3.2 Retardation factors

The above results were used to estimate dye retardation factors for the clayey soil. In general
it may be said that dye has similar adsorptive behavior like, e.g., pesticides (Sabatini and Austin,
1991) while bromide ion moves much like NO
3
-N (fertilizers) in soil (Smith and Davis, 1974;
J abro et al., 1991). Consequently, in addition to the advantages dye and bromide have in
visualizing water and solute transport, we can thus also have a rough but general idea about how
some pesticides and fertilizers may be adsorbed in the present soil types. To quantify the
retardation we calculate the retardation factors below. In these calculations we separate between
volumetric and horizontal retardation factors. Volumetric retardation factors for bromide with
respect to dye within the same soil (for clayey soil) were calculated by taking the ratio between
volume of soil containing bromide and volume of soil stained with dye in the clayey soil. The
corresponding horizontal retardation factors (for bromide with respect to dye) were calculated by
taking the ratio between average radius of the bromide and average radius of the dye. Table 2
shows a summary of the estimated retardation factors for the different plots. The volumetric
retardation factor appears rather stable 4.6-4.9. The corresponding horizontal retardation varies
from 2.2-2.3. When comparing the dye coverage in clayey and sandy soil, respectively, the dye
was retarded on average 2.4 by volume and 1.7 by average radius. This has, however, to be seen
in light of the different soil water contents for the sandy and clayey soil, respectively.

4. SUMMARY AND CONCLUSIONS

Two different dry soil types were investigated in terms of solute transport patterns from a
single dripper. At each plot about 8 L dye and bromide tagged irrigation water was infiltrated at a
constant approximate rate of 2.5 L/h. This is a typical daily application rate in the area using drip
irrigation. For the clayey soil, the dye coverage showed that substantial preferential flow occurred
below the typical tillage depth. For the soil layers above the typical tillage depth solute mixed
825
rather well even though soil peds were large and with big pores in between. Preferential flow was
detected mainly below the tillage depth in the prismatic clay structure. Also for the sandy soil,
preferential did not appear to be important. No preferential flow occurred even below the tillage
depth. Consequently, drip irrigation appears to be an efficient mean for irrigation avoiding
preferential flow even in dry soils.
The dye covered area in the sandy soil was larger than that for the clayey soil. A general
reason for this is that the adsorption of dye is greater in clay compared to that of sand. Besides,
cracks in the clayey soil below tillage depth (below 30 cm) contributed to quick movement of the
dye away from soil matrix. Another reason is the cohesive force between particles in clay that
limits movement of the dye between clay particles. Maximum depth of dye penetration was,
however, larger for the clayey soil as compared to the sandy soil due to the cracks below tillage
depth. This has important implications for possible pollutant transport to the groundwater.
The average dye volumetric retardation factor of sand with respect to clay (volume of
sandy soil stained with dye divided by volume of clayey soil stained with dye) equaled 2.36
where the horizontal retardation factor (average radius of dye in sandy soil divided by average
radius of dye in clayey soil) equaled 1.65.
The volumetric retardation factor for clayey soil regarding bromide as compared with dye
(volume of soil containing bromide/volume of soil stained with dye) are 4.6, 4.76 and 4.88 for
plots 1, 2 and 3 respectively. The horizontal retardation factors (average radius of the
bromide/average radius of the dye) are equal to 2.31, 2.18 and 2.16 respectively. Taking into
account that bromide moves like fertilizes and dye moves like pesticides, we could quantify both
horizontal spread and volume of distribution for both fertilizes and pesticides.

Acknowledgments

This work was financially supported by the Swedish Science Research Council. Equipment was
funded by the Lundberg Foundation and the Royal Physiographic Society.




















826
Tables
Table 1
Soil characteristics of the two sites.
Plots Bulk
Density
Clay %
Size<0.002
mm
Fine silt %
(0.002-0.02
mm)

Coarse silt
% (0.02-0.05
mm)
Fine sand %
(0.05-0.2
mm)
Coarse sand
% (size >0.2
mm)
Clayey soil
S1
0-10
10-20
20-30
30-40
40-50
50-60


1.27
1.21
1.44
1.44
1.68
1.68


43
43
45
45
35
35


50
40
39
39
41
41


4
11
12
12
16
16


1
1
1
1
3
3


0
0
0
0
0
0
S2
0-10
10-20
20-30
30-40
40-50
50-60

1.31
1.29
1.55
1.55
1.67
1.67

45
45
48
48
53
53

39
39
38
38
35
35

10
10
10
10
7
7

1
2
1
1
1
1

0
0
0
0
0
0
S3
0-10
10-20
20-30
30-40
40-50
50-60

1.44
1.51
1.54
1.54
1.63
1.63

46
46
41
41
48
48

37
38
37
37
34
34

12
10
14
14
13
13

1
2
2
2
1
1

0
0
0
0
0
0
Sandy soil
S4
0-10
10-20
20-30
30-40
40-50
50-60


1.65
1.64
1.64
1.64
1.66
1.66


0
0
0
0
0
0


7
8.5
10.5
10.5
14
14


3
3.5
2.5
2.5
3
3


36.5
41
21
21
31
31


53.5
47
66
66
52
52
S5
0-10
10-20
20-30
30-40
40-50
50-60

1.62
1.68
1.71
1.71
1.72
1.72

0
0
0
0
0
0

9
13
9
9
7
7

7
4.5
5
5
5.5
5.5

40
42.5
38
38
38.5
38.5

44
40
48
48
49
49
S6
0-10
10-20
20-30
30-40
40-50
50-60

1.52
1.47
1.74
1.74
1.81
1.81

0
0
0
0
0
0

12
11
10
10
10
10

3
5.5
5
5
2.5
2.5

27.5
45
43
43
17.5
17.5

57.5
38.5
42
42
70
70




827
Table 2
Volumetric and horizontal retardation factors for bromide as compared with dye for clayey soil.
S1 S2 S3
Volumetric
retardation factors
4.60 4.76 4.88
Horizontal
retardation factors
2.31 2.18 2.16




Figure captions
Fig. 1. The experimental sites (a; Kaalet El-Andalous, b; Nabeul)
Fig. 2. Average soil water content before and after infiltration (a; clayey soil, b; sandy soil)
Fig. 3. Dye covered area for the clayey and the sandy soil at all plots.
Fig. 4. Dye covered area and extent of wet area with depth for the clayey soil.
Fig. 5. Comparison between dye pattern with depth in the clayey and sandy soil, respectively.
Fig. 6. Comparison between wet areas, dye cover, and isolines for bromide concentration for the
clayey soil (plot 3).



















828










































(a)
(b)
Figure (1)
829

0 10 20 30 40
Water content %
-50
-40
-30
-20
-10
0
D
e
p
t
h

(
c
m
)
clayey soil
before
after


0 10 20 30 40
Water content %
-50
-40
-30
-20
-10
0
D
e
p
t
h

(
c
m
)
sandy soil
before
after

(a)
(b)
Figure (2)
830
0 300 600 900 1200 1500
Coverage(cm
2
)
-50
-40
-30
-20
-10
0
D
e
p
t
h

(
c
m
)
Dye coverage (cm
2
)
Sandy soil
Clayey soil


Figure (3)











831
0 500 1000 1500 2000 2500
Coverage (cm
2
)
-50
-40
-30
-20
-10
0
D
e
p
t
h

(
m
)
Plot 1
Bromide
Dye
0 500 1000 1500 2000 2500
Coverage (cm
2
)
-50
-40
-30
-20
-10
0
D
e
p
t
h

(
c
m
)
Plot 2
Bromide
Dye
0 500 1000 1500 2000 2500
Coverage (cm
2
)
-50
-40
-30
-20
-10
0
D
e
p
t
h

(
c
m
)
Plot 3
Bromide
Dye



Figure (4)








832
a: Clayey soil plot 2 b: Sandy soil plot 5 a: Clayey soil plot 2 b: Sandy soil plot 5

0 cm 30 cm

5 cm 35 cm

10cm 40 cm

15cm 45 cm

20 cm 50 cm

25 cm
Figure (5)

833
Clayey soil (plot (3))

0 cm 30 cm

5 cm 35 cm

10 cm 40 cm

15 cm 45 cm

20 cm 50 cm

25 cm
(black-gray-white images) (contour maps)
Figure (6)
(black-gray-white images) (contour maps)

834
Appendix A. List of symbols.
K
a
the complex permittivity of the bulk soil
K
w
the permittivity of the pore water
water content

a
bulk electrical conductivity

in
average natural soil conductivity

p
electrical conductivity of the applied pulse

rel
relative electrical conductivity

w
soil solution electrical conductivity

w,tot
total electrical conductivity (irrigation water and solute)





















835
References
Adam, N.K., 1963. Principles of water-repellency. In: Moillet, J .L.(Ed.), Water proofing and
Water-Repellency. Elsevier, London, pp. 1-23.
Aeby, P., Forrer, J ., Steinmeier, C., Flhler, H., 1997. Image analysis for determination of dye
tracer concentrations in sand columns. Soil Sci. Soc. Am. J ., 61, 33-35.
Bisdom, E.B.A., Dekker, L.W., Schoute, J .F., 1993. Water repellency of sieve fractions from
sandy soils and relationships with organic material and soil structure. Geoderma 56, 105-
118.
Blackwell, P.S., 1993. Improving sustainable production from water repellent sands. Western
Australia J . of Agr., 34, 160-167.
Carter, D.J ., Hetherington, R.E., 1994. Claying of water repellent soils on the South coast of
Western Australia. Proceedings of the 2
nd
National Water Repellency Workshop, 1-5
August, Perth, Western Australia. pp. 49-57.
Chan, K.Y., 1992. Development of seasonal water-repellence under direct drilling. Soil Sci. Soc.
Am., 56, 326-329.
Crockford, S., Topalidis, S., Richardson, D.P., 1991. Water repellency in a dry sclerophyll forest-
measurements and processes. Hydrol. Proc., 5, 405-420.
DeBano, L.F., 1971. The effect of hydrophobic substances on water movement in soil during
infiltration. Proceedings of the Soil Sci. Soc. Am., 35, 340-343.
DeBano, L.F., 1991. The effects of fire on soil properties. United states Department of
Agriculture, Forest Service. General Technical Report, INT-280, 151-156.
Dekker, L.W., Ritsema, C.J ., 1996b. Preferential flow paths in a water repellent clay soil with
grass cover. Water Resour. Res., 32 (5), 1239-1294.
Flury, M., Flhler, H., 1994. Brilliant Blue FCF as a dye tracer for solute transport studies A
toxicological overview. J . Environ. Qual., 23, 1108-1112.
Flury, M., Flhler, H., 1995. Tracer characteristics of Brilliant Blue FCF. Soil Sci. Soc. Am. J .,
59, 22-27.
Gee, G. W., Kincaid, T., Lenhard, R.J ., and Simmons, C.S., 1991. Recent studies of flow and
transport in the vadose zone, U.S. Natl. Rep. Int. Union Geod. Geod. Geophys., 1987-
1990, Rev. Geophys., Vol. 29, 227-239.
836
Giovannini, G., Lucchesi, S., 1983. Effect of fire on hydrophobic and cementing substances of
soil aggregates. Soil Sci., 136, 231-236.
Hilhorst, M.A., 2000. A pore water conductivity sensor. Soil Sci. Soc. Am. J ., 64, 1922-1925.
Ketelsen, H., Meyer-Windel, S., 1999. Adsorption of Brilliant Blue FCF by soils. Geoderma, 90,
131-145.
King, P.M., 1981. Comparison of methods for measuring severity of water repellence of sandy
soils and assessment of some factors that affect its measurement. Australian J . of Soil
Res., 19, 275-285.
Kung, K.-J . S., 1990. Preferential flow in a sandy vadose zone: 1. Field observation. Geoderma,
46, 51-58.
Lin, H.S., Mcinnes, K.J ., 1995. Water flow in clay soil beneath a disc tension infiltrometer. Soil
Sci., 159, 375-382.
Marion, G.M., Babcock, K.L., 1976. Predicting specific conductance and salt concentration in
dilute aqueous solutions. Soil Sci., 122, 181-187.
McGhie, D.A., Posner, A.M., 1980. Water repellence of a heavy textured western Australia
surface soil. Australian J . of Soil Res., 18, 309-323.
Nyfors, E., Vainikainen, P. (1989) Industrial Microwave Sensors. Artech Hous, Norwood, USA.
hrstrm, P., Persson, M., Albergel, J ., Zante, P., Nasri, S., Berndtsson, R., Olsson, J ., 2001.
Field-scale variation of preferential flow as indicated from dye coverage. J . Hydrol., 257,
164-173.
Rhoades, J .D., Kandiah, A., Mashali, A. M. (1992) The Use of Saline Waters For Crop
Production. FAO Irrigation and Drainage, paper 48. Food and Agriculture Organization of
the United Nations, Rome.
Roberts, F.J ., Carbon, B.A., 1971. Water repellence in sandy soils of south-western Australia: 1.
Some studies related to field occurrence. Division of Plant Industry CSIRO (Australia).
Field Station Record 10, 13-20.
Shalhevet, J ., Hoffman, G.J ., Meiri, A., Heuer, B. and Francois, L.E.,1983. Salinity tolerance of
crops in irrigated agriculture dynamic conditions, Final Report to BARD.
Shalhevet, J ., 1994. Using water of marginal quality for production: major issues. Agricultural
Water Management 25, 233-269.
837
Soil Survey Staff, Soil Taxonomy, 1975. A basic system for soil classification for making and
interpreting soil surveys, Agric. Handb. U.S. Dep. Agric. Sci. Edu. Adm., 436, 754 pp.
Wallis, M.G., Scotter, D.R., Horne, D.J ., 1991. An evaluation of the intrinsic sorptivity water
repellency index on a range of New Zealand soils. Australian J . of Soil Res. 29, 353-362.
Wilkinson, J .F., Miller, R.H., 1978. Investigation and treatment of localized dry spots on sand
golf greens. Agron. J ., 66, 299-304.
Witter, J .V., J ungerius, P.D., ten Harkel, M.J ., 1991. Modelling water erosion and the impact of
water-repellency. Catena 18, 115-124.
Yasuda, H., Berndtsson, R., Persson, H., Bahri, A., Takuma, K., 2001. Characterizing preferential
transport during flood irrigation of a heavy clay soil using the dye Vitasyn Blau. Geoderma,
100, 49-66.
Zehe, E., Flhler, H., 2001. Preferential transport of isoproturon at a plot scale and a field scale
tile-drained site. J . Hydrol., 247, 100-115.















838

FIRST
AIN SHAMS UNIVERSITY INTERNATIONAL CONFERENCE
ON
ENVIRONMENTAL ENGINEERING

April 9-11 2005


SOIL IMPROVEMENT IN NEW ARES IN EL SALAM
CANAL COMMAND AREA


Ghada G. El-Refaie

Researcher, Drainage Research Institute, National Water Research Centre,
Cairo, Egypt.

ABSTRACT

In arid and semi arid regions such as Mediterranean, population growth, rapid
urbanization and industrialization are imposing rapidly growing demands and
pressures on water resources. In addition to that water use has increased since
1940 with agricultural accounts for two thirds of all water used-mostly for
irrigation. The growing imbalance between supply and demand has led to
shortage of water, rising of pollution and other environmental negative impacts.
This increasing pressure on available fresh water resources created interest in
the use of marginal quality water for irrigation, situation that has resulted in
national policy for the reuse of drainage water in reclamation and irrigation in
new areas in Egypt.

One of the major projects that contributes with this policy is El-Salam Canal
project which divert considerable amounts of drainage water after bending with
fresh water from the Nile River to irrigate 620,000 feddans (260,400 ha) of new
lands in the northeastern Delta (in south El-Husseinia plain) and northern Sinai
Peninsula. South El-Husseinia Plain of El-Salam canal command area comprises
about 64,000 feddans, and is located in former marshlands that at one time
comprised the lake bed of Lake Manzala. The soils are therefore heavily saline
in their original state leading to numerous challenges when using water from El-
Salam canal for irrigation. Reuse of low quality drainage water in such areas
with such highly saline soil conditions require special reclamation and cropping
measures and it could also have adverse impacts on soil and crop quality.
839
This research was carried out in two pilot areas (El-Eman and El-Rowad) in
south El-Husseinia plain to assess the effects of drainage water reuse of El-
Salam canal on the soil quality improvement and on land use and cropping
pattern. A monitoring program was carried out since 1999 to monitor water, soil
and crop. Simple statistical analysis and GS+program were used to study the
salinity variation and distribution along the study period.
The results showed that using mixed water of El-Salam canal in reclamation and
irrigation caused a significant improvement in soil salinity.
2. DESCRIPTION OF STUDY AREA
The study area lies within south El-Husseinia Plain of El-Salam Canal command area
where the total land area within the plain is about 64,000 feddans. Soil texture in the area is
loamy clay-to-clay extending to more than 2.0m and extremely saline. The calcium carbonate
content is low and the high sodicity leads to wetness of the soil profile at some locations. The
water table depth is shallow and varied between 0.7 to 1.9m and it is highly saline. In general,
the soils of the study area are mainly salt-affected soils where its salinity is more than 16
dS/m and reaches 83 dS/m in about 70 percent of the area, (NAWQAM, 1999).
In winter season of 1998-1999, there were no systemic cropping patterns and about 25
percent of the land was sporadically cultivated, whereas 75 percent was still idle. In the
following years, the percentage of cultivated lands increased slightly.
In this research two pilot areas were considered in south El-Hussania plain, El-Rowad pilot
area where the farmers are graduated but they do not have the financial support which could
help in land reclamation. The second area considered is El-Eman Pilot area which is owned
by investors who have the financial help for land reclamation and use modern techniques for
different agricultural procedures.
2.1 El Rowad Pilot Area

El-Rowad pilot area is located at the far western part of El-Rowad village. It consists of
nine strips divided into 87 plots, approximately five feddans each as shown in Figure 1. The
available irrigation water is El-Salam canal water which is supplied through an intake on El-
Saeedy canal.
Soil texture was determined at 4 different depths (0-25, 25-50, 50-75, and 75-100cm)
where it is mainly silty loam in the top layers and clay in the lower layers. The hydraulic
conductivity (k) was measured by the auger holes method, its values varied between 0.003
and 0.203 m/day with an average of 0.054 m/day. The hydraulic conductivity is low, due to
the high clay content and soil chemical characteristics.
840

Figure1. Plan of El-Rowad Pilot Area
2.3 El Eman Pilot Area

El-Eman pilot area is located in the middle of El-Eman cooperative, which is located in the
middle of South El-Husseinia plain. It is surrounded by El-Salam Canal to the east, and El-
Marhala El-Thaltha branch canal to the South. Its total area is 1500 feddans and is divided
into 22 plots owned by a number of investors as shown in figure 2.


Figure 2. El-Eman Pilot Area

841
3. SOIL SALINITY DETERINATION

Soil salinity has been defined and assessed in terms of laboratory-measurements of the
electrical conductivity of the extract of a saturated soil-paste sample EC
e
. This is because
electrical conductivity is easily measured and reprehensive of the total concentration of
ionized solutes in an aqueous sample. The saturation percentage (SP) is the lowest water/soil
ratio suitable for the practical laboratory extraction of readily dissolvable salts in soils (US
Salinity Laboratory, 1969).
Soil salinity can also be determined from the measurement of the electrical conductivity
of a soil-water sample (EC
w
). This measurement can be made either in the laboratory using a
collected sample or directly in the field using in situ, imbibitions-type salinity sensors.
Theoretically, the electrical conductivity of soil solution (EC
w
) is a better index of soil
salinity than the traditional index (EC
e
), where the plant roots extract their nutrients from the
soil and absorb other solutes, and they consume this water through the process of
transpiration. However, EC
w
has not been widely used to express soil salinity for several
reasons. Firstly, it is not single-valued; it varies over the irrigation cycle as the soil/water
content changes (Rhoades, 1992). Thus, EC
w
does not lend itself to simple classification or
standard unless it is referenced to specific water content, such as field capacity. Secondly, and
probably most important EC
w
has not been widely adopted for routine appraisals of soil
salinity because method for obtaining soil water samples are not practical at typical field
water contents.
Numerous measurements are needed to characterize field salinity conditions and the
measurements often need to be updated as conditions change to study the soil salinity
variations with time. These measurements are; soil profile description, hydraulic conductivity,
moisture content, and Electromagnetic Conductivity (EM) using (EM38) as an indicator of
soil salinity. Also, soil samples were collected and analyzed periodically to determine soil
quality through monitoring program. The monitoring activities were undertaken in October
1999 at El Rowad pilot area and at El Eman pilot area in April 2000.

4- IRRIGATION WATER SALINITY
The most important factor affects the soil leaching and reclamation is the irrigation
water availability and quality, mainly salinity. The irrigation water of El-Salam canal is
available along the year and its salinity was monitored twice per month along study period.
The irrigation water salinity of El-Salam canal and the irrigation water salinity of the
two pilot areas are shown in figure 3. It could be noticed from the figure that in March 2001,
there was a jump in the irrigation water salinity in both the pilot areas especially in El-Rowad
where it reached to 14.8 dS/m while it reached to 6.0 dS/m in El-Eman pilot area. Along the
year 2002, the irrigation water salinity of El-Eman and El-Rowad pilot were very close to
each other and very close to El-Salam canal water salinity.
Simple descriptive statistical analysis was carried out as shown in Table (1) which
summarizes the irrigation water salinity variations for the two pilot areas and El-Salam Canal.
The analysis showed that the average salinity of the irrigation water of El-Rowad and El-
Eman were 2.63 and 2.23 dS/m respectively. Generally, the irrigation water salinity of El-
Rowad is higher than El-Eman where El-Eman area is the closest pilot area to El-Salam canal
and the seepage effects from agricultural fields is minimum.
842
0
2
4
6
8
10
12
14
16
10/00 01/01 04/01 07/01 11/01 02/02 05/02 09/02 12/02 03/03
Date
W
a
t
e
r

s
a
l
i
n
i
t
y

(
d
S
/
m
)
El SalamCanal El Rowad El Eman

Figure 4. Temporal Variations in Irrigation Water Salinity

Table 1. Descriptive Statistical Analysis of Irrigation Water Salinity (dS/m).
El-Salam El-Rowad El-Eman
Maximum 3.56 14.80 6.28
Minimum 1.05 1.00 1.18
Average 1.59 2.63 2.23
Standard deviation 0.76 3.37 1.39

5. RESULTS AND DISCUSSION
5.1 Land Use of El Rowad Pilot Area

At early stage of reclamation, the fields were first smoothed and leveled using
conventional methods or laser equipment, and then the land is divided into strips bordered by
10-15m spaced open ditches to enhance the leaching process. The fields are then flooded for
one or two years for salt leaching and reclamation. Rice was cultivated when soil salinity
reached to 25 dS/m in summer seasons to help in leaching and land reclamation although no
good production was predicted. After a period when soil salinity reduced and reached suitable
values different winter crops were cultivated such as; Barley, Wheat, Berseem and Sugar Beet
as shown in figure 4 which present the layout of the demonstration farm as a representative
plot of El-Rowad pilot area.
843

Crop Stri ps number
Canal
Drain
1
Irri gation pump
1
0
0


m
200 m
H
o
s
h

D
r
a
i
n
R3477 R3476
2
beet
Suger
Drainage discharge
measuring point
R3475
beet
Suger
2 1
Barly
R3474
1
R3472
2
Wheat
R3471
Wheat
1
Barly
R3473
and flow meter
Irrigation pump
H
o
s
h

R
o
a
d
H
o
s
h

D
r
a
i
n
Drainage discharge
measuring point

Figure 4. Layout of the Demonstration Farm in El Rowad Pilot area (Plot 347)
Land use in El-Rowad pilot area during the period from summer 1996 up to winter 2002 is
shown in figure 5. The figure shows that the percentage of idle (abandon) plots was very high
at the beginning where it was 90% and10 % only for leached plots and no cultivated plots due
to high soil salinity. While in winter 02/03, the percent of cultivated plots increased from 0%
to be 63% and only 2% was idle. It is clear from the graph that, there is a good improvement
in land use and soil fertility due to soil salinity reduction as a result of regular leaching,
irrigation and existence of good field drainage system.

0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
S-96 W-96 S-97 W-97 S-98 W-98 S-99 W-99 S-00 W-00 S-01 W-01 S-02 W-02
Seasons
A
r
e
a

P
e
r
c
e
n
t
Cultivated Fallow Leaching Idle

Figure 5. Land use of El-Rowad pilot area
844
5.2 Soil Salinity of El-Rowad Pilot Area
The study of the soil salinity changes in El-Rowad pilot area was carried out in the all
monitoring plot of El-Rowad pilot area, where the soil salinity was reduced gradually from
63.0 dS/m in October 1999 to 8.0 dS/m in April 2002. In this study the results of one plot
only will be presented (plot no.347, Demonstration farm) to be a representative to the whole
El-Rowad area where all the agricultural procedures is under full control.
Figure 6 shows three soil salinity profiles for soil layer extended from soil surface up to
100 cm depth below soil surface for Plot 347 at the end of years 2000, 2001 and 2002. It
could be noticed from the graph that at the at year 2000 soil salinity of surface layer (0-25 cm)
is less than the deeper layer (25-50 cm) due to leaching processes which causes salt
movement from upper layers to lower layers while the soil salinity for the deeper layer (50-
100 cm) decreased due to good drainage system which releases the saline water table. Also, it
could be noticed that at Oct.2002, the soil salinity profile along soil layer from soil surface to
100 cm below soil surface is nearly same due to the successive reclamation and cultivation.
Figure 7 shows the soil salinity variation with time for two profiles (0-50 cm and 50-
100cm) depth. It is clear that at the beginning of the reclamation the soil salinity for upper
layer (0-50) was high nearly 17 dS/m while the lower layer (50-100) was nearly 15 dS/m and
after one year the situation changed where the upper layer salinity became less than lower
layer salinity.
This result could be concluded that the continues leaching from the beginning of
reclamation and the succeeded land use by cultivating rice in summer and different crops in
winter helped in reducing soil salinity and improvement soil fertility.

0
25
50
75
100
0.0 5.0 10.0 15.0 20.0 25.0
Salinity (dS/m))
D
e
p
t
h

(
c
m

f
r
o
m

G
.
S
.
)
23/9/00 22/9/01 23/10/02

Figure 6. Soil Salinity Profile for Plot 347

845
0
5
10
15
20
25
02/99 08/99 03/00 10/00 04/01 11/01 05/02 12/02 06/03
Date

S
o
i
l

s
a
l
i
n
i
t
y

(
d
S
/
m
)
0-50 50-100

Figure 7. Soil Salinity Variations with Time for Plot 347

To study the soil salinity distribution within the plot, GS+program was used to show the
salinity distribution for four seasons started in winter 01/02 until winter 02/03 which is
presented as an example in figure 8. The distribution showed that in winter 02/03 the
maximum soil salinity was higher than 12.0 dS/m at the higher right corner of the plot and the
salinity decreased gradually towards the left side of the plot to be less than 8.0 dS/m. This is
due to existence of the irrigation canal in the left side and the water moves to the right side
carrying the leached salts. This distribution can give an indication that the continuous
leaching affect salt concentration in the soil and also the irrigation system could affect the
direction of the salt removal and the soil improvement.




Figure 8. Soil Salinity Distributions in Plot 347 for Winter 02/03

846
5.3 Land Use of El-Eman Pilot Area
The soil salinity of El-Eman pilot area improved quickly due to the availability of
irrigation water with suitable salinity, cultivating Rice in summer seasons and trying to
cultivate different crops in winter, Also, This area is wholly owned by investors, who used
fish farms as a method for leaching their fields. Therefore the improvement in productivity is
higher than the other area in both winter and summer seasons, where in winter 99/2000 only
50 % of the plots were abandon and 50 % were under leaching processes while in summer
2002, only 4 % of the plots were abandon, 58 % were cultivated and the rest were still under
leaching process.
Figure 9 shows the percentage of the cultivated crops in El-Eman area starting in summer
2000 to winter 02/03. It could be noticed from the graph that the main crop cultivated in all
summer seasons was Rice while the main crops in winter were Wheat, Barley, Berseem and
Sugar Beet. Also, it could be noticed from the graph that the number of crops cultivated in
one season was increase due to suitable soil salinity for different crops varieties.
Figure 10 shows the layout of the demonstration farm of El-Eman pilot area (plot B) where
all the agricultural procedures are under control and the different cultivated crops in winter
seasons. This plot will be consider as an example of the whole El-Eman area to study the soil
salinity changes with time along the monitoring period.
0
10
20
30
40
50
60
70
80
90
100
G
r
a
s
s
e
s
R
i
c
e

G
r
a
s
s
e
s
W
h
e
a
t

M
i
x
e
d

c
r
o
p
s
B
e
r
s
e
e
m
G
r
a
s
s
e
s
R
i
c
e

M
i
x
e
d

c
r
o
p
s
G
r
a
s
s
e
s
B
a
r
l
e
y

M
i
x
e
d

c
r
o
p
s
B
e
r
s
e
e
m
G
r
a
s
s
e
s
R
i
c
e

G
r
a
s
s
e
s
M
i
x
e
d

c
r
o
p
s

S
u
g
a
r

B
e
e
t
B
e
r
s
e
e
m
B
a
r
l
e
y
W
h
e
a
t
Summer
2000
Winter 2001 Summer
2001
Winter 2002 Summer
2002
Winter 2003
P
e
r
c
e
n
t

o
f

C
u
l
t
i
v
a
t
e
d

A
r
e
a

(
%
)

Figure 9. Crop Pattern of El-Eman Pilot Area
847
Hamed
w el l
Ob s er v at i o n
Ob s er v at i o n
w el l
w el l
St r i p n u m b er
Di v er
Ob s er v at i o n w el l
1
8
1


m
1
3
7
m
6
1


m
8
0


m
7
3


m
5
5


m
3
7


m
Ho u s i n g
Ro ad w i t h w i d t h 3 m
B er s s em
Wh eat B ar l ey
37 m
135 m
172 m
R
o
a
d

w
i
t
h

w
i
d
t
h

5

m
S m al l C a na l
Ob s er v at i o n
D
r
a
i
n

w
i
t
h

d
e
p
t
h

2
.
0
0

m
Di v er
R
oad w
i t h w
i d t h 3 m
D
r a
i n
w
i t h
d
e
p
t h
1
.2
5
m
Su g ar b eet
Su g ar b eet
4
3
2
1
Su g ar b eet
Ou t s i d e Far m
Gr as s
Sm al l Can al
Mai n Ro ad
Hag

Figure 10. Layout of the Demonstration Farm in El Eman Pilot area (Plot B)
5.4 Soil Salinity of El-Eman Pilot Area
In plot B soil salinity decreased along the monitoring period as shown in figure 11 which
presents the soil salinity variation with time for four depths (0-25 , 25-50, 50-75 and 75- 100
cm). It clear from the figure that salinity of the upper soil layer (0-25 cm) was very high at
summer 2000, nearly equal to 35 dS/m and reduced due to leaching to be less than 9 dS/m at
Oct. 2002.
Also, it is clear that the salinity of the upper layer was higher than other deeper layers at
beginning of reclamation but after continuous leaching the situation was changed where the
upper layers salinity (0-25 and 25-50 cm) was less than deeper layer salinity (50-75 and 75-
100cm), due to the downward salt movement. The concluded result is that soil salinity of all
soil layers was decreased to be less than 9.0 dS/m at end of 2002 and reached to the value
which is suitable to different crops cultivation.
0
5
10
15
20
25
30
35
40
03/00 10/00 04/01 11/01 05/02 12/02
Date
S
a
l
i
n
i
t
y

(
d
S
/
m
)
00 - 25 cm 25 - 50 cm 50 - 75 cm 75 -100 cm

Figure 11. Soil Salinity Variations with Time for Different Soil Depths for Plot B)
848


Win 2001-2002


Win 2002-2003
Figure 12. Soil Salinity Distributions in Plot B

The soil salinity distribution for four seasons started in winter 01/02 until winter 02/03 was
studied using GS+. Figure 12 shows the salinity distribution in winter 01/02 as an example of
the beginning of reclamation and for winter 02/03 as an example of the successive reclamation.
The distribution showed that in winter 01/02 the maximum soil salinity was 23.1 dS/m at the
left side of the plot and the salinity decreased gradually towards the right side of the plot to be
nearly 11.3 dS/m. This is due to existence of the irrigation canal in the right side and the water
moves to the left side carrying the leached salts. While in winter 02/03 the salinity distribution
was changed to be in most of the plot less than 9 dS/m. This distribution can give an indication
that the continuous leaching affect salt concentration in the soil and also the irrigation system
could affect the direction of the salt removal and the soil improvement.

7. CONCLUSION
1. Using low quality drainage water in irrigation has the potential of causing serious
problems of soil degradation due to salinization. Soil salinity is one of the main
problems affecting soil quality and crop production and quality. Aassessment of soil
salinity in irrigated fields is needed for determining the appropriate management
practices and land use.
2. The land use plays a very important role in soil reclamation, where in the both pilot
areas Rice was cultivated in summer although soil salinity was higher than its threshold
value but it helps in soil leaching even no production was predicted at the beginning.
3. .Also, different crops were cultivated in winter such as Berseem, Wheat, Barely and
Suger beet where the different crops cultivation helps in the new, highly saline area in
leaching through irrigation and salt movement
4. . The results showed that in both areas the soil salinity in upper layers (0-25 cm) is less
than the deeper layers (25-50 cm) due to leaching which causes salt movement from
upper layers to lower layers. The soil salinity for the much deeper layer (50-100 cm)
decreased due to good drainage system which releases the saline water table.
5. Finally, mixed water of El-Salam canal could be used successfully in land reclamation
in the new highly saline areas, and also in cultivation of different crops where soil
salinity of both areas reduced with reasonable values through the monitoring period.

849
8. REFERENCES
1. NAWQAM, 1999, Inception Report, Drainage Water Reuse and pilot Schemes, IN-
9902-004-DR,Feb 1999,.
2. Rhoads, J .D., A. Kandiah and A.M.Mashali,1992, FAO no 48, ,The use of saline waters
for crop production
3. U.S. Department of Agriculture (USDA), US Salinity Laboratory Staff, Diagnosis and
Improvement of Saline Alkali Soils, Agriculture Handbook No.60, February
1954/August 1969.
,
850