Characterizing toxicity of metal-contaminated sediments from mining areas

Applied Geochemistry
By: , and 



This paper reviews methods for testing the toxicity of metals associated with freshwater sediments, linking toxic effects with metal exposure and bioavailability, and developing sediment quality guidelines. The most broadly applicable approach for characterizing metal toxicity is whole-sediment toxicity testing, which attempts to simulate natural exposure conditions in the laboratory. Standard methods for whole-sediment testing can be adapted to test a wide variety of taxa. Chronic sediment tests that characterize effects on multiple endpoints (e.g., survival, growth, and reproduction) can be highly sensitive indicators of adverse effects on resident invertebrate taxa. Methods for testing of aqueous phases (pore water, overlying water, or elutriates) are used less frequently. Analysis of sediment toxicity data focuses on statistical comparisons between responses in sediments from the study area and responses in one or more uncontaminated reference sediments. For large or complex study areas, a greater number of reference sediments is recommended to reliably define the normal range of responses in uncontaminated sediments – the ‘reference envelope’. Data on metal concentrations and effects on test organisms across a gradient of contamination may allow development of concentration-response models, which estimate metal concentrations associated with specified levels of toxic effects (e.g. 20% effect concentration or EC20). Comparisons of toxic effects in laboratory tests with measures of impacts on resident benthic invertebrate communities can help document causal relationships between metal contamination and biological effects. Total or total-recoverable metal concentrations in sediments are the most common measure of metal contamination in sediments, but metal concentrations in labile sediment fractions (e.g., determined as part of selective sediment extraction protocols) may better represent metal bioavailability. Metals released by the weak-acid extraction of acid-volatile sulfide (AVS), termed simultaneously-extracted metals (SEM), are widely used to estimate the ‘potentially-bioavailable’ fraction of metals that is not bound to sulfides (i.e., SEM-AVS). Metal concentrations in pore water are widely considered to be direct measures of metal bioavailability, and predictions of toxicity based on pore-water metal concentrations may be further improved by modeling interactions of metals with other pore-water constituents using Biotic Ligand Models. Data from sediment toxicity tests and metal analyses has provided the basis for development of sediment quality guidelines, which estimate thresholds for toxicity of metals in sediments. Empirical guidelines such as Probable Effects Concentrations or (PECs) are based on associations between sediment metal concentrations and occurrence of toxic effects in large datasets. PECs do not model bioavailable metals, but they can be used to estimate the toxicity of metal mixtures using by calculation of probable effect quotients (PEQ = sediment metal concentration/PEC). In contrast, mechanistic guidelines, such as Equilibrium Partitioning Sediment Benchmarks (ESBs) attempt to predict both bioavailability and mixture toxicity. Application of these simple bioavailability models requires more extensive chemical characterization of sediments or pore water, compared to empirical guidelines, but may provide more reliable estimates of metal toxicity across a wide range of sediment types.

Additional publication details

Publication type Article
Publication Subtype Journal Article
Title Characterizing toxicity of metal-contaminated sediments from mining areas
Series title Applied Geochemistry
DOI 10.1016/j.apgeochem.2014.05.021
Volume 57
Year Published 2015
Language English
Publisher Elsevier
Contributing office(s) Eastern Mineral and Environmental Resources Science Center, Contaminant Biology Program
Description 12 p.
First page 73
Last page 84
Online Only (Y/N) N
Additional Online Files (Y/N) N
Google Analytic Metrics Metrics page
Additional metadata about this publication, not found in other parts of the page is in this table