Carbon stock assessment of mangroves using remote sensing and geographic information system

Abstract

Mangrove forests are among the most carbon-rich habitats on the planet. The protection of mangroves for mitigation of greenhouse gases in the atmosphere as well as for multifaceted sustainable growth of ecosystem is of great scientific concern. Mangrove vegetation in Kerala is now restricted largely to river mouths and tidal creeks with about 15 species of mangroves and 8 associated species. Kannur holds the largest mangrove area in Kerala with an area of about 32 Km2. Located at 12°15′ North and 75°13′ East, Kunhimangalam, in Kannur district, has luxuriant mangroves. A considerable part of mangroves are undergoing destruction and over exploitation. There are organised attempts to acquire all the mangrove areas under private ownership by the real estate groups. Although a large scale conversion of the mangroves was prevented by the timely intervention of local people and NGOs, long term preservation of this system is under question. The urge to establish this mangrove as a protected area is very relevant in this context. The aim of the study is to promote restoration of these mangroves through community and government participation. The study integrates field inventory data with the satellite images. Analysis involves four major steps, namely, (i) Image processing, (ii) derivation of vegetation indices using satellite imagery (iii) ground truthing through field stratification and collection of field inventory data and (iv) calculation of carbon stock. The satellite image of 18th January 2015 is used for the study. Using ERDAS IMAGINE 9.2, the layers are stacked and given standard False Color Composite (FCC) for better vegetation analysis using ArcGIS. The vegetation indices (VIs) can estimate the biomass of mangroves from remote sensing images and the most appropriate one is the normalised difference vegetation index (NDVI). The NDVI is based on absorption in the red spectrum and very strong reflectance in the near infrared spectrum. Ground truthing is done to establish relationship between NDVI from Satellite imagery and Above-ground biomass (AGB) from field observations. Reference pixels are selected as those for which actual data are known. A regression equation is developed to calculate the AGB for the entire plot from the NDVI values in the imagery. Allometric equations for Above-ground biomass (AGB) and below-ground biomass (BGB) developed by Komiyama et al. (2005) is used in this study. The overall ratio of BGB to AGB is 0.38. Total biomass is taken as the sum of AGB and BGB. Carbon content can be obtained by multiplying the total biomass by a conversion factor 0.475. AGB values ranges from 636.832 gm to 32048.5 gm per pixel and BGB ranges from 241.996 gm to 12178.4 gm per pixel. Total biomass ranges from 878.828 gm to 44226.9 gm per pixel and carbon content per pixel ranges from 417.443 gm to 21007.8 gm. Of the total 70.6% is contributed by high class vegetation which is covering 61% of the total area. Medium class contributes 16.74% of carbon with coverage of 29.55% area and low class provide 12.66% carbon sequestration with 9.44% area coverage. Comparing the area coverage, high and low class vegetation type is more effective in contributing carbon sequestration. Low class vegetation area is mainly covered with dry grass which has low reflectance and low NDVI, but has high carbon content. Rhizhophora mucronata, Excoecaria agallocha and Bruguiera cylindrical are the species that contribute more. The estimated total carbon of 12.67 tonnes and presence of eight different mangrove species in an area of 12 acres indicates this mangrove forest is in good health. Keywords: Mangrove, Carbon sequestration, GIS, NDVI, Ground-truthing