Mapping Mangrove Area Changes In Brebes, Jawa Tengah Province Using Google Earth Engine

Abstract

Fluctuations in mangrove area on the north coast of Kaliwlingi Village, Brebes Regency are attributed to abrasion and the transformation of mangrove areas into ponds. It is crucial to conduct an analysis to monitor the changes in the mangrove area between 2000 and 2023. This study integrates the Mangrove Vegetation Index (MVI) algorithm for distinguishing mangroves from non-mangroves and Random Forest (RF) for assessing area change, executed on Google Earth Engine (GEE). The research shows that the mangrove area in Kaliwlingi Village was 119,104 ha in 2000, progressively expanding to 489,474 ha by 2023. Accuracy test results show that the MVI algorithm is representative for mangrove and non-mangrove classes, achieving an OA of 90.9% and a kappa coefficient of 0.816. Similarly, RF exhibits high test accuracy with an OA of 89.39% and a kappa coefficient of 0.727. To conclude, both algorithms prove valuable for mapping mangrove distribution in Kaliwlingi Village, Brebes Regency, Jawa Tengah Province. For further research, it is advisable to distribute sample test points more evenly, take high-resolution drone photos to validate the data, and perform mangrove density analysis.