Deriving site-specific soil clean-up values for metals and metalloids: rationale for including protection of soil microbial processes.
Bottom Line: A wide range of standardized and nonstandardized methods became available for testing chemical toxicity to microbial functions in soil.Regulatory frameworks in the European Union and Australia have successfully incorporated microbial toxicity data into the derivation of soil threshold concentrations for ecological risk assessments.Although the primary focus of this article is on the development of the approach for deriving SCVs for metals and metalloids in the United States, the recommendations provided in this article may also be applicable in other jurisdictions that aim at developing ecological soil threshold values for protection of microbial processes in contaminated soils.
Affiliation: US Army Edgewood Chemical Biological Center, Aberdeen Proving Ground, Maryland.Show MeSH
Related in: MedlinePlus
Mentions: Clear concentration–response relationships were observed in most toxicity tests of microbial processes, allowing the calculation of reliable ECx values. Therefore, it was possible to relate specific microbial activities to indications of adverse and unacceptable loads of contaminants in soil. The sensitivities of microbial endpoints overlap with the sensitivities for tests with plants or soil invertebrates (Table3). There is no species or group (plants, soil invertebrates, or microorganisms) that is consistently most or least sensitive. The microbial endpoints are generally distributed across the range of the combined SSD (Figure 1). Addition of the microbial endpoints to those for plants and soil invertebrates does not significantly affect the resulting HC5 or HC50 values compared to the SSD based on data for only plants and invertebrates (Figure 2); differences in HC50 were generally smaller than differences in HC5. However, inclusion of the microbial data decreased the uncertainty for the estimated HC5 or HC50 values (i.e., smaller confidence intervals), thereby increasing the robustness of the SSD. Moreover, including microbial endpoints into the SSD, or environmental risk assessment in general, makes the assessment more relevant as this allows consideration of critical soil functions with respect to soil fertility and nutrient cycling. These soil functions are essential because they also may affect vegetation, habitat, and soil invertebrate communities. Such effects will not become apparent in standard plant and invertebrate tests because nutrient and food supply is generally optimized in these assays. The examples for Cu, Ni, Zn, and Mo show that protecting higher-order organisms (plants and invertebrates) also protects the microbial communities (i.e., lower HC5 for plants and invertebrates compared with microbial processes) (Figure 2). These comparisons were based on the EC10 and NOEC values only. Data for Mo show that at higher effect levels (e.g., EC50 values), microbes are often less sensitive than are plants (data not shown). This may be attributed to functional redundancy (i.e., changes in community composition will prevent large changes in the microbial endpoints measured), resulting in a less steep dose–response curve compared to single-species tests. However, this phenomenon was not observed for all microbial endpoints; it was much more pronounced for endpoints related to C mineralization in comparison to nitrification assays. Protection of microbial functions and processes is considered more relevant for protection of soil functions and ecosystems than protecting the individual most sensitive species (Cairns 1986). These observations warrant the inclusion of the tests for soil microbial functions into an ecological risk assessment, and in the development of SCVs for metals and metalloids.
Affiliation: US Army Edgewood Chemical Biological Center, Aberdeen Proving Ground, Maryland.