Estimation of Sardoba Reservoir breach extent with a combined MNDWI index and difference change detection approach based on SWIR2 imagery in Google Earth Engine

Using techniques for remote sensing in the Google Earth Engine (GEE) cloud platform, this study provides an in-depth assessment of the Sardoba Reservoir breach’s scope. The modified waterbody difference index (MNDWI) and shortwave infrared (SWIR2) change detection technique are used in this investigation, which is based on Sentinel-2 satellite data. The green (B3) and SWIR1 (B11) channels were used to generate the MNDWI index in order to better detect water bodies because of the spectral contrast between the water and nearby land objects. Also, the technique of detecting the difference between SWIR2 images before and after the event was used to identify dynamic changes typical floods. By combining these two techniques, a combined water object mask was produced that successfully reduces false positives and raises the accuracy of flooding detection. Subsequently, morphological processes and median filters were used to exclude small, unnecessary water bodies and noise pixels from the boundaries of flooded areas. Agricultural land and infrastructure were severely impacted by the flooding of about 64,537 hectares within the selected study area caused by the reservoir breach, according to quantitative analysis. For quick and precise flood assessment, the results show how well MNDWI and SWIR2 analysis work together in GEE. It turned out that using a platform based on cloud technology and using the above mentioned methods, it was possible to determine flood zones with high accuracy.

Ҳали таржима қилинмаган