A comparison and validation of saturated hydraulic conductivity models

Abstract

Saturated hydraulic conductivity (Ksat) is fundamental to shallow groundwater processes. There is an ongoing need for observed and model validated Ksat values. A study was initiated in a representative catchment of the Chesapeake Bay Watershed in the Northeast USA, to collect observed Ksat and validate five Ksat pedotransfer functions. Soil physical characteristics were quantified for dry bulk density (bdry), porosity, and soil texture, while Ksat was quantified using piezometric slug tests. Average bdry and porosity ranged from 1.03 to 1.30 g/cm3 and 0.51 to 0.61, respectively. Surface soil (0-5 cm) bdry and porosity were significantly (p < 0.05) lower and higher, respectively, than deeper soils (i.e., 25-30 cm; 45-50 cm). bdry and porosity were significantly different with location (p < 0.05). Average soil composition was 92% sand. Average Ksat ranged from 0.29 to 4.76 m/day and significantly differed (p < 0.05) by location. Four models showed that spatial variability in farm-scale Ksat estimates was small (CV < 0.5) and one model performed better when Ksat was 1.5 to 2.5 m/day. The two-parameter model that relied on silt/clay fractions performed best (ME = 0.78 m/day; SSE = 20.68 m2/day2; RMSE = 1.36 m/day). Results validate the use of simple, soil-property-based models to predict Ksat, thereby increasing model applicability and transferability.