Answers are grounded in this paper and cite their sources.
Floods are one of the most significant natural disasters globally, both in terms of infrastructure damage and the number of people affected. North Aceh Regency experienced severe flooding in October 2022, affecting more than 52,000 people and causing tens of thousands more to evacuate. In this study, flood mapping was conducted using Sentinel-1 SAR satellite imagery with a change detection approach based on the difference in backscatter values before and during the flood. The flood area extraction process utilised VH polarisation and was supplemented with the application of thresholds derived from sampling techniques on flood and non-flood objects. Analyses were conducted using the Google Earth Engine and ArcGIS platforms, integrating DEM data, permanent water bodies, and flood hazard indices from InaRisk for validation. The results showed that the floods on 8 and 20 October 2022 covered an area of 12,331.309 ha and 6,070.184 ha, respectively, with the main affected areas being rice fields and residential areas. Validation using a confusion matrix yielded high accuracy, at 97.22% and 98.15%, with Kappa values of 0.84 and 0.89. This study demonstrates that the use of Sentinel-1 SAR is effective for rapid mapping of flood-affected areas, particularly in open landscapes such as rice fields, although there are limitations in detecting floods in densely populated residential areas.
No takes yet. Share an insight, caveat, or question.
Saputra et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: