Groundwater Flow Model, Optimal Management Model, and Advective Transport Model for the Mississippi River Valley Alluvial Aquifer in Cache Critical Groundwater Area, Arkansas

A three dimensional finite difference model for part of the Mississippi River Valley alluvial aquifer in the cache critical groundwater area of eastern Arkansas was constructed to simulate potential future conditions of groundwater flow. The model was calibrated using Parameter Estimation code (PEST). Additional calibration was achieved using the pilot point with regularization and singular value decomposition (SVD assist). The model was calibrated using 2,322 hydraulic head measurements for the years 2000 to 2010 from 150 observation wells located in the study area. Hydraulic conductivity values from the pilot point calibration ranged between 42 and 173 m/d. Specific yield values ranged between 0.19 and 0.337. Recharge rates ranged between 0.00009 and 0.0006 m/d. Nine pumping scenarios for the years 2011 to 2020 were tested and compared to the simulated water level head from 2010. For all scenarios, the volume of water depleted ranged between 5.7 and 23.3 percent, except in Scenario 2 (minimum pumping ) where the volume increased by 2.5 percent.;The management model was constructed to estimate the optimal withdrawal rates that could be sustained for the wells located in the areas of the groundwater cone of depression using different constraints. Four different sets of management scenarios were tested (M1, M2, M3, and M4) each with four sub-scenarios (A, B , C, and D). Sub-Scenarios M4B and M1A provided the maximum and minimum optimal pumping rates respectively. The percent of volume of water increased in the aquifer for all sub-scenarios; sub-scenario M2C had the highest percent increase of volume of water (10.9 percent) compared to the other sub-scenarios, while sub-scenario M3B had the least percent increase of volume of water (1.8 percent).;An advective transport model was constructed to trace particles forward using two different porosity values and three different hydraulic conductivity values. The model also provided backward particles tracking and the ability to assign a capture zone. The results showed that the particles travels double the distance when porosity decreased from 0.35 to 0.2. Backward particle tracking resulted in shorter particle track lengths compared to forward particle tracking.