Responses of herbivores to heterogeneity in forage resources expressed at multiple spatial scales

Abstract

Understanding responses of organisms to spatial heterogeneity in resources has emerged as an important challenge in contemporary ecology. I examined responses of foraging herbivores to multi-scale heterogeneity in plants, implementing several empirical experiments to seek answers to key questions that have arisen from theoretical foraging theory. First, I sought to understand how fine scale heterogeneity controls intake rate for large herbivores. The gain function describes the amount of food consumed in a patch as a function of patch residence time. I evaluated the strength of evidence in data for alternative gain function forms of mule deer (Odocoileus hemionus) and blue duikers (Cephalophus monticola) feeding in patches composed of different plant species and plant sizes. I found that gain functions decelerated continuously with patch residence time, but the nature of the deceleration depended on patch characteristics, notably plant size. I elucidated the mechanisms causing deceleration. I demonstrated that unwarranted assumptions about the shape of gain functions can have fundamental effects on predictions of patch models. Second, I asked the question, "Is the behavior observed at coarse scales in a patch hierarchy the collective outcome of fine scale behaviors or, alternatively, does the spatial context at coarse scales entrain fine scale behavior?" I created a 2 level patch hierarchy, and examined effects of the geometry of the hierarchy on residence time of foraging grizzly bears, mule deer, and collared lemmings. I developed a set of competing models predicting residence time as a function of patch size and distance among patches, and examined the strength of evidence in data for each model. Models that included patch mass and inter-patch distance as independent variables successfully predicted observed residence times. Grizzly bears and mule deer responded to differences in patch geometry at multiple scales. In contrast, the residence times of collared lemmings were simply predicted by the mass of food in patches and were uninfluenced by differences in the surrounding spatial context. Third, I sought to understand if herbivores choose patches to maximize food intake rate or to reduce risk of starvation in variable environments. Moreover, I examined the possibility that intake rate maximization was dependent on the spatial scale of patchiness. Two currencies have been used to assess the patch preferences of herbivores—intake rate maximization and risk sensitivity. I found that collared lemmings did not consistently select patches that maximized their intake rate at either scale studied. Instead, lemmings consistently chose patches offering the least variation in food reward over the course of the experiment. I interpret these results as evidence for risk-averse foraging strategies, which are predicted for continuous foragers aiming to minimize risk of starvation. Finally, I discuss the question of how much time a foraging herbivore should spend in a patch of food. This question poses a central challenge in classical foraging theory. However, there remains uncertainty about the relevance of the patch paradigm to foraging decisions by large herbivores. I examined evidence for successfully predicting and quantifying patch departure decisions for large mammalian herbivores foraging across several spatial and temporal scales. I found strong evidence that departure decisions at fine scales are influenced by trade-offs between maximizing intake rate and food quality. In addition, classical models for departure decisions at larger spatial scales, particularly the marginal value theorem, appear inadequate. I advocate exploring alternative models for predictions of residence time at the patch scale.