The strength and prevalence of trophic cascades, defined as positive, indirect effects of natural enemies (predatory and parasitic arthropods) on plants, is highly variable in agroecosystems. This variation may in part be due to the spatial or landscape context in which hese trophic cascades occur. In 2011 and 2012, we conducted a natural enemy exclusion experiment in soybean fields along a gradient of landscape composition across southern Wisconsin and Michigan, USA. We used structural equation modeling to ask (1) whether natural enemies influence biocontrol of soybean aphids (SBA) and soybean yield and (2) whether landscape effects on natural enemies influence the strength of the trophic cascades. We found that natural enemies (NE) suppressed aphid populations in both years of our study, and, in 2011, the yield of soybean plants exposed to natural enemies was 37% higher than the yield of plants with aphid populations protected from natural enemies. The strength of the :rophic cascade was also influenced by landscape context. We found that landscapes with a higher proportion of soybean and higher diversity habitats resulted in more NE, fewer aphids, and, in some cases, a trend toward greater soybean yield. These results indicate that landscape context is important for understanding spatial variability in biocontrol and yield, but other factors, such as environmental variability and compensatory growth, might overwhelm the beneficial effects of biocontrol on crop yield.

Biomass harvest may eventually be conducted on over 100 000 000 ha of US crop and forest lands to meet federally‐mandated targets for renewable biofuels. Such large‐scale land use changes could profoundly impact working landscapes and the arthropod communities that inhabit them. We review the literature on dedicated biofuel crops and biomass harvest from forests to look for commonalities in arthropod community responses. With expanded biofuel production, existing arthropod pests of biofuel crops will likely become more important and new pests will emerge. Beneficial arthropods will also be influenced by biofuel crop habitats, potentially altering the distribution of pollination and pest control services to the surrounding landscape. Production of biofuel crops including initial crop selection, genetic improvement, agronomic practices, and harvest regimes will also influence arthropod communities. In turn, arthropods will impact the productivity and species composition of biomass production systems. Some of these processes have the potential to cause landscape‐level changes in arthropod community dynamics and insect‐vectored plant diseases. Finally, changes in arthropod populations and their spatiotemporal distribution in the landscape will have impacts on consumers of insects at higher trophic levels, potentially influencing their population and community dynamics and producing feedbacks to arthropod communities. Given that dedicated biofuel crops and intensified biomass harvest from forests are still relatively uncommon in North America, as they increase, we anticipate ‘predictably unpredictable’ shifts in arthropod communities and the ecosystem services and functions they support. We suggest that research on arthropod dynamics within biofuel crops, their spillover into adjacent habitats, and implications for the sustainability of working landscapes are critical topics for both basic and applied investigations.

No abstract