ANALYSIS OF VARIOUS VEGETATION INDICES OF LANDSAT THEMATIC MAPPER SATELLITE DATA TO IDENTIFY THE DISTRIBUTION OF CORAL REEF (CASE STUDY IN ENGGANO ISLAND, DISTRICT OF NORTH BENGKULU, BENGKULU PROVINCE)

This research was aim at performing analysis of various vegetation indices of Landsat Thematic Mapper satellite
data to identify the distribution of coral reef in Enggano Island, District of North Bengkulu, Bengkulu Province. This research is based on the realities that: 1. Vegetation indices analysis are generally only used for objects on the land; 2. Analysis of coral reef using remotely-sensed data are mainly use multispectral classification technique, Shallow Water Image Mapping, and Lyzenga‘s. Materials used were mainly Landsat Thematic
Mapper satellite data and Map of Joint Operation Group. GIS Programs used to analyze were ILWIS (Integrated
Land and Water Information Systems) version 3.4 and ArcView version 3.5. Methods applied was by performing
analysis of various vegetation indices of Landsat Thematic Mapper satellite data, which are RVI (Ratio
Vegetation Index), NDVI (Normalized Difference Vegetation Index), and SAVI (Soil Adjusted Vegetation Index).
The result of every analysis of vegetation index then was used as a base for on-screen digitizing in areas where
the boundary of coral reef existence is probably exist such that can be gained digital data of coral reef. The
accuracy of the result was checked by overlaying it (overlay analysis) onto Map of Coral Reef derived from
topographic map published in 1942 (Map of Joint Operation Group, and used as a Reference). Analysis of
vegetation index is said to be “Good” if the accuracy is ≥ 80% (according to Daels and Antrop).The result of
research shows that the overall accuracy were 67.60% (RVI), 65.40% (NDVI), and 73.16% (SAVI). It revealed
that the accuracy of coral mapping using all vegetation indices can be said as ―Not Good‖.