2023-08-03_PERMIT FILE - M2023025 (32)

Home Browse Search

Bernhardt Pit Groundwater Evaluation May 2023 Page 16 of 31 We identified 29 registered wells that have reported water level depths (when drilled) less than 10 feet deep. However, none of these wells none are located with the modeled mounding area. Conclusions and Recommendations We do not believe the Bernhardt Pit slurry wall will have any adverse effect on the wells identified in Table 1 because the modeled impacts are small and less than the seasonal groundwater change. Therefore, we do not expect any of the well owners within the area of change will notice any impacts. If they do,we recommend that ongoing monitoring occur to quantify those impacts. Model Sensitivity The modeled mound and"shadow"impacts are insensitive to the hydraulic conductivity(K)of the aquifer. A decrease in K causes a proportional increase in model inflows, but also decreases the hydraulic connection with the SPR which offsets less mounding or shadow effects. Therefore,we did not spend excessive time trying to match modeled verses observed heads by changing K values. We did conduct a model run where we dropped the K from 600 ft/day to 300 ft/day which did not significantly change the head residuals. The model results are very sensitive to the presence of the SPR, but there is no realistic chance that the river is going away any time soon. Model results are also insensitive to streambed leakance. Even using a conservatively low Ksb value of 36 ft/day, the SPR is still wide enough and the aquifer is permeable enough to rapidly respond to changes in groundwater levels caused by the slurry wall. Model Uncertainty There is uncertainty in our predictions related to measurement and modeling error. USGS 10m DEM Data was used to assign ground elevations to off-site wells. The overall accuracy of expressed as the root mean square error(RMSE) for 10m DEM data for one study in (Haneberg, 2006) is 1.87 m or approximately+/- 6 ft. The accuracy of surveyed site well elevations data is much higher, probably within a tenth of a foot, and the accuracy of hand-measured water levels can vary several inches. Our modeled predictions (Figure 10) include three types of error: 1) conceptual error (how the model is set up and what boundary conditions are used); 2) parametric error (how aquifer properties are measured and calculated); and 3) predictive error(which includes other influences such as seasonal recharge or climate change variations). It was beyond the scope of this project to quantitatively evaluate the sum of these errors. However, based on our professional opinion, we believe our model results are reasonably accurate for the purpose of this study. Sources Barkmann, Peter E., Horn, Andy, Moore, Annette, Pike, Jeremy, and Curtiss, William. 2014. Gilcrest/LaSalle Pilot Project Hydrogeologic Characterization Report. Colorado Geological Survey, prepared for the Colorado Division of Water Resources and the Colorado Water Conservation Board. CDM-Smith,April,2013a. South Platte Decision Support System Alluvial Groundwater Model Report. McGrane Water Engineering,LLC 1669 Apple Valley Rd. 0 Lyons,CO 80540 0 Phone:(303)917-1247;email:dennis@mcgranewater.com