Oceanographic Influences on Larval Dispersal and Retention and Their Consequences for Population Connectivity

The larval stage of most coral reef fishes is spent in the pelagic environment, potentially well away from the reef proper. Survival of this stage is tenuous, being mediated by factors such as food availability, predator abundance, and physical conditions. The complex biological and physical interactions of these factors can result in a seemingly stochastic larval supply that drives temporal and spatial variation in recruitment intensity (Cowen, 1985; Roughgarden et al., 1988; Choat et al., 1988). This variation can be a major determinant of adult population sizes (Cushing, 1975; Rothschild, 1986; Sinclair, 1988). A growing awareness of this perspective has placed a premium on the need to study the early life history stages of marine organisms, forcing us to peer into the black box of larval biology and ecology. The following discussion will outline where advances have occurred in our understanding of how, and to what effect, coral reef fish larvae interact with the pelagic environment. Without denying the importance of food and predation to larval survival (Houde, 1987; Bailey and Houde, 1989), it is clear that many species of coastal marine fishes are particularly dependent on directed transport to their juvenile or adult habitat. For example, coral reef fish larvae will not survive the dispersive stage of their life history without appropriate transport back to a reef (Leis, 1991a). Similar arguments can be made for many temperate species that are estuarine dependent or benthic oriented (Boehlert and Mundy, 1988). In contrast, many pelagic species do not require directed transport to complete their larval life, although some species may recruit only if their larvae are retained within a specific area [e.g., herring populations (Sinclair, 1988)]. Although fish larvae are often (erroneously) considered strict constituents of the zooplankton community, evidence suggests that many species exhibit some form of active behavior during their pelagic stage (Neilson and Perry, 1990; Leis, 1991b). Although some larval distributions can be explained purely through a combination of spawning sites and passive larval transport [i.e., constant buoyancy (Boehlert and Mundy, 1994)], other larval distributions are influenced by larval behavior interacting with the local current regime. The importance of vertical larval movements in affecting horizontal movements in estuarine systems (Boehlert and Mundy, 1988; Epifanio, 1988) or retention in bank systems (Werner et al., 1993; Tremblay et al., 1994) is well documented. There is also evidence of impressive horizontal swimming abilities in late-stage larvae (see Chapter 8). Thus there is a continuum of possible behaviors (e.g., passive to vertical to horizontal movement), as well as the timing of the onset or ontogeny of such behaviors. Such behavioral capabilities may strongly impact the survival and recruitment of coral reef fish larvae to benthic habitats. The complexity introduced into biological systems by the physics of water movement contributes significantly to the variability associated with the recruitment of marine species. Variability in larval transport will