Abstract :
This research dealt with detecting the change in land use in the The Bahr Qarun district in Fayoum Governorate during the years 2019, 2021, and 2023 using remote sensing data, represented in Sentinel 2 and Landsat OLI images. The years were chosen for analysis because of the availability of both Sentinel 2 and Landsat OLI 8 satellite imagery on the same date (in April) for each year. The study utilized the Maximum Likelihood (ML) classification method to classify the Sentinel 2 images of the study area. Approximately 30 training samples were chosen for each land cover class to implement the supervised classification process. The Kappa coefficient value over the years of study is more than 75%, which is considered accepted accuracy. Vegetation and urban covers were also detected using the Normalized Difference Vegetation Index (NDVI) and Normalized Difference Built-up Index (NDBI), respectively, through Landsat OLI 8 images. After assessing the spectral indices NDVI and NDBI relative to the supervised classification. It was found that the results of the spectral indicator NDVI were convergent to the results of the supervised classification, as the convergence percent ranged from 88.73% to 100.94%, while the results of the spectral indicator NDBI ranged from 108.47% to 149.14%, which indicates the presence of many different features within a single pixel, such as bare soil with urban areas because all convergence values for all years were higher than 100%. Therefore, it is possible to rely on the NDVI for mapping green cover of the study area.
