Abstract
To identify temporal and spatial patterns in the supply of late-stage fish larvae to the coral reefs of the upper Florida Keys, 3 replicate light traps that collect larvae in the process of settling were deployed at each of 2 reefs from May 2002 to February 2004. Traps were deployed every other night from May through October and twice monthly from November through April to examine cross-shelf (2002 and 2003) and alongshore (2002 and 2004) patterns. The nightly abundance of settling larvae was compared to concurrently collected environmental data to assess the relationships between these variables and temporal patterns of larval supply. In total, 26 185 fish larvae from 55 families were collected during 174 nights of sampling. Seasonally, the supply of larvae peaked in the late winter to early spring and reached a minimum in late fall. Within-season supply of late-stage larvae was cyclic and strongly related to the lunar cycle, and to a lesser extent, the maximum tidal amplitude cycle. Strong negative correlations between larval supply and both moon illumination and maximum tidal amplitude each night resulted in peak supply of fish larvae between the third quarter moon/minimum amplitude tides and the new moon/maximum amplitude tides. While these cyclic environmental cues provided a high degree of temporal predictability of larval pulses, the magnitude of pulses was stochastic, with some variability related to the passage of mesoscale frontal eddies by the upper Florida Keys. However, the effects of these physical features on larval supply were not consistent over time. Spatial variability (kilometer-scale) was greater in the cross-shelf direction than in the alongshore direction: the supply of larval fishes was significantly greater to the fringing reef site (French Reef) than to the inshore patch reef (White Banks), due to either active larval avoidance of inshore environments or depletion of larvae originating offshore as they pass over and settle on the fringing reef.
