An ecosystem approach to soil toxicity testing: a study of copper contamination in laboratory soil microcosms

Abstract

An ecosystem approach to soil toxicity testing allows for integration of the effects of chemical contaminants on different components of the soil food web (system structure) and ecosystem-level processes (system function). We used this approach to study copper contamination in small laboratory soil microcosms. Microcosm soils were treated with CuSO 4 at the following concentrations: 0, 50, 100, 200, 400, and 800 mg Cu kg −1 soil. Five, 10, 20 and 40 days after soil treatment, we made the following organism-level measurements: microbial biomass N, substrate-induced respiration (SIR) and soil urease activity; total nematode numbers; earthworm mortality, growth and body accumulation of Cu. Our process-level measurements were net N mineralization and litter decomposition. SIR was the most sensitive of the parameters measured with significant effects observed at Cu concentrations as low as 50 mg kg −1. Microbial biomass N and earthworm growth showed intermediate sensitivity with effects at 200 mg kg −1 Cu. The least sensitive organism-level parameters were soil urease activity and nematode abundance, both showing significant effects only at 800 mg kg −1 Cu. At the process-level, there was an inhibition of litter decomposition starting at 100 mg kg −1 Cu, and a sharp increase in net N mineralization at 800 mg kg −1 Cu. By examining both the structure and function of the soil system, we were able to link the direct effects of copper on organisms to indirect effects on ecosystem-level processes and were able to suggest mechanisms to account for our results. The release of nitrogen from microbial cells killed by direct toxicity of Cu at 800 mg kg −1 resulted in a transient increase in dissolved organic N, followed by a flush of N mineralization, resulting in large increase in NH 4N compared with the untreated control. Microbial mortality also apparently led to the inhibition of litter decomposition. Measuring the effects of contamination at different trophic levels simultaneously, and linking them to ecosystem processes, provided insights into the ecological mechanisms of the observed effects. This makes the ecosystem approach particularly valuable for analyzing the highly complex soil system. We suggest that the information obtained in a laboratory test based on the ecosystem approach is the most appropriate method for extrapolation to field situations.