Cheng Cheng
Published: 2005
Total Pages: 356
Get eBook
Riparian buffers have been introduced as a means of decreasing nitrate concentrations and other nutrients in surface and subsurface runoff from agricultural activities. The goal of this study was to use groundwater flow models to determine possible hydrogeologic controls on the effectiveness of riparian buffers with different thickness and extents of alluvial and glacial units. A steady-state, finite-difference, groundwater flow model was constructed to simulate groundwater flow in profile at the Risdal North site in the Bear Creek watershed. The simulated profile was 155 m long and 1 row wide, and discretized into 62 columns and 14 layers (646 active cells). Four generalized model zones representing unoxidized till, oxidized till, loam and sand were assigned hydraulic conductivity values obtained from previous studies. Recharge rate was assumed to be 10 percent of mean annual precipitation of 82.9 cm and applied on the top layer. The USGS finite-difference model, MODFLOW was used to simulate hydraulic head and the water table. The model was calibrated by trial-and-error, UCODE, and PEST simulations. Calibration results showed good match between observed heads and simulated heads and the mass balance difference between recharge and discharge of the model was 0.03 percent. The USGS particle-tracking code, MODPATH was used to track particles in the groundwater flow system. The results showed that the average residence time was 94 days in the buffer and the source of groundwater contributing to Bear Creek ranged from distances of 8 to 137.5 m from the creek. Eleven generic models were constructed to assess the effects of geology on groundwater flow and residence times. The models used parameters of the calibrated model and hypothetical geologic conditions based on previous studies at the buffers. Results indicated that loam beneath a buffer had long residence time and supported the shallowest water table. Till beneath a buffer showed the shortest residence time. Sand or limestone beneath a buffer produced a deep water table. Based on these results, if long residence times and shallow water table are favorable for denitrification, loam beneath a buffer might provide the best hydrogeologic setting for removal of nitrate from groundwater.