Analysis of Mangrove Vegetation and Distribution Using Landsat 8 Images In Bolaang Mongondow East, North Sulawesi

Mangrove is one of the objects that can be identified by remote sensing technology using satellite imagery. Analysis of the distribution and density of mangrove vegetation using Landsat 8 imagery was carried out in Bolaang Mongondow Timur, North Sulawesi in September 2020. This study aims to map the distribution of mangroves and determine the correlation between NDVI values, canopy cover, and mangrove density. The data analysis used Landsat 8 images with ENVI 5.3 and ArcGIS 10.1 software. Maximum likelihood classification is used to separate mangrove and non-mangrove features. The calculation of mangrove vegetation density using the NDVI algorithm and single-channel classification using the density slice method to divide mangrove density based on the range of pixel values of the NDVI image. Next, to test the accuracy of the classification results using an error matrix (confusion matrix) and the NDVI vegetation index correlation test compared with canopy cover and density data. The classification resulted in four different land cover classes with an overall accuracy of 97.70% and a kappa coefficient of 0.9688. The mangrove vegetation distribution from the classification results is 524.75 ha. The NDVI correlation with the percentage of canopy cover is very significant with a correlation coefficient (r) = 0.9516, while the NDVI correlation with density resulted in moderate correlation (r = 0.5315).