Satellite based flood mapping
This study relies on flood data indicating a recent flood event.
Synthetic-aperture radar (SAR) data from Sentinel-1 satellites provides unprecedented opportunity to evaluate inter-annual flood characteristics, although consensus on best flood detection methods is lacking. This study compared the performance of three flood detection methods to evaluate inter-annual flood characteristics for any area of interest to the MSc student. The methods are Change Detection and Thresholding (CDAT), Normalized Difference Flood Index (NDFI) and Root of Normalized Image Difference (RNID). A reference flood map needs to be prepared based on a field data from a recent flood. Inter-annual flood characteristics will be evaluated in terms of flood onset, recession and frequency of occurrence.
This study aims to (i) compare inter-annual flood characteristics of two flood prone sites in Awash River Basin of Ethiopia, using time series S-1 images and ii) to contribute to the scientific literature on comparison of accuracies of flood detection methods. Three remote sensing methods of change detection were compared to select the best performing method for spatial flood extent mapping. These methods are (i) Change Detection and Thresholding (CDAT) (Clement et al., 2017), (ii) Normalized Difference Flood Index (NDFI) (Singha et al., 2020) and (iii) Root of Normalized Image Difference (RNID) (Bekele et al., 2022).
The methods employed in this study consisted of (i) pre-processing of satellite images, (ii) comparison of three remote sensing methods for flood detection, (iii) time-series flood mapping using the algorithm that performed best, and (iv) comparison of the characteristics of the 2020 floods against those of the 2017-2022 floods.
Dubti site (Ethiopia)