Prairie restoration is widely viewed as beneficial because it can restore natural ecological and hydrologic function. Modeling of prairie restoration is essential to quantify the benefits of prairie lands. The accurate prediction of soil parameters is vital for model input to improve the model performance and for other soil related research studies. Pedotransfer functions (PTFs) generate the soil properties and variables needed to parameterize the soil water processes models from readily available soil information such as soil texture and soil organic matter. Native prairie soils are structurally different from disturbed non-prairie soils with the same soil texture, as native vegetation can improve soil physical properties due to soil aggregation and soil organic matter accumulation around the root system. We hypothesize that applying Rawls et al. (1998) PTF on native prairie soils underestimates the soil hydraulic conductivity (SOL_K). The accuracy of estimation of SOL_K was evaluated using nine existing PTFs against observed SOL_K at two watersheds with dominant land-use of prairie land. The results indicated none of the PTFs provided a better estimate. Currently, there is no PTF available explicitly designed for native prairie soils. This research developed a new PTF that incorporates root components (specific root length) to native prairie soils. Stepwise multiple linear regression was used to develop an equation with measured data at Cypress Creek watershed (Waller County, TX). Cypress Creek watershed is dominated by fine sandy loam/sandy loam soil. Percentage of silt (SI), percentage of sand (SD), soil organic matter (OM), and specific root length (SRL) are the independent variables in the new equation to estimate the SOL_K. R2 was 0.9226, indicated 92% of the variation of SOL_K could be explained by CL, OM, SD, and SRL in the regression model. The accuracy of the new equation was successfully evaluated against measured SOL_K in the Clear Creek watershed (Cooke, Montague, Wise and Denton counties, TX). The Clear Creek watershed dominant soil consists of gravelly clay/clay. The new PTF was integrated into the SWAT model and the model performances were evaluated against observed streamflow for four watersheds with dominant land-use of prairie land. The simulation results were statistically satisfactory. The newly developed PTF can be applied to native prairie soils for the better prediction of SOL_K in modeling studies and other research studies.
