Relationships between recruitment and postrecruitment processes in lagoonal populations of two coral reef fishes

Abstract

Pomacentrus amboinensis Bleeker and Dascyllus aruanus (L.) exhibit similar large scale patterns in recruitment to One Tree lagoon, southern Great Barrier Reef. They recruit in greatest numbers near the windward perimeter compared with the centre of the lagoon, and at the perimeter, recruit preferentially to deeper sites (4–8 m depth). This pattern was observed in recruitment to standard coral heads of Pocillopora damicornis (L.), a preferred substratum transplanted to a range of locations and depths. As substratum was held constant, spatial variation in recruitment was not determined by the availability of suitable habitat. Patterns in growth and mortality of the two fish species were monitored for juveniles transplanted to another set of standard coral heads, in order to determine whether post-recruitment processes would reinforce or disrupt patterns established at the time of recruitment. The two species exhibited contrasting patterns. Pomacentrus amboinensis exhibited faster growth and better survivorship at deeper sites, regardless of the location in the lagoon, reinforcing the initial pattern. Dascyllus aruanus exhibited faster growth at perimeter sites and significantly better survivorship at central sites, patterns that would tend to disrupt the initial pattern. Within locations, individuals exhibiting slower growth rates ( Pomacentrus amboinensis) or smaller initial size ( Dascyllus aruanus) had reduced life expectancy. The results suggest that location selection by settlers of Pomacentrus amboinensis results in improved growth and/or survival. This was not the case for Dascyllus aruanus, which is probably more influenced by the presence of conspecifics. However, even for Pomacentrus amboinensis, large-scale patterns probably reflect constraints on habitat selection. Recruitment in Pomacentrus amboinensis was consistently lower to sites immediately down current of other reefs, regardless of their position. I suggest that the absence of recruits of both species to central sites indicates that this region is a recruitment-shadow. Juveniles settle in preferred habitat at the first opportunity after entering the lagoon, leaving much of the preferred habitat in the centre of the lagoon unoccupied. Patterns are initially recruitment-determined, but either reinforcement or disruption mean that the numerical abundance of adults cannot be predicted on the basis of recruitment data alone.