Mapping Mangrove Species Using Hyperspectral Remote Sensing

Mangrove forests are in constant flux due to natural as well as anthropogenic forces and the changing mangroves could result in climate change; sea level rise. Therefore, observation and monitoring of the distribution and dynamics of mangroves is highly important. The advancement in remote sensing data availability, image processing methodologies, computing and information technology and human resource development have provided an opportunity to observe and monitor mangroves from local to global scales on a regular basis.

Mangroves are woody, seed bearing, highly specialized plants ranging from shrubs to tall trees which occur along sheltered intertidal coastlines in association with estuaries and lagoons. The mangroves in Sri Lanka are composed of 14 species of true mangroves and 12 species of mangrove associates. The most extensive mangroves occur in Puttalam – Kalpitiya area in association with estuaries. Dense localized stands also occur in association with estuaries in Southern, South Western and North Eastern coasts namely; Koggala Lagoon, Kalametiya Lagoon and Kokilai Lagoon. The mangrove systems covering an area of 6000-7000 ha are interspersed along the coastline of Sri Lanka.

  This project mainly uses the PSR-1100F portable Spectroradiometer for data collection and Hyperspectral satellite images obtained from EO-I satellite Hyperion sensor. The main objective of this project is to find the best method to classify hyperspectral satellite images to clearly demarcate different mangroves under species level for further monitoring activities.