Abstract
Land cover dynamics were analyzed temporally and spatially in the Al-Hubail wetland (Al-Ahsa, Kingdom of Saudi Arabia) to determine the evolution of the environmental status of this biologically and ecologically interesting area in Saudi Arabia. Using remote sensing data, land cover changes were estimated for 36 years (1985-2021). For this analysis, three images from 1985 (Landsat 5 MSS), 2003 (Landsat 7 ETM +), and 2021 (Sentinel-2) were used to classify and detect changes. A machine learning algorithm was used, and the images were classified into four main land cover classes: Water bodies, hydromorphic areas, vegetation, and open ground. Change detection was performed for the year pairs 1985 to 2003 and 2003 to 2021. The results of this classification showed a significant increase in the area of hydromorphic areas and vegetation. The results were checked with a confusion matrix indicating an overall accuracy between 89.3 and 92.8%. The qualitative trend data show that the Al-Hubail wetland has changed significantly during the study period. Thus, a significant expansion of the wetland was observed in conjunction with an increase in agricultural drainage toward the wetland. This analysis shows the strong anthropogenic pressure on the area and highlights the need to strengthen the existing laws to preserve local biodiversity in the long term. It suggests that more efforts should be made to manage the water resources of the region effectively.