14 Groundwater
317
imagery or synergistic microwave-optical imagery can be used in semi-quantitative
approaches, whereby a water balance of the upper zone is calculated. It has been
found that the method of Thomthwaite and Mather (1955) can result in recharge
estimates. Usually monthly effective rainfall (Pe = rainfall minus direct runoff) and
monthly potential evapotranspiration (ETO) is used but the time step can be
shortened. The method consists of a simple bookkeeping procedure. The soil
moisture (Sm) status during dry periods (Pe
Sm = WHC e- WHC/APWL
(2)
The water holding capacity (WHC) is based on soil texture (assumed to be
uniform in each of the terrain units) and rooting depth of vegetation classes
(derived by spectral classification). The decrease of soil moisture plus the rainfall
during dry periods equals the actual evapotranspiration in the method. When Pe >
ETa, the actual evapotranspiration is assumed to be equal to the potential one. A
delay of the calculated excess soil water can be formulated, for example by a linear
reservoir, whose parameters depend on the hydrogeological situation. This
approach can be implemented in a GIS, for simulation of the baseflow (Meijerink
et al., 1994) and can be extended with NDVI values which are correlated to crop
coefficient factors (Seevers and Ottman, 1994). It is advisable to calibrate the
values ofWHC and the delay with water level fluctuations in shallow wells.
Houston (1982) used - successfully - a somewhat similar method to estimate
recharge for an area in Zambia. In his approach, evapotranspiration takes place at
the potential rate unless the soil moisture deficit is smaller than the root constant,
which must be locally estimated from the dominant vegetation type (open
forests,200 mm; short vegetation,75 mm; and poor vegetationlbare soil 50 mm).
This leaves room to account for water to flow along preferred paths before the root
zone is saturated.
For the estimation of the direct runoff, required to determine the effective
rainfall (pe), the work of Rodier (1975) is of interest. He worked out runoff
coefficients with different frequencies of rainfall in the Sahel in small catchments
and presented aerial photographs showing the catchment conditions (permeable
and impermeable surfaces) as a description. Similar data for other regions may be
derived from local gauging data. Relative direct runoff in various units are
mentioned in the descriptions of Fig. 14.1 and Plate 14.B. Groundwater discharge
areas can be excluded from the recharge evaluation because of the upward fluxes.
A simple way of estimating the loss in vegetated discharge zones is to assume that
the loss equals ETO. The rationale behind this assumption is that the capillary rise
from the groundwater reaches the root zone, hence, the vegetation is "not short of
water". In case of crops with shallow roots or bare conditions and prolonged dry
periods, an estimate of the capillary rise may be taken, instead of the ETO loss,
based on depths of the water table and textures of the overburden. Thermal
imagery could be used as a check. It is obvious that GIS procedures are suitable
for the estimation, using combinations of depth from surface to groundwater,
317
imagery or synergistic microwave-optical imagery can be used in semi-quantitative
approaches, whereby a water balance of the upper zone is calculated. It has been
found that the method of Thomthwaite and Mather (1955) can result in recharge
estimates. Usually monthly effective rainfall (Pe = rainfall minus direct runoff) and
monthly potential evapotranspiration (ETO) is used but the time step can be
shortened. The method consists of a simple bookkeeping procedure. The soil
moisture (Sm) status during dry periods (Pe
Sm = WHC e- WHC/APWL
(2)
The water holding capacity (WHC) is based on soil texture (assumed to be
uniform in each of the terrain units) and rooting depth of vegetation classes
(derived by spectral classification). The decrease of soil moisture plus the rainfall
during dry periods equals the actual evapotranspiration in the method. When Pe >
ETa, the actual evapotranspiration is assumed to be equal to the potential one. A
delay of the calculated excess soil water can be formulated, for example by a linear
reservoir, whose parameters depend on the hydrogeological situation. This
approach can be implemented in a GIS, for simulation of the baseflow (Meijerink
et al., 1994) and can be extended with NDVI values which are correlated to crop
coefficient factors (Seevers and Ottman, 1994). It is advisable to calibrate the
values ofWHC and the delay with water level fluctuations in shallow wells.
Houston (1982) used - successfully - a somewhat similar method to estimate
recharge for an area in Zambia. In his approach, evapotranspiration takes place at
the potential rate unless the soil moisture deficit is smaller than the root constant,
which must be locally estimated from the dominant vegetation type (open
forests,200 mm; short vegetation,75 mm; and poor vegetationlbare soil 50 mm).
This leaves room to account for water to flow along preferred paths before the root
zone is saturated.
For the estimation of the direct runoff, required to determine the effective
rainfall (pe), the work of Rodier (1975) is of interest. He worked out runoff
coefficients with different frequencies of rainfall in the Sahel in small catchments
and presented aerial photographs showing the catchment conditions (permeable
and impermeable surfaces) as a description. Similar data for other regions may be
derived from local gauging data. Relative direct runoff in various units are
mentioned in the descriptions of Fig. 14.1 and Plate 14.B. Groundwater discharge
areas can be excluded from the recharge evaluation because of the upward fluxes.
A simple way of estimating the loss in vegetated discharge zones is to assume that
the loss equals ETO. The rationale behind this assumption is that the capillary rise
from the groundwater reaches the root zone, hence, the vegetation is "not short of
water". In case of crops with shallow roots or bare conditions and prolonged dry
periods, an estimate of the capillary rise may be taken, instead of the ETO loss,
based on depths of the water table and textures of the overburden. Thermal
imagery could be used as a check. It is obvious that GIS procedures are suitable
for the estimation, using combinations of depth from surface to groundwater,