Land use and land cover mapping using Landsat-8 and Sentinel-2 data in Al Qunfudhah coast, western Saudi Arabia: A comparative study
Samia S. Hasan,Omar A. Alharbi,Amr S. Fahil
DOI: https://doi.org/10.1016/j.rsma.2024.103429
IF: 2.166
2024-04-01
Regional Studies in Marine Science
Abstract:Al Qunfudhah Governorate has assumed a tourist status in the strip of Saudi coastal cities on the Red Sea coast. Agriculture, food industries, and engineering industries are among the most important economic activities in the city. As a result, Land use and land cover (LULC) detection is essential for several reasons: urban planning, sustainability, resource management and environmental preservation. This research aims to identify the LULC of the Al Qunfudhah area in western Saudi Arabia by comparing Landsat-8 and Sentinel-2 data outputs and evaluating their accuracy results. Various methods are used to process satellite image data to enhance the contrast between surface features. Some of these techniques are true colour composite (TCC) images, enhanced true colour composite (ETCC) images, false colour composite (FCC) images, principal component analysis (PCA), band ratios and Maximum Likelihood Algorithm (MLA) Supervised Classification. By utilising band combination methods, PCA methods, and band ratio methods, we were able to map a variety of features, including vegetation cover, mangroves, Sabkha and non-Sabkha deposits, wadi basins, shallow and deep shelf areas, islands, sand and mud sediments, rocky reefs and urban areas. As part of the study, ArcGIS 10.8's supervised classification tool was employed to categorise the study region's features and calculate the percentage of each class. In both Landat-8 and Sentinel-2 classification images, the same classes were identified (i.e. sand and mud sediments, sabkhas, wadi basins, aeolian sand and loess sediments, islands, urban areas, vegetation, and mangroves). The accuracy of the classification results was determined using a reliable statistical approach. The accuracy assessments of Landsat-8 and Sentinel-2 revealed accuracy rates of 87% and 89.1% respectively, along with kappa coefficients of 84.9% and 87.3%. For the investigated area, the Landsat-8 and Sentinel-2 satellites were effective in producing LULC maps.
ecology,marine & freshwater biology