Multi-decadal land cover evolution in the Sundarban, the largest mangrove forest in the world

Abstract

Having one of the most vulnerable coasts of the world to climate change, Bangladesh and India are facing new challenges to protect Sundarban, the largest mangrove in the world, from adverse impacts of climate-driven changes. In this study, unsupervised classification of mangrove density is performed on Landsat images of four time periods over 35 years (1975-77, 1989, 2000 and 2010) to identify the spatio-temporal evolution of the Sundarban mangrove, and infer the causes of observed changes. Specifically the influences of environmental and anthropogenic drivers on these changes are discriminated by correlating the observed change patterns with cyclones' tracks, erosion processes and areas of expected human influences as assessed by local stakeholders. Results highlight a very dynamic system, with rapid changes characterized by decreasing mangrove density, mainly in India, erosion and accretion. There is ∼1.2% (or 11,500 ha) net erosion in 35 years in Sundarban, which remains limited relative to the documented high rate of sea level rise of 6 mm/yr. Mangrove density changes are attributed to cyclones, increased salinity and anthropogenic influences. Based on interpretation of the spatial pattern of density changes by local stakeholders, we conclude that the zones accessible to human activities are significantly more affected by land cover changes. Management policies of Sundarban by Bangladeshi and Indian authorities should acknowledge the important role played by natural or global drivers like sea level change, cyclones in affecting the mangrove as well as manage the localized anthropogenic pressures.