Indirect estimation of near-saturated hydraulic conductivity from readily available soil information

Application of process-based water flow and solute transport models is often hampered by insufficient knowledge of soil hydraulic properties. This is certainly true for dual- or multi-porosity models that account for non-equilibrium flow of water in macropores, where the saturated 'matrix' hydraulic conductivity is a particularly critical parameter. Direct measurement is possible, but this is impractical for larger scale studies (i.e. catchment or regional), where estimation methods (pedotransfer functions) are usually required. This paper presents pedotransfer functions for hydraulic conductivity at a pressure head of -10 cm, K-10, based on measurements of near-saturated hydraulic conductivity made with tension infiltrometers in 70 soil horizons at 37 different sites in 13 different countries. Pedotransfer functions were developed using texture classes, the geometric mean particle size, organic carbon content, bulk density and effective porosity as predictor variables. The pedotransfer functions explained no more than 12% to 29% of the variation in K-10 for the complete dataset. Some important sources of unexplained variation in K-10 may include errors and uncertainty in the (indirect) method used to measure K-10, differences in the way the tension infiltrometer is used, and also temporal changes in hydraulic conductivity due to tillage and/or surface sealing. The importance of tillage was emphasized by the fact that excluding arable topsoils from the analysis gave improved predictions (r(2) values between 26% and 44%) for pedotransfer functions based on texture classes, mean particle size and effective porosity. (C) 2002 Elsevier Science B.V All rights reserved.