Concluding Remarks

Abstract

AbstractOne of the earliest results was to show a detailed understanding of soil water and heat dynamics, which demonstrated how the thermal-driven or isothermal-driven water fluxes can alternatively dominate in soil on a daily scale. Then, the diurnal soil moisture and heat dynamics were applied in the Badain Jaran Desert to determine the diurnal variation of the drying front after a rainfall event, which is subsequently applied to determine the effective infiltration. Although the traditional coupled moisture and heat transport model is able to explain the field experiment quite well, the single-phase transport mechanism limits its capability of describing the two-phase flow phenomenon. In order to discuss the uncertainty caused by such a simplification, a two-phase mass and heat transport model is proposed fully taking diffusion, convection, and dispersion mechanisms into consideration. The proposed model was then applied to analyze how the vapor transport in soil affects evaporation on the soil surface. It was found that there was a surprising underestimation error of evaporation when soil airflow was ignored. To understand the mechanism of the error, a simulation analysis of the advective effect on evaporation was implemented. Furthermore, the proposed model is implemented in a data assimilation system to check the impact of different model physics complexities on retrieving soil state variables by using an ensemble Kalman filter. The result explains that an optimal combination may exist between the model physics chosen and the data typically available for being assimilated, for gaining the best retrieval of soil state variables. KeywordsSoil MoistureLand Surface ModelData Assimilation SystemBadain Jaran DesertSoil Water DynamicThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.