Using categorical soil structure information to improve soil water retention estimates of tropical delta soils

Abstract

Models of soil water and solute transport require input data of soil hydraulic properties (e.g. soil water retention and hydraulic conductivity curves). Lack of such data, especially in tropical delta regions, has usually been the main constraint for the application of simulation models. Direct field or laboratory measurement of soil water retention is costly, laborious and time-consuming; therefore, indirect estimation from other easily measured soil properties has received great interest. However, indirect estimates are often unreliable. In this study, we hypothesise that including basic descriptive information of soil structure such as aspect of presence or absence of pedality can improve the prediction of the soil water retention characteristic (SWRC). Stepwise multiple linear regression was used to develop point pedotransfer functions (PTFs) to estimate soil water retention at eight pressure potentials (e.g. -1, -3, -6, -10, -20, -34, -100, -1500 kPa). Soil structural information was exploited as a preliminary grouping criterion to test our hypothesis. Soil samples were taken from 160 horizons distributed along the Mekong Delta, Vietnam. The results reveal that SWRC of tropical Mekong Delta soils could be satisfactorily estimated by typical predictors of PTFs (e.g. soil texture, organic carbon content and bulk density). Moreover, incorporating soil structure in developing PTFs did improve the prediction accuracy of SWRC, especially in the wet moisture range. Plastic limit was found to be a promising predictor for SWRC-PTFs of soils having a given degree of structural development.