Introduced species are having major impacts in terrestrial, freshwater, and marine ecosystems worldwide. Given that resources for management are limited and that only a small percentage of invaders are likely to cause large ecological change, management priorities should be based on the severity of immediate and anticipated impacts on native assemblages and commercial species. This paper synthesizes work on the current and predicted impacts of an introduced predatory seastar (Asterias amurensis) on soft sediment assemblages, including native species subject to commercial fishing, in the Derwent Estuary and other areas of southeast Tasmania. Due to the absence of baseline data prior to the arrival of the seastar and the presence of other anthropogenic stressors in the estuary, estimating the impact of the seastar is difficult. To help overcome the weaknesses of any single method, our assessment of impact rests on 'weight of evidence' from multiple approaches. Results from experimental manipulations at small scales, detailed observations of feeding, and field surveys over a range of spatial scales in areas with and without the seastar provide strong evidence that predation by the seastar is likely to be responsible for the decline and subsequent rarity of bivalve species that live just below or on the sediment surface in the Derwent Estuary. The data suggest that should seastar densities in other areas on the Tasmanian coast attain the current levels in the Derwent Estuary, there are likely to be large direct effects on native assemblages, particularly on populations of large surface dwelling bivalves, including several commercial species. Given the seastar's ability to exploit a broad range of food resources other than bivalves, and the functional importance of bivalves in native systems, we predict broader direct and indirect effects on native assemblages. We would be unable to reach these same conclusions from a single approach to assessing impacts. The overall picture from the combination of methods at different scales provides more information than the sum of the results of the separate lines of investigation.
