Limnol. Oceanogr., 44(3, part 2), 1999, 950–97

Invasions by non-indigenous species (NIS) are recognized as important stressors of many communities throughout the world. Here, we evaluated available data on the role of NIS in marine and estuarine communities and their interactions with other anthropogenic stressors, using an intensive analysis of the Chesapeake Bay region as a case study. First, we reviewed the reported ecological impacts of 196 species that occur in tidal waters of the bay, including species that are known invaders as well as some that are cryptogenic (i.e., of uncertain origin). Second, we compared the impacts reported in and out of the bay region for the same 54 species of plants and fish from this group that regularly occur in the region’s tidal waters. Third, we assessed the evidence for interaction in the distribution or performance of these 54 plant and fish species within the bay and other stressors. Of the 196 known and possible NIS, 39 (20%) were thought to have some significant impact on a resident population, community, habitat, or process within the bay region. However, quantitative data on impacts were found for only 12 of the 39, representing 31% of this group and 6% of all 196 species surveyed. The patterns of reported impacts in the bay for plants and fish were nearly identical: 29% were reported to have significant impacts, but quantitative impact data existed for only 7% (4/54) of these species. In contrast, 74% of the same species were reported to have significant impacts outside of the bay, and some quantitative impact data were found for 44% (24/54) of them. Although it appears that 20% of the plant and fish species in our analysis may have significant impacts in the bay region based upon impacts measured elsewhere, we suggest that studies outside the region cannot reliably predict such impacts. We surmise that quantitative impact measures for individual bays or estuaries generally exist for ,5% of the NIS present, and many of these measures are not particularly informative. Despite the increasing knowledge of marine invasions at many sites, it is evident that we understand little about the full extent and variety of the impacts they create—singly and cumulatively. Given the multiple anthropogenic stressors that overlap with NIS in estuaries, we predict NIS–stressor interactions play an important role in the pattern and impact of invasions. Non-indigenous species (NIS) are increasingly conspicuous in marine, and especially estuarine, communities as their number and impacts continue to accumulate. Marine NIS are reported for most geographic regions of the world, and the total number of these species known for some sites or regions underscores the scale and potential ecological significance of these invasions (e.g., Por 1978; Pollard and Hutchings 1990; Hutchings 1992; Boudouresque et al. 1994; Cohen and Carlton 1996; Eno 1996). For example, as many as 230 NIS have been documented for a single estuary, and about 400 NIS are known to be established in marine and estuarine habitats of the continental U.S. (Cohen and Carlton Acknowledgments We thank Denise Breitburg for the invitation to contribute this paper as well as for her comments and patience on previous drafts. We also thank two anonymous reviewers for their recommendations. This manuscript benefited greatly from discussions and various contributions from Jim Carlton, Emily Lipman, Linda McCann, Whitman Miller, Kim Philips, and Ann VonHolle. We especially acknowledge Laura Rodriguez for her perseverance, assistance, and valuable suggestions in data organization and presentation. Finally, we thank Angela Haggins and the library staff of the National Museum of Natural History for their help in obtaining the voluminous references. This research was funded in part by a grant from the U.S. Fish and Wildlife Service and funding from the Smithsonian Institution. 1998; Ruiz et al. 1997). These are only minimum estimates for the extent of NIS invasions, since many NIS will never be recognized as such (Carlton 1996a; Geller 1996). Furthermore, contemporary surveys of NIS are very incomplete, and the rate of new invasions seems to be increasing for many of the best-studied sites (Mills et al. 1993; Ruiz et al. in press; Cohen and Carlton 1998). It is evident that NIS invasions are a significant stressor and force of change in marine communities. Various recent studies have measured effects associated with invasions by plants, fish, crabs, snails, clams, mussels, bryozoans, and nudibranchs (e.g., Race 1982; Brenchley and Carlton 1983; Bertness 1984; Partridge 1987; Allmon and Sebens 1988; Goss-Custard and Moser 1988; Posey 1988; Lambert et al. 1992; Posey et al. 1993; Roberts et al. 1995; Cloern 1996; Grosholz et al. in press). Although these studies provide illustrative examples of some quantified impacts, many additional invaders are believed to have significant impacts based on a broad range of qualitative to quantitative criteria (e.g., see Mills et al. 1993, 1996; Cohen and Carlton 1996; Ruiz et al. in press). Analyses of invasion impacts (as above) have not, however, produced a framework with which to characterize and evaluate impact type, data quality, magnitude of impacts, or spatial and temporal extent of impacts. As a result, the evidence and nature of NIS impacts for particular sites are often unclear. 951 Marine invasions as stressors Interactions between NIS and other anthropogenic stressors may greatly influence colonization and distribution patterns as well as the effects of marine invaders. Estuaries, bays, and other sheltered coastal regions are frequent sites of marine invasions, probably as a result of the relatively high rate of human-mediated transfer compared to exposed coastal habitats. These sheltered coastal regions experience a broad spectrum of anthropogenic disturbances or stressors that overlap and interact with invasion processes. Estuaries are often centers of human activities that release chemical pollutants (e.g., heavy metals, hydrocarbons, chlorine, nutrients), modify thermal environment and habitat characteristics, deplete fishery stocks, increase sedimentation, and increase anoxia (e.g., Kennish 1992; Cooper and Brush 1991, 1993; Rothschild et al. 1994; Cornwell et al. 1996; Nixon 1997). These and other stressors may have a strong influence on invasions, modifying opportunities to invade as well as the subsequent effects. For example, numerous studies demonstrate that physical disturbances allow colonization and persistence by locally rare species (see Sousa 1984; Pickett and White 1985). Furthermore, episodic or chronic disturbance is thought to facilitate invasion by NIS (e.g., Crawley 1986; Fox and Fox 1986; Moyle 1986; Mills et al. 1993; Carlton 1996b), resulting especially from reduced competition or predation (e.g., Robinson and Dickerson 1984; Case 1990, 1996; Baltz and Moyle 1993). Anthropogenic change can provide or cancel a requisite condition for invasion, but many complex interactions may also exist in which anthropogenic stressors influence the probability of establishment and the characteristics of invasions. In this paper, we review and characterize the ecological impacts reported for NIS in Chesapeake Bay and evaluate evidence for the interaction of NIS with other stressors in the bay and elsewhere. More specifically, we have classified the quality of existing data for impacts according to impact type (e.g., competition, predation, bioturbation) and information type (e.g., qualitative to various quantitative measures). Because significant spatial variation may occur for both impact type and magnitude (e.g., Grosholz and Ruiz 1996), we have analyzed data from Chesapeake Bay and other regions separately to be explicit about geographic sources of information. Although we discuss the magnitude of some impacts, we did not attempt to classify impact magnitude (including both spatial scale and severity), which is the topic of a separate paper (Ruiz et al. in prep.). Finally, we have reviewed the present state of knowledge about interactions between NIS and other stressors, using examples from Chesapeake Bay and elsewhere.