Flood Mapping and Damage Assessment Using UN-SPIDER Recommended Practices in Google Earth Engine: A Case Study of the 2024 Chiang Rai Flood, Thailand

Abstract

The 2024 Chiang Rai flood in Thailand highlighted the increasing frequency and severity of flood events driven by climate change and urbanization. This study applied the United Nations Platform for Space-based Information for Disaster Management and Emergency Response (UN-SPIDER) recommended practices to map flood extents and assess damages using Synthetic Aperture Radar (SAR) data from Sentinel-1 satellites in Google Earth Engine (GEE). The methodology involved change detection between pre-flood (1–9 September 2024) and post-flood (11–20 September 2024) SAR imagery, utilizing VH polarization for enhanced sensitivity to surface water. A threshold value of 1.25, recommended by UN-SPIDER, was identified as optimal through validation with ground data from the Geo-Informatics and Space Technology Development Agency (GISTDA), achieving 93.38% accuracy. The results indicated a total flood-affected area of 36,409 hectares, primarily along the Nam Kham, Nam Kok, and Nam Mae Ing rivers, with severe impacts in Mae Sai district. The study estimated 3,246 exposed individuals, 11,107 hectares of affected cropland, and 1,944 hectares of inundated urban areas. However, discrepancies arose when compared to GISTDA reports, particularly in cropland and population estimates, due to differences in data resolution and methodologies. The UN-SPIDER approach demonstrated scalability and efficiency for near-real-time flood monitoring, though limitations included challenges in urban flood detection and reliance on Sentinel-1's acquisition frequency. The findings underscore the utility of satellite-based flood mapping for disaster management while emphasizing the need for higher-resolution data to improve accuracy. This study contributes to flood risk mitigation strategies by providing actionable insights for policymakers and disaster response teams in Chiang Rai and similar flood-prone regions globally.