stage and exclude more long-lived species, especially epibiota. The harvest impacts vary with method, habitat type, species and size of response organism(s) being studied, and scale of the harvest activity (National Research Council, 2002; Kaiser et al., 2006). In wild mollusc fisheries, dredges or scrapes are normally used to capture epifaunal species (e.g., oysters, scallops, mussels), whereas hydraulic suction dredges or pumps are used to capture infaunal species (e.g., clams). In a review of fishing impacts, Kaiser et al. (2006) found that initial impacts to biota were small and short-lived; however, recovery was slower in muddy and especially in biogenic habitats (e.g., mollusc reefs, seagrass, coral) than in sandy coarse sediments that were subject to higher frequencies of natural disturbances. This was particularly true for the use of mechanical dredges and rakes versus harvest by hand, as numerous studies have demonstrated the significant habitat and community changes caused by these methods (Dayton et al., 1995; Jennings and Kaiser, 1998; Collie et al., 2000; Cranfield et al., 2001; National Research Council, 2002). The community effects and their persistence for small benthic organisms are generally related to mobility and generation time so prolonged effects are only apparent when the benthic fauna is sessile and/or relatively long-lived or when affected areas are so large as to break connections with the surrounding undisturbed habitat.
In many but not all cases, wild-harvest impacts are not directly comparable to bivalve mariculture because culture occurs in a location (shallow and even intertidal habitats) different from that of wild harvest (often deeper subtidal areas), and culturists often transplant harvested individuals from place to place. Bivalve culture can also occur in a different form (e.g., single oysters planted on a tide flat or mussels growing on a line or rack versus an oyster or mussel reef in a wild-harvest scenario) and is typically more concentrated in local areas favorable for growth than wild-stock molluscs. Impacts to wild oyster and mussel reefs are thus potentially more severe and longer lasting than mariculture harvest impacts, and both clam and oyster harvests from these reefs have been shown to cause reef degradation and more substantial losses to oyster resources than clams (Lenihan and Micheli, 1999; Lenihan and Peterson, 2004). Secondary impacts, especially to birds and the less mobile fish and invertebrates, that use the structured habitat for food and protection are also likely to be greater in wild-stock bivalve fisheries that disturb these reefs.
Because of the importance of aquatic vegetation as habitat for other organisms, the effects of harvest activity on these plants have been most studied (Waddell, 1964; Fonseca et al., 1984; Peterson et al., 1987; Orth et al., 2002; Neckles et al., 2005; Wisehart et al., 2007; Tallis et al., 2009). In general, the disturbance to seagrass habitat by mollusc harvest activities