This paper aims to simulate land-use and land cover (LULC) changes in Mashhad metropolitan area. These changes were measured for 2016-2020 and forecast for 2020-2030. To this end, Sentinel-2A satellite imagery, Cellular Automata (CA), and the Markov chain model in ArcGIS and TerrSet software were used. LULC was classified into seven groups, using maximum likelihood estimation, KAPPA coefficient, and ROC curve. According to the study results, the area is 105243.6 hectares, and there are 177 rural settlements (2016). During the period 2016-2020 in the study area, 7845 hectares of LULC changes occurred. Most of the land-use increase is associated with barren lands with 7174 hectares (91.45%) and built-up with 436.32 hectares (5.56%). In the same period, the highest decrease in land-use is related to mass and light vegetation with 1335 hectares (17.01%) and 5241 hectares (66.81%), respectively. Predicting LULC change for the period 2020-2030 using the CA Markov chain model showed that the land-use change would occur in the study area of 2626.2 hectares. Three land uses with positive changes (barren lands 5.5%, built lands 13.5% and mountainous lands 2.32%) and four cases (light vegetation-5.70%, dense vegetation -23.35%, Road 0.34%, and water level (-23.40%) will face negative changes. To confirm these classifications\\\' accuracy, the KAPPA coefficient = 0.58 and the area under the curve =0.61 in the ROC curve were calculated. Spatial-temporal conversion in land-use from agriculture and vegetation to the barren land and built-up will occur mostly in the north and east of the city. These findings provide a basis to support decision-making and planning systems for sustainable land-use development.

Keywords