Skip to main content
Dryad logo

Plant Biomass data from: Bottom-up Herbivore-Plant Feedbacks Trump Trophic Cascades in a Wolf-Elk-Grassland System


Weeks, Trevor; Merril, Evelyn; Hebblewhite, Mark (2020), Plant Biomass data from: Bottom-up Herbivore-Plant Feedbacks Trump Trophic Cascades in a Wolf-Elk-Grassland System, Dryad, Dataset,


Top-down predator-prey effects that alter the abundance, biomass, or productivity of a population community across more than one link in a food web are referred to as trophic cascades. While these effects have been extensively studied in aquatic environments, fewer studies have examined trophic cascades in terrestrial ecosystems. And fewer still terrestrial studies have tested for trophic cascades between vertebrates and grassland vegetation. Across the globe, grassland plant biomass is driven by both precipitation and non-linear positive feedbacks between grazing and plant productivity, as predicted by the Intermediate Grazing Hypothesis. Yet little is known about the role that apex carnivores play in regard to trophic impacts on grassland biomass. We utilized a long-term dataset collected over the last two decades on a montane rough-fescue grassland adjacent to Banff National Park, Alberta, to test whether top-down effects regulate grassland biomass in a wolf-elk system. First, we measured annual growing season plant biomass from 2006 – 2018 at 61 repeat sampled plots in the grassland. Next, we measured wolf predation risk using a previously developed wolf resource selection function created from GPS radiocollar data from 5 wolf packs. Finally, we measured grazing intensity using Brownian Bridge Movement Models derived from GPS radiocollar data from 131 unique elk. We then tested top-down, bottom-up and abiotic hypotheses for grassland biomass over time in program R. The top model incorporated precipitation and positive non-linear effects of elk use, excluding predator effects and thus failing to support the trophic cascade hypothesis. This may be due to the observational nature of this study, or predation effects in this system may be obscured by human use. Alternatively, our results also support the hypothesis that intermediate grazing may outweigh the benefits of predation in grassland systems. Our study serves to help fill a gap in trophic cascade literature, and emphasizes that positive feedback between grazers and grasslands may trump top-down effects. Understanding when trophic cascade theory is or is not applicable is vital for carnivore management, conservation, and reintroduction efforts across North America.


We measured annual growing season plant biomass from 2006 – 2018 at 61 repeat sampled plots. At each plot, we measured standing biomass height at four subplots in the four corners of a 5m x 5m square (NE, SE, SW, and NW) (Fig. 2). We indirectly measured height using a disc-pasture-meter approach (Vartha & Matches 1977, Dӧrgeloh 2002) in a 0.5m x 0.5m quadrat, then visually estimated the percent cover of grasses, forbs, shrub, and bare ground within the quadrat. At this fifth subplot, in addition to the above measurements, we clipped all vegetation down to a grazed level (~ 2 cm). We sorted all clipped vegetation into live or dead grass, live or dead forb, and live or dead shrub. We weighed each vegetation type in the field and recorded wet weights. Each sample was later dried in an oven at 100 oC for 24 hours and then weighed again to determine dry weights. All samples were stored and weighed in brown paper bags, so we removed 7g (the mean weight of a dry paper bag) from the final dry weight for every sample. We recorded the dry weight for each vegetation type for each clipped subplot as well as the total dry weight of the clipped subplot. To estimate dry weight values for the drop-disc-only readings, we used the data from the clipped plots to create a model predicting dry weight values (e.g., Dӧrgeloh 2002, Hebblewhite et al. 2008).


National Science Foundation, Award: Grant DEB 1556248