A GIS-BASED MULTI-CRITERIA FRAMEWORK FOR ENVIRONMENTAL RISK ASSESSMENT AND SUSTAINABLE URBAN DEVELOPMENT USING SATELLITE-DERIVED INDICATORS AND SPATIAL VULNERABILITY MAPPING

Abstract

Rapid urban expansion increases environmental pressures and community exposure to urban heat, flooding, ecosystem degradation, and infrastructure deficiencies. This study develops a geographic information system (GIS) based multicriteria framework for environmental risk assessment and sustainable urban development in Multan, Pakistan, integrating satellite derived indicators with spatial vulnerability mapping. The dataset comprises 120 satellite scenes acquired between 2015 and 2025, 2,400 spatial assessment units, and 18 environmental, demographic, and infrastructure indicators. Landsat and Sentinel imagery were processed to derive land surface temperature, vegetation condition, built area extent, surface moisture, and land cover transitions. These indicators were integrated with elevation, drainage proximity, population density, road accessibility, and socioeconomic characteristics to characterize environmental hazards, exposure, and adaptive capacity. The methodology combines satellite image preprocessing, spatial harmonization, indicator normalization, analytical hierarchy process weighting, and GIS weighted overlay analysis. Spatial clustering identifies concentrations of environmental risk, while sensitivity analysis examines classification stability under alternative weighting assumptions. Independent reference observations evaluate agreement with the resulting risk classifications. Three development scenarios, comprising continued urban expansion, compact development, and environmentally constrained growth, were assessed to examine the environmental implications of alternative planning strategies. Results reveal a 28.4% increase in built area and a 13.7% decline in vegetation cover during the study period. High and very high environmental risk zones account for 31.6% of the study area and accommodate 42.3% of the assessed population, indicating a disproportionate concentration of residents in environmentally vulnerable locations. Risk classification achieves 86.8% overall agreement with independent reference labels and a Cohen’s kappa of 0.81. Sensitivity analysis shows that 89.2% of assessment units retain their original risk categories. Compared with continued expansion, environmentally constrained growth reduces development within high risk zones by 24.6% and improves the modeled green space accessibility index by 18.9%. The proposed framework provides an interpretable spatial basis for prioritizing urban greening, drainage improvements, infrastructure investment, and land use regulation. By connecting environmental risk patterns with alternative development scenarios, it supports evidence informed planning for more sustainable and climate resilient urban growth, while accounting for uncertainty in indicator selection and weighting.