Evaluation of Spatio-Temporal Forest Health of Bangladesh Using Google Earth Engine

Bangladesh has a significantly low and steadily declining land covered in forest compared to the required one-third. Therefore, it is essential to explore the condition of the forest cover in Bangladesh. In recent years, remote sensing techniques have become increasingly popular for assessing the health of forests. The research evaluates the forest health seasonality and spatiotemporal variability. Landsat 7 ETM+, Landsat 8 OLI, and Sentinel-2 images for 2002-2021 and seven vegetation indices are used in the Google Earth Engine platform as it is widely accepted and convenient. The results reveal the time series analysis of vegetation indices; they show a maximum value of 0.8107 for SAVI in Sundarban and a minimum value of 0.0146 for NDVI in the Dinajpur and Hill Tract areas. Also, spatial variability illustrated a maximum value of 0.8107 for SAVI in Sundarban and a minimum value of 0.0146 for NDVI in the Dinajpur area. Moreover, Seasonal patterns are also identified where forest health is best observed during the monsoon season (July - October). Furthermore, the assessment indicates that the south and southeastern portions of the research region, Sundarban, and the Hill Tract area have healthier forest cover than the others. This study could be considered a comprehensive reference for managing and planning forests.