Estimation and Analysis of Change Detection, Forest Canopy Density, and Forest Fragmentation: A Case Study of the Indian Sundarbans

Abstract

ABSTRACT In this paper, satellite imagery is classified into four categories using the fuzzy c-means algorithm of the Indian Sundarbans due to its forest density changes. The categories are dense forest, sparse forest, water bodies, and wetlands. The study reveals that net forest areas declined by 3.75% from 932 km2 in 1975 to 847 km2 in 2018 and the rate of deforestation was 1.96 km2 year−1. The correlation statistic shows that the deforested areas were converted to wetland and water bodies. The results of the forest canopy density (FCD) model show that areas, with canopy density of 60–100% gradually declined from 42% (939 km2) in 1975 to 36% (814 km2) in 2018. Moreover, we also observed that maximum canopy density was >80% in 1990 and 60–80% in 1975. The results of the forest fragmentation model show that forest patch and edge areas progressively increased by 253% and 28%, respectively, while perforated forest areas slowly decreased with 11%. We find that most forest fragmentation happened in patch, edge, perforated, and core forest with an area >4 km2. Therefore, this study may be helpful in monitoring land cover changes of the Indian Sundarbans for sustainable mangrove forests.