Scattering power components from dual-pol Sentinel-1 SLC and GRD SAR data

Abstract

Accurate land cover information is pivotal in numerous planning and management activities. Synthetic Aperture Radar (SAR) data has emerged as a valuable resource for land cover assessment. Extracting scattering power components from Polarimetric SAR (PolSAR) data provides essential insights into the characteristics of land cover targets, aiding in their detailed characterization. Various techniques in radar polarimetry have been introduced to obtain scattering power components and explore their significance in land cover applications, typically leveraging full-polarimetric (full-pol) SAR data. However, applying these methods directly to dual-polarimetric (dual-pol) HH-HV or VV-VH data presents challenges due to the limited polarimetric information available compared to full-pol SAR data. Distinguishing between the scattering mechanisms of dihedral (or double-bounce) targets and trihedral (or surface) targets based solely on conventional polarimetric information is particularly challenging in dual-pol data compared to its full-pol counterpart. To address this limitation, our study introduces two indices (DpRBI and DpRSI) designed to characterize scattering from “dihedral-like” (e.g., buildings, bridges) and “surface-like” (e.g., water bodies, bare fields) targets in dual-pol SAR data. These indices, measuring the similarity of a given pixel to either a “dihedral-like” or a “surface-like” target, are then used in the Scattering Power Factorization Framework (SPFF) to derive the proposed scattering power components. The SPFF discerns the primary scattering power contribution within a resolution cell by factorizing the total power, employing the two indices. Our method applies to both Single Look Complex (SLC) and Ground Range Detected (GRD) dual-pol SAR data from operational platforms such as Sentinel-1 and the upcoming NISAR missions. We assess the proposed methodology using Sentinel-1 SLC and GRD SAR data for two scenarios. (1) Spatial variation in the scattering power components over various land covers near Delhi, India; Milan, Italy; and Barcelona, Spain. (2) Temporal variation in the scattering power components over rice-cultivated areas near Vijayawada, India. During these assessments, we observed significant variations in scattering power components over different land cover targets and rice-cultivated fields at various phenological stages. This highlights the capability of the proposed methodology to discern scattering information from dual-pol SAR data.