Cloud-Based Monitoring of Twenty Years of Shoreline Change on Ko Phi Phi Don, Thailand, Using Google Earth Engine and Multi-Temporal Satellite Imagery

Coastal environments on tropical islands are increasingly threatened by the combined impacts of natural coastal processes, climate change, and rapidly expanding tourism. Reliable long-term monitoring of shoreline dynamics is therefore essential for supporting sustainable coastal management and protecting environmentally sensitive regions. This study investigates shoreline change on Ko Phi Phi Don, Thailand, between 2004 and 2024 using freely available satellite imagery processed within Google Earth Engine (GEE). All images were acquired during the dry season (November–May) to minimize cloud cover and ensure consistent shoreline visibility under comparable environmental conditions. The specific scenes used were 11 April 2004 (Landsat 5/USGS), 2 February 2014 (Landsat 8/USGS), and 15 February 2024 (Sentinel-2/Copernicus).

Shorelines were extracted using the Normalized Difference Water Index (NDWI), applying a fixed threshold of 0 to generate binary water–land classifications, which were then vectorized to delineate the shoreline boundary for each reference year. Geographic Information Systems (GISs) were used to quantify erosion and accretion patterns and evaluate spatial changes over time.

The analysis revealed pronounced spatial variability in shoreline dynamics across the island. The greatest shoreline retreat occurred along the western, northeastern, and southwestern coasts, with maximum retreat reaching approximately 117 m. In contrast, shoreline accretion was limited to small, localized sectors of the northwest coast. Area-based analysis showed that approximately 1.01 km² of coastal land experienced erosion during the twenty-year study period, whereas only 0.25 km² exhibited shoreline advance, indicating a clear net trend toward coastal retreat. Although natural marine processes remain the dominant drivers of shoreline change, the observed erosion hotspots coincide with areas of intensive tourism development, suggesting that human activities may increase coastal vulnerability.

Areas of intensive tourism development, identified using the Global Human Settlement Layer (GHS-BUILT-S), showed a 42.2% increase in built-up surface between 2005 and 2025, concentrated around Ao Ton Sai and Loh Dalum Bay, coinciding with the observed erosion hotspots.

Results were validated by comparison with published research on coastal erosion in karst-landscape islands in the region, supporting the reliability of the observed shoreline retreat and advance patterns.

The proposed cloud-based workflow demonstrates how freely available Earth observation data and Google Earth Engine can provide an efficient, reproducible, and cost-effective approach for long-term shoreline monitoring. The findings contribute to an improved understanding of coastal dynamics in tropical island environments and provide valuable information to support sustainable tourism planning, coastal conservation, and evidence-based environmental management.