Abstract
In Sri Lanka, mangrove forests are scattered along the north-western, north eastern, Jaffna Peninsula and eastern coastal belt. The total estimated extent of mangroves in the country is about 87 km2. Estimation of Above Ground Biomass (AGB) of mangroves is a challenging task due to field sampling difficulties. Use of satellite based remote sensing technologies is becoming popular for estimation of AGB for different vegetation types. To overcome the limitations during field sampling and to identify the possibility of using SAR data for AGB estimation, ALOS PALSAR satellite data were used to estimate AGB of mangroves and associated vegetation in Muthurajawela- Negombo wetland in Sri Lanka. Diameter at Breast Height (DBH) measurements over 5cm of eighteen (18) sampling plots (10x10 m) were collected and the relevant allometric equation was used to estimate the AGB. Backscatter coefficient values of HH and HV polarization of ALOS –PALSAR images were used to estimate the AGB of mangroves using a previously derived model. Finally, an AGB map of mangrove associated vegetation was developed for the study area using ALOS PALSAR data as a method of minimizing field work while saving time and cost. According to the results, the average AGB is observed as 65t/ha from the field sampling method (28 -135 t/ha) while it was estimated as 76 t/ha (33-155 t/ha) using PALSAR which shows an overestimation by 17%. A significant overestimation by the remote sensing method is occurred when the tree height is more than 5 m. Though it shows an overestimation, the map developed using this approach is helpful to understand the distribution of AGB within mangrove associated vegetation systems where field sampling is a challenging task.