PS 103-172 - Phorid-ant interactions affect beetle and scale densities and stability in the field: A study of cascading effect of trait-mediated interactions

Friday, August 10, 2012
Exhibit Hall, Oregon Convention Center
Hsun-Yi Hsieh, School of Natural Resources and Environment, University of Michigan, Ann Arbor, Ann Arbor, MI and Ivette Perfecto, School of Natural Resources and Environment, University of Michigan, Ann Arbor, MI
Background/Question/Methods

Realistic food webs are composed of many different interactions – density- or trait-mediated direct, indirect, predation, mutualism, etc., which could in turn be linked with one another. To understand effects at the community level, it is important to understand these linkages. Through a series of laboratory experiments we have identified a cascade of trait-mediated interactions that results in an ant phorid parasitoid facilitating a coccinellid predator. The ant provides protection against predators of scale insects, and the phorid, by triggering anti-parasitoid behavior of the ant, disrupts the protective effect of the ant on the scale insects, allowing the coccinellid to feed and oviposit in patches of high scale densities without being harassed by the ants.

Although all these interactions have been demonstrated experimentally, we lack field observations to demonstrate that this trait-mediated cascade plays out under natural conditions. Here, using a generalized linear mixed model we analyzed field data from a coffee plantation to test whether beetle density is associated with phorid attack intensity, ant density and scale density. Calculating correlation coefficients of beetle and scale insect densities, we also test how ant nest cluster size affects beetle and scale stability.   

Results/Conclusions

Results of GLMM suggest that attack intensity of phorids, cluster size of ants, and scale density significantly increase the densities of coccinellid female adults and larvae, while only scale density affects adult male coccinellid density, although inclusion of the other two variables increases model accuracy.  Since coccinellid larvae are relatively immobile, their high density in patches with ants should be a result of female’s choice of oviposition site.  Previous laboratory experiments corroborate this argument. That is, experimentally, female beetles are able to lay more eggs in treatments where both phorids and ants are present than in treatment with only ants. Previous olfactometer trials also show that female beetles are attracted to phorid-alert pheromones of ants while male beetles show no such response.  Results also show that large ant clusters increase the stability of beetle and scale insect populations.  The results suggest that the complex interactions involving these species have an effect on the density and stability of the coccinellid beetle.  Since the ants nest in shade trees within coffee plantations, agricultural intensification that eliminates shade trees may significantly affect these complex interactions, affecting the biological control of the scale insects by the coccinellid beetle.