Document Type : Research Article
Authors
1
Graduated with a PhD in Watershed Management Faculty of Natural Resources, Urmia University, Iran
2
Faculty of Natural Resources, Urmia University, IRAN.
3
Ali Akbar Rasouli, Professor Department of Environmental Sciences, Macquarie University, Sydney, Australia
10.22067/geoeh.2024.88446.1494
Abstract
Investigating the prediction of landuse changes is one of the crucial factors for understanding environmental transformations at all temporal and spatial scales. The present research aims to examine the trend of changes and predict the future landuse status in the Zolachay Watershed located in West Azerbaijan Province, one of the sub-basins of Lake Urmia, in the last 33 years. For this purpose, first, the Sentinel-2 and Landsat 5,7 images for 1990, 2020, 2016, 2010, 2005, 2000, 1995, and 2023 were acquired from their official sites. Then, needed preprocessing methods were applied in various software environments, and the relevant images were produced inside the eCognition software environment. Then, the nearest neighbor classification model was executed using the object-based method, and landuse and landcover maps were generated. Finally, using the Markov Chain Cellular Automata (CA) method, simulations of landuse changes for the year 2030 were performed. To assess the accuracy of the CA Markov model, the landuse change map for 2023 was validated against the 2023 classification map.
The final results indicate that applying knowledge-based methods, especially the nearest neighbor classification, allows for the product of landuse maps with a high accuracy coefficient (Kappa 91%), followed by the Markov (CA) model change maps with an acceptable accuracy of 87%. The final results demonstrate that by 2030, agricultural landuse will increase by 15.03%, residential areas by 9.0%, and drylands by about 14%. Soil landuse will decrease by 23.68% and pastures by 6.5%. Overall, the final models indicate the high accuracy of knowledge-based and object-based methods, as well as the satisfactory performance of the Markov model in the study of landuse changes. The findings of this research can serve as a reference in future environmental planning processes, aiming at sustainable recommendations and prudent land utilization.
Keywords
Main Subjects
)
Send comment about this article