The effect of lizards on spiders and wasps: variation with island size and marine subsidy

Abstract

AbstractIntroduced predators can have dramatic effects on island ecosystems, the magnitude of which are likely to vary with island characteristics. We investigated the influence of two important properties of islands—size and amount of resource subsidy—on the effects of an introduced predatory lizard (Anolis sagrei) on three groups of arthropod prey. Lizards were experimentally introduced to 16 islands that spanned gradients in vegetated area and seaweed deposition (a marine resource subsidy); 16 similar islands served as lizard‐free controls. The abundance of web spiders, salticid spiders, and wasps was estimated prior to lizard introduction and again four months after lizard introduction. Lizard introduction reduced the average abundance of all three groups of arthropods. The effect of lizards on salticid spiders—which was very large (94% reduction in salticid abundance)—decreased with island size. In contrast, the effect of lizards on wasps—which was also very large (88% reduction in wasp abundance)—tended to increase with island size, but with only marginal significance. There was no evidence for variation in the effect of lizards on web spiders with island size. This variation between prey taxa may be related to the relative importance of environmental stress (such as wind and wave exposure, which tend to be more pronounced on smaller islands) in determining abundance. Salticids seem to tolerate the stressful environmental conditions that characterize smaller islands, allowing for larger lizard effects; wasps seem to be limited by these conditions (either directly or indirectly via reduced prey availability), minimizing lizard effects on smaller islands. There was a marginally significant tendency for the effect of lizards on salticid spiders to be weaker on islands with more seaweed deposition, suggesting that subsidies may play a role in reducing predator effects on islands. Our results highlight the importance of ecological context in determining the top‐down effects of introduced predators and underscore the need to extend existing theories relating island area and community characteristics toward an explicit consideration of species interactions.