Document Type : Original Article

Authors

• Sajjad Tabe Bardbar ¹
• Mohammad Ebrahim Afifi ²
• Abdolrasoul Ghanbari ³

¹ PhD in Geography and Urban Planning, Larestan Branch, Islamic Azad University, Larestan, Iran.

² Associate Professor, Department of Geography, Larestan Branch, Islamic Azad University, Larestan, Iran.

³ Assistant Professor, Department of Geography, Larestan Branch, Islamic Azad University, Larestan, Iran.

Abstract

This study aims to predict the future urban expansion of Shiraz up to the year 1410 (2031 CE) using the CA‑Markov model integrated with Remote Sensing (RS) data and Geographic Information Systems (GIS) to support urban planning and sustainable land management.
Multi‑temporal land use datasets for the years 1999, 2004, and 1404 (2025 CE) were processed to generate an Urban Transition Probability Matrix. Input layers included road distance, proximity to existing urban areas, DEM elevation data, protected zones, land‑use transition matrices, and conversion probability maps. The CA‑Markov modeling was employed to simulate land‑use changes, while Binomial and Geometric probability distributions were applied to assess success rates of non‑urban to urban conversion. Model validation was conducted using indices such as Overall Accuracy (OA), Producer’s Accuracy for Urban (PAU), User’s Accuracy for Urban (UAU), Figure of Merit (FoM), Quantity Disagreement (QD), Allocation Disagreement (AD), and the Kappa coefficient.
The simulation revealed a probability of 0.28 for non‑urban areas converting to urban, with only 0.15 probability for urban areas remaining unchanged. PUU=0.89 indicates high urban stability, while PNU=0.4512 reflects significant potential for urban growth in peripheral zones. Validation metrics (OA=0.82, PAU=0.86, UAU=0.79, Kappa=0.64, FoM=0.22) confirmed satisfactory performance, although allocation errors (QD=4.9%, AD=13%) suggest spatial precision improvements are needed. Statistical analysis showed that out of every 50 non‑urban pixels, approximately 14 are likely to convert to urban, with initial changes occurring near major infrastructure.

Keywords

• Urban Expansion Prediction
• Markov Model
• Urban Transition Probability Matrix
• Remote Sensing

Main Subjects

• Remote sensing and Gis