Over the last 15 years, remote sensing has played a crucial role in mapping and understanding changes in the areal extent and spatial pattern of mangrove forests caused due to natural disasters and anthropogenic forces. While traditional pixel-based classification of multispectral imagery has been widely applied for mapping mangrove forests, more recent types of satellite imagery like hyperspectral data, taken from sensors like Hyperion combined with sub-pixel classification algorithms is expected to demonstrate the potential for reliable and detailed characterization of mangrove forests including species level classification. This paper brings forth the recent advancements in hyperspectral data and classification techniques and describes opportunities for integration of high spatial and spectral data for species level identification of mangroves. Future prospects include the application of existing methods in natural resource management and overcoming challenges in the global monitoring of mangrove forests.

Key words: Hyperspectral, Spectral Library, Ground Truthing, End Member Determination, Linear Spectral Unmixing, Non-linear Spectral Unmixing