Assessing field-scale variability of soil hydraulic conductivity at and near saturation

Saturated hydraulic conductivity (K-s) is a crucial hydraulic property for assessing soil water dynamics. Understanding the spatial variability of K-s in a field is important for site-specific resource management. However, direct measurement of hydraulic conductivity K as a function of soil water pressure head h [K(h)] is time consuming and laborious. Alternatively, pedotransfer functions (PTFs) have been developed to predict K-s indirectly based on more easily measurable soil properties. Although PTFs have been used for decades, their validity for estimating the field-scale spatial variability of K-s remains unclear. The objectives of this study were to characterize the spatial pattern of K(h) at and near saturation in an agricultural field by a coregionalization technique, and in comparison, to evaluate the performance of ROSETTA PTF in characterizing the spatial variability of K(h) at the field scale. Surface soil (7-13 cm) K(h) in the vertical direction was measured at 48 locations in a 71-m by 71-m grid within a no-till farmland. Apparent electrical conductivity was densely measured using a contact sensor Veris 3150 and used as ancillary variable in a coregionalization approach. Experimental semivariograms and cross semivariograms were derived and applied in cokriging to generate K(h) maps. Geostatistical analysis presented similarities in maps of measured K(h) with ROSETTA-predicted K(h) data for a matric potential of - 10 cm. However, the strong spatial heterogeneity of measured K-s, which was caused by macropores, observed in the field was not captured by ROSETTA estimates. The results indicated that texture dominated PTFs like ROSETTA, in which soil structure is not considered, might be useful in characterizing the spatial pattern of unsaturated hydraulic conductivity rather than K-s. Field scale K-s maps based on PTF estimates should be evaluated carefully and handled with caution.