Estimation of Global Diurnal Soil Moisture Dynamics from Various Satellite Systems using Bayesian Machine Learning, Deep Learning, and Data Assimilation
- Dr. Julianne Quinn (Chair) - ESE
- Dr. Venkataraman Lakshmi (Advisor) - ESE
- Dr. Jonathan Goodall - ESE
- Dr. Lawrence Band - EVSC & ESE
- Dr. Todd Scanlon - EVSC
- Dr. Sujay Kumar - NASA, Goddard Space Flight Center
Over the last several decades, researchers have proposed various methods of estimating near-surface soil moisture values using satellite microwave sensors. This work is critical since soil moisture estimates at regional scales are necessary for operational applications such as water resource and irrigation management, near-real-time numerical weather prediction, hydrological modeling, and understanding many other surface processes. However, satellite observations have two major limitations: they are not spatially or temporally continuous and have low spatial resolution and temporal repeat. This proposal outlines the utilization of current microwave satellite systems and global navigation satellite systems (GNSS) to obtain surface soil moisture data with high temporal repeats. It also proposes the use of Bayesian machine learning, deep learning, and data assimilation methods to understand the error characteristics of soil moisture data and to improve the quality of soil moisture data from land surface models. The ability to estimate soil moisture at high temporal resolution and accuracy would help us to better understand the role of hydrometeorological factors in extreme events and to accurately parameterize land surface variables in hydro-climatological studies.
*For Zoom information, please send an email to firstname.lastname@example.org.