Analysis of Multiple Scattering of Radiation amongst End Members in a Mixed Pixel of Hyperspectral Data for Identification of Mangrove Species in a Mixed Stand

The Sunderban Biosphere Reserve of southern West Bengal, India offers an ideal locale where Hyperspectral Remote Sensing may be applied for mapping diverse range of mangrove plantations that are unique to this pristine ecological system. This delta is characterized by both 'pure' and 'mixed' mangrove forest patches where photons interact with multiple species within the instantaneous field of view of the hyperspectral sensor, and multiple scattering between the vegetation canopies becomes significantly nonlinear. This paper attempts to apply the principle of multiple scattering of light in naturally occurring conditions and see how best they describe the resultant mixed spectra for certain endmember distributions. It is observed that natural forests, mixed mangrove forest in this case, are better suited for non-linear scattering models as compared to linear models. The non-linear model has successfully identified areas of interaction of mangrove mixtures such as ceriops-excoecaria-phoenix, excoecaria- avicennia alba avicennia officinalis etc. in the study area.