Spatiotemporal Flood Characterization and Early Warning Development in the Rokan Watershed Using Sentinel-1 SAR and Water Level Data Rijaldi R.M.(1*), Sidik R.F.(1), Liyantono.(1), Setiawan Y.(1) Faskayana Y.S.(2) and Desam R(3)
1) Center for Environmental Research, IPB University, Indonesia
2) The Graduate School of Agriculture, Tokyo University of Agriculture and Technology, Japan
3) Pertamina Hulu Rokan, Indonesia
*rizkirijaldi44[at]gmail.com
Abstract
The Rokan Watershed in Sumatra frequently experiences severe flooding, with the most prolonged inundation recorded between late 2023 and early 2024, lasting more than three months. However, the long-term spatiotemporal characteristics of floods in this region remain poorly understood, and an effective early warning developmnet has yet to be developed. This study aims to characterize flood dynamics over a 10-year period (2014-2024) and establish a foundation for a flood early warning development in the Rokan Watershed. Flood extent was derived from Sentinel-1 Synthetic Aperture Radar (SAR) imagery using a threshold-based classification applied to VH polarization. The use of SAR data enables consistent flood mapping under cloudy and low-light conditions. All processing was conducted on Google Earth Engine via Google Colab to ensure efficient and reproducible analysis. Validation using 439 ground reference points yielded an overall accuracy of 83.8%. The results indicate significant inundation events in 2014, 2018, 2019, and particularly during 2023-2024, reflecting an increasing trend in both flood frequency and spatial extent. Comparison with water level observations from BWSS III stations revealed a time lag between upstream water level rise and flood occurrence. For instance, when the Lubuk Bendahara Station records water levels above 220 cm during the late-year rainy season, downstream flooding in Bonai Village typically occurs within 4-9 days. These findings enhance understanding of flood behavior in the Rokan Watershed and contribute to the development of data-driven flood early warning systems in tropical river basins.
Keywords: flood dynamics, Sentinel-1 SAR, early warning systems, Google Earth Engine
