Cotton Storages Project 2011

Storage Seepage
& Evaporation
Final summary of results

A summary of the results from the measurement of seepage and evaporation

losses from 136 on-farm storages across the cotton industry

Dr Ehsan Tavakkoli BScAgEng, MSc, PhD

Research Fellow Environmental Soil
Science & Crop Physiology
Ph:
Fax:
Mobile
Email:

+61 7 4687 3966

+61 7 4631 1870
0421 018 075
ehsan.tavakkoli@usq.edu.au

NCEA
University of Southern Queensland

NATIONAL CENTRE FOR ENGINEERING IN AGRICULTURE

Disclaimer
The views and opinions expressed in this publication are those of the authors and do not necessarily reflect those of the Australian
Government, the Minister for Sustainability, Environment, Water, Population and Communities or the National Water Commission.
While reasonable efforts have been made to ensure that the contents of this publication are factually correct, the Commonwealth does
not accept responsibility for the accuracy or completeness of the contents, and shall not be liable for any loss or damage that may be
occasioned directly or indirectly through the use of, or reliance on, the contents of this publication.

Key Points

88% of storages had low seepage of less than 4 mm per day.

In about 20% of cases, the measured seepage was in a different category to that estimated
by the grower.

A
 nnual evaporation for individual storages (if storages held water year round) ranged from
around 1m/year to just over 2m/year.

On-farm storages can be a major source of water loss on cotton farms. Previous studies in the Macintyre valley (Dalton
2001) indicated seepage losses of between 2% and 10% and evaporation losses of between 14% and 40% of all farm water
in a year. In order to better understand the range of seepage and evaporation losses across the whole industry, the project
Measurement to Improve the Water Efficiency of On-Farm Storages in the Cotton Industry, was established in 2008 to
undertake storage seepage and evaporation measurements.
This project was specifically designed to also encourage the ongoing measurement of storage seepage and evaporation
losses using newly developed measurement technology. A network of consultants was utilised to undertake measurements
for 136 storages across all cotton regions using the Irrimate Seeepage and Evaporation Meter.

Measurement Process
The Irrimate Seepage and Evaporation Meter is able to estimate seepage and evaporation losses from an entire storage,
and is believed to be the only equipment available to achieve this. Most other methods for measuring evaporation and
seepage (such as atmospheric flux techniques or infiltrometers) rely on point source measurements and do not give a
value for the entire storage.
The meter includes a highly accurate pressure sensitive transducer (PST) which is installed under the water and is able
to measure very small changes in water level. An accurate analysis of seepage and evaporation can usually be achieved
with approximately 20 days of quality data. As periods of rainfall and storage inflow/outflow cannot be used, the equipment
usually needs to be deployed for at least 5 weeks to ensure enough quality data is collected.
Data analysis is achieved by using regression techniques to compare measured water level changes and local
evapotranspiration data. This process allows the evaporation and seepage components of the total loss to be separated,
thus determining an average daily seepage rate and a dam evaporation factor (kdam), which can be used to convert a local
estimate of evaporation to an actual rate of evaporation for a specific water storage.

storage seepage & evaporation loss

 Copyright Cotton Catchment Communities CRC 2011

This publication is copyright. Except as permitted under the Copyright Act 1968, no part of this publication may be reproduced by any process,
electronic or otherwise, without the specific written permission of the copyright owner. Neither may information be stored electronically in
anyway whatever with out such permission.

Results
Results
Table
1 presents
a summary
of the
seepage
evaporation
figures
or astorages.
ll storages.
It isuggested
s suggested
Table
1 presents
a summary
of the
seepage
andand
evaporation
figures
for fall
It is
that the range of
storage
sizes
evaluated,
from
75
ML
to
14,000
ML,
would
encompass
the
full
size
range
of
irrigation
that the range of storage sizes evaluated, from 75 ML to 14,000 ML, would encompass the full size storages found
throughout the cotton industry.
range of irrigation storages found throughout the cotton industry.

1 Summary
of key data
TableTable
1 Summary
of key data

Seepage (mm/day)

Mean
2.3

Minimum
0

Maximum
38.1

Evaporation m/year

1.52

1.03

2.18

Dam Factor (kdam)

0.97

0.67

1.31

Storage Size (ML)

1,950

75

14,000

Storage Size (ha)1

44

303

3.5

1.0

9.1

Water Depth (m)

1
2

Area data not available for the 4 storages located in Central Queensland
Depth of water in storage at the time of equipment installation, not the depth of the storage

Storages
were
located
across
all cotton
growing
regions
(Figure
1), although
seasonal
conditions
nd distribution of
Storages
were
located
across
all cotton
growing
regions
(Figure
1), although
seasonal
conditions
and athe
measurement
equipment
resulted in more
measurements
Condamine,
Lowerin
Balonne
and Namoi catchments than
the distribution
of measurement
equipment
resulted in
in the
more
measurements
the Condamine,
in some
other
regions.
Whilst
the
total
number
of
storages
within
the
industry
is
not
accurately
known,
Lower Balonne and Namoi catchments than in some other regions. Whilst the total number
of it is likely that the
sample size (136) represents no more than 10% of all storages in use.
storages within the industry is not accurately known, it is likely that the sample size (136) represents
no more than 10% of all storages in use.

Insert map of storage locations here (provided separately)
Figure 1 Location of measured storages

Seepage

Figure 2 shows the distribution of seepage results obtained. Significantly, 88% of storages (120) had
seepage of less than 4 mm/day, a rate which could be considered low, with most of these (89)
indicating extremely low seepage of less than 2 mm/day. A single outlier exists for a storage that was
known to leak very badly and was confirmed to have seepage of 38 mm/day. This storage contained
water during the measurement period for the first time in over 6 years (since being purchased by the
current owner) and was drained within a matter of weeks due to the excessive loss being
experienced.

COTTON STORAGES PROJECT 2011

Figure 1 Location of measured storages

storage seepage & evaporation loss

Seepage and Evaporation Measurements

Catchment Boundaries

Seepage
Figure 2 shows the distribution of seepage results obtained. Significantly, 88% of storages (120) had seepage of less than
4 mm/day, a rate which could be considered low, with most of these (89) indicating extremely low seepage of less than
2 mm/day. A single outlier exists for a storage that was known to leak very badly and was confirmed to have seepage of
38 mm/day. This storage contained water during the measurement period for the first time in over 6 years (since being
purchased by the current owner) and was drained within a matter of weeks due to the excessive loss being experienced.

Figure 2 Histogram of all seepage results

As part of the measurement process, growers were asked
Figure
Figure
2

2

istogram
H
istogram
o
f
o
a
f
ll
a
s
ll
eepage
s
eepage
r
esults
r
esults
to estimate their level of seepage before the evaluation
was conducted (Figure 3). Whilst most grower estimates
were reasonably close toAs
the
there
were
As
pmeasured
art
part
of otf results,
he
the
mm
easurement
easurement
process,
process,
growers
growers
ww
ere
ere
asked
asked
to to
estimate
estimate
their
theirl
about 20% of cases where the measured seepage could be
classified differently to the
estimate
provided.ww
the
the
evaluation
evaluation
as
as
conducted
conducted
(Figure
(Figure
3).
3).
WW
hilst
hilst
mm
ost
ost
grower
grower
estimates
estimates
ww
ere

For example, of those growers who estimated their seepage

measured
measured
results,
results,
there
there
ww
ere
ere
about
about
20%
20%
of ocf ases
cases
ww
here
here
the
the
mm
easured
easured
seep
see
as low, two had a measured rate above 7 mm/day which
could be classified as high,
whilst another
had
a
differently
differently
to tthree
o
the
the
estimate
estimate
provided.
provided.

measured rate of above 4 mm/day which could be classified
as medium. However it should be noted that numerical
guidance was not provided
to growers,
therefore
individual
For
For
example,
example,
of otf hose
those
growers
growers
ww
ho
ho
estimated
estimated
their
their
seepage
seepage
as
as
low,
low,
two
two
had
ha
growers could have a different concept of low, medium or
7 7
mm
m/day
m/day
ww
hich
hich
could
could
be
be
classified
classified
as
as
high,
high,
ww
hilst
hilst
another
another
three
three
had
had
a am m
eae
high seepage.

In one case where the grower

estimated
very
high
4 4
mm
m/day
m/day
waw
hich
hich
could
could
be
be
classified
classified
as
as
mm
edium.
edium.
HH
owever
owever
it ist hould
should
be
be
noted
note
seepage rate, the measured seepage rate was quite
low at less than 3.5 mm/day.
Such
ap
case
illustrates
the
was
was
not
not
rovided
provided
to to
growers,
growers,
therefore
therefore
individual
individual
growers
growers
could
could
have
have
a ad iffe
dif
importance of objective measurement before taking action
medium
medium
oh
r igh
high
seepage.
seepage.
to address perceived seepage
loss. or

In In
one
one
case
case
ww
here
here
the
the
grower
grower
estimated
estimated
a av ery
very
high
high
seepage
seepage
rate,
rate,
the
the
mm
eas
ea
Figure
2low
low
Histogram
f a3
ll .5
results
quite
quite
at alt ess
less
than
tohan
3seepage
.5
mm
m/day.
m/day.
Such
Such
a ac ase
case
illustrates
illustrates
the
the
importance
importance
ofo
Figure 3 Histogram of measured seepage results for each
Seepage was not found to vary with soil type, storage shape
category of grower predicted seepage. The outlier (38mm/day) has
or between regions. However,
evaluations
were
unevenly
before
before
taking
taking
action
action
to to
address
address
p
erceived
perceived
seepage
seepage
oss.
loss.
estimated by
been
removed for improved
claritylbut
was correctly

distributed across theseAs

categories,
part omaking
f the itmdifficult
easurement pthe
rocess,
growers
were asked to estimate
grower as very
high.
to draw accurate conclusions. The range of soil types
encountered and the number
storages constructed
the eofvaluation
was con
onducted (Figure 3). Whilst most grower estimate
lighter soil types was unexpected (Table 2). However higher
seepage rates were notmeasured
solely associated
with lighter
results,
there were about 20% of cases where the measured
soil types. It is possible that the rudimentary surface soil
classification achievabledifferently
within the resources
of this
project
to the
estimate
provided.
may not accurately represent subsoil conditions (including
compaction) that could significantly influence seepage.

For example, of those growers who estimated their seepage as low, tw

7 mm/day which could be classified as high, whilst another three had a
4 mm/day which could be classified as medium. However it should be
was not provided to growers, therefore individual growers could have
medium or high seepage.

In one case where the grower estimated a very high seepage rate, the
COTTON STORAGES PROJECT 2011
5
quite
l3ow
aHt
less tohan
3.5 msm/day.
uch
ecach
the piredicted
mportan
Figure
Figure
3
H
istogram
istogram
f om
f m
easured
easured
eepage
seepage
results
rSesults
for
faor
ease
ach
category
cillustrates
ategory
of ogf rower
grower
p
redicted
ses

resources of this project may not accurately represent subsoil conditions (including compaction)
could significantly influence seepage.
Table 2 Measured seepage for a range of different surface soil types

Soil Type

Measured Seepage Rate (mm/day)

Mean
Minimum
Maximum

Sample size

Heavy Clay

23

2.0

0.1

5.0

resources of this project may not accurately represent subsoil conditions (including compaction) that
Medium Clay
57
2.7
0.1
38.1
could significantly influence seepage.

Light Medium Clay

29

Light Clay

13

1.5

0.0

4.5

3.2

0.5

11.5

Table for
2 a M
easured
seepage
for asoil
range
of different surface soil types
Table 2 Measured seepage
range
of different
surface
types

Soil
Type
Clay
Loam

Sample
4 size

Measured Seepage Rate (mm/day)

1.4
0.5
2.2
Mean
Minimum
Maximum

Sandy
Clay
Heavy
Clay
Loam

23 5

2.0

1.4

0.1

0.5

5.0

3.7

Medium
Sandy CClay
lay

57 6

2.7

2.2

0.1

0.5

38.1

3.5

Light Medium Clay

29

1.5

0.0

4.5

Light Clay

13

3.2

0.5

11.5

In terms of storage size, it might be expected that larger storages, with less compaction over the
Clay Loam
4
1.4
0.5
2.2
base during the construction process and with greater potential for soil variability, might have
Sandy Clay Loam
5
1.4
0.5
3.7
typically higher seepage losses. However, all of the higher seepage results were obtained from
Sandy
Clay of smaller volumes
6
2.2 areas. 0.5
3.5
storages
or surface

The data from four storages was also analysed to determine the effect of water depth on seepag
In rate.
terms Wohilst
f storage
size, it might
be expected
that tlhat
arger
storages,
with dless
compaction
over
conventional
wisdom
suggests
greater
water
epths
will result
in hthe
igher rates of
In terms of storage
size,
it
might
be
expected
that
larger
storages,
with
less
compaction
over
the
base
during
the
base
d
uring
t
he
c
onstruction
p
rocess
a
nd
w
ith
g
reater
p
otential
f
or
s
oil
v
ariability,
m
ight
h
ave
oil hydraulic
conductivity
and the
complicated
nature
of seepage
some loss
pathways
will also h
construction processseepage,
and with sgreater
potential
for soil variability,
might
have typically
higher
losses.
However,
typically
h
igher
s
eepage
l
osses.
H
owever,
a
ll
o
f
t
he
h
igher
s
eepage
r
esults
w
ere
o
btained
f
rom
all of the higher seepage
results
were obtained from storages of smaller volumes or surface areas.
a major
effect.
storages of smaller volumes or surface areas.

The data from four storages was also analysed to determine the effect of water depth on seepage rate. Whilst conventional
wisdom suggests that
greater
water b
depths
result
higher
seepage,
conductivity
This
is reflected
y the will
results
in Tinable
3, wrates
hich ofshow
that tsoil
wo hydraulic
of the four
storages and
had the
lower
The
ata from
our storages
as ahave
lso aanalysed
to determine the effect of water depth on seepage
complicated nature
of d
some
loss fpathways
will w
also
major effect.

measured
seepage when
the swuggests
ater depth
as greater.
torage
D rsesult
howed
igher
seepage
rate.
Whilst conventional
wisdom
that gw
reater
water dSepths
will
in hh
igher
rates
of when

This is reflected by the results in Table 3, which show that two of the four storages had lower measured seepage when the
water
dsepth
was 5 m
chigher
ompared
to t4he
mc, omplicated
but
no further
increase
in
seepage
when
w
ater
depth wa
seepage,
oil hydraulic
conductivity
and
nature
some
loss
pathways
will no
also
have
water depth was greater.
Storage
D showed
seepage
when
water depth
was o5f m
compared
to 4 m, but
further
6 when
m. Tehis
limited
mThis
ost likely
suggests
hat for
storages
with
low
ariations
increase in seepage
water
was 6 m.
limited
analysistmost
likely
suggests
that
forseepage,
storages vwith
low in w
a m
ajor
ffect.
depthanalysis
seepage, variations in water depth cause changes in seepage that are within the bounds of measurement error.

depth cause changes in seepage that are within the bounds of measurement error.

This
is reflected
by the results
in Table fact
3, wsheet
hich savailable
how that on
two
f the four
storages Communities
had lower CRC
Further detail of this
analysis
is contained
in a separate
theoCotton
Catchment
website (www.cottoncrc.org.au).
measured
s
eepage
w
hen
t
he
w
ater
d
epth
w
as
g
reater.
S
torage
D

s
howed
h
igher
s
eepage
when
Further detail of this analysis is contained in a separate fact sheet available on the
Cotton Catchm
water
depth was C
5RC
m cw
ompared
to 4 m, but no further increase
Communities
ebsite (www.cottoncrc.org.au).
in seepage when water depth was
6 m. This limited analysis most likely suggests that for storages with low seepage, variations in water
Table
on
Tseepage
he
effect
wsater
depth
on asre
eepage
rate
depth
c3ause
changes
eepage
that
within
the bounds of measurement error.
Table 3 The effect of water
depth
rateoif n

Further
detail
of this analysis W
is cater
ontained
in (am)
separate
fact sheet
available
on the Cotton Catchment
Storage
Approximate
Depth
Seepage
Rate
(mm/day)
Communities
A CRC website (www.cottoncrc.org.au).
2.5
3.9
Table 3 The effect of water depth on seepage rate

Storage Approximate Water Depth (m)

A C
2.5
5.0
B D
1.0
1.6
C
1.7
2.0
D
4.0
5.0
6.0
6

storage seepage & evaporation loss

5.0
1.0
1.6
Seepage
Rate (mm/day)
1.7
3.9
2.0
2.6
1.7
4.0
2.2
5.0
0.8
6.0
0.5
1.5
2.4
2.4

2.6
1.7
2.2
0.8
0.5
1.5
2.4
2.4

water bodies, etc. Therefore, a dam factor is used to estimate e

This dam factor (Kdam) is measured during the period of deploym
data recorded over one or more years.

The average annual evaporation from all storages (following app

storage) is presented in Figure 4.

Evaporation

Figure 4 Histogram of potential annual evaporation for all

sites after application of individually-determined dam factors

Comparison of evaporation measurements is not straightforward

because the measurement technology was typically only deployed
for a period of 1 to 2 months. Therefore, measured evaporation
during the period of deployment will depend entirely on the climatic
conditions experienced at that time. Hence it is most appropriate
to present typical annual evaporation figures that take seasonal
conditions into account.
Evaporation from the water surface will be influenced by a range of
site specific variables including wind, surface water temperature,
surrounding features (trees, hills), proximity to other water large
water bodies, etc. Therefore, a dam factor is used to estimate
evaporation for individual storages. This dam factor (Kdam) is
measured during the period of deployment and can be applied to
climatic data recorded over one or more years.
The average annual evaporation from all storages (following
application of the relevant kdam for each storage) is presented in
Figure 4.

Figure 4 Histogram of potential annual evaporation for all sites after applica
factors
For consistency,
all dam factors reported within this project relate to SILO FAO56 ETo data. The range of dam factor values

is indicated in Figure 5. The majority (82%) of dam factors lie between 0.8 and 1.2.

Dam factor was compared to a number of storage characteristics such as local average rainfall, water depth, surface area,
storage location (latitude) and the characteristics of the surrounding area. However, no correlation between dam factor
and any of these parameters was found.

For consistency, all dam factors reported within this project rela
Figure 5 The range of calculated dam factor values
range of dam factor values is indicated in Figure 5. The majority
and 1.2.

Dam factor was compared to a number of storage characteristics

depth, surface area, storage location (latitude) and the character
However, no correlation between dam factor and any of these p

Figure 5 The range of calculated dam factor values

COTTON STORAGES PROJECT 2011

References
Dalton, P., Raine, S. and Broadfoot, K. (2001). Best management practices for maximising whole farm irrigation efficiency
in the cotton industry. Final Report for CRDC Project NEC2C. National Centre for Engineering in Agriculture Publication
179707/2, USQ, Toowoomba.

Acknowledgements
This project is funded by the National Water Commission through its Raising National Water Standards Program. This
Australian Government program supports the implementation of the National Water Initiative by funding projects that are
improving Australias national capacity to measure, monitor and manage its water resources.

storage seepage & evaporation loss