Monitoring Flood Dynamics in Hoa Vang District, Da Nang City Using SAR Remote Sensing and Google Earth Engine

Abstract This study aimed to assess the flood dynamics in Hoa Vang District, Da Nang City using Sentinel 1 radar satellite images on the Google Earth Engine (GEE) platform from 2015 to 2022. Within the study, various functions were established in GEE to accomplish tasks such as selecting the study area, pre-processing data, filtering images, geometric corrections, and employing the Otsu thresholding method to identify flooded and non-flooded regions for each image. The results showed that the Otsu threshold values ranged from −11.663 to −10.874. Additionally, the study has taken the advantage of Digital Elevation Models and hydrological network to extract flooded areas at different time intervals from 2015 to 2022. In order evaluate the accuracy of flood results, the study uses field survey flood data and calculates the Kappa coefficient. Results show that the Kappa index (0.815) is highly reliable. The results show that the communes in Hoa Vang with the largest inundated area and flooding rate are Hoa Phong with 384.5 hectares, accounting for 20.7% of the area and Hoa Tien with 366 hectares, accounting for 25% of the commune’s area. The research findings provide the scientific background for evaluating the socio-economic impacts of flooding, particularly in the agricultural sector of the southern region of Hoa Vang District, Da Nang City.