The Blue River basin in south-central Oklahoma has been the focus of considerable hydrologic research in recent years, including the National Weather Service's Distributed Model Intercomparison Project (DMIP) and the Oklahoma Water Resources Board's Arbuckle-Simpson Hydrology Study. The basin is dominantly composed of loam and clay soils that have a tendency to shrink or swell as a function of soil water content and temperature. It is hypothesized that after prolonged dry periods, the soil shrinks and forms macropores that act as conduits for enhanced water flows. These changes in the soil matrix yield increased values of saturated hydraulic conductivity (Ksat), which affect runoff volumes and seasonal water budget calculations. A study in Pennsylvania showed that Ksat increased by several orders of magnitude as soil water content decreased for a soil type similar to those found in the Blue River basin. In hydrologic modeling, the identification of a single optimal parameter set becomes a challenge if one of the most sensitive parameters, Ksat, varies significantly with time.

Field measurements of Ksat have been collected in the Blue River basin since June 2008. These measurements have enabled us to develop a parameter estimation method for the TREX distributed hydrologic model that adjusts the Ksat input as a function of initial volumetric water content derived from Oklahoma Mesonet soil moisture observations. Results show streamflow responses during wet and dry periods are better simulated by TREX if the temporally-variable Ksat field is used rather than applying a static Ksat field.