Integrating water indices and cloud-based engine for change detection of aquaculture areas in Lampung, Indonesia

Abstract

Population expansion and climate change have significantly affected the coastal environment in Lampung, Indonesia, mainly through the conversion of mangroves into shrimp-farming ponds. This transformation requires effective monitoring to evaluate its impacts on coastal ecosystems and local livelihoods, as shrimp farming is a major income source in East Lampung. This research improves aquaculture detection and monitoring along the eastern coast of Lampung by integrating several waterindices such as the normalized difference water index (NDWI), modified NDWI (MNDWI), water ratio index (WRI), and a newly developed water index (WI), within the cloud-based Google Earth Engine (GEE) platform to capture spatial and temporal variations. Reference data were derived from the 2019 Regional Medium-Term Development Planning Document (RPJMD) and high-resolution Google Earth imagery for accuracy assessment. Results showed that WRI combined with the Otsu’s thresholding method achieved the highest performance, with an overall accuracy (OA) of 93.3% and a kappa coefficient (κ) of 86.7%. Analysis from 2018 to 2022 showed a decline in aquaculture area from 8,407.35 ha to 3,415.50 ha, aligned with statistical data on shrimp production, which decreased from 24,202 t to 8,041 t. These results indicate that the method provides a rapid and effective tool for detecting aquaculture changes, enabling local authorities to strengthen coastal management for sustainable development, ecosystem protection, and livelihood support.