10/10/2025
16:00 - 16:30
Collabotorium
, Hybrid
Session with interpretation
Why attend
This session explores how AI-powered geospatial intelligence integrates hyperspectral and multispectral satellite imagery with machine learning to revolutionize habitat mapping and soil classification. Attendees will discover how this technology enhances conservation decision-making, enabling real-time environmental monitoring and predictive analysis for land-use planning, ecological restoration, and biodiversity management.
Session Description
Traditional biodiversity conservation efforts struggle with fragmented and outdated ecosystem data, limiting the effectiveness of conservation strategies. This session introduces an AI-driven geospatial intelligence system that integrates hyperspectral (EMIT) and multispectral (Sentinel-2) satellite imagery with machine learning models such as XGBoost and Random Forest to achieve highly accurate soil classification (93%) and habitat delineation (94%). By leveraging this innovative approach, conservationists, land managers, and policymakers gain access to a scalable and predictive tool for land-use planning, climate adaptation, and biodiversity management. The session will present real-world case studies, including mangrove conservation and afforestation planning, demonstrating how AI-driven conservation tools can be applied to local and global conservation efforts. Attendees will leave with insights into how cutting-edge AI models can transform biodiversity conservation from a reactive practice into a proactive, data-driven discipline.Organised by
Commission on Ecosystem Management
Partners
Speaker
