Spatial heterogeneity of environmental degradation in Iran: A stochastic dominance-based composite index at the county level

Environmental degradation is a multidimensional, spatially heterogeneous, and policy-relevant phenomenon that requires robust, reproducible, and spatially explicit assessment frameworks. This study constructs a Composite Environmental Degradation Index (CEDI) for 397 counties in Iran (2018–2022) by integrating multi-source satellite observations with a stochastic dominance (SD) algorithm and mixed-integer programming (MIP)-based weighting. Seven environmental dimensions were incorporated: climate change, greenhouse gases, land use, biodiversity, water resources, air pollution, and soil erosion. Quantitative results indicate that the SD–MIP approach assigns the highest contributions to climate change (0.279), greenhouse gases (0.226), and land use (0.223), while air pollution and soil erosion exert relatively lower impacts. Spatial autocorrelation analysis (Getis–Ord Gi∗) reveals statistically significant hot spots of environmental degradation across the western, southern, southeastern, and central regions, whereas cold spots are concentrated along the northern belt, particularly within the Hyrcanian forests, indicating greater ecological resilience. Temporally, although year-to-year fluctuations were observed, the overall trend suggests a gradual intensification of environmental stress, with the peak average CEDI reaching 0.58 in 2018 and 2020. Methodologically, applying stochastic dominance to composite index construction eliminates the bias of arbitrary weighting, incorporates the full distribution of each indicator, and quantifies uncertainty through Monte Carlo sensitivity analysis (n = 1000). The results demonstrate high robustness (mean Spearman's ρ = 0.91; SD = 0.03; mean Jaccard hotspot overlap = 85.2 %), confirming the model's stability and reliability. From a policy perspective, the CEDI functions as an early-warning and decision-support system, enabling region-specific interventions in high-risk areas while reinforcing sustainable practices in resilient regions. The proposed framework directly supports Sustainable Development Goals 13 (Climate Action) and 15 (Life on Land) and can be adapted to other developing-country contexts with diverse environmental and data conditions. • Novel SD-based composite Environmental Degradation Index for 397 counties in Iran. • Integrates MODIS, Sentinel-5, GRACE & Landsat-8 with MIP-derived weights. • Hotspots in west, south and southeast; Hyrcanian forests form a resilient cold-spot. • Climate change, greenhouse gases and land-use are the dominant degradation drivers. • Provides a robust, policy-ready early-warning tool aligned with SDG 13 and SDG 15.