Mapping mangrove forest density distribution and deforestation with machine learning on Google Earth Engine in West Nusa Tenggara Province

Mangrove forests play a very important role in reducing greenhouse gas emissions as part of efforts to mitigate climate change in Indonesia. The existence of mangrove forests is sensitive to extreme environmental changes, and high levels of utilization increase the risk of deforestation. The purpose of this study is to provide information regarding distribution, density, and deforestation of mangrove forests in 2023 in West Nusa Tenggara Province. Data processing was carried out based on cloud computing using Google Earth Engine with a random forest algorithm and supervised classification, utilizing Landsat satellite imagery data from 2000 and 2023. The classification of mangrove forest density was performed by interpreting the satellite imagery from 2023 using the NDVI (Normalized Difference Vegetation Index) method. Field surveys were conducted at 40 sample points to observe the presence and condition of mangrove forests that experienced deforestation in 2023. The results of image data processing were validated using field data and high-resolution images from Google Earth, with an accuracy assessment method presented in the error matrix. The results showed that mangrove forests have a sparse density of 4968.10 ha, a medium density of 2516 ha, and a dense density of 4928 ha. The total deforestation area in West Nusa Tenggara Province is 610 ha, with Dompu Regency having the largest percentage of deforestation, accounting for 68% of the total mangrove deforestation in West Nusa Tenggara Pronvince. The destruction of mangrove forests is mostly caused by the fisheries sector, namely the conversion of mangrove forests into pond sites.Keywords: Deforestation; Google Earth Engine; Mangrove Forest; NDVI; West Nusa Tenggara