Monitoring Mangrove Health: Analyzing Degradation and Biomass Carbon in the Sundarban Mangrove Forest Using Satellite Imagery and GEOAI.

Keywords: Mangroves, biomass, GEE, GeoAI, Sundarban
Abstract Type: Poster Abstract

Abstract

Mangroves are typically a focal point of interest to environmental scholars due to their rich biodiversity, natural resources, climate change resilience and also because of their contribution as a buffer against coastal hazards. The sundarban mangrove forest of Bangladesh and India is considered to be the largest contiguous mangrove forest in the world which is threatened by climate change impacts and anthropogenic disturbances. Monitoring mangrove degradation and changes in biomass carbon is essential for understanding forest health and informed conservation efforts. This study incorporates Google Earth Engine (GEE) and GeoAI techniques to analyze the spatiotemporal changes in mangrove cover and biomass carbon storage in the Sundarbans. Machine learning classifiers like Random Forest and Support Vector Machines (SVM), and high-resolution Sentinel imagery will deliver highly accurate forest cover change with time. Moreover, biomass carbon analysis will be conducted within GEE platform, utilizing GEDI dataset and vegetation indices to estimate the variations. By combining machine learning, remote sensing, and cloud-based geospatial analysis, this research will provide a scalable methodology for assessing mangrove ecosystem vulnerability. Efficient monitoring of mangrove ecosystems will provide insights into degradation trends and carbon sequestration dynamics. The findings will contribute to conservation strategies, climate adaptation planning, and carbon offset initiatives. Understanding biomass carbon changes in degraded areas can support policy-making efforts to protect and restore these critical coastal ecosystems. This study highlights the power of GEOAI and satellite-based techniques in advancing sustainable environmental monitoring and management.