Effects of Vertebrate Herbivores on Forest Structure, Diversity, and Biogeography

One hypothesized outcome of both modern defaunation and past megafaunal extinctions is the reduction of herbivory (including plant consumption as well as trampling, rooting, and digging) by large vertebrates.  Large mammalian herbivores (especially megafauna) often physically damage plants during their foraging activities, creating trails, reducing seedling density, and possibly altering forest structure and diversity on a broader level.  These factors may combine to drive some of the biogeographic differences in forest structure and diversity between Neotropical and Paleotropical forest regions.  In a collaborative effort (with John Poulsen, Varun Swamy, and Alys Granados), we're studying this process in both protected and hunted/logged forests in Peru, Guyana, Gabon, and Malaysian Borneo.

Assessing the role of physical damage to seedlings can be difficult, given that other causes of damage – herbivory, pathogens, withering, etc. – result in visually similar outcomes for the seedling.  Clark and Clark (1989) developed an inert “artificial seedling” model to isolate the effects of physical damage from other causes of mortality.  The model is constructed of two plastic drinking straws attached to a stiff wire “root”, and replicates natural seedlings in general shape, flexibility, and resilience to damage.

Using this model gives us a rare opportunity to compare sites with strikingly different animal communities, to understand both natural biogeographic variation and the impacts of defaunation on seedling mortality and forest structure.  We’ve been particularly interested to see how the process of physical damage varies in forests with megafauna such as elephants versus those without.  Check out the video at right to see elephant trampling in action in Gabon - keep an eye on the artificial seedling circled in red.  

We've found that vertebrate physical damage is a consistent force in forests across the tropics, and that hunting significantly alters its strength.  We documented a ~70% decrease in damage in hunted vs. intact sites that resulted in a ~3.5-fold (350%) increase in artificial seedling survival.  Physical damage by terrestrial vertebrates may be a driver of divergent patterns of forest structure in Afrotropical vs. Neotropical forests, but continued comparative research is needed to understand this process.  You can read more about this project in our paper here.

Along with the artificial seedling model, we have studied the effects of elephant foraging on the structure and diversity of the forest in Ivindo, Gabon.  We set up plots around "elephant trees" that were likely to have high elephant activity (they produce attractive fruits and had visible elephant damage on their trunks), and for all woody stems within (>26,000 in total!), we identified, measured, and assessed them for damage.  There were some clear elephant effects, and other effects that were not so clear.  One significant result is that elephant activity close to high-traffic trees reduced sapling diversity (figure at right).   We are now working to expand on this research theme and establish a collaborative project studying the effects of megafauna (or lack thereof) in Afrotropical vs. Neotropical forests, with work in Gabon and Guyana.  Stay tuned for more info soon.

Our project sites, representing the three main regions of tropical forest. Adapted from a map by Mongabay.com