100-RG-PNC-00000-900008 | Summer 2010

Appendix B
Report on Approaches to UWWTD Compliance in Relation to CSOs in major cities across the EU

THAMES TUNNEL

NEEDS REPORT APPENDIX B

LIST OF CONTENTS

Executive Summary 1. Introduction 1.1. Background 1.2. Scope of work 1.3. Methodology 1.4. EU City selection process 2. City Responses 2.1. Austria 2.1.1. Vienna 2.2. Croatia 2.2.1. Zagreb 2.3. Czech Republic 2.3.1. Prague 2.4. Denmark 2.4.1. Copenhagen 2.5. Finland 2.5.1. Helsinki 2.6. France 2.6.1. Lyon 2.6.2. Marseille 2.6.3. Paris 2.7. Germany 2.7.1. Berlin 2.7.2. Hamburg 2.7.3. North Rhine-Westphalia/Rhine-Ruhr (Emscher) 2.8. Greece 2.8.1. Athens 2.8.2. Thessaloniki 2.9. Italy 2.9.1. Naples 2.10. Netherlands 2.10.1. Rotterdam 2.11. Portugal 2.11.1. Lisbon 2.12. Spain 2.12.1. Barcelona 2.12.2. Madrid 2.13. Sweden

6 8 8 8 8 9 12 12 12 14 14 16 16 19 19 23 23 26 26 26 28 31 31 34 40 45 45 46 47 47 48 48 48 48 50 50 53 53

2.13.1. Stockholm 3. Summary and Conclusions 3.1. Large projects within the EU 3.2. Small projects within the EU 4. References

53 56 56 56 59 68 68 69 70 71 71 72 72

Appendix A CSO consenting/design approaches in the EU A.1. A.2. A.3. A.4. A.5. A.6. A.7. General Czech Republic Flanders Germany Italy Netherlands Spain

List of Figures
Figure 1. Location and schematic layout of the Wiental-Kanal, Vienna (Wien International, 2006) ...................................................................................................... 12 Figure 2. Schematic of RTC data flow for the sewer Network of Vienna (Teufel, 2007) ............................................................................................................................ 13 Figure 3. Schematic of the Vienna sewer network RTC system (Teufel, 2007) ......... 14 Figure 4. Flood gate on the central sewer of Prague (Hrabak et al., 2005) ................. 17 Figure 5. Measured and predicted reductions in CSO volumes released to the Copenhagen harbour areas (Sorensen and Kofod-Andersen, 2005) ............................ 21 Figure 6. Sewer network of the Helsinki metropolitan area (Helsinki Water, 2009) .. 24 Figure 7. Water quality in Helsinki and Espoo (Helsinki Water, 2009) ...................... 25 Figure 8. The Marseille sewer network (Laplace et al., 2007) .................................... 27 Figure 9. WWTPs on the Paris sewer network (new/planned WWTPs shown with dashed lines), (Even et al., 2007)................................................................................. 30 Figure 10. Storage volumes within the Berlin combined sewer system (Schroeder, 2009) ............................................................................................................................ 33 Figure 11. The Rhine-Ruhr Metropolitan Area, Germany (Wikipedia, 2009b) .......... 41 Figure 12. Location of treatment works and restructuring of the Emscher sewer system (Frehmann et al., 2008).................................................................................... 41 Figure 13. The 'Rainwater Route' of the Emscher Region (Becker and Raasch, 2001) ...................................................................................................................................... 42 Figure 14. Rainwater and infiltration projects in the Emscher Region (Becker and Raasch, 2001) ............................................................................................................... 43 Figure 15. Longitudinal section of the Emscher trunk sewer (Frehmann et al., 2008)44 Figure 16. COD for 4 WWTP control scenarios on the Emscher sewer system (Frehmann et al., 2008) ................................................................................................ 45 Figure 17. Sand removal at Impianto di Coroglio, Naples (Gisonni, 2009) ................ 48 Figure 18. Roto sieving at the Impianto di Coroglio, Naples (Gisonni, 2009) ............ 48 Figure 19. The sewer network of Lisbon ..................................................................... 49 Figure 20. Vortex flow control valve in the Lisbon sewer network (Almeida, 2009). 49 Figure 21. Construction of the Alcantara WWTP in Lisbon (Almeida, 2009) ............ 49

Figure 22. Real Time Control of the sewer network of Barcelona (Escaler Puigoriol, 2009) ............................................................................................................................ 51 Figure 23. Detention tank in the sewer network of Barcelona (Escaler Puigoriol, 2009) ............................................................................................................................ 52 Figure 24. Gates in the sewer network of Barcelona (Escaler Puigoriol, 2009).......... 52 Figure 25. The Real Time Control Process in Barcelona (Escaler Puigoriol, 2009) ... 52 Figure 26. Faecal coliform numbers at various sampling locations along Sant Sebastia beach, Barcelona (CFU/100ml), (Escaler Puigoriol, 2009) ......................................... 52 Figure 27. Location of WWTPs in Stockholm (Stenroos and Katko, 2006) ............... 53 Figure 28. Profile and plan of the Ormen Project, Stockholm (Nordmark, 2002) ...... 54

List of Tables
Table 1. Wastewater collection systems in EU Member States (EWWG, 1995) Table 2. Large European Union cities (ranked by metropolitan population) Table 3. CSOs permitting back flow to the Prague sewer network (Hrabak et al., 2005) Table 4. Modifications to CSO83 on the Botic Stream, Prague (Kabelkova et al., 2007) Table 5. Investment costs () for the three scenarios (Clauson Kaas et al, 2008) 9 11 18 18 22

Table 6. Operation and maintenance costs () for the three scenarios (Clauson Kaas et al, 2008) 22 Table 7. Simulated annual discharge of E. coli under three different discharge scenarios (Clauson Kaas et al, 2008) 23

Table 8. Simulated annual discharge of pollutants from the WWTP in under three different discharge scenarios (Clauson Kaas et al, 2008) 23 Table 9. Statistics of the Helsinki sewer network (Helsinki Water, 2009c) Table 10. Sources of BOD and suspended solids in the Hamburg sewer network (Abraham, 2009) 24 37

Table 11. Overflow volume reductions achieved by facilities of the sewer network in Hamburg (Abraham, 2009) 38 Table 12. Odour management regime of the Schdlerstrae retention basin, Hamburg (Abraham, 2009) 40 Table 13. Wastewater volumes by sector in Madrid (GHK, 2006) Table 14. Investment in wastewater treatment in Madrid (GHK, 2006) Table 15. Summary of CSO abatement approaches in large cities of the EU (alphabetically by country) Table 16. Summary of CSO abatement approaches in small cities of the EU (alphabetically by country) 53 53 57 58

Table 17. CSO permitting arrangements in selected member states (Adapted from Zabel et al., 2001) 68 Table 18. Overview of emission requirements for CSOs in EU member states (Adapted from Dirckx, 2009) Table 19. Investment priorities for wastewater in Spain (GHK, 2006) 69 72

Executive Summary
The report identifies and documents the way in which large European Union (EU) cities in several Member States aim to achieve or are achieving compliance with the Urban Wastewater Treatment Directive (UWWTD). Emphasis is placed on how these cities are complying with respect to combined sewer overflow (CSO) discharges into key watercourses, rather than wastewater treatment schemes (although some overlap has been identified). Large EU cities were ranked based on their metropolitan population size. Cities with a population equivalents in excess of 1 million were prioritised for investigation. A thorough review of publications from journals, conference proceedings and other sources was undertaken and key experts from across Europe were contacted, in order to provide the most up to date information. The main drivers in dealing with problematic CSOs are not only restricted to their compliance with the UWWTD but also include complying with the Bathing Water Directive and reducing urban flooding. Additionally, responses to either the UWWTD or National Laws transposed from the UWWTD vary quite widely between Member States depending on the region or water/wastewater management structure/organisation. Several cities have established a Master Plan for Urban Drainage as a vehicle by which to review and address the existing and future issues associated with parts of their sewer networks (Barcelona, Hamburg, Prague, Zagreb). The most common approach to resolving CSO issues was identified to be the addition of extra capacity, whether by the construction of detention tanks and/or trunk or interceptor sewers (Athens, Thessaloniki). Within some cities the use of these approaches was complemented by the use of RTC (Barcelona, Lisbon, Marseille, Vienna, Zagreb). Several cities also combined both of these approaches with WWTP expansion (Copenhagen, Lisbon, Paris, Prague) and/or sewer separation (Copenhagen, Hamburg). Two German cities were identified as utilising source control techniques (SUDS/disconnection-infiltration, retention basins) alongside some of the more traditional approaches (Berlin, North Rhine-Westphalia). Only two recent projects, located in Naples and Vienna, utilised tunnels in combination with WWTP expansion and RTC, respectively, although the use of tunnels is currently being assessed in Paris. Helsinki and Stockholm were also identified as utilising tunnels, but this was due to a range of historic reasons, not UWWTD compliance. Within smaller cities, a range of approaches was also identified, ranging from interceptor sewers (Granollers, Spain and Steinkjer, Norway) and off-line storage basins (Cosenza, Italy), through to source control and SUDS (Baerum and Bergen, Norway and Lund and Malmo, Sweden) and local or pre-treatment techniques (Oeiras, Portugal). RTC was also popular at this scale, especially in Germany (Bochum, Dresden, Leipzig, Obere Iller). Within all approaches identified, data collection and modelling were key components of design and comparisons of several options using feasibility assessments were common. Operationally, Scandinavian and Western European cities were identified as being further ahead with implementation than Eastern European cities. These tended to be in the data collection and modelling phase, rather than the construction phase. Vienna was the most advanced in utilising RTC, but even so was not 100% operational. In

conclusion, it is apparent that there is not a one size fits all intervention in dealing with problematic CSOs, when trying to comply with the UWWTD or other drivers.

1.

Introduction

1.1. Background
The overall aim of this report is to examine the schemes that European cities are planning or have implemented to comply with the Urban Wastewater Treatment Directive (UWWTD). Specifically, the report will outline: The UWWTD implementation actions of selected EU Member States; A description of major city schemes for selected EU Member States.

Emphasis will be placed on how the major EU cities are complying with the requirements of the UWWTD with respect to CSO discharges into key watercourses, rather than wastewater treatment schemes (although there is clearly some overlap).

1.2. Scope of work

The report will concentrate on specific schemes and key cities/rivers where CSOs are problematic. The focus will be primarily the responses that have been investigated and implemented to resolve these issues. The degree to which each member state has complied with the UWWTD will not be covered in detail.

1.3. Methodology
Information has been gathered for this report using the following methods: Internet search to produce a list of the largest cities in the EU, together with their associated water bodies; Preliminary internet literature search for documents relating to the UWWTD implementation and CSOs; Efforts to obtain Eureau document B4-3040/96/000173/DI, titled Stormwater pollution control in the EU member states, via contacting Eureau direct and also personal contacts the document could not be located; Review of the Pennine Water Group Thames Tideway Study-Overview of Findings report (Ashley and Stovin, 2005); Electronic journal search using EBSCO-EJS to identify relevant academic papers; Review of appropriate conference proceedings, including:
o o o o International Conference on Urban Drainage via Internet, 2000; 2nd International Conference on Interactions Between Sewer, Treatment Plants and Receiving Waters in Urban Areas (Interurba II), 2001 (Lisbon); Speciality Symposium of the World Water and Environment Congress Urban Drainage Modelling, 2001 (Orlando); International Symposium on Frontiers in Urban Water Management: Deadlock or Hope?, 2001 (Marseille);

o o o o o o o o o

3rd International Conference on Sewer Processes and Networks, 2002 (Paris); International Conference on Sewer Operation and Maintenance, 2002 (Bradford); 9th International Conference on Urban Drainage, 2002 (Portland); 6th International Conference on Urban Drainage Modelling, 2004 (Dresden); 10th International Conference on Urban Drainage, 2005 (Copenhagen); 7th International Conference on Urban Drainage Modelling and the 4th International Conference on Water Sensitive Urban Design, 2006 (Melbourne); International Symposium on New Directions in Urban Water Management, 2007 (Paris); Novatech, 2007 (Lyon); 11th International Conference on Urban Drainage, 2008 (Edinburgh);

Identifying and contacting appropriate authors from papers derived from the above conference proceedings; Generating a database of primary and secondary personal contacts to approach for information for particular member states/cities.

The latter three sources have yielded the most relevant and detailed documents and information. These have been collated and ordered by country/city to form the main body of this report (Section B).

1.4. EU City selection process

The scale and age of wastewater collection systems varies across the EU and are summarised alphabetically in Table 1. As can be seen, the UK has one of the oldest systems, which is similar in age to those in Germany. The UK system also serves the joint second highest percentage of population. Table 1. Wastewater collection systems in EU Member States (EWWG, 1995)
% of population % of urban areas served by Age profile of collection served combined systems (estimated) systems (where known) Belgium 58 70 Denmark 94 45-50 50% after 1960 20% after 1980 France 74 70-80 Germany 90 67 74% after 1945 60% after 1963 Greece 45 20 60% after 1970 Ireland 67 60-80 Italy 82 60-70 40% after 1965 Luxembourg 96 80-90 50% after 1965 Netherlands 97 74 50% after 1955 Portugal 62 40-50 70% after 1960 Spain 82 70 UK 96 70 50% after 1945 Country

Additionally, approaches to the issuing of consents and permits to discharge, as well as the design of CSOs, varies considerably within the EU. A review of approaches used across member states is given in Appendix A. In order to identify cities to focus on within this report, various criteria for comparison were selected. This was primarily the city and metropolitan population (person equivalent, p.e.), or agglomeration size, as this approximately determines the type and scale of schemes considered. London is the largest EU city, with a city population of 7,556,900 (Table 2). It was therefore decided to focus on EU cities with a city and/or metropolitan p.e. comparable to London, but of no less than 1 million. Cities outside the EU were not included in the in-depth study, as the UWWTD would not apply. Table 2 summarises the largest European Union cities and their associated water bodies; the agglomerations highlighted in bold are included within this report. The latest European Commission (EC) Report in the authors possession (CEC, 2004), determines that only Austria, Denmark and Germany had fully complied with the directive (the last questionnaire was conducted in 2007 and the next report is due in 2009 (Lenz et al. 2007; CEC, 2008)). In 2005, Spain was being taken to the European Court of Justice by the EC for failing to fulfil its obligations (GHK, 2006). As recently as 2008, France was under a similar threat (IHS, 2008) and in 2009 Italy also faced a written warning from the EC for non-compliance (Web4water, 2009). As such, activities being undertaken within large cities of these member states will be included in this report. Section C of the report summarises the overall main findings and Section B provides greater detail on each of the previously identified large EU cities.

10

Table 2. Large European Union cities (ranked by metropolitan population) Country Russia UK Germany France Spain Germany Spain France Greece Italy Italy Italy Germany Portugal UK Poland Hungary Austria Sweden Netherlands Belgium France Denmark Finland Czech Rep. Greece Croatia** City name Moscow* London Rhine-Ruhr area Paris Madrid Berlin Barcelona Lyon Athens Milan Rome Naples Hamburg Lisbon Manchester Warsaw Budapest Vienna Stockholm Amsterdam Brussels Marseille Copenhagen Helsinki Prague Thessaloniki Zagreb Water bodies Rr Moskva Rr Thames Rrs Ruhr to south, Rhine to west, Lippe to north Rr Seine Rr Manzanares Rr Elbe/Spree/Havel Rr Llobregat/Bess Rhne and Sane Rivers Saronikos Bay Rr Ticino/Adda/Po Rr Tiber/Aniene Tyrrhenian Sea Rr Elbe/Alster/Bille Rr Tagus Irwell/Medlock/Irk/Merse y Rr Vistula/Baltic Sea Rr Danube Rr Danube Lake Malaren, Baltic Sea Rr Amstel (& IJ Bay) Rr Senne Mediterranean/ Canal de Marseille Various waterways Gulf of Finland Rrs Vltava/Elbe Rr Axios Rr Sava City population 10,382,754 8,278,251 2,188,500 3,213,271 3,425,000 1,615,908 472,305 745,514 1,294,305 2,724,347 966,209 1,773,218 499,700 458,100 1,706,624 1,702,297 1,697,937 814,418 752,911 1,080,790 839,043 613,603 578,126 1,233,211 (L) 360,000 786,000 Metropolitan population 13,500,000 12,300,000(A) 11,800,000 (B) 10,000,000 (C) 6,100,000 (D) 4,275,000 (B) 4,250,000 (D) 4,415,000 (C) 3,750,000 (E) 3,550,000 3,500,000 3,075,000 (F) 2,575,000 (B) 2,550,000 (G) 2,475,000 2,375,000 2,300,000 2,000,000 (H) 1,989,422 (I) 1,930,000 1,850,000 1,500,000 (C) 1,390,000 (J) 1,299,541 (K) 1,200,000 (E) 1,108,000 (M)

* = non European Union shown for comparison only; ** = EU member candidate planning to join EU and is trying to implement EU legislation; A = UK National Statistics (2009); B = SBD (2009); C = INSEE (2009); D = INEE (2009); E = NSSG (2009); F = INSI (2009); G = INEP (2009); H = SA (2009); I = SCB (2009); J = DS (2009); K = SF (2009); L = CZSO (2009; M = CBS (2009)

11

2.

City Responses

2.1. Austria
2.1.1. Vienna Viennas sewer network is approximately 2200 km long (53,000 individual pipes), has five main catchment areas draining approximately 260 km2 with an average imperviousness ratio of 50% and serves a population of 1.8 million p.e. (Nowak, 2007). Recently, Viennas RTC activity has become one of the best documented case studies of urban drainage management introduced to comply with water protection regulations. Recent developments within the network include large storage sewers along the river banks of the Danube, Donaukanal, Wein and Liesing (Fuchs and Beeneken, 2005). Additionally, the construction of a detention basin close to the main WWTP of Vienna is planned, with a total volume of approximately 255,000m3 (Teufel, 2007). These are designed to minimise CSO spills to receiving water bodies during storm water episodes, as well as moderating the inflow to the WWTP. At the end of 2006, a 3 km long, 30 m deep wastewater tunnel along the River Wien, the Weintal-Kanal, was completed. The Wiental-Kanal is capable of storing up to 110,000 m3 of waste water. In the event of heavy rainfall the Vienna Sewer Network Control sets the rate at which wastewater is discharged to the main WWTP. Additionally RTC is used to control the distribution and pumping systems, which regulate the discharge of water and prevent pollution of the River Wien with diluted wastewater (Wien International, 2006).

Figure 1. Location and schematic layout of the Wiental-Kanal, Vienna (Wien International, 2006) The Sewer Management System Real Time Control (SeMaSys RTC) will activate the storage capacity of these relief sewers and detention basin (Nowak, 2007). In total 123 million is being invested in these activities (Teufel, 2007). SeMaSys RTC consists of: o Measurement devices to measure flow (55 points), levels (20 points), point and areal rainfall (25 stations and rainfall radar);

12

o Local control devices to regulate flow and levels; o SCADA system to collect and display measured data; o Central control system to facilitate decision making based on measurements and forecasts. The process of data collation and control is illustrated in Figure 2 and Figure 3 (Teufel, 2007). Within SeMaSys RTC the sewer network is reduced to approximately 2200 pipes, in order to increase computational time. The model was calibrated using measured data and uses Hystem-Extran software to generate rainfall-runoff. The control system is rule-based, but evaluated with the aid of fuzzy logic and the input data. Results from simulations (undertaken using ITWH-CONTTROL software) and real situations are stored in a database for upgrading and self-learning (Fuchs and Beeneken, 2005). Simulations for the left bank of the River Danube (known as LDS), using all available rain gauges revealed that for constant areal rainfall the mean reduction in overflow volume achievable would be approximately 43,000 m3 (Fuchs and Beeneken, 2005). Further analysis with loads simulated for an extensive rainfall event show that although nearly the whole system capacity was required for the diversion of flows, a reduction in CSO volume of 2.4% was possible. The LDS phase of the RTC became operational in 2005 and the total simulated reduction in accumulated CSO volumes for one year was estimated at around 30% (Nowak, 2007). The second phase comprises the Donaukanal right bank main collector (known as RHSK-E), which came into operation in 2006. Evaluation results showed the retention volume amounted to slightly over 40% of the previously discharged volumes (Nowak, 2007). The third and final phase, to be realised by 2015, comprises the integration of the Liesing and Wien storage sewers, as well as system optimisation. Currently, however, the system is not fully operational due to some failures in data measurement (Fuchs, 2009c).

Figure 2. Schematic of RTC data flow for the sewer Network of Vienna (Teufel, 2007)

13

Figure 3. Schematic of the Vienna sewer network RTC system (Teufel, 2007)

2.2. Croatia
At the time of writing, Croatia had not ascended to the EU. However it is a candidate country, after applying in 2003 and is expected to ascend sometime in 2010. In response to this, the country is already implementing EU Directives and undertaking related activities in order to comply with relevant legislation. 2.2.1. Zagreb The Zagreb sewer system serves approximately 1 million inhabitants and a large industrial sector. It contains more than 1300 km of public and approximately 1000 km of industrial sewers. No WWTP currently exists, but this is being considered in future plans. The sewer system serves various types of catchment (from natural to urban) across a range of slopes. Creeks from the neighbouring mountain, Medvednica, feed into the sewer system. The network has very large conduits, but almost no CSOs (Dedus and Pavlekovic, 2005). The main recipient is the Sava River, which divides the city and the sewer system into two separate systems it is anticipated that the WWTP will be constructed at the confluence of these, in order to treat flows from both systems. In and around the city centre, a system was built parallel to the Sava (west-east), while the main collectors (north south) transport wastewater towards the GOK (main outlying open channel) and Sava. On the northern bank of the Sava the sewerage system is gravitational, but the southern system must be pumped during high river flows. These systems were built as combined ones and as such, flows are constantly increasing. One of the largest issues today in Zagreb is the number of creeks that lead to the sewerage system, which cause problems for a large trunk sewer that does not have any CSOs. An international consortium, consisting of four local and three international companies, was established to achieve accepted environmental goals. Due to a growing concern about water quality (including groundwater), local funds are being directed to the revision of existing sewerage master plans. City drainage managers

14

realised that any further investment in the sewer system or the WWTP, without first conducting research, would be unwise. 'The Zagreb Sewerage System Optimization Project' (ZSSO), initiated at the end of 1994, has established new standards in urban drainage modelling within Croatia. Using a Scandinavian methodology, based on continuous hydrological modelling, hydrodynamic description of flows in the system and a statistical evaluation of the main parameters representing sewer capacity and flow effects, the project utilises the modelling tools MOUSE and MIKE I (Dedus and Pavlekovic, 2005). At the beginning of 1995 the first phase of ZSSO set out to collect all available data on the present state of the sewerage system and to create tools for the development of a master plan. The final goal of the project is to implement an engineering and decision making tool for real time control (RTC), in order to provide an integrated flood protection system, reconstruction of the GOK and the construction of a WWTP. Basic ZSSO project goals include: o Comprehensive network data collection; o Definition of remedial measures to improve the operation of the system; o Establishment of principles for a master plan for rehabilitation and further sewerage development; o Definition of the standards for the protection receiving waters and the environment; o Input for the future WWTP definition; o Creation of a set of tools to collect data on system performance and for RTC; o Establishment of a Zagreb Sewerage Real Time Control and Management Centre. Achieving these goals relies on the following tools and principles: o Distributed modelling of the surface storm runoff (FRC) and wastewater loads; o Distributed modelling of continuous infiltration of storm and groundwater (SRC); o Full dynamic modelling of the processes in principal collectors; o A comprehensive rain and flow survey in the catchment area and the sewer network for the runoff and network model calibration;

15

o Long-term runoff and pipe flow simulations, based on historical rainfall records, statistical evaluation of the hydraulic performance and environmental impact of the present situation. In later phases of the project, feasible structural or operational alternative solutions would be modelled and evaluated through the comparison with the present situation; o Feasibility assessment of the selected upgrade strategies, aimed at a cost/benefit analysis. Recommendations resulting from the ZSSO project to date include: o Build up the retention volume of collectors; o Achieve active outflow/retention control of the creeks from Medvednica; o Design several overflows to the Sava River; o Exclude the Bliznec and Vuger creeks from the GOK; o Implement a strict maintenance schedule; o Create a SCADA-based RTC system (Dedus and Pavlekovic, 2005).

2.3. Czech Republic

2.3.1. Prague The main sewer system of Prague is combined and covers about 60% of the total city area. It consists of 2,360 km of sewers, 54,000 manholes, 140 CSOs and 19 pumping stations. The sewer system is connected with the central WWTP, which has a design flow rate of Q = 6m3/s. Due to the large amount of inter-connections, the type of storage structures and the regular extension of the system, its behaviour is complex. Problem areas identified include the pollution of receiving waters (Vltava River) by combined sewer overflows and the deteriorating condition of the system itself. Another problem is connected with several local separate sewer systems located on the outskirts of the city (Hrabak et al., 2005). In 1995 the Municipality of Prague initiated a feasibility study to define a new Urban Drainage Master Plan (UDMP). There were several reasons for a new study, but in particular the previous UDMP ended in 1988 and did not take into account impacts on receiving waters (Gustafsson et al., 2000). The fundamental principle of the UDMP is based on an integrated approach to rainfall-runoff transport and treatment processes in the catchment, sewer, treatment plant and the receiving waters. The conceptual phase of the UDMP began in 1999 and included modelling work undertaken to: o Evaluate the functionality of a backbone collector ' K ': a sewer tunnel of 3.8 m in diameter collecting wastewater from principal parts of the city;

16

o Evaluate the main trunk sewers in catchments using MOUSE-HD, MOUSE-TRAP and MOUSE-NAM (Dedus and Pavlekovic, 2005); o Evaluate the long term impact of the main CSOs on the Vltava river water quality; o Evaluate a final strategy for the time horizon 2015-2030: specify a final urban drainage system in a digital form. Results indicated primarily that the Troja Island WWTP would require expansion, but that the existing structure of the sewer network (combined in the centre and separated in the suburbs) could remain unmodified (Gustafsson et al., 2000). However, in August 2002 serious flooding hit Prague, which challenged this conclusion (Hrabak et al., 2005). The Prague sewerage network, as well as the main WWTP, was hit by the floods. Rising water in the Vltava River flooded the WWTP and prevented the outflow of wastewater. The system became highly overloaded due to the high water level in the river, the closing of flood defence caps, the overflowing of outlets from the main sewers beneath the city and direct runoff caused by the storm. Measures to remedy this disaster began immediately in autumn 2002. A great deal of attention was focused not only on protective measures against floods caused by rainfall volumes, but also on measures to prevent the self-flooding of the city, caused by the combination of flood and urban waters forced into the sewer system (Hrabak et al., 2005). With repect to this, there are a number of projects being executed to improve the performance of the sewer system. The following districts have been identified as the priorities: o Karln; o Holeovice; o Old Town, Lesser Town, Smchov and Podol; o Troja and Libe; o Other endangered areas (Zbraslav, Radotn valley). Flood gates have been installed within the sewers (Figure 4) and CSOs are receiving modification in these areas, including slight changes in tank geometry and the installation of pre-treatment devices. As far as could be identified, there is also a large retention tank being constructed in the Karlin area (Pryl, 2009). Table 3 summarises CSOs identified as having implications for sewer flooding and these have been incorporated into the UDMP for future consideration.

Figure 4. Flood gate on the central sewer of Prague (Hrabak et al., 2005)

17

Table 3. CSOs permitting back flow to the Prague sewer network (Hrabak et al., 2005)

Further work has been undertaken to fully understand the impacts of CSO remodelling on receiving water bodies in Prague, but from a water quality perspective. CSO83 on the Botic Stream (a drain in the central area of Prague) was heavily reconstructed between 2001 and 2005 (Table 4), (Kabelkova et al., 2007). Changes to throttling, increases to the critical discharge when an overflow occurs and changes in the catchment, as well as overflow hydraulic characteristics have been monitored and simulated by the Prague Water Supply and Sewerage Company and the UDMP consultants (Hydroprojekt and DHI Hydroinform). Table 4. Modifications to CSO83 on the Botic Stream, Prague (Kabelkova et al., 2007)

In order to assess benefits of the individual CSO reconstruction phases, pollution and hydraulic impacts of overflow were simulated by a computer model (REBEKA) using a 70 year rainfall time series for Prague. REBEKA simulates both emissions from the sewer system (discharge of water, total suspended solids and NH4-N) and water quality standards in the receiving water (frequency of acute water pollution events caused by unionised ammonia NH3). Additionally, the ecological status of the

18

Botic Stream has been monitored (heavy metal concentrations in sediments and macrozoobenthos) since 1998. Simulation results for the reconstruction phases revealed a gradual decrease in the average number of overflows per year, overflow volume and duration and the amount of suspended solids discharged in the individual reconstruction phases. The average number of overflows possibly causing ammonia toxicity was reduced by half (from 9 to 4.9) in phase II and nearly eliminated in phase IV (0.4). However, the overflow volume, amount of suspended solids correlating to heavy metals and hydraulic stress were not significantly reduced until phase V in 2005. In addition to this, research on new types of CSO is being undertaken (Pollert et al., 2008). Within Prague, there is currently much of research being undertaken into redesigning the sewer network and subsequently CSOs. The main drivers are the creation of a new UDMP, reducing urban flooding and increasing receiving water quality. However, a current limiting factor in the development of a final technical model for the drainage system is a gap in legislation in terms of CSO overflows, although this is planned to be addressed in 2010.

2.4. Denmark
2.4.1. Copenhagen The City of Copenhagen covers an area of approximately 90 km and has a population of approximately 501,000. The Copenhagen sewerage system was established in 1857, with most of the present day system having been built between 1860-1910 The sewer network now totals approximately 68 km, of which 90% is covered by a combined sewer system (the rest is covered by a separate system). The total annual amount of treated rain- and wastewater in the two main treatment plants (Lynetten and Damhusen) is approximately 90-100 million m3. In addition, there are 43 overflow structures in action between 0-33 times a year the average discharge being 13.7 times annually. However, this varies between CSOs near the harbour, where interventions have been implemented, and those areas awaiting development (Sorensen et al, 2005). The extension of the Copenhagen sewer system has taken place in stages, each with their specific target: The large sewers leading wastewater and rainwater into the harbour were constructed from 1860 to 1885 in order to improve the sanitary conditions of the city; The sanitary conditions of both the city and the harbour were improved when intercepting sewers were installed along the harbour from 1893 to 1903 to pump most of the wastewater into the Sound; The sewer system of the western part of Copenhagen was established from 1904 to 1920. The wastewater was treated biologically before the outlet at the marine recipient in the south of Copenhagen, and the plan was to merge all outlets in a joint treatment plant Damhusen - (at the outlet of the Damhus

19

Brook to the southern marine recipient) and to lead a joint outlet sewer across the island of Amager to the Sound. These considerations took into account the receiving waters capacity to receive the wastewater; The Lynetten treatment plant in the north of the island of Amager was built and commissioned in 1980, when it was identified that the receiving waters were unable to receive the wastewater. The extension, including installations to remove nutrient salts at the Damhusen and Lynetten treatment plants, was planned and implemented between 1987 and 1997. Sewer system plans now clearly allow for the capacity of receiving waters and the environmental authority defines the requirements; A number of new retention basin facilities were built from 1994 in order to avoid or reduce wastewater discharge to marine and freshwater areas during high intensity rainfall events. The plants were dimensioned according to the specific requirements of individual receiving waters set out by the environmental authority. The Sydhavnen facility (15,000 m3) utilises comprehensive online management of gates and meters, which has improved conditions in the inner harbour. The Utterslev Marsh facility (limited to the northern neighbouring municipalities) was supplemented with a constructed wetland system in 1998 allowing the mixture of wastewater and rainwater to be treated before discharge to the nearby marsh. The East Amager facility (40,000 m3) ensures the requirement for bathing water quality can be met. It is the first time that this has been made a primary requirement. During this period, attention was increasingly directed to the sewer system's impact on ground water and on conditions in lakes and streams; From 2000, the frequency of CSOs to marine receiving water has been defined in terms of bathing water quality. The timeframe for objectives concerning sewer rehabilitation and measures to meet a certain quality with regard to lakes, streams and marine receiving waters is 2009.

The water quality has increased in the harbour, permitting the reopening of several public swimming baths, 50 years after the last one was closed due to the levels of pollution reaching the harbour. As some areas are still to be improved, a SCADAbased alarm system was implemented within the harbour, to warn of CSO incidences likely to cause high E. coli concentrations. The limit was established at less than 1000 E. coli per 100 ml for less than 5% of the bathing season. This was estimated to correspond to 2-3 overflow events during the bathing season, which is of 92 days. Between the 15th of July and end of September 2002, 5 rainfall events caused the closure of the bathing facility for a total of 91 hours, equal to 11 days. Figure 5 illustrates the reduction in CSO volumes released to the harbour since 1995 (Sorensen and Kofod-Andersen, 2005).

20

1000 x m3
1800 1600 1400 1200 1000 800 600 400 200 0 1995

CSO Volume Reduction (Measured and Predicted)

1996

1998

1999

2000

2010

Figure 5. Measured and predicted reductions in CSO volumes released to the Copenhagen harbour areas (Sorensen and Kofod-Andersen, 2005) Every combined sewer overflow structure has at least one level transmitter that indicates when there is (risk of) an overflow event. The transmitters continuously send their information to the central computer in the Sewerage Department, and every 15 minutes data of overflow volumes are transmitted to the Environmental Protection Agency (EPA) and Danish Hydraulic Institute (DHI). If an overflow occurs, an SMS is automatically sent to the guard duties of the EPA, Copenhagen Energy and DHI. The EPA decides on whether a warning should be issued. Once a warning is issued the DHI uses a MIKE-model of the harbour area with inputs from the SCADA system and predictions of direction of the current in the harbour. The model runs with conservative scenarios and on-line data of overflow and current for as long it takes to obtain a lower value than the critical level (500 E-coli per 100 ml for 12 hours). When this level is reached the warning is removed. Data on the actual situation of the bathing facilities can be seen by the public at any time during the bathing season on the internet (http://www.miljoe.kk.dk/badeudsigt), (Sorensen and Kofod-Andersen, 2005). The next phase of intervention is rehabilitation to the sewer system of the new restad district of Copenhagen and involves implementation of a separate sewer system. The water will be divided into three streams: wastewater, rainwater from contaminated surfaces (roads, parking areas, etc.) and rainwater from clean surfaces (roofs, parks, etc.). The wastewater will be led to the treatment plant, while the rainwater from clean surfaces will be used recreationally in the established canals. The contaminated rainwater will be treated in a specially designed rainwater treatment plant, which will utilise dual porosity filtration (a pilot was undertaken in 2005), before being led into the canals. The future townscape of this district will include open drains containing the uncontaminated rainwater from roofs (Sorensen et al, 2005). Additionally, in 2008, an investigation was made as to whether to treat surface run-off and discharge directly into the harbour or convey the surface run-off with the wastewater to a centralised wastewater treatment plant and discharge into the sea, outside Copenhagen (Clauson Kaas et al, 2008). The study included testing roof

21

water and the efficiency of a newly developed mechanical treatment unit for urban run-off. A test area in the city centre was selected for hydraulic modelling using MOUSE and the effect on bathing water quality was investigated using the Copenhagen Harbour MIKE-model. A 0.6 km area of the harbour area which is being completely renewed was selected for conducting the modelling of three scenarios: 1) Interception of all surface runoff with the centralised wastewater system; 2) Interception of poor quality surface runoff to the centralised wastewater system and local discharge of good quality surface runoff; 3) Local discharge of all surface runoff. For each scenario, the investment, operation and maintenance costs were calculated and are given in Table 5 and Table 6. Table 5. Investment costs () for the three scenarios (Clauson Kaas et al, 2008) Size Retention basin (m3) Pumping station (#) Pipeline (m) Local runoff treatment (#) Sum 16,500 14 8,200 0 Scenario 1 (M ) 21.5 0.2 0.5 0.0 22.1 Size 5,300 14 8,200 0 Scenario 2 (M ) 4.0 0.2 0.5 0.0 4.7 Size 0 23 15,600 12 Scenario 3 (M ) 0.0 0.3 1.3 0.1 1.6

Table 6. Operation and maintenance costs () for the three scenarios (Clauson Kaas et al, 2008) Size Retention basin (m3) Pumping station (#) Pipeline (m) Local runoff treatment (#) Sum 16,500 14 8,200 0 Scenario 1 (M ) 0.02 0.05 0.02 0.00 0.10 Size 5,300 14 8,200 0 Scenario 2 (M ) 0.01 0.05 0.02 0.00 0.08 Size 0 23 15,600 12 Scenario 3 (M ) 0.00 0.07 0.04 0.04 0.15

Combining investment, operation and maintenance costs, the interception of poor quality surface runoff alone is, financially, comparable to discharging all surface runoff locally. Environmentally, Table 7 shows the simulated annual discharge of E.Coli in total numbers for the three alternatives. It is estimated that for scenario 2 more than ten times the number of E.Coli would be potentially discharged compared to scenario 3. For scenario 1 the discharge is about ten times higher than for scenario

22

2. In terms of bathing water quality, scenario 3 (local discharge of all runoff) is the best solution. Furthermore, Table 8 shows the simulated discharge of pollutants for the scenarios. Table 7. Simulated annual discharge of E. coli under three different discharge scenarios (Clauson Kaas et al, 2008)
Local discharge of runoff (no. of cfu x 1012) Discharge from CSOs (no. of cfu x 1012) Scenario 1 None 10.63 Scenario 2 0.044 1.33 Scenario 3 0.085 None

Table 8. Simulated annual discharge of pollutants from the WWTP in under three different discharge scenarios (Clauson Kaas et al, 2008)
COD (kg/year) Total nitrogen (kg/year) Total phosphorus (kg/year) Scenario 1 5,724 774 159 Scenario 2 1,404 190 39 Scenario 3 0 0 0

Considering the economic and environmental assessment, it has been decided to implement scenario 2 (local discharge of good quality surface run-off and intercepting of poor quality surface run-off) for the whole harbour area. An overall deciding factor is the treatment efficiency of local surface run-off treatment units, which cannot ensure sufficient removal of ecotoxic compounds (metals) to protect aquatic life. Presently, all new urban developments in the area will have to comply with this system. In future, the issue will have to be approached comprehensively for example by reducing this diffuse pollution from vehicles and improving the removal of ecotoxic compounds. At the same time building permits will only be given if the roof materials do not include ecotoxic compounds such as Zn, Pb, Cr, Cu and PAH. This solution also supports the policy of the management of Copenhagen who want to disconnect as much rainwater as possible from its wastewater system to reduce operational costs on pumping stations and wastewater treatment plant and also to prepare for a change in rainfall pattern as a consequence of climate change (Clauson Kaas et al, 2008).

2.5. Finland
2.5.1. Helsinki Water and sewerage services in Helsinki are provided by Helsinki Water, a commercially operated municipal utility. One of the primary targets of water utility operation in Helsinki is to improve the quality of effluent discharged into the Gulf of Finland, in order to protect the environment and facilitate its use as a bathing area (Helsinki Water, 2009). The Helsinki sewer system consists of 1,800 km of sewers, over 110 wastewater pumping stations, 60 km of wastewater tunnels through bedrock, the Viikinmki WWTP and a 17 km effluent outfall tunnel, as well as the 0.18 km2 composting site at Metspirtti. Figure 6 illustrates the structure of the sewer network of Helsinki, including the use of tunnels and pressure sewers.

23

Helsinki has both combined and separate sewer systems. The combined system covers most of the downtown area with both wastewater and rainwater conveyed to the Viikinmki WWTP. The combined system accounts for approximately 14% of the total length of the sewer network. In the separate sewer system, which covers the rest of the city area, wastewater is conveyed via foul sewers (44%) to the Viikinmki WWTP and rain water is conveyed in storm sewers (42%) directly to the nearest watercourse (Helsinki Water, 2009c). The network consists of a range of different sized sewers and tunnels, the lengths of which are summarised in Table 9.

Figure 6. Sewer network of the Helsinki metropolitan area (Helsinki Water, 2009) Table 9. Statistics of the Helsinki sewer network (Helsinki Water, 2009c) Diameter (mm) 200 225 250 300 350 400 500 600 Length of Sewer (m) 101,248 89,898 76,211 857,707 28,396 191,983 93,680 92,555 % 5.7 5.1 4.3 48.4 1.6 10.8 5.3 5.2 Diameter (mm) 800 1000 1200 1400 Pressure sewers Sewer tunnels Total Length of Sewer (m) 60,626 35,420 22,036 22,001 45,624 55,320 1,772,705 % 3.4 2.0 1.2 1.2 2.6 3.1 100

24

A major contribution to the protection of the Gulf of Finland (and consequently the Baltic Sea) is the Viikinmki WWTP, which has been operational since 1994. Viikinmki treats all the wastewater from Helsinki, as well as runoff from the downtown area, (Helsinki Water, 2009b). Most of the 120 million m3 capacity plant is located underground in rock caverns. Centralised wastewater treatment has completely eliminated previous nuisances including noxious odours and noise, caused by now decommissioned decentralised WWTPs. The ventilation system from the treatment process area leads all emissions to a tall chimney where they are rapidly diluted in the atmosphere, leaving no disagreeable odours at ground level. Wastewater treatment removes 95% of the oxygen-consuming substances, as well as 95% of phosphorus. The current nitrogen removal rate is 80% in accordance with EU directives and the highest national standards. The effluent from the plant flows through an outfall tunnel to a discharge area 8 km off Helsinki. Regular monitoring programs of the sea approved by government authorities and conducted by the City of Helsinki Environment Centre, include water, plankton, and bottom sampling. Sea water quality in the area has improved over the last 20 years, so that a stretch of coastline of more than 40 km, as well as the coastal archipelago, continues to be a recreational area of good standard (Figure 7), (Helsinki Water, 2009). Both the sewer network and the Viikinmki WWTP are controlled using RTC from the main control centre at the Vanhakaupunki raw water treatment plant (Helsinki Water, 2009b). RTC of the sewer network is based on flow and pressure measurements in the network and over 110 wastewater pumping stations. The monitoring system enables disturbances in the network to be quickly rectified and energy consumption to be optimised (Helsinki Water, 2009c).

Figure 7. Water quality in Helsinki and Espoo (Helsinki Water, 2009)

25

2.6. France
2.6.1. Lyon Historically in Lyon, all wastewater and rainwater was conveyed within a combined sewer network. Increasing development led to unchecked expansion of this system, creating a dense sewer network in the centre of Lyon increasingly under pressure from the volumes of rainwater being conveyed. Consequently, approximately twenty years ago, political changes occurred to the way in which rainwater was managed. A separate sewer network was constructed for approximately 90% of the 2700km network, along with 8 retention basins and the initiation of the implementation of source control practises (such as infiltration). Such practises are applied to all new urban developments in the Lyon area. There are 8 WWTPs on the network, along with 380 CSOs. An information gathering and modelling project was initiated in 1990 and further developed in response to the UWWTD and WFD. The project aims to identify and limit flooding in inhabited areas and to reduce CSOs to the Rhone and Saone rivers (Volte et al., 2007). The project consists of 6 stages: Stage 1: Stage 2: Stage 3: Stage 4: Stage 5: Stage 6: Gathering sewer network data within the greater Lyon area (including sewer flooding incidents); Archiving the data in a GIS (CIGNET) and building a sewer network model; Calibrating the model using rainfall data and historic flooding occurrences; Simulating and assessing historic flooding events of 10-30 year return periods using CANOE modelling software for 650 km of network; Prioritising locations requiring intervention (creation of the black point list) to reduce spills to approximately one per month; Using RTC to manage the CSOs (control of approximately 30 CSOs representing 70% of major overflows).

Stages 1 to 5 have been completed and stage 6 (RTC) is due to be completed in 2009 (Volte, 2009). 2.6.2. Marseille Marseille and its suburbs cover 230 km of which 150 km are urbanised: therefore a large proportion is impermeable. The city is in the form of a basin, surrounded by hills and a hydrometric network composed of 50 km of water courses and 500 km of underground rainwater drainage pipes, which convey polluted urban surface runoff. The sewer network is composed of 1000 km of foul sewers and 500 km of combined sewers (Figure 8). 64 CSOs divert water from the combined network should rainfall exceed 5mm/hour. The main receiving water body is the Mediterranean Sea and more

26

particularly, beach areas. Therefore preserving good quality bathing water is of primary importance and the main driver is compliance with the revised 2006 Bathing Water Directive (2006/7/EC). It is noted that the Marseille WWTP discharges its water outside the bathing area and local studies of the currents have shown that the bathing area is not affected. Consequently, the quality of bathing water in Marseille is particularly dependant on the effective functioning of the sewer network, rather than the effectivity of the WWTP (Laplace et al., 2007). For several years the city of Marseille and currently, the Marseille Provence Metropolis urban community, has implemented an ambitious programme in order to protect the bathing areas from pollution from overflows from the foul and rainwater drainage networks (not pollution from the WWTP). The principal sources of pollution were identified about 30 years ago: o Outflow of the small coastal river Huveaune arrived in the middle of the southern bathing area, transporting polluted water; o Coastal development built below drainage networks discharged effluents directly into the sea; o Rainwater pipes collecting water used for cleaning urban surfaces during dry weather discharged polluted water into the sea; o Problems with effluents from pumping stations; o Rainfall entering the main discharge pipe of the combined sewer network bought with it pollution.

Figure 8. The Marseille sewer network (Laplace et al., 2007)

27

During the last 30 years, numerous interventions have been implemented, which are now realising positive results: o The river Huveaune has been re-directed from its natural outflow by constructing a 6 km underground trunk sewer that arrives at the same place as the waste discharged from the treatment plant; o Systematic surveys were carried out on all developments bordering the coast in order to remove all waste water discharges, either directly in the sea or in rainwater networks. A foul pipe was laid along the coast and all connections to it from private properties had to conform to municipal standards; o Surface water was redirected towards the foul network by establishing links between the rainwater and foul systems. These links are equipped with automatic remote controlled gates, enabling the link to be closed in the event of heavy rain thereby avoiding saturating the foul network. 26 installations of this type have been installed upstream of discharges into the sea from rainwater outlets in bathing areas; o The 3 largest discharge pipes of the combined network, initially static, have been replaced by remote controlled gates, allowing water to be maintained in the network until risk of flooding occurs. These will then open when this level is reached; o The overall electromechanical equipment is equipped with an electronic surveillance system that enables rapid detection of malfunctions resulting in improved RTC. Remote control of the same equipment enables the optimisation of storage capacities of polluted water in the sewerage network to avoid all effluent overflows into bathing areas; o The radar images of Mto-France and a network of 24 rain gauges spread throughout the town and connected to the remote control centre enable rainfall events to be anticipated and managed using RTC. Approximately 100 preventive closures occurred in 2004 and 2005 due to this dynamic management process, ranging from half a day to one day each season on the 22 beaches. On average, there were 5 closures per beach during the season, mainly due to rainfall events (Laplace et al., 2007). 2.6.3. Paris Wastewater from Greater Paris (10 million inhabitants) is collected mostly by a combined sewer system and treated in four WWTPs. The largest one is the Seine Aval plant, located 30 km downstream from Paris (2M m3/d). Previously, during rainfall events, runoff that could not transit through the drainage system or be treated in a WWTP would overflow from more than 200 outlets distributed throughout Paris and its suburbs. The largest one, located at Clichy, experiences instantaneous overflows greater than 40 m3/s (on average, five events per year). Most discharges exceed 20 m3/s, but in summer the flow of the Seine can be less than 100 m3/s in Paris (Even et

28

al., 2007). In response to this the capacity has been raised to 2.8m3/day to cover a population increase to 2015, as well as a 10% reserve capacity in case of a blockage (SIAAP, 2007). The first evidence of CSO impacts on the receiving water bodies came to light in the 1960s. However, it was not until the 1990s that reducing CSOs became a concern, as the most visible dry-weather pollution had been reduced by a systematic construction of WWTPs. In the Paris area, during rainfall events the daily flux of organic pollution dominates the fluxes from WWTPs. Until the late 1990s, the 50 km long reach of the Seine inside Paris was permanently affected by high oxygen consumption accounting for 112% of the flux upstream of the city. 20% of this demand resulted from CSOs (Even et al., 2007). In 1995, the Syndicate of Sanitation of Paris (SIAAP) and the Water Agency of SEINE Normandie established the Master Plan for the restructure of the central Parisian zone (Ile de France). A multi-criteria analysis (MCA) was conducted in 1997, which resulted in a management plan time-scaled to 2015 covering WWTPs, storage facilities and other treatment facilities discharging to the Seine. In 2003, SIAAP updated the plan to consider the evolution of technical and regulatory practices in anticipation of the application of the WFD. The master plan for remediation was completed in SIAAP by February 2007 and divided into two phases. The first phase, conducted up to 2004, was to understand the current state of the system. The second phase, from 2004 to 2006, consisted of planning several alternatives based on previous experience and establishing a further master plan for the period 2007-2021. The objective of this plan is to minimise the impact on the natural environment of discharges of urban waste water with l rainfall events of 16 mm or greater. Within the MCA interventions examined are subject to the following criteria: natural environment - weighting of 40%; economic considerations - 35% weighting and stresses - 25% weighting (SIAAP, 2007). The CSOs of the Paris sewer network have been widely studied within the PIREN SEINE program. The composition and fate of the urban effluents have been characterized through sampling, laboratory experiments and modelling studies. The PROSE model (based on the SaintVenant equations solved by a finite difference method, the Preissmann scheme, with bottom friction calculated by a Strickler formulation) was designed to simulate the impact on the river of both permanent dryweather effluents and of highly transient CSOs. It was also used to represent the impact of Paris at large spatial and temporal scales. Using PROSE, interventions to the treatment capacity of the existing WWTP were simulated. The BOD concentration in effluent from the WWTPs inside Paris is currently between 2 and 6 mg/l, but was shown to increase to between 11 and 16 mg/l when doubling the Seine Amont WWTP capacity and introducing construction of the planned La More WWTP (Figure 9), prior to the main CSO at Clichy. The purpose of these extensions, in addition to the building of another plant (Grsillons), is to reduce the permanent discharge at the Seine Aval WWTP (25%). In the future, this and the Grsillons plant will be dedicated to the treatment of runoff providing a 90% reduction in CSOs.

29

Figure 9. WWTPs on the Paris sewer network (new/planned WWTPs shown with dashed lines), (Even et al., 2007) As well as focusing on the capacity of the existing WWTPs, 14 storage devices and 5 WWTP specifically for the retention and treatment of rainwater are being considered. The technical details for the 14 storage devices are: 10 reservoirs: 220K, 140K, 100K, 30K, 7.5K, 65K, 53K, 95K, 25K, 23K (a total of 758.5K m3); 4 Tunnels: 148k, 35K, 50K, 27K cubic meters (a total of 260K).

The storage devices have been sized according to the following criteria: Rainfall of 16 mm for a duration of 4 hours, representing a uniform depth of water across all of the SIAAP catchment areas or with a return period between 9 and 12 months; A policy of limiting runoff to be established in 2015.

Regarding the tunnels, there are many unanswered questions in relation to problems of silting, odour and aerodynamics, due to a lack of Parisian experience with these techniques. For this reason storage reservoirs are currently preferred in comparison to tunnels. However, feasibility studies will be undertaken based on operation of the tunnel Ivry Massena Austerlitz, currently under construction. This will determine whether this type of intervention can be considered more extensively or whether it requires further research and refinement of the design (SIAAP, 2007). The cost of both WWTP and storage interventions is estimated to be 2159 million and 22 million for annual operating costs. In addition there is the cost of works for

30

upgrading treatment ( 1598 million) and the costs of new network (170 million), bringing the total investment to 3927 million. Finally, including the evaluation project costs (12 million for flow management, 51 million for reductions in microbiology), as well as other operating costs (36 million), the overall master plan cost could be in excess of 4000 million (SIAAP, 2007). At the same time, RTC of parts of the Paris sewer system has been under development. RTC requires huge data acquisition and modelling undertakings in order to build a comprehensive understanding of the sewer network and to optimise the function of the networks features (gates, pumping stations) for any rainfall event. In terms of water quality, several scientists have supported the integration of water quality modelling into the RTC objective function. However, realistically, the practical implementation of RTC only minimizes discharged water volumes (Even et al., 2007).

2.7. Germany
2.7.1. Berlin In Berlin it is argued that compliance with the UWWTD (91/271/EEC) will be reached if measures are implemented according to the German engineering standard ATV-A 128. In Berlin, drinking water is almost completely supplied by groundwater located within the municipal area, with wastewater disposal occurring within the same municipal area. This creates an inevitable usage conflict between urban development and pollution control and groundwater protection (Schroeder, 2009). The receiving waters of the Berlin sewer network, the Spree and Havel, are small and sensitive due to their small catchment areas (approximately 13500 km2 in total) and low gradients (0.1% and 0.012%, respectively) generating low runoff rates. The resultant low flow velocities create low self-purification conditions and thus are at risk from pollution. During low flow periods, backwater accumulation can regrade the rivers stage from flowing to stagnant, leading to heavy algae formation in summer. These natural conditions are exacerbated by anthropogenic effects, such as urban agglomerations. The sewer network consists of 12 individual but combined drainage arms, arranged radially to the watercourses and separated from each other by natural catchments. Currently, Berlin is drained by both separate and combined sewer systems the inner city (167 km2) is still combined. The sewer network serves an area of approximately 900 km2 with 3.5 million inhabitants. The total length of collectors is around 9000 km. Three quarters of the area of Berlin is drained via the separate system, whereas in the inner city, a quarter of the total area drains into the combined system. Wastewater is pumped by 149 pump stations and over 1000 km of pressurized pipelines to six WWTPs for mechanical and biological treatment. On average, a total of approximately 635000 m3 of wastewater are delivered and cleaned per day. Upon introduction of the UWWTD, the federal state of Berlin called for a noticeable reduction of discharges (volumes and pollution loads) from the combined sewer system. It was determined that the volumes and discharge rates of CSOs and storm

31

water tanks would be below 25% of the average annual rainfall runoff volume. Additionally, the pollution loads of COD, BOD5 and TSS of the discharge would be below 20% of the average annual load of the rainfall runoff. These requirements are based on the German engineering standard ATV-A 128, but are even more stringent, due to the previously identified sensitive nature of the receiving waters. To achieve these levels, Berliner Wasserbetriebe (the local utility for water and wastewater services) has established a programme of measures to upgrade the combined sewer system. The programme will ensure compliance with the legislation until 2020. It includes the following measures (in order of priority): o Monetary incentives for green roofs and SUDSs on private properties; o Unsealing of impervious surfaces and implementation of decentralised SUDS (where possible); o Full utilization of the static in-pipe storage capacities by heightening CSO crests (taking into consideration flood prevention); o Implementation of actuators (weirs, sluices, throttles) for local RTC; o Construction of additional stormwater tanks; o Increasing flow (pumping) to WWTP above usual wet weather inflow (if the treatment capacity is available). Due to the utilisation of a combined sewerage network and low topographic gradients in Berlin, all outflow from the network and consequently in-pipe storage capacities, are controlled by pumping stations. These stations deliver the wastewater to the treatment plants and act simultaneously as variable throttles in case of rainfall events. When exceeding the maximum pump capacities the combined water is retained within the sewer network until a critical level is reached and a CSO occurs. However, due to the large size of the conduits and the very low sewer gradients, a high storage volume is present. Depending on the size of the sewer, between 1,200 m and 12,500 m of storage space can be utilised. These figures correspond to a minimum specific in-line storage volume of 3 m per 0.01 km of impervious area and a maximum of 65 m per 0.01 km of impervious area. The total in-pipe storage volume accumulates to a dimension of 125,000 m, which corresponds to an average specific volume of 15 m per 0.01 km of impervious area. In addition, 25,750 m of storage is available in the form of storm water tanks and sewers with storage capacity and overflow. Figure 10 gives an overview of available storage capacities before and after implementation of the expansion programme.

32

Figure 10. Storage volumes within the Berlin combined sewer system (Schroeder, 2009) Further activation of in-pipe storage is realised by local actuators within the sewer networks. Since 1996 a movable weir has operated within the main collector of the catchment Berlin IX. The sewer is 3.07 m wide and 2.30 m high. The weir is activated in case of a high intensity rainfall event and retains approximately 3000 m of combined water. A position sensor within the hydraulic cylinder of the weir, as well as upstream and downstream level meters, measures the necessary process data. The control strategy limits the flow towards the pump station to the design flow (twice the dry weather peak flow) by activating the weir. If the critical upstream water level is reached the weir is lowered to maintain this level. When the critical water level is exceeded, the sewer cross-section is completely opened to avoid flooding. At the end of the rainfall event the flow towards the pump station is once again restricted to the design flow until the collectors have been emptied. A further alternative store for combined water has resulted from the modification of sewer overflow conduits. Originally the conduits only functioned to convey discharges from CSOs to receiving waters. Two modification projects have been implemented in Berlin. Firstly, they are operated as sewers with storage capacity and overflow. Due to the wide and flat catchments, these overflow conduits can be very long and therefore provide high storage capacities. In the centre of Berlin (the district of Tiergarten) there is one sewer where the overflow crest was installed as a movable sluice board. This local actuator is used to control the water level within the overflow sewer. Dynamic operation facilitates the activation of an additional storage space of 11,000 m. Secondly, reconstruction of an overflow sewer in the district of Friedrichshain led to an increase of storage by 12,000 m. Here, a pump is used to control the emptying of the overflow sewer and the return of combined water towards the main pump station. These measures strongly contribute to compliance with the emission targets set by the water authority. Further local control objectives are planned and will be implemented during the next stages of the rehabilitation works (to be finished in 2020). Scenarios of global (rather than local) RTC have been evaluated on the basis of numerical modelling and a decision support system was tested to assess the possibilities and

33

benefit of a fully automated global pump station control (Schroeder and PawlowskyReusing, 2005; Rouault, 2008). However, due to operational boundary conditions, the current potential of applying global RTC is rated as low. Currently, in the case of clearly regional storm events, manual global control of the pump stations is applied by the operators in the central control room. 2.7.2. Hamburg The city of Hamburg is located on the banks of the river Elbe and its small tributary the Alster. It covers an area of 750 km2. The metropolitan area has a total population of 2.4 million people. Urban drainage in the city dates back to 1843 and includes centralized sewerage and water supplies. Today the city operates a sewer system of about 5240 km. There are 1300 km of combined sewers, 1850 km of foul sewers and 1600 km of storm sewers. Most of the combined sewer overflow structures are located along the small river Alster and its canalized tributaries, although the main outlet structure of the system is situated on the Elbe (the Hafenstrae pumping station, discharging 11 million m3 in a wet year). The river Elbe is the main waste-load carrying stream in the area and the only water body suitable for taking treatment plant effluents and greater combined sewer overflow discharges. However, most of the pollutant load of the river Elbe originates from third party states or countries, therefore for decision making in Hamburg the Elbe ranks lower, in terms of construction priorities and quality goals, than the Alster (Abraham, 2009). Due to the flat topography in the Elbe estuary, 190 pumping stations are operated for sewage collection. Smaller WWTP have operated in Hamburg since the early 1900s. However, wastewater disposal in terms of sewage treatment was first provided for the downtown combined area at the end of the 1950s. This began with the construction of the Hafenstrae pumping station, located on the river Elbe and the KhlbrandhftNord WWTP, situated on a peninsula on the south bank of the river. In the 1960s, after post-war reconstruction and sewer system repair, the sewerage department embarked on a staged long term sewerage improvement program, known as the drainage Masterplan. A total of 767 million has been invested in treatment plant expansion and interceptor construction. The main objectives of the master plan were to: o Improve the total sewer system performance in terms of: Public health aspects (foul flows); Flood control aspects (storm flows); o Improve the ecological status of the citys water bodies through: Complete interception and treatment of the foul flows; Reduction of the pollutant load discharged during wet weather from the sewer system (combined and storm sewers); Reduction of the effluent load from the waste water treatment plant during wet weather; o Improve the overall security for the dry weather flow transport and treatment process to an optimum;

34

o Adjust the sewer system so that urban flooding is avoided at least on a 5year frequency interval; o Reduce the pollutant load to the receiving streams in order to balance the system at a eutrophic level (thus particular control of phosphorus emissions). These objectives were defined as being achievable by undertaking: o The construction of new interceptors for sewage transport to centralize waste water treatment to the Khlbrandhft-Nord WWTP; o The shutdown of smaller treatment plants discharging into small water bodies with insufficient waste load carrying capacity and the expansion of the Khlbrandhft-Nord WWTP to meet requirements of incoming legislation; o The development of a wet weather pollution control programme, focussing on CSOs, treatment plant efficiency during and after a high intensity rainfall event and storm sewer runoff. Activities undertaken during the master plan evaluation stage included the use of data collection and computer analysis. The actual capacity of the sewer system and the existing water quality situation in the receiving waters were evaluated with the help of mathematical models. A computer model (based on the US EPA SWMM) was used for the flow calculations and sewer system capacity analysis, due to the large and complex structure of the network, as well as the flat gradients, which produced backwater situations even under dry-weather flow conditions. The whole project, including sewer system data collection, took about 6 years to complete. Rainfall records, collected over decades, had to be analysed to derive rainfall-durationfrequency relations. The records were digitized and processed on a computer for design storm evaluation and later continuous flow simulation. A flow measuring programme was established in the sewer system to facilitate model calibration and verification. Two distinct events were identified from this data: the first a first flush event and the second a dilution event. Eventually, the mathematical flow model was installed on the city computer and staff of the planning department trained to use the model for sewer planning work. Within the combined sewer sections, the downtown system was most problematic. Therefore the evaluation of the master plan for this area, comprising about 900 km of sewers, started in the early 1970s. Similar plans were undertaken for the combined systems in Altona, situated west of the downtown area, in the harbour area and in Bergedorf in the south-eastern part of the city. The downtown combined sewer system covers an area of approximately 80 km2 (900 km in length) with 620,000 inhabitants. The area is mostly residential mixed with substantial commercial districts. 45 % of the sewers are more than 77 years old, with only 10 % constructed during the last 50 years. The average per capita water consumption in the downtown area is between 200 and 300 l/d. The dry weather flow from the downtown sewer district equals 180,000 m3/d, including infiltration. The total average dry weather flow of the city is

35

360,000 m3/d. In comparison with other cities this could be considered small and is mainly due to heavy industry and the oil refineries in the harbour area operating independent treatment plants. Wet weather flow in the downtown district is generated from an impervious area of about 30 km2, consisting of house roofs, street surfaces and car parks. For 85 overflow chambers, flow that cannot be handled by the trunk sewers is discharged to the river Alster and its tributaries. The flow measuring programme revealed that approximately 4 million m3 of combined sewage was spilled by the 85 overflow points into the Alster and its tributaries each year. The old trunk sewers of the combined system are laid with grades of 1:3000 or less and lead to the main Hafenstrae pumping station. Dry weather flow and up to 7 m3 during wet weather is pumped over to the Khlbrandhft-Nord treatment plant. To avoid flooding in the downtown area, during wet weather a storm water outlet to the river Elbe, protected by tide gates, is operated in addition to the pumping station. Construction of the interceptor system began in the 1970s and in 1982 major parts became operational, with the foul flows from the northern districts of the city being conveyed directly to the Khlbrandhft-Sd treatment plant. Dry-weather flow levels in the trunk sewers of the combined system dropped, resulting in online storage available for storm flow. Subsequently the Hafenstrae pumping station became more effective in handling combined flows originating from the downtown area. As such, the pollution control benefits of the 15 year interceptor-construction-program became apparent. Based on computer runs for the 1-year and 5- year design storms a flow capacity plan was prepared. The plans include, for each sewer reach in the system: o The actual through flow relative to the flow capacity for the reach (calculated from the cross sectional area of the profile and the sewer slope); o For each node in the system, the maximum water level relative to the sewer diameter when flowing free, or relative to the difference in height between the sewer crown and ground elevation when under a surcharge condition. These plans serve as valuable instruments in all kinds of sewer planning and are regularly updated. The sewer capacity analysis revealed that the following deficiencies existed in the historic system: o o o o o Excessive dry-weather water levels; Low flow velocities; Large amounts of deposition; Insufficient flow capacity to prevent flooding; An unacceptably large overflow volume.

In terms of pollution load, it was found that BOD and suspended solids originate from the sources summarised in Table 10.

36

Table 10. Sources of BOD and suspended solids in the Hamburg sewer network (Abraham, 2009)
BOD (% ) 48 25 15 12 SS (%) 18 9 44 29 Source Residual sewage in the sewers during DWF Foul flow during high intensity rainfall events Surface runoff Sewer deposits

To overcome the above areas of deficit and to meet the master planning objectives, relief flow capacity was installed, 80 flow restrictions eliminated and the overflow volume of a 1-year storm abandoned. In addition, it was determined that the master plan objectives could be achieved by assessing a range of additional options, including: o o o o o Advanced sewer separation; Relief trunk sewer construction; Retention basin construction; Pumping station and force mains instead of trunk sewers; A combination of the above.

Advanced separation collects the less polluted flow from roof surfaces separately and discharges it using: o Storm drains into receiving waters; o On-site detention and infiltration facilities; o Storm water reuse facilities for toilet-flushing or lawn-sprinkling. Advanced separation is effective where the water quality goal cannot be met by treatment and storage of the combined sewage flows alone. In Hamburg this is being utilised for some of the canalized tributaries of the Alster. Relief trunk sewer construction can prove advantageous when: o The WWTP can handle additional flows; o Selective overflowing is utilised to convey peak flows away from sensitive receiving waters to water bodies with a larger carrying capacity. In consideration of the trunk sewer networks existing features, such as poor slope, a large number of flow obstacles and existing channels and subways, as well as the ease construction, deep tunnel relief sewers were also chosen in Hamburg. Retention basins serving as storage and primary treatment devices can provide: o An adequate solution for pollution control. However, an intervention solely based on the construction of retention basin facilities was rejected due to the poor slope within Hamburgs sewer network. In large flat sewer systems, retention basins can only alleviate the effects of poor system behaviour; they will have a limited impact on the improvement of the hydraulic

37

performance of the system. Therefore, it was identified that a combination of additional retention and discharge capacity would best fit the needs of the Hamburg sewer system. A solution was developed based mainly on gravity flow in a new system of transport sewers, laid about 10 m below the existing trunks and interconnected using drop-shafts. The transport sewers serve as conveyance and storage elements. In addition to the storage volume of the transport sewers retention basins were also constructed in the vicinity of the drop-shafts (so that they could also be used for dewatering the basins). The resultant highly interconnected combined sewer area is subdivided into smaller catchments with an approximately 60 minute flow time between drop shafts. A comparison based on dry-weather computer simulations between the old and the new system determined that the total volume of sewage resulting from the dry weather flow in the sewers would be reduced from 110,000 m3 to 40,000 m3, thus reducing the total pollution-causing potential of the system. Contributions to overflow volume reduction from facilities on the entire network are broken down in Table 11. A detailed computer simulation revealed minor overflows in the region of 15,000 m3, due to an uneven distribution of flow and storage capacities over the whole system. It is anticipated that system optimisation will eliminate these overflows. By introducing RTC it may be possible to collect and transport the bulk pollution load to the WWTP. Analysis of BOD/COD data from 1981 (prior to the start of operation of the first interceptors) and 1988 identified that without the interceptors in operation the wetweather input load is generally less than the dry-weather load, whereas with the interceptors in effect the opposite is true. Not only the waste load transported in the sewer system improves, but also the overflow volume is drastically reduced. In 1988 84 % of the annual rainfall-runoff of the downtown combined area was intercepted and conveyed for complete treatment. Table 11. Overflow volume reductions achieved by facilities of the sewer network in Hamburg (Abraham, 2009)
Facility Hafenstrae pumping capacity Khlbrandhft-Sd (WWTP) pumping capacity Safe in-system storage Retention basin storage Interceptor sewer storage Transport sewer storage Retention sewer storage Roweg retention basin Overflow volume reduction (m3) 7 10 150,000 125,000 80,000 100,000 60,000 40,000

The first retention basin of the master plan was the retention facility Schdlerstrae, situated in the watershed of the Wandse/Eilbek channel, a tributary of the Alster. The direct catchment area of the basin comprises two sub-areas, one of which was assigned for sewer system separation in the 1960s, which was approximately 50% complete in 2002. Analysis of overflow data revealed that the Schdlerstrae structure ranked first in terms of overflow volume and second in overflow frequency among all overflow points in the system.

38

Standard procedures for retention basin design in Germany call for area specific rectangular storage volumes of 20-30 m3 per 10,000m2 of impervious area. To ensure a sufficient settling and primary treatment rate the basins surface hydraulic load should be limited to a value not greater than 10 m/h and the horizontal through-flow velocity of the basin should not exceed 0.05 m/s. Additionally, to achieve reasonable sedimentation rates, the residence time in the basin should exceed at least 20 minutes. Based on these design criteria Schdlerstrae would have had a volume of 4000 m3, which would have violated water quality objectives (overflow frequency in the range of one event per year). Consequently, simulations were undertaken for the 1-year design storm and determined an overflow volume of 6250 m3. Analysis of monitored overflow data between 1966 and 1977 yielded 18000 m3. In consideration of the fact that the total impervious area would be reduced by applying advanced sewer system separation, a continuous flow simulation, covering a rainfall period of 10 years, was conducted using an estimated retention basin volume of 7400 m3. Simulations (incorporating separation) identified that a storage volume of 70,000 m3 was a good estimate. The retention basin site is located in a commercial area which was already being utilised as a parking facility. Therefore the basin was designed to be a deep circular shaft with an inner cylinder housing dewatering pumps. Operation of the basin is as follows. When the capacity of the trunk sewer draining the basins catchment area to the interceptor (Kuhmhle) is fully utilised, the retention basin begins filling. The flow enters the basin tangentially on the bottom of the circular tank from a flow dividing structure on the trunk via a vortex drop-shaft. The basin fills with a swiftly circulating flow, which concentrates sediments on the bottom of the tank as a result of the because of the so-called tea-cup effect. When the tank is full, it overflows to a former small natural ditch (the Gehlzgraben), which is now utilised as a storm sewer to the river Wandse. The basins input and overflow structures are designed to handle a through-flow of 30 l/s per 10,000m2 of impervious area, giving a flow of about 4 m3/s; for this flow the retention basin provides primary treatment. During very high intensity rainfall events the surplus flow conveyed in the trunk sewer will be spilled directly via a 16 m long trunk sewer overflow to the Gehlzgraben. At the end of the event, when sufficient capacity is again available in the trunk sewer, pumps start the dewatering process. Three pumps of 0.3 m3/s are provided and it takes about 7 hours to dewater the tank. During the end-phase of dewatering one or both of the stand-by pumps are activated for recirculating the residual tank volume and for scouring the bottom sediments. After dewatering is complete an additional cleansing procedure can be activated using groundwater from the wells installed during basin construction or using surface water from the Gehlzgraben. Odour control for the retention basin is dealt with depending on a range of conditions and factors. A ventilation system with a capacity of about 30,000m3/h has been implemented, along with an odour monitoring program, which uses hydrogen sulphide as the primary odour indicating parameter. Table 12 summarises the odour management regime.

39

Table 12. Odour management regime of the Schdlerstrae retention basin, Hamburg (Abraham, 2009)
Basin Status Empty Filling Filling Partially full/Full Dewatering Situation No odour expected due to high cleansing efficiency and natural ventilation Expected odour dependent on scouring potential of the rainfall event Complaints received Slight emission expected Strong emission from drawn sewage in basin and at discharge to trunk sewer Action None None anticipated wash-off effect of rainfall on odorous substances Forced ventilation to maintain temperature-humidity relationship for odour dilution Forced ventilation (including aeration with circulation) Forced ventilation and deodorisation with activated carbon filtration (optional)

In terms of cost, a cubic metre of combined sewage stored in the Schdlerstrae retention basin (rather than being spilled to the Gehlzgraben) costs approximately 7.7. The standard sewer charge in Hamburg is approximately 1.6. Considering additional storage to achieve a 1-year quality goal for the Isebek Channel could increase the storage cost to 28.6/m3 (Abraham, 2009). 2.7.3. North Rhine-Westphalia/Rhine-Ruhr (Emscher) The Rhine-Ruhr metropolitan area (RRMA) is the largest metropolitan area of Germany with about 11,800,000 inhabitants. It is the only megacity in Germany. It covers an area of 10,820 m and lies entirely within the federal state of North RhineWestphalia (Wikipedia, 2009a). The RRMA is bordered to the west by the River Rhine, to the north by the River Lippe and to the south by the River Ruhr (Figure 11). Within the RRMA is a fourth river, the River Emscher, a tributary of the Rhine. The Emscher flows through Dortmund, Castrop-Rauxel, Herne, Gelsenkirchen, Essen, Bottrop, Recklinghausen, Oberhausen and Dinslaken. The Emschers location, in the geographical centre of a vast industrial area of 5 million inhabitants, has meant that the river was used as an open waste water canal since the end of the 19th century until recently. As heavy industry and mining have shifted further north, land subsidence has almost ceased and therefore the installation of underground sewers has become feasible and possible. In 1990 the decision was made to reconstruct the Emscher system with restored watercourses - the Emscher project. An entire river with a total length of 85 km within a huge metropolitan region will be rehabilitated (Frehmann et al., 2008). Currently, approximately 99% of the population is connected to the wastewater collection and treatment system. This has three modernised wastewater treatment plants, which achieve tertiary treatment. Furthermore, the entire Emscher is currently treated at the WWTP furthest west (Dinslaken), before entering the Rhine (Figure 12). However, this will change in the future, as the WWTPs will be operated as regional WWTPs. The domestic urban wastewater discharges of a total of 2.3 million inhabitants plus industrial wastewaters will be fed directly to these three WWTPs. However, after reconstruction of the Emscher system, the maximum inflow during

40

heavy rainfall events will be 16.5 m/s instead of the 30 m/s at present (Frehmann et al., 2008).

Figure 11. The Rhine-Ruhr Metropolitan Area, Germany (Wikipedia, 2009b)

Figure 12. Location of treatment works and restructuring of the Emscher sewer system (Frehmann et al., 2008) In 1990 the Emscher project was initiated, designed to be implemented as a decentralised sewerage system, built as a new separated underground trunk sewer system (parallel to the Emscher). Rainwater is disconnected from the sewer and infiltrated to enrich the groundwater and to re-establish the water cycle. Additionally, the restoration of watercourses is being undertaken and several industrial enterprises are constructing their own pre-treatment plants. Additionally, in October, 2005 the Emscher Association (Emschergenossenschaft), the local authorities in the Emscher region and the Environmental Ministry of the Federal Land of North RhineWestphalia signed the Agreement for the Future Management of Stormwater. The target of this agreement is that within the next 15 years 15% of the stormwater generated in the Emscher catchment will be decoupled from the sewer system, with a

41

new sewer network ready by 2014/15 (Geretshauser and Wessels, 2007). This will substantially reduce the number and frequency of CSOs required. Future plans for the area also include: o Improvement of rainwater treatment (for example, infiltration of rainwater at source, where possible); o Reduction of municipal, industrial, trade, mining and polluted rainwater discharges. (Herbke et al., 2006) In terms of charging, a membership fee is paid, which is dependent on the quantity and quality of sewerage and rainwater being discharged. For example direct dischargers pay directly to the Emscher Association, whereas households are represented through municipalities (democratic legitimisation). As such the charging system already incorporates the polluter-pays principle and cost recovery in the Emscher River Basin ranges from 96.9% to 107.2% (Herbke, 2006). It is estimated that the overall wastewater management restructuring of the 865 km Emscher catchment area will cost around 4.5 billion (Frehmann et al., 2008). Subsidies of 4.5 million were provided for the Emscher region for the implementation of rainwater harvesting and infiltration projects up to 1999 known as the Rainwater Route (Figure 13). The subsidies amounted to 5/m of impervious area disconnected from the drainage system. Since 1994, 18 towns have participated with a total of 82 different projects and 47 projects have been or are being implemented (Figure 14). One such project is covered in Gruning and Hoppe (2007), but is not reviewed in detail here. These pilot disconnection programmes are centralised into a GIS known as the Stormwater Management Information System, SMIS, based on the Open Geospatial Consortium (OGC) standard. SMIS allows local authorities to easily identify feasible disconnection measures to implement in other areas, as well as calculating the percentage of an area with potential for disconnection (Geretshauser and Wessels, 2007).

Figure 13. The 'Rainwater Route' of the Emscher Region (Becker and Raasch, 2001)

42

Figure 14. Rainwater and infiltration projects in the Emscher Region (Becker and Raasch, 2001) The rates of disconnection achieved within the various schemes exhibit extreme variability. In densely populated areas, in the case of multi-storey apartments for example, the rate of disconnection is only 2 to 3 %. For many schemes involving commercial projects, site pollution still exists and therefore resulting additional costs make disconnection financially infeasible. Higher levels of disconnection can be achieved, however, in the case of larger individual commercial sites and in the case of land owned by housing associations. Here, disconnection rates of 50 % or more can be achieved. In addition, the provisions implemented are concentrated on a significantly smaller area than in residential areas involving scattered individual schemes. A disconnection potential of 10% 21 km2 on average of the impervious areas in the Emscher region appears to be realistic in the long term when the complete range of schemes is implemented. With this disconnection potential, the flood peak flow in the tributaries of the Emscher could be reduced by as much as 40 %, in the case of minor floods with a two-year return period. The resulting significant reduction in the flow-erosion of the bed of the watercourse is of immense ecological importance for the tributaries of the Emscher. As well as these disconnection activities, the main component of the restoration scheme is the large underground trunk sewer, which has been under construction since 2001. The sewer will have a total length of 51 km and runs alongside the Emscher from Dortmund to Dinslaken. The wastewater which had previously been discharged to the Emscher will be collected and subsequently fed to the WWTPs in Bottrop and Dinslaken (also known as KLEM), (Frehmann et al., 2008). Consideration of sediment-free flow and pumping costs has resulted in an average installation slope of 1.5 and depth of up to 40 m (Figure 15). In order to transport flows of up to 16.5 m3/s, diameters of the (mainly) double-pipe sewer vary from 1.6 m to 2.8 m. Due to the increasing depth of the sewer over its length, there are two intermediate pumping stations (pumping station Gelsenkirchen and pumping station Bottrop I + II). Finally, at the KLEM WWTP, wastewater will have to be pumped up to 40 m in a third planned pumping station. Pumping station Gelsenkirchen is

43

important operationally as different pumps allow the wastewater flows to be diverted to different sewers: the incoming wastewater flow from Dortmund to Gelsenkirchen will be distributed to the existing Bottrop sewer and/or the new Emscher sewer, which conveys flows to the Bottrop and KLEM wastewater treatment plants, respectively. This results in a number of options for flow control. Operation of such a critical infrastructure, serving several millions of people poses particular challenges. As such, Frehmann et al. (2008) undertook an integrated simulation of the entire system. Treatment plants were implemented using the SIMBA simulation for wastewater systems, rainfall-runoff modelling was provided by MOMENT (previously established by the Emscher Association) and the trunk sewer was modelled using the US EPA SWMM5. The integration of these models permitted a range of automated operational control scenarios to be investigated across the entire Emscher wastewater system. An example of this is illustrated in Figure 16, where the impact of WWTP management scenarios on COD have been simulated.

Figure 15. Longitudinal section of the Emscher trunk sewer (Frehmann et al., 2008)

44

Figure 16. COD for 4 WWTP control scenarios on the Emscher sewer system (Frehmann et al., 2008) Further work will be undertaken to include a multitude of rainfall scenarios (for example, the non-uniform distribution of rainfall of different characteristics), as well as scenarios covering the future development of the population in this highly populated area. The value of increasing base flows for CSO abatement using hydropower gate operation via RTC has also been researched for the Westphalia area (Achleitner and Rauch, 2007).

2.8. Greece
Most cities in Greece have separate sewer systems, so CSOs are rare. A common problem is the illegal connection of stormwater sewers to foul sewers. However, both Athens and Thessaloniki have combined systems in the old city centres. There are no plans to change this, as any reversal would be economically and technically unrealistic (any excavation triggers interest or opposition from archaeologists that would severely delay implementation). However, certain interventions are proposed for each city. These aim to separate the initial runoff of each rainfall event and drive it to the existing WWTP. This is being undertaken in order to reduce pollution loads ending in the sea; the final recipient in both cases (Bensasson, 2009). These interventions are discussed below for each city. 2.8.1. Athens The sewerage system of the basin of Athens covers an area of 270 km2 and serves a population of approximately 4,500,000. The network consists mainly of separate sewerage and stormwater systems, plus a WWTP on the island of Psyttalia. In the centre of Athens there is a combined sewerage system of an area of 13 km2. The combined sewer and stormwater flows for Q/Qfull 72%, discharge to the Main Sewerage Collector (Kyklovoros or KAA). This discharges to the existing WWTP on the island of Psyttalia, where the wastewater of the whole basin of Athens is

45

treated. For flows Q/Qfull > 72%, the overflows discharge through two different routes: i. Through existing overflow discharge structures (, , and ) to the Kifissos river (at present this is covered for several kilometres upstream the river mouth, with a major road running on top of it); Through the existing overflow discharge structures (1, 2 and H) to the Profitis Daniil stream which discharges to the Kifissos River.

ii.

However, one overflow discharge structure (H) and the connection pipeline to the Profitis Daniil stream do not operate adequately due to a lack of capacity. Moreover, this overflow discharge structure serves the existing combined sewer system of the central areas of Athens (Syntagma square, Akropolis etc). Therefore, the Ministry of Environment and Public Works is undertaking (with ENM) a project to design a diversion of the Kyklovoros. This will lead the overflows directly to the Kifissos River (instead of through the Profitis Daniil stream). The diversion of a Q = 170 - 270 m3/sec from the Kyklovoros is considered very significant for flood control and pollution alleviation in the centre of Athens (Bensasson, 2009). 2.8.2. Thessaloniki The sewerage system of Thessaloniki covers an area of 77.5 km2 and serves a population approximately 1,200,000. The network comprises separate sewerage and stormwater systems, as well as some combined sections (in the centre of the town covering 24 km2) and a WWTP at Kalohori. The combined sewerage system discharges to the Main Sewerage Collector (Kyklovoros or KAA), which is 16 km in length and was constructed in the 1980s. The Kyklovoros discharges to the Wastewater Treatment Plant. During intense rainfalls, the overflows discharge to the sea (Gulf of Thessaloniki / Thermaikos). There is one single overflow spillway (2 m in length) on the Kyklovoros and so overflows discharge through an existing submarine pipeline to the sea (next to the touristic area of the White Tower). When extreme overflows occur, a second pipeline is used which discharges to the sea at another location (New Beach). The sewerage system of Thessaloniki, however, cannot cope with very intense rainfall events. In addition, due to the landscape/basins of the area, stormwater overflows have to be discharged to the Kyklovoros. Therefore, the design of a second main collector was undertaken. The second collector is 4.5 km long (DN 2800 mm) and connects to the Kyklovoros (at manhole no 50), with the two running in parallel up to the WWTP. The second collector operates: a) during periods of medium-intense rainfalls (in parallel with Kyklovoros); b) when there are maintenance works in the Kyklovoros; and c) for anticipated future capacity requirements. The objectives of the construction of the second sewerage collector are: o Preventing overflows during mild rainfall events; o Reducing overflows in the cases of medium - intense rainfall events; o Accommodating the combined sewer system overflows;

46

o Alleviating pollution/discharges to the sea. o Facilitating maintenance works on the Kyklovoros or the second collector. In practice, the second collector also serves as a tank for the interception of stormwater flows and the attenuation of peak flows to the waste water treatment plant (Bensasson, 2009).

2.9. Italy
2.9.1. Naples The UWWTD was transposed into Italian law through the National Directive 152/1999, which was published on May 11th, 1999. However, its complete fulfilment is still far from being achieved, particularly in Southern Italy. However, there are a number of features of the Neapolitan sewer network designed to deal with CSO spills. Most of the main sewer system is comprised of tunnels, due to the particular urban context (steep slopes, dense urbanization and a large underground infrastructure). The principles of sustainable drainage are not heavily utilised by the Neapolitan City Council. Within the tunnels, drop structures are quite frequent with drop heights as large as 80 m. For this reason most of the structures are vortex drop shafts. Tunnel ventilation was not carefully addressed during design. Instead, the installation of airtight inlets to prevent the odour from escaping from the sewer during dry weather condition is common. Additionally, in relation to odour, only the most recent installations have addressed this issue. Odour control was generally realised by inducing slightly reduced pressure within the WWTP. Odour neutralization is essentially achieved through a two-stage chemical process (oxidation + neutralization of residual) to deal with large concentrations of organic gas (i.e. HS or H2S). When applied, the odour control operates during the entire working cycle. One of the latest installations on the sewer network in Naples is the Impianto di Coroglio. Its main features include preliminary wastewater treatment, such as screens, sand traps, roto-sieves (illustrated in Figure 17 and Figure 18), which then pump the effluent to the main WWTP via a 12 km long tunnel. The design discharge is 22,000 m3/per hour (approximately 500,000 p.e.) and the plant is equipped with noise and odour control. Within the tunnels the maximum head under pressurized flow is 0.5 bar and pneumatic wave conditions under rapid filling during extreme storms are not considered at all. Furthermore, hydraulic design is conducted in accordance with the National Directive, which dictates that a design return period should not be lower than 5 years (however, in practise this is normally assumed to be 10 years). CSO structures (such as side weirs, bottom openings and baffled weirs) are generally designed so that the sewer flow is directed to the WWTP for up to five times the average dry weather flow. Discharges exceeding this may then be conveyed directly towards the receiving water bodies.

47

Figure 17. Sand removal at Impianto di Coroglio, Naples (Gisonni, 2009)

Figure 18. Roto sieving at the Impianto di Coroglio, Naples (Gisonni, 2009)

RTC is not currently being applied extensively. A few pumping stations are equipped with automatic systems aimed at managing emergency conditions. Additionally, debris screens and sediment traps are often built, but they suffer from a systematic lack of maintenance. Furthermore, no flushing systems are in operation due to: (i) steep slopes that guarantee adequate self-cleaning velocities; and (ii) water shortages, meaning there is little spare capacity to flush the sewer channels. Flushing gates were not installed within the system (Gisonni, 2009).

2.10. Netherlands
2.10.1. Rotterdam The main drivers for CSO reduction in the Netherlands are national legislation for sewer systems, termed basis inspanning (basic level of reduction) and Waterplan 2 (plans from the Municipality and the Waterboards). These focus on working with water for an attractive city. The main technique used is RTC, to reach the basic level of reduction of CSO spillages, costing approximately 1 million. However, in the centre of Rotterdam a 10,000m3 storage tank has been constructed, at a cost of 10.5 million. Smaller tanks are also being planned but are not yet under construction. Additionally, small scale SUDS techniques are used. Odour problems are not addressed under the schemes, but if future odour problems did occur these would be dealt with by additional measures. Implementation of the above schemes has resulted in a certain degree of good surface water quality and less flooding (Goedbloed, 2009).

2.11. Portugal
2.11.1. Lisbon In Lisbon the management of the sewerage network falls under the jurisdiction of SIMTEJO, a company representing state-owned company Aguas de Portugal and several municipalities. SIMTEJO is tasked with the collection, treatment and disposal of wastewater and reducing pollution to the Trancao river basin and Tagus River (and estuary). Within Lisbon, there are 339 km of sewers, 91 pumping stations and 34 WWTPs, with an estimated treated flow of 233,782 m3/day (Jan-Oct 2008), (Figure 19). Problems associated with combined sewers, such as debris and pollution carried by inflows

48

during first flush effect, have led to the development and implementation of specific solutions. Storage structures have frequently been built into the network (off-line or in-line storage) and flow control devices have been adopted for drainage system inflows (Figure 20). These consider the acceptable average dilution for discharges in water resources receiving treated effluents or the maximum number of allowed discharges per year.

Figure 19. The sewer network of Lisbon Drainage sub-systems in Lisbon (Alcntara, Beirolas and Chelas), particularly the Alcntara sub-system, are large combined sewers, making the transformation of existing combined sewers into separate sewer systems a virtually impossible task. Thus, the Alcntara sub-system constitutes a representative example of the large combined sewer paradigm, with inflows to the WWTP from the Alcntara sewer channel and remaining inflows arriving at waterfront station of Largo Chafariz de Dentro Algs, through pumping carried out at pumping stations. At present the existing systems are inadequate and inefficient.

Figure 20. Vortex flow control valve in the Lisbon sewer network (Almeida, 2009)

Figure 21. Construction of the Alcantara WWTP in Lisbon (Almeida, 2009)

49

As a result, an estimated 160 million is to be invested in the upgrading and extension of sewer networks and WWTPs up to 2012. Interesting challenges arise from the types of construction, cross-sectional range, tidal influence, number of CSOs and the areas topographic profile. The main investment activities include remodelling the Alcantara WWTP (Figure 21) and the construction of the Largo Chafariz de DentroTerreiro do Paco-Cais do Sodre-Alcantara interceptor sewer system and six associated pumping stations, using trench-less techniques (tunnel boring). In addition to this a WWTP water reuse project is planned (similar to that already in operation in the municipality of Mafra within Lisbon, where gardens are watered using treated effluents from the Mafra WWTP). A final project covered by the investment will be the implementation of RTC to facilitate performance modelling, receiving water impact assessment and maintenance cost optimisation (Almeida, 2009).

2.12. Spain
2.12.1. Barcelona The drainage system within Barcelona experiences a rainfall regime with high intensity events. In addition the catchment contains both high mountains and flat coastal areas with a high population density and a high percentage of impervious land. This combination has resulted in historic floods and fluvial and coastal pollution during rainfall events. The Great Olympic sewerage works was completed in 1992, but the sewer network was still insufficient, inflexible and under conventional management. At this time the municipality became aware of the need for a new approach towards the sewer system and its management. This resulted in the council creating a new company, CLABSA, which is a public-private partnership tasked with transforming the drainage system. CLABSA is focused on the planning, control and exploitation of technology to be more effective against flooding and pollution (Salamero et al., 2002). The main drivers for improvement have been fulfilment of EU directives, increased demand on the system, increased environmental awareness and coordination problems with wastewater treatment plants (WWTP). With this in mind a new Advanced Management of Urban Drainage (GADU) approach has been utilised. The main principles of this integrated approach are: o o o o Having a detailed knowledge of the system; Using integrated planning; Utilising RTC; Taking an environmental and sustainable approach.

These principles have been used to develop a Master Drainage Plan and decision support systems (DSS), such as the territorial information system (SITE), a modelling system (SIMO) and a RTC system (SITCO), with an emphasis on data quality and reliability to facilitate full sewerage management. As such, the sewer network now comprises: 1,650 km of sewers, 41,000 manholes, 60,000 inlets, 69,000 connections, 500 control instruments, 405 km of fibre optics, a 150km vacuum waste collection network, 146 flow control points, 24 rain gauges, 9 CSO control points and 2 water quality control points (some of these are illustrated in Figure 22). Actuators allow

50

automatic local/global control of 10 tanks (Figure 23), 19 pumping stations and 36 gates (Figure 24). The system is strictly maintained and uses appropriate control algorithms and the extensive SITCO database within SIMO to simulate levels and flows in the network, as well as CSO spills and their effect on receiving waters. The Master Drainage Plan also identified a need for a 70% reduction in CSOs. Therefore between 1997 and 2005 a range of interventions were implemented, including: o 10 x 500,000 m3 tanks (Figure 23); o 1 storage gate and 5 diverting gates (Figure 24); o 25 km of sewers with large dimensions. In 2002, a pilot RTC scheme was conducted in the Bac the Roda catchment to develop a methodology for Barcelona-wide implementation (Barro et al., 2002), which was then fully implemented. During a rainfall event the infrastructure is managed using RTC across a range of emergency levels, which are summarised in Figure 25. RTC of the detention tanks has permitted the regulation of 2,700,000 m3 of discharges per year (including industrial discharges), preventing 470 tonne of suspended solids being spilled. This has led to an improvement in the quality of the receiving water evidenced by a reduction in faecal coliform numbers (Figure 26) and a decrease in anoxia zones in the harbour bottom. The infrastructure interventions and control techniques undertaken have been heavily publicised within Spain, both at national conferences and in local media. Education and awareness raising activities were undertaken, including tours of the control rooms and underground tours of the detention tanks (Escaler Puigoriol, 2009).

Figure 22. Real Time Control of the sewer network of Barcelona (Escaler Puigoriol, 2009)

51

Figure 23. Detention tank in the sewer network of Barcelona (Escaler Puigoriol, 2009)

Figure 24. Gates in the sewer network of Barcelona (Escaler Puigoriol, 2009)

Figure 25. The Real Time Control Process in Barcelona (Escaler Puigoriol, 2009)

Figure 26. Faecal coliform numbers at various sampling locations along Sant Sebastia beach, Barcelona (CFU/100ml), (Escaler Puigoriol, 2009)

52

2.12.2. Madrid With respect to the city of Madrid, wastewater volumes across the urban, tourism and industrial sectors are anticipated to increase by a total of 59% between 2001 and 2015 (Table 13). Accordingly a total investment of 600 million has been allocated to Madrid between 2005 and 2010 (Table 14). However, no documents on the schemes being planned or implemented to utilise this investment have so far been identified. Table 13. Wastewater volumes by sector in Madrid (GHK, 2006)
Sector Urban and tourism Industrial m3/year 2001 254,956,983 69,932,947 m3/year 2015 315,490,466 116,273,171 % increase 19 40

Table 14. Investment in wastewater treatment in Madrid (GHK, 2006)

Plan Clean-up/reuse water 2005-2010 Total investment (M) 600 Average investment /yr (M) 100 Total investment per Inhabitant per year () 17

2.13. Sweden
2.13.1. Stockholm In Stockholm, wastewater is transmitted to several WWTPs by gravity through large underground wastewater tunnels, excavated from granite bedrock. The principal receiving water is Lake Malaren and the Baltic Sea, although there are many others. In addition to the tunnels the sewer network consists of 192 pumping stations, 3031 km of standard sewers and 3 WWTPs, illustrated in Figure 27 (Stenroos and Katko, 2006).

Figure 27. Location of WWTPs in Stockholm (Stenroos and Katko, 2006)

53

The tunnel network is approximately 50 km in length and most parts were constructed sometime ago. The tunnels operate as a combined system, conveying both domestic and industrial wastewaters as well as stormwater from catchment surfaces. The average wastewater flow in the tunnels can be in the region of 110,000 m3/day (Jerome, 2006). The Underverket (wonder) project was constructed in 19961998 and consists of a 3.25 km long tunnel at an average depth of 15-20 m, with a 15 m2 cross-sectional area and a maximum capacity of 8 m3/s. It includes a 25 000 m3 storage tank. Underverket is designed to minimise sewers leaking into groundwater. The Regnbagen (rainbow) project, constructed in 19931995, has a 21,000 m3 storage tank, which is 1895 m in length and 10.5 m3 in cross-sectional area (3 m wide and 3.5 m high). The Ormen (snake) project constructed in 19911993 is a combined stormwater-foul sewer tunnel having an overall length of 3.2 km with an excavation diameter of 3.5, at a depth of 40 m below sea level (Figure 28). At times of flooding surplus water is temporarily stored until the load on the sewers and nearby WWTP is reduced. Ormen includes a storage tank of 32,000 m3. It was excavated at a depth of 1040 m using tunnel boring machines (TBMs). The ground, composed of granite and gneiss, was excavated at an average speed of 300 m/month. The Saltsjo (a Stockholm bay area) project, constructed in 19861989, is a sewer tunnel 7.5 km in length and 3.5 m in diameter, at a depth of 70 m below sea level. The tunnel is used to transfer effluent from WWTPs under Stockholm to the Baltic Sea. En route the heat from the treated effluent is recovered at a nearby heating plant. This has gradually improved the quality of the lake water, while keeping the amount of dissolved oxygen at an acceptable level and allowing swimming in the lake. A special specification TBM was developed for noise and vibration control to excavate hard granite and gneiss under densely populated areas (Nordmark, 2002).

Figure 28. Profile and plan of the Ormen Project, Stockholm (Nordmark, 2002) The discharge of untreated wastewater has largely ceased in Stockholm. However, diluted and untreated wastewater is still discharged from the sewer system by CSOs

54

when flows are large. The overflow is discharged to several Mlaren and Saltsj bays the largest quantities (relative to the size of the lake or watercourse) being discharged into Klara Sj. The only small lakes that receive overflow are Lngsjn and Lillsjn (Stockholm Vatten, 2006). The UWWTD was not the main driver for the tunnels; reasons for underground construction included land shortages, elimination of groundwater seepage and for the control of flooding (Nordmark, 2002). More recently, the Stockholm Vatten Water Programme: 2006 to 2015 has been implemented, underpinned by WFD compliance. Within the programme stormwater is considered for local treatment. Where this is not possible, it will require transportation to a less sensitive recipient or a WWTP (moderately and severely polluted stormwater must be transported away from very sensitive recipients), (Stockholm Vatten, 2006).

55

3.

Summary and Conclusions

3.1. Large projects within the EU

A range of approaches to CSO abatement have been identified for cities outlined in Section A. Drivers include, but are not limited to, facilitating compliance of CSOs with the UWWTD. Other drivers include complying with the Bathing Water Directive and reducing urban flooding. Additionally, within Member States responses to either the UWWTD or National Laws transposed from the UWWTD varied widely, depending on region or water/wastewater management structure/organisation. Several cities have established a Master Plan for Urban Drainage as a vehicle by which to review and address existing and future issues associated with parts of their sewer networks (Barcelona, Hamburg, Prague, Zagreb). Table 15 summarises the main approaches identified, along with the main drivers and current status of each project. The most common approach to resolving CSO issues is the addition of extra capacity, whether by the construction of detention tanks and/or trunk or interceptor sewers (Athens, Thessaloniki). Within some cities the use of these approaches is complemented by the use of RTC (Barcelona, Lisbon, Marseille, Vienna, Zagreb). Several cities also combine these approaches with WWTP expansion (Copenhagen, Lisbon, Paris, Prague) and/or sewer separation (Copenhagen, Hamburg). Two German cities were identified as utilising source control techniques (SUDS/disconnection-infiltration, retention basins) alongside some of the more traditional approaches (Berlin, North Rhine-Westphalia). Only two recent projects, located in Naples and Vienna, utilised tunnels in combination with WWTP expansion and RTC, respectively, although the use of tunnels is currently being assessed in Paris. Helsinki and Stockholm were also identified as utilising tunnels, but this was due to historic reasons, not UWWTD compliance. In terms of operationality, Scandinavian and Western European cities were identified as being further ahead with implementation than Eastern European cities. These tended to be in data collection and modelling phases, rather than construction phase. Vienna was the most advanced in utilising RTC, but even so was not 100% operational.

3.2. Small projects within the EU

A number of other studies into CSO abatement/compliance with the UWWTD were identified for smaller cities within the EU. These are not covered in detail, as they fall outside of the selection criteria. However, Table 16 summarises the approaches used in order to give a broader picture of activities being undertaken references are provided so that further research can be undertaken if required. Again, a range of approaches to dealing with CSOs was identified, ranging from interceptor sewers (Granollers, Spain and Steinkjer, Norway) and off-line storage basins (Cosenza, Italy), through to source control and SUDS (Baerum and Bergen, Norway and Lund and Malmo in Sweden) and local or pre-treatment techniques (Oeiras, Portugal). RTC was also popular at this scale, especially in Germany (Bochum, Dresden, Leipzig, Obere Iller). Within all approaches identified (for both large and small cities), data collection and modelling were key components of any design activity and comparisons of several options using feasibility assessments were common.

56

Table 15. Summary of CSO abatement approaches in large cities of the EU (alphabetically by country)
Country Austria Croatia Czech Rep. Denmark Finland France City Vienna Zagreb Prague Copenhagen Helsinki Lyon Marseille Paris Germany Berlin Driver National guidelines Water quality Environment Flooding Water quality Bathing water quality Bathing water quality Environment WFD Flooding Bathing water quality WFD National guidelines (via UWWTD) National guidelines (+ flooding) Environment Flood and pollution control Flood and pollution control UWWTD Flood and pollution control Water quality (pollution) Flood and pollution control Various Approach/es RTC, additional detention tunnel/basins. Tunnel 3km long, 30m deep, 7.0m dia.; 7 shafts RTC, expand collectors Pre-treatment, expand interceptor, add retention tanks, expand WWTP RTC, retention basins, WWTP expansion, sewer separation RTC, WWTP, sewer tunnels, separate sewers RTC, data collection, modelling RTC, trunk sewers RTC, new/expanded WWTP, storage (reservoirs/tunnels) RTC (local), SUDS, heightening CSO crests, storm tanks Sewer separation, retention basins, interceptors, WWTP expansion Trunk sewer, disconnection of impervious areas, infiltration Interceptor sewer diversion/expansion Additional interceptor Tunnel 12 km length up to 80m deep, vortex drop shafts; new WWTP; RTC; odour control RTC, detention tanks and small scale SUDS RTC, on/off-line storage, interceptor sewer, WWTP upgrade RTC, detention tanks Tunnels, WWTP Cost 123 M (RTC only) 4000 M 767 M 7.7/m3 Status Mostly operational Data collection/modelling Part implemented, part planned Operational Operational Operational Operational Mostly operational Operational

Hamburg

Operational

North RhineWestphalia Greece Athens Thessaloniki

4.5 B

Part operational, part ongoing construction Planned Under construction

Italy

Naples

11 M (tanks & RTC) 160 M (to 2012) -

Operational

Netherlands Portugal Spain Sweden

Rotterdam Lisbon Barcelona Stockholm

Operational Part operational, part ongoing construction Operational Operational

RTC = Real-Time Control; WWTP = Wastewater Treatment Plant; WFD = Water Framework Directive; UWWTD = Urban Wastewater Treatment Directive; SUDS = Sustainable Urban Drainage Systems; CSO = combined sewer overflow

57

Table 16. Summary of CSO abatement approaches in small cities of the EU (alphabetically by country)
Country Denmark City Albertslund (Copenhagen) Various Estonia France Tallinn Approach/es Source control (retention ponds), oil separators Enhanced clarification technologies (local treatment) Water quality monitoring/modelling Simulated RTC of storm basin and WWTP Modelling for instrumentation Interceptor sewer/RTC RTC RTC of sewers and retention ponds RTC/WWTP rehabilitation RTC of sewer and WWTP Simulated RTC Off-line storage basin; Constituent-Index Relationships Application of RTC to theoretical sewer network Sewer separation Distributed Stormwater Management Practices Stormwater solutions without pipes and the BlueGreen concept - planning flood ways within urban areas Interceptor sewer Pre-treatment of CSOs Disconnection/source control Interceptors, source control, detention basins RTC, sewer system rehabilitation, retention pond construction, WWTP expansion SUDS SUDS Source control (modelled) Optimal Pollution Control (OPC simulated RTC) Reference Hog et al. (2005) Johansen et al. (2007) Hood et al. (2007) Beck et al. (2005) Vazquez et al. (2006) Gogien et al. (2002) Frehmann et al. (2001) Fuchs et al. (2004) Rohlfing et al. (2005) Scheer et al. (2004) Risholt et al. (2001) Piro et al. (2008); Garofalo et al. (2007) Cenacchi and Maglionico (2005) Ahmedi and Kusari (2008) Freni et al. (2002) Thorolfsson et al. (2007) Thorolfsson and Ystad (2001) Ferreira et al. (2001) Smith et al. (2008) Gomez et al. (2005) Hernebring et al. (2000) Villarreal et al. (2002) Stahre (2002) Ahlman and Svensson (2002) Thomas et al. (2004)

Rosheim
Selestat St Malo Bochum Dresden Leipzig Obere Iller Various Cosenza N/A

Germany

Germany/ Norway Italy

Kosova Norway

Prishtina Baerum Bergen Steinkjer

Portugal Scotland Spain Sweden

Oeiras Various cities Granollers, Besos Basin Helsingborg Lund Malmo Vasastaden

UK

Liverpool

58

4.

References

Abraham, C. (2009) Hamburg Drainage Masterplan 2002. [Email] (Personal Communication from Hans-Reinhard Verworn, 15-25 May 2009). Achleitner, S. and Rauch, W. (2007) Increase of river base flow by hydropower gate operation for mitigation of CSO impacts potential and limitations. Water Resources Management, Vol. 21, pp 1487-1503. Ahlman, S. and Svensson, G. (2002) Substance flow analysis of the stormwater system in Vasastaden, Goteborg. Proceedings of the 9th International Conference on Urban Drainage, Portland, Oregon, 8-13th September 2002. Ahmedi, F. and Kusari, L. (2008) Possibilities of converting the existing wastewater collection system to a separate or combined system. Proceedings of the 11th International Conference on Urban Drainage, Edinburgh, Scotland, 31st August-5th September 2008. Almeida, M. (2009) Tejo (Tagus) and Tranco Multi-municipal Sanitation System. PowerPoint presentation, Ncleo de Engenharia Sanitria (on behalf of SIMTEJO). ATV (1992) ATV Rules and Standards: Standards for the Dimensioning and Design of Stormwater Overflows in Combined Wastewater Sewers. Gesellschaft zur Frderung der Abwassertechnik e.V. (GFA), Hennef, North Rhine-Westphalia, Germany. ISBN 3-934984-17-7. Barro, J. R., Ortega, L., Salamero, M. and Escaler, I. (2002) The Spanish National CSO Measurement Programme (Promedsu) as a base for real time sewer management depending on water quality Barcelonas case. Proceedings of the 9th International Conference on Urban Drainage, Portland, Oregon, 8-13th September 2002. Beck, C., Le Roy, K. and Sadowski, A-G. (2005) A coupled sewer system and WWTP modelling approach to minimise annual discharge: a case study. Proceedings of the 10th International Conference on Urban Drainage, Copenhagen, Denmark, 2126 August 2005. Becker, M. and Raasch, U. (2001) Sustainable rainwater management in the Emscher river catchment area. Proceedings of the 2nd International Conference on Interactions between Sewers, Treatment Plants and Receiving Waters in Urban Areas (Interurba II), Lisbon, Portugal, 19-22 February 2001. Bensasson, L. (2009) The combined sewerage systems of Athens and Thessaloniki. [Email] (Personal Communication, 25-29 May 2009). Capodaglio, A. (2009) Discussion on CSO compliance in Italy. [Email] (Personal Communication 27 May 2009).

59

CBS (2009) Central Bureau of Statistics. http://www.dzs.hr/default_e.htm (accessed August 2009). CEC (2004) Implementation of Council Directive 91/271/EEC of 21 May 1991 concerning urban waste water treatment, as amended by Commission Directive 98/15/EC of 27 February 1998. http://eur-lex.europa.eu/LexUriServ/site/en/com/2004/com2004_0248en01.pdf (accessed 23 March 2009). CEC (2008) Common Implementation Strategy for the Water Framework Directive Progress Report. http://www.naturvardsverket.se/upload/04_arbete_med_naturvard/vattenforvaltning/A rbetet_inom_EU/CIS_progress_report.pdf (accessed 24 March 2009). Cenacchi, C. A. and Maglionico, M. (2005) Design and RTC of tanks for the control of pollution due to CSOs. Proceedings of the 10th International Conference on Urban Drainage, Copenhagen, Denmark, 21-26 August 2005. Clauston-Kaas, J., Sorensen, S., Bent-Johansen, N. and Nielsen, J. B. (2008) Run-off management in Copenhagen Harbour. Proceedings of the 11th International Conference on Urban Drainage, Edinburgh, Scotland, 31st August-5th September 2008. Clemens, F. (2009) Discussion on CSO compliance in the Netherlands. [Email] (Personal Communication 15 May 2009). CZSO (2009) Czech Statistical Office. http://www.czso.cz/ (accessed August 2009). Dedus, B. and Pavlekovic, M. (2005) Application of a novel approach to sewer system analysis. http://www.proning-dhi.hr/dokumenti/doc03.pdf (accessed 14 May 2009). Dirckx, G. (2009) Indicators for the CSO performance for compliance with ambient water quality targets. CD4WC report. [Email] (Personal Communication 15 May 2009). DS (2009) Denmark Statistik (StatBank). http://www.statbank.dk/statbank5a/default.asp?w=1600 (accessed August 2009). Escaler-Puigoriol, I. (2009) CLABSA: The transformation of the urban drainage of Barcelona. PowerPoint Presentation, CETaqua, Agbar (on behalf of CLABSA). Even, S., Mouchel, J-m., Servais, P., Flipo, N., Poulin, M., Blanc, S., Chabanel, M. and Paffoni, C. (2007) Modelling the impacts of combined sewer overflows on the river Seine water quality. Science of the Total Environment, vol. 375, pp 140-151. EWWG (1995) International Source Book On Environmentally Sound Technologies for Wastewater and Stormwater Management. http://www.unep.or.jp/Ietc/Publications/TechPublications/TechPub-15/36EuropeWest/6-2-1.asp (accessed 03 April 2009).

60

Ferreira, F. M., Matos, J. S., Rodrigues, A. C. and do Monte, H. M. (2001) Performance of partially separate sewer systems and impacts on receiving waters. Proceedings of the 2nd International Conference on Interactions between Sewers, Treatment Plants and Receiving Waters in Urban Areas (Interurba II), Lisbon, Portugal, 19-22 February 2001. Frehmann, T., Ustohal, P., Geiger, W.F. and Niemann, A. (2001) Effects of real time control of sewer systems on treatment plant performance and receiving water quality. Proceedings of the 2nd International Conference on Interactions between Sewers, Treatment Plants and Receiving Waters in Urban Areas (Interurba II), Lisbon, Portugal, 19-22 February 2001. Frehmann, T., Althoff, H., Hetschel, M., Schutze, J. A. (2008) Integrated simulation of the Emscher trunk sewer for an analysis of coordinated control concepts for two large wastewater treatment plants. Proceedings of the 11th International Conference on Urban Drainage, Edinburgh, Scotland, 31 August to 5th September 2008. Freni, G., Schilling, W., Saegrov, S., Milina, J. and Konig, A. (2002) Catchment-wide efficiency analaysis of Distributed Stormwater Management Practices: the case study of Baerum (Norway). Proceedings of the 9th International Conference on Urban Drainage, Portland, Oregon, 8-13th September 2002. Fuchs, L., Gunther, . and Lindenberg, M. (2004) Minimizing the water pollution load by means of real-time control (RTC) the Dresden example. Proceedings of the 6th International Conference on Urban Drainage Modelling, Dresden, Germany, 15-17th September 2004. Fuchs, L. and Beeneken, T. (2005) Development and implementation of a real-time control strategy for the sewer system of the city of Vienna. Water Science and Technology. Vol. 52 (5), pp 187-194. Fuchs, L. (2009a) Regulations on sewer overflow. http://isi.tudresden.de/twiki/bin/view/CD4WC/RegulationsOnSewerOverflowsD41?TWIKISID= ca1 (accessed 25 May 2009) [Email] (Personal Communication, 18 May 2009). Fuchs, L. (2009b) Limits proposed by Guidelines. http://isi.tudresden.de/twiki/bin/view/CD4WC/LimitsProposedByGuidelinesD71?TWIKISID=0c b63 (accessed 25 May 2009) [Email] (Personal Communication, 18 May 2009). Fuchs, L. (2009c) Vienna RTC. [Email] (Personal Communication, 11 June 2009). Garofalo, G., Carbone, M., Piro, P. and Sansalone, J. (2007) The role of constituent index relationships for CSO treatment process decisions in the highly urbanized Liguori catchment of Cosenza, Italy. Proceedings of the 6th International Conference on Sustainable Techniques and Strategies in Urban Water Management (Novatech), 24-28 June 2007, Lyon, France.

61

Geretshauser, G. and Wessels, K. (2007) The stormwater management information system a GIS portal for the close-to-nature management of stormwater in the Emscher region. Proceedings of the 6th International Conference on Sustainable Techniques and Strategies in Urban Water Management (Novatech), 24-28 June 2007, Lyon, France. GHK (2006) Strategic evaluation on environment and risk prevention under structural and cohesion funds for the period 2007-2013: National Evaluation Report for Spain. http://ec.europa.eu/regional_policy/sources/docgener/evaluation/pdf/evalstrat_env/es_ main.pdf (accessed 01 April 2009). Gisonni, G. (2009) Discussion on the sewer system of Naples. [Email] (Personal Communication, 12-19 May 2009). Goedbloed, D. (2009) Questions relating to CSO/sewer management in Rotterdam. [Email] (Personal Communication, 14 August 2009). Gogien, F., Zug, M. and Le Luherne, J. (2002) Diagnosis of the real time controlled sewage system of Saint Malo using modelling. Proceedings of the 9th International Conference on Urban Drainage, Portland, Oregon, 8-13th September 2002. Gomez, M., Navarro, J., Serra, P., Castillo, F. and Menendez, S. (2005) The CSO problem in the Besos basin: a general approach for a Mediterranean basin. Proceedings of the 10th International Conference on Urban Drainage, Copenhagen, Denmark, 21-26 August 2005. Gruning, H. and Hoppe, H. (2007) Innovative alternatives for the treatment of storm water in separate system catchments. Proceedings of the 6th International Conference on Sustainable Techniques and Strategies in Urban Water Management (Novatech), 24-28 June 2007, Lyon, France. Gustafsson, L. G., Kuby, R., Metelka, T., Mucha, A. and Pryl, K. (2000) The urban drainage master plan for the city of Prague. Proceedings of the International Conference on Urban Drainage via Internet, 2000 (CD ROM). Helsinki Water (2009c) Cleaner Wastewaters (Last updated 20.04.2009). http://www.helsinginvesi.fi/general/Uploads_files/Puhtaammat_jatevedet_englanti.pd f (accessed 24 June 2009). Helsinki Water (2009b) Quality to the last drop (Last updated 20.04.2009). http://www.helsinginvesi.fi/general/Uploads_files/Nauti_vedesta_englanti.pdf (accessed 24 June 2009). Helsinki Water (2009c) Reliable Water Services (Last updated 20.04.2009). http://www.helsinginvesi.fi/general/Uploads_files/Vetta_varmasti_englanti.pdf (accessed 24 June 2009).

62

Herbke, N., Pielen, B., ward, J. and Kraemer, R. A. (2007) Urban Water Management, Case Study II: Emscher Region. PowerPoint Presentation. http://ecologic.eu/download/vortrag/2006/herbke_milan_emscher.pdf (accessed 03 April 2009). Hernebring, C., Yde, L., Magnusson, P. and Gustafsson, L. (2000) Helsingborg integrated sewer modelling goes online. Proceedings of the International Conference on Urban Drainage via Internet, 2000 (CD ROM). Hog, H. Gabriel, S. Henze, A. and Krayer von Krauss, M. (2005) Sustainable urban drainage and public participation in Albertslund, Denmark. Proceedings of the 10th International Conference on Urban Drainage, Copenhagen, Denmark, 21-26 August 2005. Hood, M., Reihan, A. and Loigu, E. (2007) Modelling urban stormwater runoff pollution in Tallinn, Estonia. Proceedings of the 6th International Conference on Sustainable Techniques and Strategies in Urban Water Management (Novatech), 2428 June 2007, Lyon, France. Hrabak, D., Hanak, S., Pryl, K., Kuba, P. and Metelka, T. (2005) Sewer system protection against large scale floods modelling, evaluation and design Prague case study. Proceedings of the 10th International Conference on Urban Drainage, Copenhagen, Denmark, 21-26 August 2005. IHS (2008) EC Gives France Final Warning on Wastewater Treatment Violations. http://engineers.ihs.com/news/2008/eu-en-wastewater-treatment-france-1-08.htm (accessed 23 March 2009). INEE (2009) Instituto Nacional de Estadstica, Espaa. http://www.ine.es/ (accessed August 2009). INEP (2009) Instituto Nacional de Estatstica Portugal. http://www.ine.pt/ (accessed August 2009). INSEE (2009) Institut National de la Statistique et des tudes conomiques, Paris, France. http://www.insee.fr/fr/default.asp (accessed August 2009). INSI (2009) Istituto Nazionale di Statistica, Italia. http://www.istat.it/ (accessed August 2009). Jerome, M (2006) Flow Equalisation by Temporal Waste Water Detention in Tunnels at Himmerfjrdsverket, Stockholm. http://www.chemeng.lth.se/exjobb/E309.pdf (acccessed 24 June 2009) Johansen, N. B., Hallager, P., Laustsen, A., Nielsen, J. B., Kristensen, N. T. D., Nordemann, P., Nielsen, G. G. and Arnbjerg-Nielsen, K. (2007) Enhanced local treatment of combined sewer overflows enabling the implementation of the Water Framework Directive. Proceedings of the 6th International Conference on Sustainable Techniques and Strategies in Urban Water Management (Novatech), 24-28 June 2007, Lyon, France.

63

Kabelkova, I., Stastna, G., Nabelkova, J. And Stransky, D. (2007) Effect of reconstruction of the combined sewer overflow CSO83 on the ecological status of the Botic Stream, Czech Republic. Proceedings of the 6th International Conference on Sustainable Techniques and Strategies in Urban Water Management (Novatech), 2428 June 2007, Lyon, France. Kabelkova, I. (2009) Discussion on CSO control in the Czech Republic. [Email] (Personal Communication, 21 May 2009). Krebs, P. (2009) Discussion on UWWTD in Germany. [Email] (Personal Communication, 19-20 May 2009). Laplace, D., Garro, Y., Rovera, A. Jiannetti, S. And Guivarch, J. Y. (2007) Sewer network management and the quality of bathing water in Marseille. Proceedings of the 6th International Conference on Sustainable Techniques and Strategies in Urban Water Management (Novatech), 24-28 June 2007, Lyon, France. Lenz, K., Nagy, M. And Weigl, J. (2007) The electronic reporting system for UWWTD Questionnaire 2007. http://www.anpm.ro/Files/twining/6000/09_6017_EN.pdf (accessed 23 March 2009). Lau, K.T., Butler, D. & Schutze, M. (2002) Is combined sewer overflow spill frequency/volume a good indicator of receiving water quality impact? Urban Water, Vol. 4, No. 2, pp 181-189. Nordmark, A. (2002) Overview on survey of water installations underground: underground water-conveyance and storage facilities. Tunnelling and Underground Space Technology, Vol. 17, pp 163178. Nowak, R. (2007) Real-time control of Viennas sewer system. Water21, April 2007, pp 36-37. NSSG (2009) National Statistical Service of Greece. http://www.statistics.gr/portal/page/portal/ESYE (accessed August 2009). Piro, P., Carbone, M., Garofalo, G. And Sansalone, J.J (2008) Management of CSOs based on observations from the urbanised Liguori catchment of Cosenza, Italy. Proceedings of the 11th International Conference on Urban Drainage, Edinburgh, Scotland, 31 August to 5th September 2008. Pollert, J., Srnicek, P., Bares, V. and Kunc, J. (2008) New type of CSO chamber responds to economic and ecological needs. Proceedings of the 11th International Conference on Urban Drainage, Edinburgh, Scotland, 31 August to 5th September 2008. Pryl, K. (2009) Discussion on CSO control in the Czech Republic. [Email] (Personal Communication, 20 May-03 June 2009).

64

Risholt, L. P., Schilling, W., Volker, E. and Alex, J. (2001) Pollution based real time control of wastewater systems. Proceedings of the 2nd International Conference on Interactions between Sewers, Treatment Plants and Receiving Waters in Urban Areas (Interurba II), Lisbon, Portugal, 19-22 February 2001. Rohlfing, R., Nietzschmann, C. and Wiemann, M. (2005) Real-time control of the integrated urban drainage system in the city of Leipzig. Proceedings of the 10th International Conference on Urban Drainage, Copenhagen, Denmark, 21-26 August 2005. Rouault, P., Schroeder, K., Pawlowsky-Reusing, E. and Reimer, E. (2008). Consideration of online rainfall measurement and nowcasting for RTC of the combined sewage system. Water Science & Technology, (57) 11, 2008, S. 1799-1804. SA (2009) Statistics Austria. http://www.statistik.at/web_en/ (accessed August 2009). Salamero, M., Cembrano, G., Escaler, I., Quevedo, J., Puig, V. and Figueras, J. (2002) Real time optimization of the no flooding/no CSO binomium: the Barcelona sewer network case. Proceedings of the 9th International Conference on Urban Drainage, Portland, Oregon, 8-13th September 2002. SBD (2009) Statistisches Bundesamt Deutschland - Gemeindeverzeichnis. http://www.destatis.de/gv/suche_gv2000.htm (accessed August 2009). SCB (2009) Statistiska Centralbyrn (SCB), Sverige. http://www.scb.se/ (accessed August 2009). Scheer, M., Heppeler, D., Krapp, G., Nusch, S. and Messmer, A. (2004) Real time control of an integrated system sewer system and wastewater treatment plant of the Wastewater Association of Obere Iller. Proceedings of the 6th International Conference on Urban Drainage Modelling, Dresden, Germany, 15-17th September 2004. Schroeder, K. (2009)CSO management in Berlin, Germany. [Email] (Personal Communication, 19 May 2009). Schroeder, K. and Pawlowsky-Reusing, E. (2005). Current State and Development Of The Real-Time Control Of The Berlin Sewage System. Water Science & Technology, (52) 12, 2005, S. 181-187. SF (2009) Statistics Finland. http://www.stat.fi/index_en.html (accessed August 2009). SIAAP (2007) Synthese de I'etude d'actualisatiou du scenario C Validee en comito de pilotage du 26 avril 2007. Ile de France, Agencie de Eau Seine-Normandie and SIAAP.

65

Smith, K. W. S., Jefferies, C. and Blackwood, D. J. (2008) Session-management of drainage near source disconnection/minimise surface water entering public drainage. Proceedings of the 11th International Conference on Urban Drainage, Edinburgh, Scotland, 31 August to 5th September 2008. Sorensen, S., Petersen, B., Kofod, N. and Jacobsen, P. (2005) Historical overview of the Copenhagen sewerage system. Proceedings of the 10th International Conference on Urban Drainage, Copenhagen, Denmark, 21-26 August 2005. Sorensen, S. and Kofod-Andersen, N. (2005) Alarm system for bathing water in the harbour of Copenhagen. Proceedings of the 10th International Conference on Urban Drainage, Copenhagen, Denmark, 21-26 August 2005. Stahre, P. (2002) Integrated planning of sustainable stormwater management in the City of Malmo, Sweden. Proceedings of the 9th International Conference on Urban Drainage, Portland, Oregon, 8-13th September 2002. Stenroos, M. and Katko, T. S. (2006) D38: WaterTime case study - Stockholm, Sweden. A research project supported by the European Commission. http://www.watertime.net/docs/WP2/D38_Stockholm.doc (accessed 24 June 2009). Stockholm Vatten (2006) Stockholm Water Programme 2006 2015. http://miljobarometern.stockholm.se/content/docs/gc/9/Stockholm_Water_Programme %5Beng%5D.pdf (accessed 24 June 2009). Teufel, M. (2007) Volume flow rate determination for a sewer management system with real time control in the city Vienna. PowerPoint Presentation, CIWEM/WaPUG, Spring 2007. Thomas, N. S., Burrows, R., Templeman, A. B. and Najafian, G. (2004) Optimal pollution control for management of large interceptor sewer systems. Urban Water Journal, Vol. 1, No. 3, pp 235-250. Thorolfsson, S. T. and Ystad, D. (2001) The Steinkjer System a system for intercepting combined sewers in cold climate areas. Proceedings for the Speciality Symposium Held in Conjunction with The World Water and Environmental Resources Congress: Urban Drainage Modeling, 20-24 May 2001, Orlando, Florida. Thorolfsson, S. T., Aasen, A., Sekse, M. and Totland, O. (2007) Investigations and research on stormwater management in the city of Bergen, Norway, 1981-2006. Proceedings of the 6th International Conference on Sustainable Techniques and Strategies in Urban Water Management (Novatech), 24-28 June 2007, Lyon, France. UK National Statistics (2009) http://www.statistics.gov.uk/census2001/default.asp (accessed August 2009). Vazquez, J., Kouyi, G. L. And Zug, M. (2006) Modelling and instrumentation of the storm sewer overflows of the combined sewer system of Selestat. Urban Water Journal, Vol. 3, No. 2, pp 91-110.

66

Villarreal, E., Bengtsson, L., Niemczynowicz, J. and Zhang, L. (2002) Extension of storm water system at Ostra Torn, Lund. Proceedings of the 9th International Conference on Urban Drainage, Portland, Oregon, 8-13th September 2002. Volte, E., Varnier, J-C. And Pecoraro, J. (2007) Overflows of the sewer system of the urban community of Lyon. Proceedings of the 6th International Conference on Sustainable Techniques and Strategies in Urban Water Management (Novatech), 2428 June 2007, Lyon, France (translated from the French). Web4water (2009) EC sends Italy warning on wastewater. http://www.web4water.com/news/news_story2.asp?id=4530 (accessed 23 March 2009). Wien International (2006) New Sewage System Protects Viennas River Water. http://www.wieninternational.at/en/node/1490 (accessed 24 June 2009) Wikipedia (2009a) Rhine-Ruhr. http://en.wikipedia.org/wiki/Rhine-Ruhr (accessed 24 March 2009). Wikipedia (2009b) Ruhr area map. http://en.wikipedia.org/wiki/File:Ruhr_area-map.png (accessed 24 March 2009). Zabel, T., Milne, I. and McKay, G. (2001) Approaches adopted by the European Union and selected Member States for the control of urban pollution. Urban Water, Vol.3, No. 1-2, pp 25-32.

67

Appendix A CSO consenting/design approaches in the EU

A.1. General

It is common for continuous discharges to receiving waters to be regulated with permits or discharge licences and to be routinely monitored. The degree of legal control over intermittent discharges, such as CSOs is variable throughout the EU, with some member states having more defined as well as stringent design parameters than others. This is partially due to the non-prescriptive way in which the UWWTD gives guidance on CSO management. The technical fact sheet M3 developed for North Rhine-Westphalia in Germany, for example, includes a simplified procedure that provides an assessment of the ecological status without modelling. But most approaches are based on more or less detailed modelling of the water quality and the urban catchment. The permitting arrangements of selected EU member states are summarised in Table 17. In most member states where CSOs do require authorisation, the permits are usually related to allowable spill frequency (Zabel et al., 2001). Even so, the particular benefits of this approach are open to question (Lau et al., 2002). Additionally, only a small proportion of overflows are actually monitored. Two common approaches are emission standards and receiving water quality standards (sometimes referred to as immission standards). Table 17. CSO permitting arrangements in selected member states (Adapted from Zabel et al., 2001)
Country Belgium (Flanders) Denmark France Germany Regulatory body Environmental agency (VLAREM) Regional authority Departments States (Lander) Comments Discharge permit may specify overflow frequency Discharge permit specifies overflow frequency, but is rarely checked. Some municipalities monitor problem CSOs. Permit required if polluting load exceeds 500PE. 1-2% of CSOs are monitored, mainly near bathing and shellfish waters Permits are required for all wastewater discharges, including CSOs. Monitoring, regulation and sampling procedures vary between individual states. Some states require new CSO structures to be equipped with monitoring/telemetry facilities. Legislation proposed which will require discharge licences New CSOs require authorisation (approval permits). Existing CSOs require registration. Discharge permit sets limit on overflow frequency. This is rarely checked except for problem CSOs causing public complaint (< 5%). Monitoring facilities are being added to many systems. All CSOs require registration but formal permits (with conditions) are not issued at present. Discharge consents are required for all CSOs. Monitoring of problem CSOs only at present. Spill frequencies assessed by short- or long-term monitoring plus modelling studies. Major new CSO structures may include permanent monitoring facilities. As above.

Ireland Luxembourg Netherlands Spain UK England and Wales UK Scotland

EPA Ministry of environmental affairs Water boards National river authority, regional authorities Environment agency

Scottish EPA (SEPA)

EPA = Environmental Protection Agency

There are only a limited number of approaches to emission standards for CSOs. A commonly used approach is a minimum CSO setting depending on the mean dry

68

weather flow, as applied in many European member states. In the USA the percentage of wastewater that has to be treated is defined. Another possibility is to limit the number of overflows per year, which can be dependent on the receiving water (for example, Belgium, Denmark or Netherlands). The volume of the CSO storage basin can be prescribed depending on the impervious area connected to the CSO, by a critical rainfall event or a certain dilution (ATV, 1992), (Dirckx, 2009). Table 18 summarises the emission requirements for a range of EU member states. Table 18. Overview of emission requirements for CSOs in EU member states (Adapted from Dirckx, 2009)
Country Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden UK % combined sewer system 75-80 70 45-50 10-15 75-80 67 20 70 60-70 80-90 74 40-50 90-95 25-40 70 Qm (WWTP) 2 QDWs 3-5 QDWm 2 QDWs 2 QDWs 2-3 QDWm 2 QDWs 2 QDWm 3 QDWm 2 QDWm 2-3 QDWm 3 QDWs 2 QDWm 2 QDWm 3-4 QDWm 3 QDWm Qd (CSO) <15 1/(s haAred) 5-10 QDWm N=7/a 5 QDWs N=2-10/a 6-7 QDWs 3 QDWs 7.5-15 1/(s haAred) 3-6 QDWm 6-9 QDWm 5 QDWm 7.5-15 1/(s haAred) 5 QDWm N=3-10/a 6 QDWm 5 QDWm 5-20 QDWm 6-9 QDWm Storage volume 15m3/ haAred N=7/a N=2-10/a

10-40m3/haAred)

10-40m3/haAred) N=3-10/a; ca. 70m3/ha Aimp

tA=2h at 3 QDW

Qm (l/s) = Wastewater flow to wastewater treatment plant; Qd (l/s) = Throttle discharge at CSOs; QDWm (l/s) = Mean dry weather flow; QDWs (l/s) = Dry weather flow (daily maximum); N (l/a) = Yearly overflow frequency; tA (h) = Detention time; Area (ha) = Impervious area connected to the combined sewer system.

Approaches based on water quality standards are more complicated than emissionbased approaches. Their goal is the protection of human health related to the use of the receiving water, or its ecological quality. Indicators can be both biotic and abiotic and have set limits which must be complied with. For example, to ensure bathing water quality the number of colony forming units (CFUs) of pathogenic microorganisms per litre of water is restricted (under the Bathing Water Directive (76/160/EEC). (Dirckx, 2009).

A.2.

Czech Republic

69

Czech legislation dealing with CSOs is vague, as water discharged from CSOs is not considered to be wastewater if the CSO meets the preconditions determined by the water authority. However, there are no guidelines for the meaningful setting of the preconditions and the criteria applied are decided upon by the respective water authority (Kabelkova, 2009). In some cases there is an immission approach (Prague, Brno), but more often an emission approach is used. Sometimes just qualitative limits are requested, but more frequently a dilution ratio ranging from 1:1 to 1:8 (approximately) is used (Pryl, 2009). At present, Swiss, German and Austrian approaches to the assessment of CSO impacts on urban streams are being tested. These are based on the application of the combined approach during rainfall events, to prepare a background for the new Czech legislation concerning CSOs. Additionally, a study into a new design of CSO has been conducted in the Czech Republic (Pollert et al., 2008). The current work being undertaken stems from the transitional period outlined in the Accession Treaty for the Czech Republic, which states that:

By way of derogation from Articles 3, 4 and 5(2) of Directive 91/271/EEC, the requirements for collecting systems and treatment of urban waste water shall not apply in the Czech Republic until 31 December 2010 in accordance with the following intermediate target: as regards agglomerations with a population equivalent of more than 10000, the Czech Republic shall ensure compliance with the provisions of the Directive for 18 agglomerations by the date of accession at the latest and for 36 further agglomerations with the same population equivalent by 31 December 2006. In relation to this, in 2010 a new Czech water law is due to be introduced, which contains several paragraphs regarding CSO impacts on receiving waters (Pryl, 2009).

A.3.

Flanders

In Flanders the vulnerability of the receiving water is important, ranging from extremely vulnerable to not vulnerable. The vulnerability determines the overflow frequency that is allowed. For extremely vulnerable receiving waters, no overflows are permitted. For other receiving waters, the basic rule is that a CSO basin has to retain an event that occurs seven times per year. The number of overflows per year is assessed using simulations. Until recently there have been no real guidelines or legally binding design criteria on CSOs. However, the recent VLAREM legislation, which stems from the translation of the UWWTD into Flemish law, states that best available practise must be applied. This is known as the Code of Good Practise, which is circulated by Ministerial Order. The Code states: o The throttle device after the CSO is to work to 6Q14. Which means that throttle flows are limited to six times the (peak) dry weather flow (Q14), where Q14 equals 1.7 times the average dry weather flow (150 l/day/p.e. multiplied by the number of p.e.); o The overflow layer should not exceed 20 cm, according to which the length of the CSO can be altered;

70

o The crest level is designed according to the local situation and is therefore site dependant. For example, care should be taken not to cause basement or surface flooding. In spite of the design rules, operational CSO control is still in its infancy. Recently however, CSOs have been linked to Performance Indicators, one of which applies to the first working CSO in a catchment. The VMM (Flemish Environmental Agency), the ecological and economic regulator, has set up a network of level sensors monitoring these particular CSOs. Additionally, the Code of Good Practise is currently under revision and a new guidance document on CSOs will be published soon (Dirckx, 2009).

A.4.

Germany

In Germany, the most important regulations on CSOs are ATV-A128 (ATV, 1992), ATV-M153 and the guidelines given by BWK-M3. BWK-M3 was established by the BWK task group in order to address the issue of integrated urban water management in Germany. BWK-M3 is currently being revised and will be updated to form guideline BWK-M7, which uses a more simulation based procedure (Krebs, 2009). Directive ATV-128 includes proposed measures for mitigating the magnitude of the impacts on receiving waters (catchment area: infiltration, surface detention, diversion of rivers into other catchment areas), sewer systems storage and retardation of sewerage water) and in river measures improvement of the bed and bank stability, continuity of the river and reduction flow. BWK-M3 is one of the first data sheets in Germany to take the receiving water into account and uses the river as a limitation for overflow. Measures to improve the receiving water according to the Water Framework Directive are also included within the BWK-M3 (Fuchs, 2009a). BWK-M3 supports the development of requirements (i.e. limit values) for river water quality influenced by urban drainage. The guideline is based on a review of several studies undertaken on hydraulic and pollution based assessment of river water and therefore represents a synthesis of existing knowledge in this field. The proposed methodology including limit values serve as guidance for the assessment of ecological conditions in the river water being influenced by urban drainage. In the meantime, the approach has been established in some federal states of Germany; the application of the proposed methodology is compulsory in a few federal states. Regarding the target conditions in the river water, requirements are fulfilled if a) an annual potentially natural flood flow in the river (HQ1) is not exceeded by CSO discharges by 10%; b) a minimum DO concentration is reached and c) if a maximum NH3-N concentration is not exceeded. The consideration of duration of impact is neglected since this has been implicitly taken into account during the development of limit values. Limit values are rather high; critical conditions may not occur even for higher return intervals. The guideline outlines the permitted ecological impact of discharges, depending on the return interval and the impact duration (Fuchs, 2009b).

A.5.

Italy

71

Within Italy the UWWTD has been translated as National law 152/99. National law 152/06 was more recently translated from the WFD. CSOs, however, are regulated by regional rather than national laws. The regions of Piemonte, Lombardia and EmiliaRomagna have established regional laws concerning CSOs, but for other regions the situation is unknown (Capodaglio, 2009). However, these regulations are only available in Italian and were not available to the authors.

A.6.

Netherlands

In the Netherlands, CSO control is under the management of the municipalities, while river basin management is under the management of the Government or the water boards (depending on the type of river basin). In Rotterdam RTC was implemented in the early 1990s, with the main goal of controlling CSO discharges along basic rules for that municipality (Clemens, 2009): Avoid flooding; Discharge in the following order of priority: i. ii. iii. iv. Discharge to large water bodies after passing a settling basin; Discharge to large water bodies without passing a settling basin; Discharge to small (vulnerable) water bodies after passing a settling basin; Discharge to small water bodies without passing a settling basin.

A.7.

Spain

Whilst the provision of wastewater treatment is a municipal responsibility, most Spanish regions have prepared regional strategies for wastewater treatment aimed at alignment with the EU Directives on their territory. In total, some 14 regional plans coincide (partly) with the 2007-2013 programming period, with a total investment of more than 5,500 million. Whilst most regional strategies and plans focus mainly on compliance with the UWWTD, there are differences depending on the regions degree of compliance with the directive. More advanced regions concentrate on achieving full compliance and the extension of treatment coverage to smaller municipalities. Infrastructure investments focus on the construction of smaller wastewater treatment plants or the renovation of older already existing plants as well as infrastructure for treatment of waste water sludge. Less advanced regions focus mainly on closing the conformity gap and this is done on the basis of infrastructure investments for new medium or large waste water treatment plants (GHK, 2006). Investment focus is on full compliance with the UWWTD and field activities have been ranked by importance. The evaluators assessment of the ranking of types of investment in the area of waste water treatment are summarised in Table 19. As can be seen, CSO upgrading is second from last, indicating low investment in this activity.

Table 19. Investment priorities for wastewater in Spain (GHK, 2006)

Type of investment Ranking

72

New WWTPs Renovation / upgrade WWTPs Sludge treatment Wastewater pumping stations New sewerage Renovation / upgrade sewerage CSO upgrading Sludge disposal

1 2 3 4 5 6 7 8

73

Approaches to UWWTD compliance in relation to CSOs in major cities across the EU

Country % combined sewer system 75-80 70 City name Receiving Water bodies City population Metropolitan population 2,000,000 1,108,000 Driver for CSO abatement National guidelines Water quality; Environment Flooding; Water quality Bathing water quality National Standards / guidelines Approach/es Status Regulatory body CSO Permitting Approach Qm (WWTP) 2 QDWs 3-5 QDWm Qd (CSO) <15 1/(s haAred) 5-10 QDWm; N=7/a Storage volume 15m3/ haAred N=7/a -

Austria Belgium (Flanders) Croatia (candidate EU member) Czech Rep.

Vienna Bruxelles Zagreb

River Danube River Sava

1,697,937 786,000
(E)

RTC, additional detention tunnels / basins. Mostly operational Tunnel 3km long, 30m deep, 7.0m dia.; 7 shafts RTC, expand collectors Pre-treatment, expand interceptor, additional retention tanks, expand WWTP RTC, retention basins, WWTP expansion, sewer separation Data collection/modelling

Environmental agency (VLAREM) -

Discharge permit may specify overflow frequency Current approach is variable as are set by each water authority ; New legislation due in 2010 Discharge permit specifies overflow frequency, but is rarely checked. Some municipalities monitor problem CSOs. Permit required if polluting load exceeds 500PE. 12% of CSOs are monitored, mainly near bathing and shellfish water

45-50 10-15

Prague Copenhagen Helsinki Lyon

Rivers Vltava / Elbe Various waterways Gulf of Finland Rhne and Sane Rivers Mediterranean / Canal de Marseille River Seine Rivers Elbe / Spree / Havel Rivers Elbe / Alster Rivers Ruhr to south, Rhine to west, Lippe to north Saronikos Bay River Axios -

1,233,211

1,390,000 1,299,541 4,415,000 1,500,000 10,000,000 4,275,000 2,575,000 11,800,000 3,750,000 1,200,000 3,500,000 3,550,000 3,075,000 1,499,665 1,831,496 2,550,000 4,250,000 6,100,000 1,989,422

Part implemented, part planned Operational Operational Operational Operational Mostly operational Operational Operational Part operational, part ongoing construction Planned

Water authority Regional authority -

2 QDWs -

5 QDWs ; N=2-10/a -

N=2-10/a -

Denmark Finland

613,603 578,126 472,305 839,043 2,188,500 3,425,000 1,773,218 745,514 360,000 2,718,768 1,299,633

RTC, WWTP, sewer tunnels, separate Bathing water quality; Environment sewers WFD; Flooding RTC, data collection, modelling

France

75-80

Marseille Paris Berlin

Bathing water quality RTC, trunk sewers WFD National guidelines (via UWWTD) National guidelines (+ flooding) Environment Flood and pollution control Flood and pollution control RTC, new/expanded WWTP, storage (reservoirs/tunnels) RTC (local), SUDS, heightening CSO crests, storm tanks Sewer separation, retention basins, interceptors, WWTP expansion Trunk sewer, disconnection of impervious areas, infiltration Interceptor sewer diversion/expansion Additional interceptor (Scheme not developed) (Scheme not developed) Tunnel 12 km length up to 80m deep, vortex drop shafts; new WWTP; RTC; odour control; (Information awaited)

Departments

2-3 QDWm

3 QDWs

Germany

67

Hamburg Rhine-Ruhr area

States (Lander)

Permits are required for all wastewater discharges, including CSOs. Monitoring, regulation and sampling procedures vary between individual states. Some states require new CSO structures to be equipped with monitoring/telemetry facilities.

2 QDWs

7.5-15 1/(s haAred) 10-40m3/haAred)

Greece

20 -

Athens Thessaloniki Rome Milan Naples

Under construction Environmental Protection Agency

Legislation proposed which will require discharge licences

2 QDWm 3 QDWm

3-6 QDWm 6-9 QDWm

Ireland

70

Italy

60-70

Regions Operational Ministry of Environmental Affairs

Variable, depending on region

2 QDWm

5 QDWm

Tyrrhenian Sea -

966,209 1,041,157

UWWTD -

Luxembourg

80-90

Amsterdam

New CSOs require authorisation (approval permits). Existing CSOs require registration. Discharge permit sets limit on overflow frequency. This is rarely checked except for problem CSOs causing public complaint (< 5%). Monitoring facilities are being added to many systems. All CSOs require registration but formal permits (with conditions) are not issued at present. -

2-3 QDWm

7.5-15 1/(s haAred) 10-40m3/haAred) 5 QDWm ; N=310/a 6 QDWm 5 QDWm 5-20 QDWm N=3-10/a; ca. 3 70m /ha Aimp -

Netherlands

74 Rotterdam River Tagus Rivers Llobregat / Bess River Manzanares Lake Malaren, Baltic Sea 986,557 499,700 1,615,908 3,213,271 814,418 Water quality (pollution) Flood and pollution control Various (Information awaited) RTC, on/off-line storage, interceptor sewer, WWTP upgrade RTC, detention tanks Part operational, part ongoing construction Operational

Water boards

3 QDWs

Portugal Spain Sweden

40-50 70 25-40

Lisbon Barcelona Madrid Stockholm

National river authority, regional authorities -

2 QDWm 2 QDWm 3-4 QDWm

London Liverpool (MEPAS) UK 70 Belfast

Thames Tideway

8,278,251

12,300,000

UWWTD

Mersey Estuary River Lagan / Irish Sea

477,600 268,100

579,554

UWWTD UWWTD UWWTD Bathing Water Directive UWWTD

Tunnels, WWTP Tunnels, 39 km total length, 7.2 internal dia, 50 - 75 m depth; approx 40 shafts; cascade and vortex drops; max 3 spills per annum (typical year); RTC; STW expansion and improvements 4.7km of 3.2m pipe. 14 shafts varying between 5 and 18m depth. 9.4km tunnel length, 30 -40m depth, main tunnel 4m diameter. 5 shafts inc. 37m dia shaft for TPS. Also upgrade to STW. 3.5km pipe providing approx 30,000m3 storage. Up to 26m below ground. Tunnel, 5.1 km length, 6 m dia, 25-43 m depth, vortex drop shafts; RTC

Operational Lee Tunnel construction started 2010, target completion in 2015. Thames Tunnel planning ongoing; construction from 2014, target completion 2020. Operational Under construction Completed 2009 Construction to start Spring 2010, target completion 2012 Operational since 1998

Discharge consents are required for all CSOs. Monitoring of problem CSOs only at present. Environment Agency (SEPA Spill frequencies assessed by short- or long-term in Scotland) monitoring plus modelling studies. Major new CSO structures may include permanent monitoring facilities.

3 QDWm

6-9 QDWm

tA=2h at 3 QDW

Preston Brighton NOTES Sources of population data Country AUT BEL CZE DEU DNK ESP FIN FRA GBR

River Ribble / Irish Sea English Channel

131,900 461,181

187,200 -

National Statistical Office Statistik Austria Statistics Belgium, NIS-INS, Ministry of Economic Affairs Czech Statistical Office Statistisches Bundesamt Deutschland - Gemeindeverzeichnis Denmark Statistik (StatBank) Instituto Nacional de Estadstica, Espaa Statistics Finland Institut National de la Statistique et des tudes conomiques, Paris, France UK National Statistics

Country GRC HRV IRL ITA LUX NLD PRT SWE

National Statistical Office General Secretariat of National Statistical Service of Greece Croatian Bureau of Statistics Central Statistics Office, Ireland Istituto Nazionale di Statistica, Italia Le Portail des Statistiques du Luxembourg Statistics Netherlands Instituto Nacional de Estatstica Portugal Statistiska Centralbyrn (SCB), Sverige