New Sediment-Toxicity Benchmarks Available for Pesticides in Whole Sediment

Tools such as sediment benchmarks are needed to evaluate the likelihood that pesticides detected in stream sediments may adversely affect stream invertebrate communities. Some pesticides used today in agricultural and urban applications are "hydrophobic" (that is, have low water solubility and tend to bind to particles). Such pesticides may enter streams in surface runoff and with eroded soil, and accumulate in bed sediments. Some hydrophobic insecticides (for example, pyrethroids and fipronil) that are toxic to invertebrates in laboratory tests have been detected in bed sediments of streams in agricultural and urban areas across the United States.

Sediment-toxicity benchmarks were determined for 129 pesticides based on an analysis of compiled literature and database information on laboratory toxicity tests—either spiked-sediment bioassays (preferred) or water toxicity tests (used along with equilibrium partitioning theory if sediment data were not available). Two types of freshwater whole-sediment benchmarks were developed for each pesticide:

Sources/Usage: Some content may have restrictions. View Media Details

• Likely Effect Benchmark — a concentration above which there is a high likelihood of adverse effects on benthic invertebrates, and
• Threshold Effect Benchmark — a concentration below which adverse effects are unlikely.

These benchmarks can be used to

1. interpret which pesticides in a mixture are expected to contribute to observed sediment toxicity; or
2. predict the potential toxicity of pesticides detected in whole sediment in monitoring studies that did not measure toxicity.

In a test to see how well these benchmarks predicted toxicity in field-collected sediments, the new benchmarks were applied to data from two case studies that collected sediment samples from a total of 197 streams across the United States. These case studies measured concentrations of pesticides in whole sediment, and also conducted laboratory whole-sediment-toxicity tests with invertebrates (amphipods and midge) using the same sediments.

In these case studies, the new benchmarks predicted toxicity to amphipods quite well. Toxicity to amphipods occurred in 100 percent of the few (three) samples that exceeded the Likely Effect Benchmark for one or more pesticides, and in 72 percent of samples that were above one or more Threshold Effect Benchmarks. In contrast, there was relatively lower (18 percent) toxicity in samples with pesticide concentrations below all benchmarks. None of the sediment benchmarks were effective in predicting toxicity or non-toxicity to midges, possibly because midges are less sensitive than amphipods or because of insufficient toxicity data for midges.

Factors affecting toxicity when sediment concentrations are below benchmarks may include the presence of other contaminants besides the pesticides being measured and physical or chemical characteristics of sediment that affect pesticide bioavailability or toxicity.

Although there is uncertainty in the benchmark values because of the scarcity of spiked-sediment bioassay data available for many pesticides, the new toxicity-based sediment benchmarks utilize the best available approaches for evaluating pesticides in whole sediments, and they can be used to assess the biological significance of pesticide concentrations measured in whole-sediment monitoring studies.

This research was funded by the USGS Ecosystems Mission Area’s Environmental Health Program (Contaminant Biology and Toxic Substances Hydrology) and the USGS National Water-Quality Assessment (NAWQA) Project of the National Water Quality Program.