Recent models suggest that herbivores optimize nutrient intake by selecting patches of low to intermediate vegetation biomass. We assessed the application of this hypothesis to plains bison (Bison bison) in an experimental grassland managed with fire by estimating daily rates of nutrient intake in relation to grass biomass and by measuring patch selection in experimental watersheds in which grass biomass was manipulated by prescribed burning. Digestible crude protein content of grass declined linearly with increasing biomass, and the mean digestible protein content relative to grass biomass was greater in burned watersheds than watersheds not burned that spring (intercept; F1,251 = 50.57, P < 0.0001). Linking these values to published functional response parameters, ad libitum protein intake, and protein expenditure parameters, Fryxell's (Am. Nat., 1991, 138, 478) model predicted that the daily rate of protein intake should be highest when bison feed in grasslands with 400–600 kg/ha. In burned grassland sites, where bison spend most of their time, availability of grass biomass ranged between 40 and 3650 kg/ha, bison selected foraging areas of roughly 690 kg/ha, close to the value for protein intake maximization predicted by the model. The seasonal net protein intake predicted for large grazers in this study suggest feeding in burned grassland can be more beneficial for nutrient uptake relative to unburned grassland as long as grass regrowth is possible. Foraging site selection for grass patches of low to intermediate biomass help explain patterns of uniform space use reported previously for large grazers in fire-prone systems.
This data set was used to test the forage maturation hypothesis in the Konza Prairie bison enclosure from 2012-2013. Our objectives were to quantify foraging site selection of Plains bison in order to determine if bison in a fire-prone grassland selected sites of low-to-intermediate forage biomass as posited by Fryxell’s (1991) forage maturation hypothesis. Additionally, to understand how foraging patterns shifted when grass regrowth was not possible, we quantified the annual diet of four GPS-collared bison via stable isotope analysis of tail hair plucked during roundup.
Foraging site vegetation data: Data set contains estimates of vegetation characteristics at bison foraging sites and paired random non-grazed sites nearby and live graminoid foliar crude protein and adjusted crude protein content at these paired sites for watersheds where bison grazing occurred in 2012-2013.
Bison tail hair stable isotope data: Data set contains d13C of yearly composite hair profiles in relation to hair follicle length (cm) for four matriarchal female bison. Values represent averages of d13C every 5 mm over a 4-year period (2010– 2013) per individual. Hair follicle length represents the distance from the base of the follicle (collection in late October each year) to older portions of the hair closest to the hair tip.
This was Graduate Student Research, a part of research related to KNZ LTER project. For additional metadata and method information, please contact firstname.lastname@example.org or see Raynor et al. 2016 paper.