3D geospatially driven predictive modeling for sustainable management of heavy metal pollution (Pb, Cd, As) in rapidly developing urban regions

This study presents a novel 3D geospatial modeling framework designed to assess the spatiotemporal dynamics of dissolved heavy-metal contamination of lead (Pb), cadmium (Cd), and arsenic (As) in the Shafa Badran corridor (Zarqa River Basin, Jordan). The approach integrates Geographic Information System (GIS)–based spatial interpolation (inverse distance weighting and ordinary kriging), hydrologic flow-path analysis from Digital Elevation Models (DEMs), and a calibrated one-dimensional advection–dispersion equation (ADE) for a long-channel transport. Field measurements indicate pronounced clustering of Pb and Cd within industrial reaches and broader As plumes aligned with irrigation return flows. Compared with World Health Organization (WHO) guideline values, Pb shows frequent exceedances and Cd occasional exceedances, while As is generally at or below 10 µg L −1 . Source attribution is routing-consistent: industrial hotspots co-locate with mapped outfalls; diffuse signatures align with irrigated belts. Forecasts suggest that absent mitigation measures, elevated concentrations persist through 2035 along priority segments. Uncertainty is quantified via kriging variance, cross-validation diagnostics, and ADE prediction envelopes from parameter perturbation; these are propagated into decision-oriented risk layers to distinguish screening from management priorities. Scenario analysis indicates that combining industrial pre-treatment at discharge nodes with nature-based polishing at return-flow confluences can substantially reduce along-channel concentrations and guideline exceedance probabilities. The study therefore provides a routing-consistent fusion of 3D voxel-based geostatistics with a calibrated 1D ADE, yielding two-year exceedance-probability maps, end-to-end uncertainty envelopes, and scenario-based risk reductions (S0–S3) resolved along hydrologic flow paths. The workflow converts technical forecasts into operational management surfaces that can guide stepwise interventions in similar semi-arid, data-constrained urban–industrial watersheds.