Efficient Prediction of Root Zone Soil Moisture Content with Improved Force-Restore Method.

Root zoon soil moisture is considered as a key variable controlling surface water and energy balances. Numerous researches have been done to estimate the soil moisture variation as a time series from meteorological data using the so-called force-restore method (FRM). The difficulty to estimate the variation is attributable to the hardness of accurate evaluation of the soil water flux passing through the lower plane of a root zone under various conditions. We have investigated the distribution of soil water flux during drainage and evapotranspiration on numerical solutions. Approximately, the flux in the root zone was described by a product of two functions of time and depth, the time function expressed the flux variation at a lower reference plane and the function of depth described the distribution of relative flux to reference one. The coefficients contained in the functions were determined on soil texture and soil moisture variation data. Incorporating this approximate expression into FRM, a simple formulation was obtained for prediction of soil moisture variation including the case of existence of upward flux from the lower layer. Assimilation of soil moisture data with this model framework was also examined.