Urban development and rapid changes in land use and land cover have been one of the most important environmental and urban planning challenges in recent decades. These changes, which are mainly caused by urbanization and human activities, have widespread impacts on ecosystems, natural resources, and environmental sustainability. The present study aimed to analyze the trend of land use changes and predict future developments in the city of Gorgan. In this research, the advanced GeoSOS-FLUS model as integrated of artificial neural network and cellurar automata - markov chain, was used as a powerful tool to simulate land use changes and predict its future trend. The research data included multi-temporal landsat satellite images (TM, ETM+, OLI) and spatial data of different parameters. The research results showed that urban growth in Gorgan city has changed from a centralized and nuclear pattern to a dispersed and non-integrated pattern. This trend has led to a significant reduction in agricultural, forest, and pasture lands. The model predictions indicate that in the next 25 years, residential areas will increase significantly and the expansion of the city to peripheral areas, especially natural lands, will intensify. These changes can increase pressure on natural resources, destroy ecosystems, and reduce biodiversity. Evaluation of the GeoSOS-FLUS model showed that this tool has a very high accuracy in simulating land use changes which the value of the kappa coefficient and other evaluation criteria indicated the appropriate accuracy of the model. Also, the agreement of the model results with real data indicates the high ability of this model to predict future changes. The findings of this study, in addition to identifying factors affecting land use changes, can be used as a practical tool for urban managers and decision-makers in the field of land use planning and policy making.