Exceedance Probability and Toxicity-Value Sensitivity in Health-Risk Assessment of a Pb–Zn Slag Reprocessing Site Under the 2025 IRIS Arsenic Values

Screening-level risk assessments report single-point estimates of carcinogenic risk and hazard index. These convey neither the probability that a threshold is exceeded nor the sensitivity of the conclusion to revision of toxicity values. We examined both at an active lead–zinc slag reprocessing site in Shymkent, Southern Kazakhstan, by reanalysing a published geochemical dataset (24 transect soil samples, ICP-OES) using Monte Carlo simulation (100,000 iterations, US EPA RAGS adult outdoor-worker scenario) under the toxicity values applied in the source assessment and, separately, under the values established by the January 2025 IRIS revision of inorganic arsenic. Under the original values the probability of exceeding the 10⁻⁴ carcinogenic threshold was 69.9% and of exceeding HI = 1 was 82.1%; the reported site-mean and worst-case point estimates fell at the 76.9th and 97.2nd percentiles of the simulated distribution. Under the 2025 values mean risk was 21-fold higher and the exceedance probabilities rose to 99.3% and 97.3% respectively. The qualitative conclusion — that risk at this site is unacceptable — therefore survives both sources of uncertainty, while the quantitative estimate does not. Sensitivity analysis identified arsenic concentration (ρ = +0.867) as the binding constraint on precision under either parameterization.