|
چکیده
|
Urban expansion and land use changes are critical issues for sustainable development, necessitating robust monitoring and prediction methods. This study addresses this need by analyzing Landsat satellite imagery from 2000 to 2023, employing advanced classification techniques and predictive models to assess and forecast urban growth in Sulaymaniyah, Iraq. The methodology involved the use of various spectral indices (ABEI, EBBI, UI, NDBI, IBI, NBI) and classification methods (SVM, ML, MD, MN) to identify urban areas. SVM was determined to be the most accurate classification method. Additionally, spatial change detection was performed to visualize urban expansion over time. Predictive models (ANN, LR, Markov) were utilized to forecast land use changes up to 2046, with validation against actual 2020 data to ensure accuracy. The results indicate significant urban growth, with urban areas expanding from 28.45 km² in 2000 to 94.51 km² in 2023. Concurrently, pasture areas decreased from 531.54 km² to 440.21 km². Among the spectral indices, the Urban Index (UI) proved most effective in identifying urban areas. Predictive models forecast further urban expansion, with urban areas potentially reaching up to 134.30 km² by 2046. The study concludes that SVM is the most reliable classification method for this context, and the employed predictive models are useful for future urban planning. Recommendations include using higher-resolution imagery and integrating socio-economic data for more comprehensive future studies. This research provides valuable insights and tools for sustainable urban development in Sulaymaniyah and similar regions
|