Environmental Engineering Reference
In-Depth Information
values for model input (Brunner
et al
.,
2007
).
These data could include, for example, land-
surface elevation, land use, vegetation type and
growth state, surface and air temperature, and
soil-water content. Hendricks Franssen
et al
.
(
2006
) demonstrated that uncertainty in esti-
mates of recharge and other parameters of a
groundwater-flow model for the Chobe region
in Botswana (where data are sparse) could be
substantially reduced by inclusion of what
the authors referred to as recharge potential.
Recharge potential is the difference between
precipitation and evapotranspiration, and for
that study both of these parameters were esti-
mated solely from satellite data.
study area. The interpolating equation for one
subbasin was:
D
−
=× −
7.7
10
5
(
P
27 100)
1.65
for
P
>
27 100
(3.15)
where
D
(drainage) and
P
(annual precipitation)
are in terms of acre-feet for the entire study
area, and drainage is assumed to be 0 when
P
is
less than or equal to 27 100. Equation (
3.15
) was
developed by regression analysis using results
of the BCM for 1970 to 2004. Average annual
recharge for all the watersheds for the 112-year
period was estimated at about 18 mm, or 6% of
precipitation.
3.8 Aquifer vulnerability analysis
Example: Basin and Range carbonate
rock aquifer system
The Basin Characterization Model (BCM) is a
watershed model that requires as input GIS
coverages of geology, soils, vegetation, and
elevation and monthly air temperature and
precipitation (Flint
et al
.,
2004
). Water budg-
ets for 13 hydrographic areas in the Basin and
Range carbonate rock system in Nevada and
Utah were simulated to generate estimates of
drainage through the bottom of the root zone
(Flint and Flint,
2007
). The BCM was applied
on a 270-m grid by using monthly climatic
data for the period 1970 to 2004. Precipitation
values for the model grid were obtained by
interpolation of PRISM (Daly
et al
.,
1994
) data
that are on a 1.8-km grid. Submodels within
the BCM calculated potential evapotranspira-
tion and snow accumulation and melt. Most
recharge is predicted to occur in the moun-
tainous areas of the study area. Averaged over
the entire study area, drainage for the 35-year
period was about 20 mm, or 7% of precipita-
tion and about 47% greater than estimates
generated using the Maxey-Eakin method
(Flint and Flint,
2007
).
Precipitation data for the area are available
from PRISM for 1895 to 2006. Flint and Flint
(
2007
) developed power function interpolation
equations to extend drainage estimates from
the 35-year BCM simulated period to a 112-
year period for each of 30 subbasins within the
Many methods have been proposed for deter-
mining the vulnerability of groundwater to
contamination from surface sources (Gogu and
Dassargues,
2000
). The underlying assumption
of these methods is that contaminants are trans-
ported to the water table by recharging water -
areas of high recharge rate are more vulnerable
than areas of low recharge rate. Two types of
vulnerability are usually discussed: intrinsic
and specific. Intrinsic vulnerability refers to
the susceptibility of contamination based on
physical features, such as geology, depth to
water table, and climate, and is independent
of any particular contaminant. Specific vulner-
ability refers to susceptibility to one particular
contaminant. In addition to physical features of
the groundwater system, specific vulnerability
analysis must consider the chemical and biolog-
ical properties of the system and the contami-
nant. This section provides a brief overview of
the link between model estimates of recharge
and assessment of intrinsic vulnerability of
aquifers.
There is no universally accepted standard by
which aquifer vulnerability is measured. Most
methods for assessing vulnerability provide
overlay maps, or indices, or scores of vulner-
ability; these methods include models such as
DRASTIC (Aller
e t a l
.,
1985
), AVI (van Stempvoort
et al
.,
1993
), EPIK (Doerfliger
et al
.,
2000
), and PI
(Goldscheider,
2005
). Although recharge rate is
