Background to channel seepage
Earthen channels in Australia have mostly been constructed using local materials,
often with poor water-retaining characteristics. Despite attempts to reduce permeability,
construction methods have often failed to achieve a watertight barrier, particularly
in older channels. The importation of better- quality soils is often limited
by availability or cost. Seepage from open channels is therefore an Australia-wide
issue.
Channel seepage involves the relatively uniform passage of water through the
wetted perimeter of the channel profile (bed and batters inclusive) due to poor-quality
substrate material. It does not refer to leakage that occurs due to localised
cracks, holes or bank failures, although the remediation techniques explained
on this website could be used to reduce water losses from these sources.
Channel seepage is influenced by the permeability of the layers forming or adjacent
to the wetted perimeter of the channel. Seepage is also affected by the hydraulic
characteristics of the channel and surrounding area. Seepage losses generally
increase with greater water depth in the channel and as the difference in elevation
between water level in the channel and local watertable increases. The quality
of the channel water can also influence seepage rates because suspended particles
seal soil pores and create a sediment lining, thereby reducing seepage.
Seepage mechanisms from earthen channels are dominantly horizontal or vertical,
or a combination of the two. The dominant mechanism at a site affects the rate
of seepage, the impact and the determination of the most appropriate remediation
approach.
Seepage rates from irrigation channels vary from site to site depending on local
conditions. Unlined channels might lose about 150L/m2/day in clay loam, about
250L/m2/day in sandy loam, and 750L/m2/day or more in sandy or gravelly soil
(Swan, 1978). A comprehensive review by the Victorian Rural Water Commission
of channel seepage measurements taken between 1962 and 1983 across the Goulburn-Murray
Water region found seepage ranged from 2.4 to 116L/m2/day (Dunstone, 1998). A
summary of seepage rates reported in the literature is provided in the table
below.
|
Classification
|
Material
|
Seepage rates (L/m2/day)
|
Reference
|
|
|
|
Clays and clay loams
|
Alluvium (unspecified)
|
82
|
Lovas, 1970
|
|
Cemented gravel and hardpan with clay loam
|
104
|
Houk, 1956
|
|
Impervious clay loam
|
76-107
|
Davis, 1952
|
|
Medium clay underlaid with hardpan
|
107-152
|
Davis, 1952
|
|
Clay and clay loam
|
125
|
Houk, 1956
|
|
Ordinary clay loam, silt soil or lava ash loam
|
152-229
|
Davis, 1952
|
|
Gravelly clay loam or sandy loam, sand and clay
|
229-305
|
Davis, 1952
|
|
Silts and silty loams
|
Silty loam
|
341
|
ICID, 1967
|
|
Sands and sandy loams
|
Sandy loam
|
201
|
Houk, 1956
|
|
Volcanic ash
|
207
|
Houk, 1956
|
|
Fine to medium sand
|
216
|
USBR, 1965
|
|
Volcanic ash with some sand
|
299
|
Houk, 1956
|
|
Sandy loam
|
305-457
|
Davis, 1952
|
|
Sand and volcanic ash or clay
|
366
|
Houk, 1956
|
|
Loose sandy soils
|
457-533
|
Davis, 1952
|
|
Sandy soil with some rock
|
512
|
Houk, 1956
|
|
Gravels
|
Gravelly sandy soil
|
610-762
|
Davis, 1952
|
|
Sandy and gravelly soil
|
671
|
Houk, 1956
|
|
Very gravelly soil
|
914-1,829
|
Davis, 1952
|
|
|
Table 1 Seepage rates for unlined channels
Visual inspection and flow records
Seepage identification and measurement is often undertaken in situations
where there is a known or suspected seepage problem requiring investigation.
A particular section of channel with seepage loss (from hundreds
of metres to tens of kilometres) is then targeted for testing.
RWA staff might know which particular sections of channel have seepage
issues through visual observation or from channel flow records. However,
these methods do not usually produce a systematic assessment of channel
seepage risk.
Limitations of visual inspection - Not all channels (e.g. smaller
channels) are subject to visual inspection, so seepage may go unnoticed.
Further, visual inspection only addresses lateral seepage, not vertical
seepage (see General issues in channel seepage identification and
measurement). A channel may have high vertical losses with no surface
expression.
Limitations of flow records - Measurements are not always sufficiently
accurate to assess seepage losses and may be within the error bounds
of the gauging sites.
| Risk
management |
 |
Channel seepage can also be considered in a risk
management framework. This may help the place channel seepage
into context according to its level of likelihood and the consequences
of the impact of seepage.
Risk management may involve reviewing channel seepage risk across
an entire authority or particular region.
| Reasons for investigation |
|
The reasons for the investigation are a key issue
to consider when planning a seepage risk investigation. This
will affect the seepage assessment method employed. There are
three main reasons for seepage investigations:
- Water loss concerns (value of lost water)
- Local land degradation
concerns
- Regional environmental concerns
| Methods to assess seepage risk |
|
Methods to assess seepage risk for each of these
different reasons are presented below.
Water loss concerns
For water loss concerns there are two main methods that can be
employed to assess the regional seepage risk:
- Regional water balance analysis
- Spatial analysis
Regional water balance analysis
Regional water balance analysis involves a system-wide analysis
of flow data, taking account of all diversions and inflows (including
rainfall and evaporation). In some respects this is similar to
inflow-outflow testing but
differs in that it does not focus on one particular section of
channel and is not likely to involve
specific testing. A regional water balance analysis is based
on
careful analysis of available flow records, with the investigation
conducted over a much longer period that a typical inflow-outflow
test. Assumptions may need to be made regarding unmeasured diversions
and, to a lesser extent, inflows, so the final result may be
subject to some uncertainty. Use of inflow-outflow testing on
particular
sections of channel can be used to clarify uncertainty in ungauged
or unsatisfactorily gauged sections of channel.
The advantage of the regional water balance analysis method is
that it enables an estimate of the magnitude of seepage loss
(e.g. estimate of megalitres lost per kilometre). This may be
broken
down into channel or sub-channel sections depending on the location
of gauges.
Spatial analysis
Spatial analysis involves the following steps:
- Identifying key factors affecting channel seepage
in the region (many will be common to all regions). These factors
include soil
and geological information, channel information (capacity,
age, elevation of water surface, wetted perimeter, maintenance
record,
channel bed condition, etc.) and groundwater elevation data.
- Collecting
data on each of the key factors. Many RWAs have basic channel
information in GIS format (e.g. channel alignment,
capacity). Basic soil and geological information (at a large scale) can be
obtained in GIS format from various sources (National
Land
and Water Resources Audit, Geoscience
Australia,
Basin in a Box
Series from Murray
Darling Basin Commission,
State Government Agencies). Individual RWAs may have local information available
at a smaller
scale. Collection of soil
and geological data using remote
sensing techniques could be considered, depending on
the budget.
- Weighting of key factors, which involves determining
their relative importance. The weighting of factors will vary from region
to region. Determining the relative importance of each of the factors should
involve personnel with a background in earth sciences and
channel seepage and operation issues. Soils and geology are likely to be
weighted highly. The elevation of the channel supply level
compared to the elevation of the adjacent groundwater is also an important
parameter, although this information is less likely to
be uniformly available along the channel. The impact of silt
layers in the base
of channel can be very influential and, if data is available,
this factor should also be given a strong weighting.
- Combining factors to determine
overall seepage risk. GIS is the most suitable tool for combining
factors to determine the
overall seepage risk. At its most basic level, this might involve simply
overlaying the channel system onto a soil or geology
map. The information is combined to assign areas into categories
of seepage risk.
Given
the broad-scale nature and level of uncertainty inherent
in this approach, assignment into only a few categories is suggested, (e.g.
high, medium, low). This information is best conveyed
as a map.
Spatial analysis determines areas at risk of seepage without
quantifying seepage rates, although it does provide precise
spatial definition
of areas at risk. Spatial data sets of suitable resolution
enable identification of areas at risk of channel seepage
to a much
finer scale than the regional water balance analysis.
Spatial information gained from the regional water balance
analysis is directly related to the density of gauging
stations. Spatial
analysis of key factors is suitable for assessing seepage
risk before channel construction as well as in operating
channels.
Combined approach
A combination of regional water balance and spatial analysis
can provide greater benefits than either method on its
own as it can
provide information on both the magnitude and spatial
variability of seepage.
Local land degradation concerns
Three methods can be employed to assess the risk of local
land degradation (e.g. salinity and waterlogging) caused
by channel seepage across a large region:
- Visual inspection
- Spatial analysis of key factors
affecting seepage
- Remote sensing
Visual inspection
This is the most basic method of assessing the risk of
local land degradation adjacent to the channel. The
main disadvantages of this method are that it is quite labour
intensive (this may be reduced if incorporated into
regular channel maintenance routines) and that it does not identify
areas not yet showing visible signs of degradation
but that could be susceptible in the future. Despite these
disadvantages, cataloguing visual observations of seepage
on GIS is a valuable first step towards assessing local
land degradation risks.
Spatial analysis of key factors affecting seepage
The key factors and their weighting will vary (see
spatial analysis method above).
Depth to watertable is an example. A channel with a
deep watertable adjacent the channel, even with a relatively
high seepage loss rate, is unlikely to be ranked as
a
high-risk site. Conversely, a channel with groundwater
levels very close to the surface is likely to be ranked
as a high-risk site, even if only moderate seepage
rates are anticipated.
Groundwater salinity could be introduced as an additional
factor. The higher the groundwater salinity, the faster
the degradation that can occur from an elevated watertable
adjacent to the channel.
Land use (or land value) adjacent to the channel could
be incorporated as a further variation. The likelihood
of seepage causing degradation, such as salinity adjacent
to the channel, could then be combined with the consequences
if this did occur, to produce an overall seepage risk
map. Such an output could help prioritise sites for
remediation. For example, a particular site may be
classified as having
a high risk of seepage-induced damage adjacent to the
channel. However, if the land use near the channel
indicates minimal consequences of impact, the overall
rating for
the site might be low to moderate risk, and remediation
would be of lower priority.
Remote sensing to detect actual channel seepage
Remote sensing to
detect lateral seepage adjacent to a channel seepage
is described in the ‘Seepage
identification and measurement, Techniques’ section
of this website. The words of caution described in
the remote sensing section regarding the inability
of this
technique to detect deep (i.e. vertical) channel seepage
are not relevant when the only concern is seepage-induced
degradation adjacent to the channel. Note that the
difference between this and the use of remote sensing
as an input
layer to Spatial Analysis of Key Factors (described
above) is that the aim here is the actual detection
of seepage.
In the above description remote sensing is a tool to
collect information which could indicate the likelihood
of seepage (i.e. soils and geology). In this application
the remote sensing technique is used to actually directly
detect seepage, or the effects of seepage.
Regional environmental concerns
The primary regional environmental concern associated
with seepage is the contribution of seepage to groundwater
recharge and hence regional salination. There are
two related methods that can be employed to assess
the risk of channel seepage to regional salination:
- Regional water balance analysis
- Regional water balance
combined with groundwater flow system data
Regional water balance analysis
Regional water balance analysis is discussed above
(see description under ‘Water loss concerns’).
This approach provides an overall estimate of seepage
to groundwater recharge. These volumes can be compared
with other sources of recharge to determine the relative
significance of channel seepage to total recharge
and hence regional salinity. The greater the degree
to
which the water balance can be broken down, the better
will be the indication as to the elements of the
system contributing most to groundwater recharge.
Regional water balance combined with groundwater
flow system data
Groundwater flow system data includes spatial data
of intermediate and regional groundwater flow systems,
and depth to groundwater information. Combining the
water balance approach with groundwater flow system
data enables an estimate of the impact of recharge
rates on groundwater systems. For example, a high-seepage
section with a groundwater system containing a deep
watertable may not necessarily be ranked as a high-risk
site. Conversely, a system with groundwater levels
already close to the surface might be ranked as a
high-risk site even if only low to moderate seepage
rates are
anticipated.
Regional assets at risk from rising groundwater levels
could be incorporated in the analysis. The likelihood
of seepage causing elevated regional watertables
could then be combined with the consequences. This
can help
to prioritise sites for remediation. For example,
it might assist with the prioritisation between two
areas
with a elevated groundwater levels: one area in which
no key assets have been identified, but with moderate
to high seepage, might be prioritised lower than
a site with only moderate seepage but containing
high-value
assets (e.g. high-value river, towns). |
