Nghiên cứu xác định diện tích ngập nước sử dụng ảnh Sentinel-1 trên nền Google Earth Engine: áp dụng cho tỉnh Đồng Tháp, đồng bằng sông Cửu Long

Abstract

This paper presents the spatio-temporal variability of surface water area for Dong Thap province, Vietnamese Mekong Delta by using a time-series of 114 Sentinel-1 images. The image process was implemented on the Google Earth Engine cloud computing platform using Java Script language. The results showed that (i) the temporal variation of water surface area consists with the change in water elevation, (ii) the largest value of water surface area often occurs in July and August, with a value ​​ranging from 1584 to 1892 km² (approximately from 45.8 to 54.7% of the province's area), while the smallest value ​​appears in April and equals from 4.65 to 6.18% of the area of ​​the province, (iii) water surface area strongly correlates with the change in water elevation at Tan Chau station, with a correlation coefficient of 0.75. The time for processing each Sentinel-1 image is less than 15 seconds, thus saving a lot of time for processing the selected time-series Sentinel-1 images.