PERMFILE49068 - Laserfiche WebLink

Home Browse Search

Mr. Alex Schatz March 27, 2006 Page 3 of 6 The model was run as a transient model for 5,000 days for each simulation. One hundred time steps were used for each run with a time step multiplier of 1.1. The PCG2 solver package was used with head closure set to 0.05 feet and the residual set to 0.05. Physical Characteristics Though the material of the alluvial aquifer vanes throughout the model domain, it was modeled as a homogeneous, isotropic aquifer due to the unavailability of reliable hydrogeologic data. Based on the nature of the material described in the well data we reviewed, hydraulic conductivity of 1,000 feet per day and specific yield of 22 percent was used throughout the model domain, except when the gravel pits were converted to lakes following mining (see "Modeling of Pits" below). Sensitivity analyses showed that varying the specific yield within the range of published values (23-28 percent) did not appear to have any noticeable effect on the model results. However, varying the hydraulic conductivity did have an effect on stream depletion and drawdown surrounding the pit. The model was run with hydraulic conductivity ranging from 500 to 5,000 feet per day. Model runs with K = 250 feet per day and K = 10,000 feet per day did not converge, even when head closure and residua] values were set at much larger values. The net effect on the river for K = 500, 1,000 and 5,000 feet per day is presented in Table 1. Note that the effect on the river is not significantly different when values of 500 feet per day were input than when 1,000 feet per day aze used, but the effect on the river is cut in half when a value of 5,000 feet per day is used for K. Based on the published ranges of values for typical sands and gravels and the results of the sensitivity analysis, a hydraulic conductivity of 1,000 feet per day was chosen as appropriate for the remaining model runs for steady state and pit dewatering. Note that this is a more conservative value than 5,000 feet per day and that if a higher value were used, effects of dewatering on the river and neazby wells would be reduced. If field data later show the hydraulic conductivity to be higher, the models could easily be re-run and new data produced. The model was initially run to steady state without any pits in place and calibrated so that the water table level in the alluvium fell just below the ground surface elevations. This is consistent with the static water level data available for nearby wells. These heads were then used as the starting heads in the remaining model runs simulating pit dewatering. The Colorado River was modeled using MODFLOW's river package with the hydraulic conductance of the riverbed set to 6000 feet squared per day, and the river bottom four feet below the head in the river, which ranged from an elevation of 5,370 feet on the west end of the model to 5,420 feet in the east. This is consistent with river head elevations on the USGS topographic map. Martin and Wood Water Consultants, Inc.