Predicting land use/land cover changes using CA-Markov and LCM models in the metropolitan area of Mashhad, Iran

• Published in: Modeling earth systems and environment Vol. 10; no. 6; pp. 7079 - 7096
• Main Authors: Aghajani, Hossein, Sarkari, Farnaz, Fattahi Moghaddam, Mehdi
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.12.2024; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Chemistry and Earth Sciences; Computer Science; Earth and Environmental Science; Earth Sciences; Earth System Sciences; Ecosystems; Environment; Land cover; Land use; Land use management; Landsat; Markov chains; Math. Appl. in Environmental Science; Mathematical Applications in the Physical Sciences; Metropolitan areas; Multilayer perceptrons; Multilayers; Neural networks; Original Article; Physics; Population growth; Remote sensing; Rural areas; Satellite imagery; Simulation; Social organization; Statistics for Engineering; Suburbs; Trends; Urban areas; Urban planning; Urban sprawl; Mashhad; Land use/Land cover change; Simulation and prediction; CA-Markov model; LCM
• ISSN: 2363-6203; 2363-6211
• DOI: 10.1007/s40808-024-02051-x

One of the main objectives of urban planning is to effectively manage the growth and development of urban and peri-urban settlements. Simulating and forecasting changes in land use and land cover (LULC) around metropolitan areas is crucial in order to improve intervention and control over unplanned growth in these settlements. The purpose of this research is to simulate LULC changes in the metropolitan area of Mashhad from 2020 to 2050.this paper utilizes the CA-Markov model and Land Change Modeler (LCM) with two algorithms: multi-layer perceptron (MLP) neural network and logistic regression (LR), along with natural and social driving factors, to accurately simulate LULC changes in the metropolitan area of Mashhad. The simulated LULC changes for the year 2020 were then compared with actual land use data of 2020 obtained through satellite images from Landsat 8. The model with higher accuracy was subsequently used to simulate future years. The results of the model show that urban land use has had significant changes compared to other Land uses and has increased from 19% of the total area in 2000 to 27% in 2020. The Markov model, with a Kappa coefficient of 0.71 and an overall accuracy coefficient of 0.81, showed more accurate predictions in terms of location for 2020 compared to the other two models, and was therefore used to simulate future years. In addition, the simulation results show that future urban growth in Mashhad will be scattered and discontinuous, mainly occurring in the outskirts of the city instead of the center. The greatest growth is expected near the primary transportation axes and large rural settlements in the suburbs along the northern and northwestern axes.