Estimation of Moisture Retention Curve in Some Saline and Saline-sodic Soils with Pedotransfer Functions

Salt-affected soils cover up to 20 million hectares of our country. The objective of this study was to derive PTFs of some salt-affected soils to predict the parameters of van Genuchten equation and water retention at a pre-defined potential from readily/easily obtainable soil characteristics. For this purpose، 68 soil samples were chosen from south part of Tehran (saline and saline-sodic soils). Moisture retention curve for these samples were determined at matric potential of 0، -10، -33، -100، -300، -500، -1000 and -1500 kPa. Particle size distribution، calcium carbonate، bulk density، organic carbon، EC and SAR were measured and RETC program was used to obtain the Van Genuchten equation parameters. As to the EC and SAR effect on pore size distribution of soil، these parameters were used in pedotransfer functions as predictor parameters. Also، performance of derived functions in estimation of moisture retention curve was compared with a number of functions (13 functions) that previously were provided by different researchers. Generally، the result of this study showed: 1- Water retention at a pre-defined potential and parameters of van Genuchten equation can be estimated by pedotransfer functions with reasonable accuracy (point transfer functions: 0. 844 ≤ R2Adj ≤ 0. 577 and 0. 0296 ≤ RMSE(cm3/cm3) ≤ 0. 0121)، 2- Sensitivity analysis results showed that the sensitivity of most point functions to changes in bulk density was more than the rest of the parameters and clay-silt ratio، EC، and sand were at the next levels، 3- Considering the coefficient of EC in the equations، partial coefficients of correlation، and results of sensitivity analysis، higher amounts of EC could somewhat affect pore size distribution and، consequently، the shape of moisture retention curve of the soils. This change، which is caused by swelling and shrinkage، to some extent can be predicted via introduction of EC and SAR parameters in developing the equations،4- Comparison of previous functions with those obtained in this study show that derived functions in this research are the best functions for estimation of soil moisture retention curve.