PS 21-50
Trophic cascades in agricultural landscapes

Tuesday, August 6, 2013
Exhibit Hall B, Minneapolis Convention Center
Tania N. Kim, Great Lakes Bioenergy Research Center, University of Wisconsin-Madison, Madison, WI
Heidi Liere, Department of Entomology, University of Wisconsin - Madison, Madison, WI
Ben P. Werling, Entomology, Michigan State University, East Lansing, MI
Timothy D. Meehan, The Nature Conservancy, Boulder, CO
Doug A. Landis, Entomology, Michigan State University, East Lansing, MI
Claudio Gratton, Department of Entomology, University of Wisconsin - Madison, Madison, WI
Background/Question/Methods

Trophic cascades occur when variation in predator communities cascade down the food chain to influence communities at lower trophic levels. While the existence of trophic cascades has been well documented, the prevalence of trophic cascades is largely dependent on food web structure. Because landscape context has been known to influence food web structure, we asked whether landscape composition can affect plant performance through cascading, indirect effects on crop pests and their natural enemies. In 2012, sites varying in landscape composition (percent soy, corn, and grassland within 1500 m radius around site centers) were established in two regions (southern Wisconsin and central Michigan). Sites were surveyed for arthropod predators and herbivores throughout the growing season using sweep net, pitfalls, and sticky cards. To examine predator effects on herbivore population growth and plant performance, focal plants were caged to exclude predators (control plants were surrounded by partial cages) and plants were stocked with herbivores (aphids for soy and wheat, diamondback moths for collards). Throughout the growing season, plants were surveyed for arthropods. At the end of the season, plant performance was assessed (e.g. seed mass for wheat, biomass for soy, and leaf damage for collards).

Results/Conclusions

Landscape composition affected predator communities, however the magnitude of cascading effects to plant performance varied with plant species and region. In soy fields, landscape composition influenced pest communities but did not influence soy biomass. On the other hand, landscape composition affected wheat seed mass in WI only, but landscape effects were not mediated through changes in pest abundances. For collards, the effects of landscape on predators cascaded down the food chain to influence pest abundance and plant damage; these effects were consistent across WI and MI. Our results indicate that the existence and magnitude of trophic cascades can vary with landscape context; landscape to plant performance cascades were weaker and less prevalent than expected.