Enhancing Forest Canopy Height Mapping in Kaziranga National Park, Assam, by Integrating LISS IV and SAR data with GEDI LiDAR data Using Machine Learning

A significant increase in forest canopy height in Kaziranga National Park, Assam, was observed between 2018 and 2022, with mean heights rising from approximately 8.58 meters to 9.07 meters. This study pioneered the integration of GEDI LiDAR data with LISS IV optical imagery, Sentinel-1 SAR, and PALSAR data using machine learning models to map canopy height continuously. Extreme Gradient Boosting (XGB) emerged as the top-performing model, achieving an RMSE of 5.47 meters and an R2 of 0.55, outperforming Random Forest, Support Vector Machine, and k-Nearest Neighbors. The analysis revealed that over 35% of the park's area had canopy heights below 5 meters, while taller canopies beyond 20 meters were found in less than 5%. This novel approach not only enhances canopy height mapping but also lays the groundwork for future research on biomass assessment and carbon sequestration in this critical biodiversity hotspot.

Bhuyan, M., Jeganathan, C., & Pujar, G. S. (2024). Enhancing Forest Canopy Height Mapping in Kaziranga National Park, Assam, by Integrating LISS IV and SAR data with GEDI LiDAR data Using Machine Learning. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, X-3-2024, 39–44. https://doi.org/10.5194/isprs-annals-x-3-2024-39-2024