Integrating plant stoichiometry and feeding experiments: state-dependent forage choice and its implications on body mass

Intraspecific feeding choices comprise a large portion of herbivore foraging decisions. Plant resource quality is heterogeneously distributed, affected by nutrient availability and growing conditions. Herbivores navigate landscapes, foraging not only according to food qualities, but also energetic and nutritional demands. We test three non-exclusive foraging hypotheses using the snowshoe hare (Lepus americanus): (1) herbivore feeding choices and body conditions respond to intraspecific plant quality variation; (2) high energetic demands mitigate feeding responses; and (3) feeding responses are inflated when nutritional demands are high. We measured black spruce (Picea mariana) nitrogen, phosphorus and terpene compositions, as indicators of quality, within a snowshoe hare trapping grid and found plant growing conditions to explain spruce quality variation (R² < 0.36). We then offered two qualities of spruce (H1) from the trapping grid to hares in cafeteria-style experiments and measured their feeding and body condition responses (n = 75). We proxied energetic demands (H2) with ambient temperature and coat insulation (% white coat) and nutritional demands (H3) with the spruce quality (nitrogen and phosphorus content) in home ranges. Hares with the strongest preference for high-quality spruce lost on average 2.2% less weight than hares who ate the least high-quality spruce relative to low-quality spruce. The results supported our energetic predictions as follows: hares in colder temperatures and with less-insulative coats (lower % white) consumed more spruce and were less selective towards high-quality spruce. Collectively, we found variation in plant growing conditions within herbivore home ranges substantial enough to affect herbivore body conditions, but energetic stats mediate plant-herbivore interactions.

Discrimination of intra- and inter-specific forage quality by collared pikas (Ochotona collaris)

The specific nutritional characteristics by which herbivores evaluate their foraging options are complex. We experimentally manipulated the crude protein and water content of two forage species ( Carex consimilis Holm. (= Carex bigelowii Torr. ex Schwein.) and Polygonum bistorta L.) commonly cached by collared pikas ( Ochotona collaris (Nelson, 1893)) to determine their influence on inter- and intra-specific forage selection. Preference data were collected for 27 pikas using cafeteria-style feeding trials in a randomized block design. A three-way interaction (species × protein × water) suggested that pikas made conditional forage selection decisions while caching these plants. The interaction was driven by greater selection for fresh rather than dried C. consimilis when both were not fertilized. Water content had no effect on the selection of either fertilized C. consimilis or fertilized P. bistorta. Overall, our results indicate that pikas made subtle decisions about their selection of vegetation during caching, based on variation in nitrogen and water content in addition to species-specific selection criteria. Further, our results imply that tests of foraging theory may need to consider intra-specific variation in forage characteristics, as well as inter-specific ranking of forage species.

Tree use by koalas in a chemically complex landscape

Although defence against herbivores is often argued to be the main action of plant secondary metabolites (PSMs), very few examples have demonstrated that intraspecific variation in PSM concentrations influences foraging by wild vertebrate herbivores. Experiments with captive animals often indicate that PSM concentrations influence how much herbivores eat from individual plants, but these experiments do not replicate the subtle trade-offs in diet selection faced by wild animals, which must avoid predators and extremes of weather, interact with conspecifics, and achieve a balanced, nutritious diet, while avoiding intoxication by PSMs. We characterized the foliar chemistry of every tree from two Eucalyptus species available to a population of koalas (Phascolarctos cinereus) and considered rates of tree visitation over a ten-year period. We show that visitation rate was most strongly influenced by tree size, but that koalas also visited trees less frequently if the foliage contained either high concentrations of deterrent PSMs known as formylated phloroglucinol compounds, or low concentrations of nitrogen. Consequently, plant chemistry restricts the use of trees by this herbivore, and thus limits the food available to koalas and potentially influences koala populations.

Does predation risk cause snowshoe hares to modify their diets?

Snowshoe hares (Lepus americanus) undergo a 10-year population cycle with several years of low densities. Several authors have suggested that snowshoe hares modify their foraging behaviour to reduce predation risk during the low phase, resulting in protein-poor diets and poor body condition. We test that idea by using a factorial manipulation of food supplementation and predator reduction and by examining the species composition, browse size, and nutritional quality of snowshoe hare diets during 3 years of low snowshoe hare abundance in southwestern Yukon. Our results negate the hypothesis that snowshoe hares change their diets in response to mammalian predators during the cyclic low phase. Snowshoe hares on the different treatments had diets that differed in species composition and twig sizes, but protected hares did not have higher protein diets than unprotected hares. Snowshoe hares with access to supplemental food ate more fibrous and lower protein natural browse than unfed hares, showing that they did not choose diets primarily for protein content. Instead, snowshoe hares converted a wide range of forage availabilities into similar intakes of protein and fibre, despite variation in predator presence. Our results suggest that snowshoe hares select their diets to balance the protein and fibre contents. Although sublethal effects of predators may influence cyclic dynamics, our results show that such a feedback does not occur via a nutritional mechanism, counter to previous suggestions.

How good must models and data be in ecology?

Linear programming models of diet selection (LP) have been criticized as being too sensitive to variations in parameter values that have not been or may not be able to be measured with a high degree of precision (small standard error). Therefore, LP's predictions have been questioned, even though the predicted diet choices agree very well with observations in 400 published tests. The philosophical and statistical aspects of this criticism of LP are reviewed in light of the ability to test any nontrivial ecological theory. It is argued that measures of error in field data may not meet simple statistical definitions, and thereby, may make sensitivity analyses that use the error measures overly conservative. Furthermore, the important issue in testing ecological theory may not be the statistical confidence in a single test, but whether or not the theory withstands repeated tests.