Boundary fluxes in the models balanced within 0.01%. Total recharge
into the 50x50x3 cell grid was 18,016 m**3/d, which averages to
8.59e-5 m/d (1.24 in/yr) per unit surface area, and represents 98.8%
of water input to the model. Cells are 260x326.51 m (
x,
y), with variable thickness ( z). Note that field
estimates of recharge for this area are approximately 1 in/yr
[Hansen, 1991]. Outputs were partitioned as follows: Constant
head nodes 31.3%, river leakage 17%, wells 51.5%. In the model,
some recharge falls on the eroded Kc upland (Layer 3), moves
underground to the valley, and is discharged upward from Kc into the
Qal and Qt layers. Recharge into Layers 1 and 2 is concentrated
adjacent to the Kc contact, to simulate overland flow on the upland
which then recharges the alluvial aquifers. The concentration factor
is 8 times the model average recharge. A factor of 10 times
concentration was required in the Chetolah Creek area in northeastern
Hays to match observed head. This assumption is justified by the
presence of storm-flow retention basins near the Kc-Qt contact in
Chetolah Creek, which might be reasonably supposed to enhance recharge
of any ponded surface runoff.
