Validation of site-specific soil Ni toxicity thresholds with independent ecotoxicity and biogeochemistry data for...

The Existing Substances Regulation Risk Assessments by the European Union (EU RA) generated new toxicity data for soil organisms exposed to Ni added to sixteen field-collected soils with low background concentration of metals and varying physico-chemical soil characteristics. Using only effective cation exchange capacity (eCEC) as a bioavailability correction, chronic toxicity of Ni in soils with a wide range of characteristics could be predicted within a factor of two. The objective of the present study was to determine whether this was also the case for three independent data sets of Ni toxicity thresholds. Two of the data sets were from Community Based Risk Assessments in Port Colborne ON, and Sudbury ON (Canada) for soils containing elevated concentrations of Ni, Co and Cu arising from many decades of Ni mining, smelting and refining. The third data set was the Metals in Asia study of soluble Ni added to field soils in China. These data yielded 72 leached and aged EC₁₀/NOEC values for soil Ni, for arthropods, higher plants and woodlot structure and function. These were reduced to nine most sensitive single or geometric mean species/function endpoints, none of which were lower than the HC₅ predicted for a soil with an eCEC of 20 cmol/kg. Most of these leached and aged EC₁₀/NOEC values were from soils co-contaminated with Cu, in some cases at its median HC₅ as predicted by the EU RA from soil characteristics. We conclude that the EU RA is protective of Ni toxicity to higher-tier ecological endpoints, including in mixture with Cu, before the assessment factor of 2 is applied. We suggest that for prospective risk assessment, the bioavailability based PNEC (HC₅/2) be used as a conservative screen, but for retrospective and site-specific risk assessment, the bioavailability based HC₅ is sufficient.