Partitioning species variability from soil property effects on phytotoxicity: ECx normalization using a plant contaminant sensitivity index

Soil properties mitigate hazardous effects of contaminants through soil chemical sequestration and should be considered when evaluating ecological risk from terrestrial contamination. Empirical models that quantify relationships between soil properties and toxicity to ecological receptors are necessary for site-specific adjustments to ecological risk assessments. However, differential sensitivities of test organisms in dose-response studies may limit the utility of such models. We present a novel approach to toxicity estimation that partitions the effect of differential sensitivities of test organisms from that of soil chemical/physical properties. Five soils that ranged in selected properties were spiked with five concentrations of sodium arsenate. Bioassays were conducted where above ground dry matter growth and the corresponding tissue arsenic concentrations were evaluated for three terrestrial plants (Alfalfa, Medicago sativa L.; Perennial ryegrass, Lolium perrene L.; and Japanese millet, Echinochloa crusgalli L.). Estimates were combined into a plant contaminant sensitivity index (PCSI) and used to normalize phytotoxicity parameters to the most sensitive species (i.e., alfalfa) where necessary. Simple linear regression and ANCOVA indicated a 36.5% increase in the explanatory power of the modifying effects of soil properties on phytotoxicity when differential arsenate sensitivities were accounted for by PCSI (r(2) = 0.477-0.833). Normalization of ecotoxicity parameters by PCSI is a seemingly effective approach to quantify the modifying effects of soil properties on phytotoxicity endpoints when it is of interest to consider multiple plant species (or varieties within a species) with differential sensitivities to experimental contaminants.