Assessing controls on diffuse groundwater recharge using unsaturated flow modeling

Understanding climate, vegetation, and soil controls on recharge is essential for estimating potential impacts of climate variability and land use/land cover change on recharge. Recharge controls were evaluated by simulating drainage in 5-m-thick profiles using a one-dimensional (1-D) unsaturated flow code (UNSAT-H), climate data, and vegetation and soil coverages from online sources. Soil hydraulic properties were estimated from STATSGO/SSURGO soils data using pedotransfer functions. Vegetation parameters were obtained from the literature. Long-term (1961-1990) simulations were conducted for 13 county-scale regions representing arid to humid climates and different vegetation and soil types, using data for Texas. Areally averaged recharge rates are most appropriate for water resources; therefore Geographic Information Systems were used to determine spatial weighting of recharge results from 1-D models for the combination of vegetation and soils in each region. Simulated 30-year mean annual recharge in bare sand is high (51-709 mm/yr) and represents 23-60% (arid-humid) of mean annual precipitation (MAP). Adding vegetation reduced recharge by factors of 2-30 (humid-arid), and soil textural variability reduced recharge by factors of 2-11 relative to recharge in bare sand. Vegetation and soil textural variability both resulted in a large range of recharge rates within each region; however, spatially weighted, long-term recharge rates were much less variable and were positively correlated with MAP (r2 = 0.85 for vegetated sand; r2 = 0.62 for variably textured soils). The most realistic simulations included vegetation and variably textured soils, which resulted in recharge rates from 0.2 to 118 mm/yr (0.1-10% of MAP). Mean annual precipitation explains 80% of the variation in recharge and can be used to map recharge.