Multi-temporal Remote Sensing Data Analysis for Devastating Flood Study in Northern Libya

Floods are considered the most dangerous and destructive geohazards, leading to human victims and severe economic outcomes. Yearly, a large number of regions around the world suffer from devasting floods. The estimation of flood aftermaths is one of the high priorities for the global community. One such flood took place in northern Libya in September 2023. The presented study is aimed at the Libyan flood aftermath evaluation using Google Earth Engine opportunities. The primary task is obtaining and analyzing data that provide high accuracy and detail for the study region. Among those data that are of importance for flood and its aftermath assessment are remote sensing images. The last decades have seen an increased interest in remote sensing data thanks to the data accessibility and variety of free, open-source computational platforms. The data from Landsat, Sentinel, and similar missions are ubiquitous and well-studied. On the other hand, such software as Google Earth Engine or QGIS has a powerful toolbox for different solutions. The goal stated in the paper is related to image classification and change detection problems. The mentioned software provides various solutions based on machine-learning approaches for image classification and change detection. Miscellaneous data have been used to reach the paper’s goal. The first stage of the study was the development of a workflow for data analysis. This workflow includes three parallel processes of data analysis. The study comprised Sentinel 2 data for image classification using multispectral bands. Different supervised classification methods were examined, including random forest, support vector machines, naïve-Bayes, and CART. The different sets of hyperparameters for classification were considered. GEOEYE-1 and WorldView-2 images of four cities, Dernah, Susah, Al-Bayda, and Brega, were investigated for change detection algorithms. In addition, different NDVIs were calculated to facilitate the recognition of damaged regions. At the final stage, the analysis results were fused using the QGIS platform to obtain the damage estimation for the studied regions. As the main output, the area changes for the primary classes, and the maps that portray these changes were obtained. The recommendations for data usage and further processing in Google Earth Engine were developed.