1887 Environmental Toxicology and Chemistry, Vol. 25, No. 7, pp. 1887–1898, 2006  2006 SETAC Printed in the USA 0730-7268/06 $12.00 + .00 CHRONIC SUBLETHAL SEDIMENT TOXICITY TESTING USING THE ESTUARINE AMPHIPOD, MELITA PLUMULOSA (ZEIDLER): EVALUATION USING METAL-SPIKED AND FIELD-CONTAMINATED SEDIMENTS SHARYN A. GALE,² C ATHERINE K. KING,‡ and ROSS V. HYNE*§ ²Centre for Ecotoxicology, Department of Environmental Sciences, University of Technology-Sydney, P.O. Box 29, Lidcombe, New South Wales 1825, Australia ‡Centre for Environmental Contaminants Research, Energy Technology, CSIRO, PMB 7, Bangor, New South Wales 2234, Australia §Centre for Ecotoxicology, Ecotoxicology and Environmental Contaminants Section, Department of Environment and Conservation, P.O. Box 29, Lidcombe, New South Wales 1825, Australia ( Received 9 August 2005; Accepted 20 December 2005) Abstract—The present study describes the development of a 42-d chronic sublethal sediment toxicity test using the estuarine amphipod Melita plumulosa (Zeilder). This test was shown to predict the toxicity of metal-contaminated sediments previously found to adversely affect benthic community structure. Metals initially were tested individually by spiking reference sediment under conditions that ensured low metal concentrations in pore waters. Fertility was the most sensitive sublethal endpoint for copper- and zinc-spiked sediments, whereas cadmium-spiked sediments were not toxic to M. plumulosa, despite their high bioaccumulation of cadmium. The 42-d chronic sediment test was reproducible; however, variation between reference sediments collected from the same field location over time or from different locations did affect the reproduction of M. plumulosa. Sensitivity of M. plumulosa to metal-spiked sediments suggested that the interim sediment-quality guidelines (ISQGs) were too conservative. However, toxicity testing of sediments collected from field sites known to affect community assemblages significantly ( p  0.001) reduced the fertility of M. plumulosa, reflecting benthic community survey results and supporting the ISQGs. Bioaccumulation of cadmium and copper by M. plumulosa was elevated following chronic exposure to both laboratory and field-contaminated sediments; however, zinc bioaccumulation could be measured only in M. plumulosa exposed to field-contaminated sediments. Keywords—Sediment Toxicity test Sublethal Amphipod Metals INTRODUCTION Metal-contaminated sediments pose a threat to aquatic eco- systems as well as human health (http://www.environment. nsw.gov.au/soe/soe2003/chapter5/chp5.7.htm). In New South Wales (NSW), Australia, the most severely contaminated sed- iments typically are located in the upper reaches of estuaries, bay ends, and coastal saltwater lagoons that have industrial activities and intense urbanization in their catchments, includ- ing Port Jackson, Lake Macquarie, and the Hunter River [1– 3]. Typical metal contaminants in sediments include cadmium, copper, zinc, and lead, which often exceed background con- centrations by several orders of magnitude [1–3]. The NSW Environment Protection Authority (http://www.environment. nsw.gov.au/soe/soe2003/chapter5/chp5.7.htm) highlighted the need for systematic identification and documentation of contaminated sediments in NSW to assist the management of potential health and ecological impacts. Toxicity tests are an important tool in predicting the effects of contaminated sediments on biota [4]. In situ tests are con- sidered to be the most environmentally relevant, because test organisms are exposed to environmental variables. Often, how- ever, the results are confounded, making interpretation diffi- cult, and tests are demanding in terms of both time and re- sources. In comparison, uncertainties are associated with the ability of laboratory sediment toxicity tests to predict the im- * To whom correspondence may be addressed (ross.hyne@environment.nsw.gov.au). pact of contaminated sediments on benthic communities in the field. Pore-water toxicity tests are susceptible to changes in chemical parameters that may confound results [5,6]. Also, pore-water toxicity tests do not test the toxicity caused by sediment ingestion, which is a significant exposure route for some organisms [7,8]. Whole-sediment toxicity tests are more environmentally realistic than pore-water toxicity tests, be- cause the sediment structure is relatively maintained and all exposure routes, including overlying water, pore water, sedi- ment particles, and food, are present [6,7]. Chronic whole- sediment toxicity tests can predict the long-term impacts of contaminated sediments, providing valuable information to as- sist regulators and environmental managers. In Australasia, few chronic whole-sediment marine toxicity tests have been published, and many of those have limited environmental relevance and/or rigor. Simpson et al. [9] (http://www.clw.csiro.au/cecr/publications.html) listed the chronic estuarine and marine whole-sediment toxicity tests de- veloped in Australia, all of which measure growth as the end- point except for the present study, which also measures re- production. Reproduction is likely to have more environmental relevance than growth, because it is directly linked to changes in population size and the ability to repopulate after stress. Many of the commonly used test species in Australia, such as Corophiurm colo [10–12], must be collected from field pop- ulations, because culturing protocols have not been developed. However, field collection of test species can be unreliable as a result of temporal variability in population density, repro-