Environmental Engineering Reference
In-Depth Information
Table 3.3. Degrees of vulnerability, simulated travel
time through the unsaturated zone, and propor-
tional area of coverage for the country of Namibia
(after Schwartz,
2006
).
an important factor in predicting aquifer vul-
nerability, explicit estimates of recharge rates
are not required in some assessment methods.
Of the four vulnerability models cited above,
only DR ASTIC requires an estimate of recharge
for model input. The other models use parame-
ters such as depth to water table, precipitation,
and soil properties as surrogates for recharge.
Schwartz (
2006
) and Neukem
et al
. (
2008
)
took a different approach for quantifying vul-
nerability; classes of vulnerability were defined
solely on the basis of the estimated travel time
of water from land surface to the water table.
Many of the models described in this chapter,
including unsaturated zone water-budget mod-
els, watershed models, and empirical models,
could be used to estimate these travel times.
Large-scale recharge maps, such as those devel-
oped by Holtschlag (
1997
), Keese
et al
. (
2005
),
and Lorenz and Delin (
2007
;
Figure 3.17
), are
becoming more available. A first-cut vulnerabil-
ity assessment might equate areas of relatively
high recharge rates on those maps with areas of
high vulnerability to contamination.
Another aspect of aquifer vulnerability
mapping in which models play an important
role is identification of areas that contribute
to a water-supply well. Aquifers are susceptible
to contamination from recharge that occurs in
these areas. Well-head protection rules have
been implemented in many areas to safeguard
groundwater supplies. Contributing areas are
often identified by a particle-tracking routine,
such as MODPATH, that is capable of backtrack-
ing along the path of a water particle from the
point on the well screen where the particle is
removed from the aquifer back to the location
where that particle first entered the aquifer
as recharge (Franke
et al
.,
1998
; McMahon
et
al
.,
2008
; Moutsopoulos
et al
.,
2008
). MODPATH
requires estimates of groundwater velocity at
all points in the aquifer; these velocities are
usually determined with a groundwater-flow
model such as MODFLOW.
Vulnerability
classification
Travel time
(yr)
Percent areal
coverage
Very low
>2 5
84.5
Low
10 -25
10.6
Medium
3-10
4.3
High
1-3
0.5
Very high
<1
0.1
water percolating downward from land surface
in the southwest African country of Namibia.
Five classes of vulnerability were identified on
the basis of travel time for infiltrating water to
reach the water table (
Table 3.3
). A map of travel
time was developed for the entire country. The
approach used a combination of numerical mod-
eling, regression analysis, and new and exist-
ing GIS data coverages. Flow through the soil
zone was simulated with the MACRO4.3 model
(Jarvis,
2002
), a model based on the Richards
equation, for four generic lithological units.
Simulations were run for each of six assumed
soil-zone thicknesses, ranging from 0 to 1 m.
Potential evapotranspiration rates required
by the model were estimated on the basis of
monthly temperature measurements. Net infil-
tration (flow out the bottom of the soil zone)
was simulated for a 5-year period with meas-
ured daily precipitation.
Regression equations were developed to
relate net infiltration,
I
net
, to annual precipita-
tion and soil-zone thickness for each lithological
unit. Travel time through the unsaturated zone,
t
, at any location was then calculated as:
tZ
net
/
θ
(3.16)
where
Z
is depth to the water table and θ is the
average volumetric water content (assumed uni-
form within each lithological unit). GIS cover-
ages of annual precipitation, lithological unit,
and depth to water table were then used to
determine and map travel times (
Figure 3.18
).
Results showed that 95% of the country falls
Example: groundwater vulnerability
in Namibia
Schwartz (
2006
) described a study of vulner-
ability of groundwater to contamination from
