Applicability of Bioavailability Correction

Physico-Chemical Ranges

The ranges of soil physico-chemical conditions used to develop the bioavailability models for all soil organisms/functions represent the physico-chemical boundaries of the models. An overview of the range of physico-chemical parameters for which the chronic bioavailability models for Ni were developed is provided in Table 2.

Table 2: Overview of the Ranges of Soil Physico-Chemical Conditions Used to Develop the Bioavailability Models

Physico-Chemical Parameter Range
pH 3.6 – 7.7
Organic carbon 2.5 – 330.5 g/kg
Organic matter 0.4 – 56.8%
Effective cation exchange capacity 1.8 – 52.8 cmolc/kg
Clay content 0 – 55%
Ni background content 1 – 113 mg/kg

Applicability to Different Species

As mentioned above, the bioavailability models have been developed for only a limited number of species/functions, and they therefore do not cover all soil species/functions included in the chronic Ni database. The chronic Ni aquatic toxicity database contains data for 43 different species/functions (see Fact Sheet 2) while chronic Ni bioavailability models are available for only 4 species (i.e., F. candida, E. fetida, H. vulgare, L. esculentum) and 3 microbial functions (nitrification, substrate induced respiration, maize induced respiration).

The application of bioavailability models to species for which no model exists requires a cross-species extrapolation.  A cross-species extrapolation is justified based on the following evidence:

  • A reduction in intra-species variability after normalization with the regression equations was achieved; and
  • The regressions for the different species/functions are similar (see Table 1).

For all species and functions the eCEC of the soil is the primary driver for Ni bioavailability and the slopes for all the regression equations are similar (vary between 0.95 and 1.34), indicating a cross-species extrapolation is warranted. Moreover, the eCEC relationship has a robust mechanistic explanation, i.e., the higher the eCEC, the lower the proportion of free Ni3+ in soil pore water, which is the assumed most toxicologically relevant nickel species.

The following approach can be used for the normalization of all soil Ni toxicity data:

  • for higher plants other than L. esculentum, the H. vulgare model can be used;
  • for hard-bodied invertebrates, the F. candida model can be used;
  • for soft-bodied invertebrates, the E. fetida model can be used;
  • for microbial processes related to the N-cycle, the model for nitrifying micro-organism can be used;
  • for all respiration processes using natural substrate or basal soil respiration, the model for maize respiration model can be used;
  • for microbial biomass, the substrate induced respiration model can be used; and
  • for all other indicators of microbial assays, the model for nitrifying micro-organism can be used.