Soil moisture plays a critical role in many environmental and hydroclimatic applications. In fact, soil moisture is so central to understanding land and atmosphere exchange that the 2007 Phoenix Mars Mission included devices specifically designed to measure soil temperature, thermal properties, and water content. Recognizing the importance of soil moisture as a vital state variable, the 2007 NRC Earth Science Decadal Survey report recommended the satellite mission Soil Moisture Active/Passive (SMAP) for global soil moisture mapping. There are other ongoing global missions, such as the C-ban AMSR-E sensor on the AQUA satellite, which provides soil moisture of the top 1cm of the earth?s skin with two orbital daily coverages in the Southern Great Plains of the U.S. Routine measurements by remote sensing are unable to capture rapidly shifting near-surface soil, heat, and water transport processes despite their importance to understanding land surface hydrology. Because of that, the need for mathematical models to estimate such variabilities and to predict large-scale behavior is more imperative than ever.
The proposed work focuses on developing predictive multiscale models for soil moisture dynamics based on liquid and vapor flow coupled with heat transport. These models take into account complex nonlinear interactions at different spatial and temporal scales. Uncertainties at the fine scale are modeled via probabilistic models. The proposed methodologies are designed to readily incorporate available dynamic data at different scales and, thus, will provide an improved guidance for ground validations of upcoming space missions concerning measurements of soil moisture and temperature for heterogeneous subsurface environments.
