Land-Use Land-Cover Dynamics and Future Projections Using GEE, ML, and QGIS-MOLUSCE: A Case Study in Manisa
Top Cited Papers
Open Access
- 7 February 2025
- journal article
- research article
- Published by MDPI AG in Sustainability
- Vol. 17 (4) , 1363
Abstract
Urban expansion reshapes spatial patterns over time, leading to complex challenges such as environmental degradation, resource scarcity, and socio-economic inequality. It is critical to anticipate these transformations in order to devise proactive urban policies and implement sustainable planning practices that minimize negative impacts on ecosystems and human livelihoods. This study investigates LULC changes in the rapidly urbanizing Manisa metropolitan area of Turkey using Sentinel-2 satellite imagery and advanced machine learning algorithms. High-accuracy LULC maps were generated for 2018, 2021, and 2024 using Random Forest, Support Vector Machine, k-Nearest Neighbors, and Classification and Regression Trees algorithms. Among these, the Random Forest algorithm demonstrated superior accuracy and consistency in distinguishing complex land-cover classes. Future LULC scenarios for 2027 and 2030 were simulated using the Cellular Automata–Artificial Neural Network model and the QGIS MOLUSCE plugin. The results indicate significant urban growth, with built-up areas projected to increase by 23.67% between 2024 and 2030, accompanied by declines in natural resources such as bare land and water bodies. This study highlights the implications of urban expansion regarding ecological balance and demonstrates the importance of integrating machine learning and simulation models to forecast land use changes, enabling sustainable urban planning and resource management. Overall, effective policies must be developed to manage the negative environmental impacts of urbanization and conduct land use planning in a balanced manner.Keywords
This publication has 76 references indexed in Scilit:
- Evaluating Landsat-8, Landsat-9 and Sentinel-2 imageries in land use and land cover (LULC) classification in a heterogeneous urban areaGeoJournal, 2023
- Modelling land use/land cover (LULC) change dynamics, future prospects, and its environmental impacts based on geospatial data models and remote sensing dataEnvironmental Science and Pollution Research, 2022
- Use of support vector machine and cellular automata methods to evaluate impact of irrigation project on LULCEnvironmental Monitoring and Assessment, 2022
- Monitoring of land use land cover dynamics and prediction of urban growth using Land Change Modeler in Delhi and its environs, IndiaEnvironmental Science and Pollution Research, 2022
- Transforming big data into smart data: An insight on the use of the k‐nearest neighbors algorithm to obtain quality dataWIREs Data Mining and Knowledge Discovery, 2018
- Evaluation of spatio-temporal variability in Land Surface Temperature: A case study of Zonguldak, TurkeyEnvironmental Monitoring and Assessment, 2015
- Machine Learning Feature Selection Methods for Landslide Susceptibility MappingMathematical Geosciences, 2013
- A Methodological Approach for Designating Management Zones in Mount Spil National Park, TurkeyEnvironmental Management, 2000
- Bagging predictorsMachine Learning, 1996
- Support-vector networksMachine Learning, 1995