PS 15-128 - Macro-ecological scale geographic variation of a keystone predator in a common garden environment

Monday, August 12, 2019
Exhibit Hall, Kentucky International Convention Center
Kenzi M. Stemp1,2, Thomas L. Anderson1 and Jon M. Davenport1,2, (1)Biology, Appalachian State University, Boone, NC, (2)Biology, Southeast Missouri State University, Cape Girardeau, MO
Background/Question/Methods

Geographic variation in species behavior and life history has been well documented in biology. Species with wide geographic distributions (i.e. across a continent) and small home ranges (i.e. <2km2) are especially likely to experience variability in abiotic environments across the entirety of their range, and possibly strong local adaptation. This can result in potential differences in species interactions. Understanding variation on a geographic scale is especially important when considering species of ecological importance, such as keystone species. Yet, few studies have compared the potential cascading ecological effects of keystone predators on a macro-geographic scale. To understand how keystone ability in pond food webs can vary across a large geographic range, we conducted an artificial pond experiment with a known keystone predator, the marbled salamander (Ambystoma opacum). To do so, we collected size-matched salamander larvae from three geographically distant populations (>650km apart) in Ohio, Mississippi, and North Carolina and placed them in mesocosms with a suite of spring-breeding amphibian prey species. We hypothesized that larval A. opacum would have similar keystone effects on amphibian prey communities. We predicted that predatory effects on amphibian prey communities would positively affect lower trophic levels (zooplankton and periphyton).

Results/Conclusions

Our results indicate differential predation on prey species, leading to differences in prey diversity among the three predator populations. Simpson’s diversity index of the prey community was lowest in the Ohio populations of salamanders (F3,11=4.811, p=0.022) indicating that keystone function may vary at a geographic scale. These differences in prey diversity were a result of leopard frog and chorus frog survival being significantly lower in North Carolina and Ohio populations, relative to the predator-free control (F3,11 > 7.0, p> 0.016). However, leopard frog larval period was shorter in Ohio populations than in control or North Carolina populations (F3,11=4.569, p=0.026. We did not find a significant effect of predator population on zooplankton abundance or primary productivity. Spatial variation in ecological processes is a widespread phenomenon, driven by variation in prey assemblages, behavioral mechanisms of predators, and climactic factors. In order to move forward in making widespread generalizations, an understanding of this variation over a large spatial scale, and its resulting effects, is important. Further understanding of large-scale variability in ecologically significant predators is needed to direct conservation efforts and potentially preserve biodiversity.