Early postsettlement predation on three reef fishes: Effects on spatial patterns of recruitment

Document Type


Date of Original Version



Marine organisms suffer extensive mortality just after settling from the plankton, yet, little is known about the role that predators play in causing this mortality. We estimated the rates of predation in the first 24 h, and the first week, after settlement in three species of reef fish. To estimate these rates we compared the accumulation of recent settlers on plots from which predators were excluded (by caging) to settlement on unmanipulated control plots. The magnitude of predation varied greatly among our three focal species, even though they are ecologically similar (all are small gobies that inhabit the reef/sand interface). Within 24 h of settlement, predators apparently killed 92% of settled blackeye gobies, 26% of bridled gobies, and 6% of goldspot gobies. Within a week, predation had significantly reduced recruitment of all three species and was the main cause of death during this period. We also tested whether predation on new settlers affected spatial patterns of abundance at two scales. At small scales (areas separated by tens of meters) the distortion of settlement patterns increased as the magnitude of postsettlement predation increased. Blackeye gobies suffered intense, density-dependent, postsettlement predation that completely obscured spatial patterns of settlement and larval supply. Moderate, density-independent predation on bridled gobies only moderately distorted spatial patterns of settlement and larval supply. Low, density-independent predation on goldspot gobies did not greatly alter patterns of settlement of this species, but unlike the other two species, settlement of goldspot gobies was not correlated with larval supply at small scales. At a larger spatial scale (sites separated by hundreds to thousands of meters), predation failed to distort significantly the input rates of any of the three gobies, and differences between sites in larval supply and settlement were still apparent up to a week after settlement. Overall, our results indicate that predation immediately following settlement can distort spatial patterns of input extremely rapidly, but the degree to which settlement patterns are distorted may be highly species- and scale-dependent.

Publication Title, e.g., Journal