Exploring Vegetation Cover Maps with Artificial Intelligence (EOSDA) Technology: A Pilot Study of Ramadi District, Iraq

Abstract

This research explores a novel approach using interactive platforms integrated with artificial intelligence to explore multi-vegetation indicator maps using Earth Observing System Data Analytics (EOSDA) technology. The research area, comprising 95,538 dunams of agricultural land in the Ramadi district (33.5183°N 43.0214°E), was selected for direct application of multiple analytical methods (VI), namely NDVI, NDMI, MSAVI, and NDRE, using data from the Sentinel-2A satellite at a resolution of 10 meters. The research compared maps from the 2022/2023 agricultural season and analyzed weather phenomena affecting crops. The findings demonstrated the platform's effectiveness in producing maps and monitoring vegetation cover across years. Variations in vegetation density were observed, indicating the impact of atmospheric conditions on agricultural lands. The platform provides Interactive maps feature a split view that allows for data comparison across different dates. This provides an opportunity to track the dynamics of crop changes in the field over time. Key recommendations include the necessity for research institutions to collaborate with companies producing these websites to obtain an up-to-date database for Iraqi regions. This will enhance scientific research resources and include a 20-year-old weather data archive. The report also recommends transitioning from software to geographic programming and algorithms, as well as monitoring agricultural lands using vegetation cover indicators based on climatic conditions, soil type, and crop requirements