RESUMO
Sediment quality values (SQV) are commonly used-and misused-to characterize the need for investigation, understand causes of observed effects, and derive management strategies to protect benthic invertebrates from direct toxic effects. The authors compiled more than 40 SQVs for mercury, nearly all of which are "co-occurrence" SQVs derived from databases of paired chemistry and benthic invertebrate effects data obtained from field-collected sediment. Co-occurrence SQVs are not derived in a manner that reflects cause-effect, concentration-response relationships for individual chemicals such as mercury, because multiple potential stressors often co-occur in the data sets used to derive SQVs. The authors assembled alternative data to characterize mercury-specific effect thresholds, including results of 7 laboratory studies with mercury-spiked sediments and 23 studies at mercury-contaminated sites (e.g., chloralkali facilities, mercury mines). The median (± interquartile range) co-occurrence SQVs associated with a lack of effects (0.16 mg/kg [0.13-0.20 mg/kg]) or a potential for effects (0.88 mg/kg [0.50-1.4 mg/kg]) were orders of magnitude lower than no-observed-effect concentrations reported in mercury-spiked toxicity studies (3.3 mg/kg [1.1-9.4 mg/kg]) and mercury site investigations (22 mg/kg [3.8-66 mg/kg]). Additionally, there was a high degree of overlap between co-occurrence SQVs and background mercury levels. Although SQVs are appropriate only for initial screening, they are commonly misused for characterizing or managing risks at mercury-contaminated sites. Spiked sediment and site data provide more appropriate and useful alternative information for characterization and management purposes. Further research is recommended to refine mercury effect thresholds for sediment that address the bioavailability and causal effects of mercury exposure. Environ Toxicol Chem 2015;34:6-21. © 2014 SETAC.
