Comparison of vegetation indices for mangrove mapping using THEOS data

Abstract

The increasing application of remote sensing for mangrove mapping and monitoring is practical for sustainable management of the biological resources. Over the past few decades, the emergence of several vegetation indices (VIs) has certainly given significant impacts on mangrove and other forest mappings. In this study, five different vegetation indices including Normalized Different Vegetation Index (NDVI), Simple Ratio (SR), Soil Adjusted Vegetation Index (SAVI), Perpendicular Vegetation Index (PVI) and Triangular Vegetation Index (TVI) were compared to discover a suitable vegetation index for identifying mangrove area in Pa Khlok sub-district, Phuket, Thailand. THEOS imagery with 15-m resolution from 2010 was utilized. Maximum Likelihood Classifier (MLC) was used to classify Mangrove and Non-Mangrove area. The results demonstrated that the best accuracy (96.78%) was from combination between 4 THEOS's spectral bands and some vegetation indices including NDVI, SR and SAVI.