Pedotransfer functions for predicting points on the moisture retention curve of Indian soils

Limited numbers of pedotransfer functions (PTFs) for predicting the soil moisture transmission parameters are available for the Indian soils. These PTFs have been developed from limited datasets and are area specific, limiting their applicability elsewhere. Secondary data from various sources often have non-systematic errors and lacking in uniformity. The present study makes an attempt to explore the possibility of developing PTFs from wide textural range of Indian soils for four points on the moisture retention curve, –33, –100, –500 and –1 500 kPa. Moisture content at these four matric potentials was significantly related to clay and sand but little or no correlation was observed with bulk density and organic C content in soils. For each potential, six sets of regression equations were used to predict the moisture content from textural composition, bulk density and organic C content. The predictive potential is better by considering only clay and silt at –33 and –500 kPa while clay, silt and bulk density gave good moisture predictability at –100 and –1 500 kPa potential values. The developed PTFs for moisture contents at –33 and –1 500 kPa were evaluated with existing PTFs and were found to perform satisfactorily. The process-based simulation models require inputparameters often difficult to measure directly; consumes large time and resource; and the spatial variability associated with the measured values. Efforts have been made to relate this difficultly measurable parameters, viz soil moisture retention and transmission characteristics with easily measurable and routinely available parameters like soil textural separates (mainly sand, silt and clay) through development of several functions, commonly known as pedotransfer functions or PTFs. The application of these PTFs has now become an important part for discipline of soil science and its associated database management (Singh 2004).