A Dynamic Model to Recognize Changes in Mangrove Species in Sunderban Delta Using Hyperspectral Image Analysis

Remote sensing and hyperspectral image data play an important role to map and monitor changes in mangrove forest mostly caused by natural disasters, anthropogenic forces and uncontrolled population growth. These have resulted in changes in the availability of natural resources and the physicochemical environment leading to a competition between the existing mangrove species for their survival. This paper applies hyperspectral data to obtain the fractional abundance of pure and mixed mangrove species at sub-pixel level and extracts the data to analyze inter-species competition in the study area over a period of time. The influence of different species traits such as rate of growth, rate of reproduction, mortality rate and processes that control coexistence of plant species that form the basis of competition analysis has been extracted from Hyperion time series image data of the study area. The image-extracted results have been verified through field visits. This study determines the dominance of certain mangrove species in a pixel area and examines its state of equilibrium or disequilibrium over a fixed time frame. This study has attempted to predict mangrove species dynamics within a mixed mangrove patch where they are interacting for coexistence or mutual exclusion.