Geoscience Reference
In-Depth Information
Table 19.1
Calculated (Renger and Wessolek
1990
) annual groundwater recharge
Predicted change in precipitation
-1%
Today
+1%
+5%
+10%
Climate scenario
HCB2
-
MPIA2
HCA2
MPIB2
P
(mm/a)
594
600
606
630
660
ET
r
(mm/a)
453
455
457
464
473
R
GW
(mm/a)
141
145
150
166
187
Change in estimated
groundwater recharge
-3%
-
+3%
+14%
+29%
of -1, +1, +5 and +10%, respectively, groundwater recharge was estimated to change
accordingly (Table
19.1
).
For the given conditions of soils and land use, an increase in precipitation by 10%
leads to an increase in groundwater recharge of 30% relative to today. Similarly
any percentage change in precipitation results in a threefold change in percentual
groundwater recharge. Note, however, that the calculation of real evapotranspiration
in Renger and Wessolek (
1990
) is based on estimates of potential evapotranspiration
not corrected for temperature rise.
19.4.2 Finite-Element Model: 'Catchment'
Recent and predicted groundwater flow conditions are modelled for steady-state
conditions by means of Feflow Wasy
R
.
The area is subdivided into triangular prisms (finite elements) and the steady-
state flow equation is calculated at each node. Groundwater recharge is implemented
as an areally varying second kind boundary condition (constant flux Neumann con-
dition, special case flux
0, no flow), whereas the sea level is modelled as a first kind
(prescribed head Dirichlet condition). The water exchange between surface water
bodies and groundwater is modelled as leakage (third kind Cauchy condition), i.e.
the exchange rate depends on the head gradient and distance between surface water
body and groundwater table and on the hydraulic conductivity of the bottom layer
of the surface water body.
The model consists of about 65,075 elements and 40,368 nodes. The model is
three-dimensional comprising five layers, i.e. three aquifers separated by two confin-
ing layers, which locally pinch out, thus providing direct hydraulic contact between
the aquifers. Groundwater recharge is incorporated as given in Table
19.1
. Hydraulic
conductivity and effective porosity is derived from lithological information from
drilling logs (Table
19.2
). The average total thickness of the modelled aquifer
system is 100 m; the three sandy aquifers comprise about 25-30% of the total
thickness.
The base model (Fig.
19.2
) reflects the recent spatially varying hydrogeo-
logical conditions, e.g. groundwater recharge, hydraulic conductivity and surface
water bodies, reasonably well. The modelled hydraulic heads have been calibrated
=
