Field scale computer modeling of soil moisture with dynamic nudging assimilation algorithm

Abstract

Soil moisture analysis is widely used in numerous practical cases, from weather forecasts to precise agriculture. Recently, availability of moisture data increased due to the rapid development of satellite image processing. However, satellite retrievals mostly provide low-resolution surface data. In this study, we attempt to retrieve surface soil moisture on the field scale using a decomposition algorithm. Furthermore, we add a mathematical model based on Richards equation to evaluate soil moisture in the root zone. To combine the results of both models, we employ a nudging data assimilation technique. Also, a dynamical variation of the method is proposed which makes it more adaptive to the soil type and provides improvement to modeling results. Two types of numerical experiments are conducted. Simulation results show reasonably good convergence with the measurements. The model performs with average correlation of 0.58 on the whole root zone, reaching 0.85 on top soil layers.