Cygnss Soil Moisture Estimation Using Machine Learning Regression

Global Navigation Satellite System-Reflectometry (GNSS-R) can retrieve Earth's surface parameters, such as soil moisture (SM) using the reflected signals transmitted from GNSS constellations. GNSS-R has advantages of non-contact, large coverage area, real-time, and continuity. The CYclone GNSS (CYGNSS) data used for SM retrieval have generated considerable interests. In this paper, estimating SM on a global scale is performed using machine learning (ML) regression. The the optimal XGBoost predicted model with root mean square error (RMSE) of 0.064 cm3/cm3 is adopted. In addition, satisfactory daily SM estimation outcome with an overall correlation coefficient value of 0.86 is achieved at a global scale.