Print PageInsectivorous birds are among the most common top predators around the world, yet studies of their trophic impact have produced equivocal results (e.g. birds control insects in some conditions but not others), leaving open the debate about the strength of terrestrial trophic cascades. One reason this question remains is that all studies to date have been small-scale experimental comparisons (typically single branch bird exclosures); we lack a landscape-level example that demonstrates the impact of insectivorous birds. Here, we take advantage of a unique situation, the loss of all forest birds from the island of Guam after the introduction of the invasive brown tree snake, to determine whether insectivorous birds exert top-down control of a forest food web. Three nearby islands (Rota, Tinian and Saipan) have similar bird communities to Guam prior to the snake invasion, and thus serve as suitable controls.
We set up bird exclosure areas (3.5 x 5.5 m), with adjacent paired control areas, on Guam (8 sites), and Saipan, Tinian and Rota (7 sites per island). In each exclosure and control area, we planted 7-14 seedlings each from six common forest species and measured height, number of leaves and herbivory on each seedling for 5-10 months. We also conducted two comparative studies in native forest on all four islands, one comparing spider abundances and the other comparing leaf herbivory, to determine whether the results found in the experimental study scale up to the landscape level.
Results/Conclusions
Our results do not support the presence of a strong trophic cascade caused by the loss of birds on Guam. The exclosure experiment showed no difference in seedling survival or growth between the exclosure and the control area on Guam, as expected, but also no difference in growth or survival between treatments on islands with birds. There were significantly more spiders on Guam relative to islands with birds, but the leaf herbivory comparison did not show differences between Guam and other islands. In conclusion, our investigation of the first landscape-level bird exclosure experiment (i.e. the island of Guam), we show that birds control large arthropods, but that their effect is attenuated in lower trophic levels. These results suggest that bird-driven trophic cascades may not be as ubiquitous or strong as cascades in other systems.
