Remote sensing-based empirical modeling using sentinel-1 for monitoring vegetation and agricultural dynamics in semi-arid areas

Estimating vegetation parameters is essential for comprehending and managing ecosystems in scientific and environmental fields. The accurate evaluation of crop parameters will also greatly facilitate the optimum use of resources along with increasing productivity in crop production for sustainable agricultural management. Through a new dual-polarization vegetation-edge technique that relies on $${\sigma }_{VV}$$ and $${\upsigma }_{VH}$$, this research estimates crop biomass from Sentinel-1 SAR data in semi-arid regions. Analysis can be made quantitatively by comparing biomass predicted by DPBVI against the actual in-situ biomass and NDVI, which yield R2 of 0.72 and 0.52, respectively. A backscatter responses shows that, vertical crops yield stronger orientation backscatter than horizontal types, whereas density responses are highest for Peanut, followed by Sesamum Indicum and Oryza sativa.A study found that, biomass estimation method using DPBVI is developed for semi-arid regions, distinguishing vegetation structures: horizontal crops have small surface backscatter while there would be more scattering associated with the canopy density and moisture of vertical crops. The Google Earth RGB visualization was used to visually confirm global model spatial patterns. The visible texture variation was consistently observed in DPBVI responses. This biomass estimation guarantees that the developed model is both pertinent and useful for various global semi-arid agricultural systems, addressing the unique characteristics of vegetation growth patterns across diverse terrains.