Predicting bite size selection of mammalian herbivores: a test of a general model of diet optimization

Citation
La. Shipley et al., Predicting bite size selection of mammalian herbivores: a test of a general model of diet optimization, OIKOS, 84(1), 1999, pp. 55-68
Citations number
72
Categorie Soggetti
Environment/Ecology
Journal title
OIKOS
ISSN journal
00301299 → ACNP
Volume
84
Issue
1
Year of publication
1999
Pages
55 - 68
Database
ISI
SICI code
0030-1299(199901)84:1<55:PBSSOM>2.0.ZU;2-T
Abstract
The architecture of woody food plants forces mammalian herbivores to make c ompromises in their food choices. Rapid rates of dry matter intake can be a chieved by choosing large bites. For woody plants, however, such bites are low in nutritive quality relative to small bites taken from leaves or twigs near the growing point of the plant. This trade-off between food quality a nd food intake rate is central to diet optimization in browsing herbivores. We developed a model that predicts a quantitative solution to 'optimal bit e size' (i.e.; the bite that results in the greatest daily net energy intak e) based on constraints in harvesting and digesting foods. This model respo nds to the chemistry and morphology of plants, and the size and digestive s trategy (ruminant versus hindgut fermenter) of the herbivore. We tested the model by conducting a set of experiments in which we offered six species o f dormant deciduous trees common to the boreal forests of Sweden to captive roe deer (Capreolus capreolus), red deer (Cervus elaphus), and moose (Alce s alces). We also tested alternative hypotheses that animals crop bites mer ely in response to the morphological structure of twigs, or the distributio n of twig sizes on trees. Twig diameters cropped by these animals were posi tively correlated with the diameter at current annual growth. However, stru ctural measures of the trees alone were not sufficient to predict differenc es in choices of twig diameters among animal species. In contrast, the opti mal bite size model accounted for the different bite sizes selected by anim als of different sizes and explained 86% of the variation in twig diameter cropped for all plant and animal species. Hence. we concluded that our mode l is useful for predicting, a priori, the twig diameters selected by herbiv ores on dormant deciduous trees. We suggest possible ways to enhance the mo del and how it can be used to assess forage availability, potential diet qu ality. and the vulnerability of trees to herbivory.