A Unifying Platform for Water Resources Management Using Physically-Based Model and Remote Sensing Data

In recent years, physically-based hydrological models provided a robust approach to better understand the cause-effect relationships of effective hydraulic properties in soil hydrology. These have increased the flexibility of studying the behavior of a soil system under various environmental conditions. One disadvantage of physical models is their inability to model the vertical and horizontal heterogeneity of hydraulic properties in a soil system at the regional scale. In order to overcome this limitation, inverse modeling may be used. Near surface soil moisture, which has been collected routinely by remote sensing (RS) platforms, and evapotranspiration, that is also a pivotal key for water balance near the land surface can be used as alternatives for quantifying the effective soil hydraulic parameters through inverse modeling. However, the new approach suffers from not only the scale discrepancy between RS pixel resolution and model grid resolution, but also its application in complex terrains. Furthermore, hydrological models require a number of required input parameters. Hence, this dissertation focuses on developing a methodology for addressing these problems. The field-scale Soil-Water-Atmosphere-Plant model (SWAP) was extended to regional application, and then coupled with a Genetic Algorithm (GA), to operate as the core of the developed decision support system at the regional level. Also, various stochastic processes were developed and applied to the GA for improving the searching ability of optimization algorithms. The computational simulation-optimization approach was tested and evaluated under various synthetic and field validation experiments demonstrating that the methodology provided satisfactory results. In this dissertation, the proposed methodologies analyzed the spatio-temporal root zone soil moisture with RS and in-situ soil moisture data at the multiple scales. Also, these approaches could provide better input parameters for hydro-climatic models, resulting in better understanding of the hydrologic cycle. Thus, a better understanding of water cycle would help us to be better prepared for efficient water resources management, agriculture, and devastating natural disasters in the real world