Modelling the effect of spatial variation in postlarval supply and habitat structure on recruitment of Caribbean spiny lobster

Abstract

Many field studies have shown that recruitment of the Caribbean spiny lobster, Panulirus argus, is sensitive to variation in both postlarval supply and local nursery-habitat structure. We used a spatially explicit individual-based model, to investigate the effects of (i) spatio-temporal variation in postlarval supply and (ii) changes in the spatial structure of the nursery habitat on lobster recruitment to the Florida Keys, Florida (USA). By simulating eight different regional scenarios describing postlarval supply, we investigated whether differences in the spatio-temporal delivery of postlarvae to the Florida Keys alters recruitment of subadult lobsters. Our results indicate that random geographical variation in postlarval supply yields the highest predicted recruitment, whereas persistently patchy settlement yields the lowest. Field observations of postlarval supply suggest that the random model is the most realistic. In separate simulations, we determined the sensitivity of the model to changes in the geographic arrangement of nursery habitat and the spatial resolution of habitat structure. The most spatially explicit depictions of habitat structure yielded small, but marginally significant differences in lobster recruitment as compared with more generalized spatial scenarios. These differences may well be magnified when more detailed depictions of postlarval settlement are implemented in the model.