نوع مقاله : مقاله مستخرج از پایان نامه کارشناسی ارشد
نویسندگان
1 استاد گروه برنامه ریزی شهری و منطقه ای دانشکده برنامه ریزی و علوم محیطی دانشگاه تبریز
2 دانشیار گروه برنامه ریزی شهری و منطقه ای دانشکده برنامه ریزی و علوم محیطی دانشگاه تبریز
3 کارشناسی ارشد گروه برنامه ریزی شهری و منطقه ای دانشکده برنامه ریزی و علوم محیطی دانشگاه تبریز
کلیدواژهها
موضوعات
عنوان مقاله English
نویسندگان English
Abstract
Background and Objective:
Rapid urbanization in developing countries, including Iran, has led to uncontrolled urban sprawl, resulting in the loss of agricultural lands, green spaces, and environmental degradation. In Iran, cities have experienced outward growth since the 1960s, exacerbating land speculation and horizontal expansion. Salmas, a medium-sized city in West Azerbaijan Province, has seen its built-up area nearly double over the past four decades, making it a prime case for studying spatio-temporal land use changes and sprawl impacts. The main objective of this study is to analyze land use changes in Salmas from 1368 to 1402 using Landsat satellite imagery, simulate future changes up to 1422 with the CA-Markov model, and evaluate sprawl effects using the Holdren Index. Hypotheses include: (1) land use changes have occurred in a sprawl pattern over the past three decades; (2) this development has caused specific environmental effects, such as reduced green spaces.
Methodology:
Primary data included Landsat 5 (ETM+) imagery for 1368 (June 13, 1989) and Landsat 8 (OLI/TIRS) for 1402 (June 27, 2023), downloaded from USGS. Secondary data comprised land use maps from Salmas Municipality, population data from Iran's Statistical Center (1368-1402), and 1:25,000 topographic maps. Field visits validated training samples. Images were preprocessed in ENVI 5.3 (ROI clipping, radiometric calibration, QUAC atmospheric correction, geometric correction with GCPs achieving <0.5 pixel accuracy). Supervised Maximum Likelihood Classification (MLC) identified four classes: built-up areas (residential/commercial), barren lands, parks/green spaces, and agricultural lands. Post-classification, Majority/Minority filtering removed isolated pixels, with accuracy assessed via confusion matrix (Kappa >0.85). Changes were quantified using cross-tabulation matrices in ArcGIS 10.8 and annual change rates. The CA-Markov model in TerrSet 2020 generated transition probability matrices from Markov chains, suitability maps via Multi-Layer Perceptron with drivers (distance to urban centers, slope, rivers), and simulations for 1422 using Moore neighborhood (8 cells). Sprawl was measured with the Holdren Index, comparing urban area and population growth.
Findings and Conclusion:
From 1368 to 1402, built-up areas increased from 538.29 ha (23%) to 1187.64 ha (50.8%), barren lands decreased from 1205.19 ha (51.5%) to 838.17 ha (35.8%), agricultural lands dropped from 525.96 ha to 182.7 ha (-65.26%), and green spaces rose from 69.21 ha to 128.97 ha (+86.34%). Change detection revealed 367.01 ha of barren lands and 343.26 ha of agricultural lands converted to built-up. Simulations for 1422 predict further sprawl: built-up to 1401.66 ha (+18.02%), barren to 747.24 ha (-10.84%), green spaces to 107 ha (-27.15%), and agricultural to 93.95 ha (-48.57%). The Holdren Index indicates 37% of Salmas' growth from 1368-1402 was due to sprawl, with population rising from 60,624 to 99,681 while urban area expanded disproportionately, destroying 343.26 ha of farmland. Conclusions emphasize promoting compact city models for sustainable development, integrating RS/GIS with predictive models to guide policy in medium-sized Iranian border cities like Salmas.
کلیدواژهها English
