Measurement and prediction of soil moisture in a medium-sized catchment

Abstract

Catchments with a small elevation range and relatively long dry periods in high radiation conditions may be described as an array of vertical one-dimensional pathways for water and energy. Such a representation makes it possible for Soil-Vegetation-Atmosphere Transfer (SVAT) models to simulate mass exchange across the catchment. This note reports on a comparison of a Soil-Vegetation-Atmosphere Transfer model and a stochastic hydrological model. We have used a stochastically distributed one-dimensional SVAT model (SiSPAT), which transfers any runoff generated at a particular location instantaneously to the catchment outlet. The hydrological model used here is the quasi-distributed Variable Infiltration Capacity model of Wood et al. (1992). Our results indicate that the uncalibrated SiSPAT model does not yield better results than the uncalibrated VIC model. However, even the uncalibrated VIC model requires knowledge of the maximum and minimum observed storage capacity to derive soil water status. The results also show that a calibrated VIC model performs better than the uncalibrated SiSPAT, but that minimum storage calibrated against the observed runoff does not retain its physical significance, since it is larger than the mean storage capacity.