This study evaluates the agreement and structural consistency of three atmospheric dispersion modeling approaches applied across four U.S. urban environments for two chemically distinct hazardous materials. ALOHA (Areal Locations of Hazardous Atmospheres), an analytic Gaussian plume model, and OpenFOAM computational fluid dynamics (CFD) passive scalar transport are applied to continuous release scenarios for chlorine and ammonia at Colorado Springs (Colorado), Newark (New Jersey; serving as the proxy location for the New York City area, as Manhattan is not available in ALOHA's location database), Pittsburgh (Pennsylvania), and Chicago (Illinois). Chlorine is modeled using ALOHA's heavy gas module and ammonia using ALOHA's Gaussian module, reflecting the distinct dispersion physics of each material. Scenario parameters—release rate, source height, wind speed, stability class, and wind direction—are held constant within each chemical scenario across all four cities; ALOHA's site dependent ambient pressure, which varies with elevation and cannot be overridden, represents a documented but unavoidable source of cross-city non-equivalence. Cross model evaluation is performed at a canonical set of 11 receptor offsets, defined as the intersection of downwind and off centerline positions present across all eight ALOHA exports. The 12th receptor offset differed by location: Colorado Springs exports included a 1,000 m downwind, 200 m off centerline receptor, while Newark and Pittsburgh exports included a 1,000 m downwind, 250 m off centerline receptor; this asymmetry precluded inclusion of either variant in the canonical set. Because OpenFOAM's passive scalar tracer field is dimensionless and not calibrated to physical concentration units, no quantitative ppm comparison is claimed; all OpenFOAM comparisons are rank-based and expressed as Spearman correlation coefficients. Results demonstrate that ALOHA and the analytic Gaussian model maintain consistent rank ordering across all cities and both chemicals (ρ = .786–.821), confirming their shared simplified physics basis. OpenFOAM agreement is strongly city dependent: Colorado Springs exhibits strong structural concordance (Gaussian–OpenFOAM ρ = .857–.905), Newark/Manhattan shows near-zero to negative correlation (ρ = .297 to −.714), and Pittsburgh shows moderate concordance (Gaussian–OpenFOAM ρ = .433, n = 11; ALOHA–OpenFOAM ρ = .667–.685, n = 7) with a detectable off-axis channeling signal, and Chicago shows high concordance (Gaussian–OpenFOAM ρ = .909–.927, n = 11; ALOHA–OpenFOAM ρ = .786–.964, n = 7), consistent with its open grid street morphology. These results complete a physically coherent urban morphology gradient across all four sites. These findings indicate that simplified operational models reliably rank receptor hazard across diverse urban settings, while CFD reveals site specific plume redistribution that simplified models cannot reproduce.
Devin Peters (Sun,) studied this question.