<![CDATA[This study aims to analyze the dynamics of mangrove cover in the Kuala Lupak Nature Reserve, South Kalimantan, using multitemporal Landsat imagery and Random Forest classification on Google Earth Engine. Two vegetation indices, Normalized Difference Vegetation Index (NDVI) and Mangrove Vegetation Index (MVI), were applied to improve mangrove discrimination across observation years (1990, 2000, 2010, and 2020). The results indicate that MVI tends to provide more stable classification performance compared to NDVI, with overall accuracy values generally exceeding 0,90. The analysis of mangrove change shows a decreasing trend over the 30-year period, with a total loss of 1,857.15 ha and a gain of 318.15 ha, resulting in a net loss of approximately 1,539 ha. However, these results are derived from classified satellite data and may be influenced by classification uncertainty, sensor differences, and environmental variability. Therefore, the findings should be interpreted as indicative trends rather than exact measurements. This study highlights the potential of integrating remote sensing and machine learning approaches for long-term monitoring of mangrove ecosystems and provides useful information to support mangrove conservation and management strategies.]]>
Copyrights © 2026
