Monitoring of Land Use Changes Using Google Earth Engine for Agriculture: Indramayu Regency

Indramayu is pivotal for national food security, acting as a buffer with designated sustainable agricultural land. Amidst industrial development in the Rebana Area, including six Industrial Designated Areas, potential threats to sustainable agricultural land arise from land conversion. This study utilizes big spatial data and machine learning on time-series satellite imagery to analyze past, current, and future prediction analyses. The data used are sentinel-2 imagery and NDVI, NDWI, and BSI index derivatives. Using Google Earth Engine (GEE), a dashboard is created to visualize, create, and store datasets, and the dataset results will be classified for land use and land cover (LULC) using the Random Forest algorithm. On the GEE dashboard, a dataset was efficiently generated through the implementation of several processes: timing window selection, cloud filtering, spectral index calculation, imagery integration, and classification. The classification result shows that the best model for Sentinel-2 in Indramayu Regency uses random forest pixel-based analysis compared to object-based analysis. Using GEE, monitoring LULC changes can support Indramayu's vision to see the potential and prospect of a national food reserve, contributing to informed agricultural development decisions for enhanced food security.