JoWUA

Abstract

Satellite-based NDVI requires optical satellite data, and the main challenge lies in acquiring clear-sky (cloud-free) images for regional agricultural crop monitoring during the monsoon season over the Indian subcontinent. Unfortunately, persistent cloud cover during that time frequently results in significant data gaps that reduce optical-based monitoring to varying degrees. Overcoming this aspect of NDVI, a study emphasized the alternative to NDVI for cotton crop growth monitoring, in the name of Radar Vegetation Index (RVI). Multi-temporal Sentinel-1 SAR and Sentinel-2 optical data were processed using the cloud-based geospatial processing platform Google Earth Engine (GEE) for the Musi watershed in Telangana, India, over the period from March 1, 2020, to March 31, 2021. Using GEE for scalable processing, we were able to generate time-series NDVI and RVI automatically without requiring dedicated local storage or high-end computing resources. From time series analysis, it was possible to notice that NDVI values were stable after the flowering cotton stage by the second week of October; however, despite stabilization of NDVI until November, coincident with boll formation in cotton. This longer-range sensitivity makes RVI a critical metric to track for later-stage growth. The study concluded that RVI can provide a robust and fast substitute for cotton monitoring concerning NDVI, particularly in cloudy regions. Operational large-scale crop monitoring will perform flawlessly, even in the peak monsoon season, by utilizing the GEE cloud computing environment.