Predatory birds and ants partition caterpillar prey by body size and diet breadth

Abstract

The effects of predator assemblages on herbivores are predicted to depend critically on predator-predator interactions and the extent to which predators partition prey resources. The role of prey heterogeneity in generating such multiple predator effects has received limited attention. Vertebrate and arthropod insectivores constitute two co-dominant predatory taxa in many ecosystems, and the emergent properties of their joint effects on insect herbivores inform theory on multiple predator effects as well as biological control of insect herbivores. Here we use a large-scale factorial manipulation to assess the extent to which birds and ants engage in antagonistic predator-predator interactions and the consequences of heterogeneity in herbivore body size and diet breadth (i.e. the diversity of host plants used) for prey partitioning. We excluded birds and reduced ant density (by 60%) in the canopies of eight northeastern USA deciduous tree species during two consecutive years and measured the community composition and traits of lepidopteran larvae (caterpillars). Birds did not affect ant density, implying limited intraguild predation between these taxa in this system. Birds preyed selectively upon large-bodied caterpillars (reducing mean caterpillar length by 12%) and ants preyed selectively upon small-bodied caterpillars (increasing mean caterpillar length by 6%). Birds and ants also partitioned caterpillar prey by diet breadth. Birds reduced the frequency dietary generalist caterpillars by 24%, while ants had no effect. In contrast, ants reduced the frequency of dietary specialists by 20%, while birds had no effect, but these effects were non-additive; under bird exclusion, ants had no detectable effect, while in the presence of birds, they reduced the frequency of specialists by 40%. As a likely result of prey partitioning by body size and diet breadth, the combined effects of birds and ants on total caterpillar density were additive, with birds and ants reducing caterpillar density by 44% and 20% respectively. These results show evidence for the role of prey heterogeneity in driving functional complementarity among predators and enhanced top-down control. Heterogeneity in herbivore body size and diet breadth, as well as other prey traits, may represent key predictors of the strength of top-down control from predator communities.

Dietary specialization is conditionally associated with increased ant predation risk in a temperate forest caterpillar community

“…The findings here confirm those of Singer et al () that ant predation is detectably biased toward dietary specialist caterpillar species in the context of ambient bird predation, but not when birds are experimentally excluded. While there are several possible explanations for this result (Singer et al, ), new evidence from the present study shows that behavioral defenses of caterpillars, in conjunction with diet breadth, are associated with reduced ant predation risk only in the context of bird predation.…”

“…The findings here confirm those of Singer et al () that ant predation is detectably biased toward dietary specialist caterpillar species in the context of ambient bird predation, but not when birds are experimentally excluded. While there are several possible explanations for this result (Singer et al, ), new evidence from the present study shows that behavioral defenses of caterpillars, in conjunction with diet breadth, are associated with reduced ant predation risk only in the context of bird predation. We think this new evidence points toward the second hypothesis outlined in Singer et al (): that bird predation reduces the abundance of alternative arthropod predators (e.g., salticid and thomisid spiders, pentatomid and reduviid bugs), thus enhancing the role of ants as the primary invertebrate predators of caterpillars.…”

“…The factorial exclusion of birds and ants in Singer et al () also revealed a predator–predator interaction that had not been predicted by prior theory or evidence: predation effects of ants on dietary specialist caterpillars were detectable only when birds were not excluded. Although unknown, some possible mechanisms underlying this predator–predator interaction include: (a) bird effects on caterpillar behavior that specifically enhance ant predation risk of dietary specialist caterpillars, (b) bird predation of invertebrate predators, such as spiders, thus indirectly changing ant–caterpillar interactions, and (c) exploitation competition between birds and ants, thus relegating ant predation to caterpillar species avoided by birds (i.e., dietary specialist caterpillars) (Singer et al, ). Regardless of the mechanism, bird and ant partitioning of caterpillars by diet breadth and body size has the potential to generate life‐history trade‐offs among caterpillar species rather than merely select for dietary specialization as the EFSH proposes.…”

“…Bird‐exclusion treatments were applied at the same time as the ant‐exclusion treatments. When analyses showed that some of the ant effects on caterpillars depended on bird exclusion (Singer et al, ) and the analyses for the present study demanded more data from dietary specialist caterpillar species from bird‐exclusion branches (which were lacking in 2012), we set up an additional trial of the ant‐exclusion experiment with all experimental (treatment and control) branches outfitted with bird‐exclusion bags (May–July 2016) using the same methods as we used in 2011. This additional trial was more modest in scope than the ant‐exclusion experiment trials of 2011 and 2012, as it included only a subset of host plant species ( H. virginiana, P. serotina, and Q. alba ) that host the focal dietary specialist caterpillar species for which additional data were needed.…”

“…However, predation at a community level is also subject to combinatorial effects of multiple‐predator types (Sih, Englund, & Wooster, ), and a multiple‐predator perspective has mostly been lacking in tests of the EFSH. A recent study of bird and ant predation on the suppression of forest caterpillars in Connecticut showed evidence for functional complementarity between these two predator groups (Singer, Johnson, Lichter‐Marck, Clark, & Mooney, ). In this case, ant predation was stronger on specialist caterpillars (contrary to the EFSH), while bird predation suppressed generalist caterpillars in support of previous findings (Singer et al, ; Zvereva & Kozlov, ).…”