Modelling Techniques in Synthetic Aperture Radar (SAR) for Geophysical Parameter Retrieval

Abstract

Abstract When cloud cover or poor light impair optical data, Synthetic Aperture Radar (SAR) modelling is crucial for deriving geophysical information from radar observations. The main SAR backscatter modelling techniques are examined in this article, including theoretical, semi-empirical, and empirical models including Oh, Dubois, and the Integral Equation Model (IEM). They are compared in terms of their assumptions, input specifications, retrieval efficiency, and suitability for soil moisture, surface roughness, and dielectric characteristics. Limitations pertaining to surface heterogeneity, model calibration, incidence angle, and vegetation effects are also covered in the paper. Future studies should concentrate on combining machine learning and multi-source SAR data with hybrid, site-adaptive, and physically limited models.