Food has a larger effect on bird abundances where thermoregulatory costs are high

C. Lane Scher and James S. Clark, Nicholas School of the Environment, Duke University, Durham, NC, James S. Clark, LESSEM, University of Grenoble Alpes, INRAE, St-Martin-d'Heres, France

Background/Question/Methods
Winter bird abundances are influenced by environmental conditions that determine their survival and can affect their success throughout the full annual cycle. As endotherms, birds can raise their metabolic rate to increase their body temperature when ambient temperatures are low, but at a high energetic cost. The cost of thermoregulating depends on environmental conditions: the cost decreases with warmer temperatures and higher solar radiation. Winter food, which varies across space and time, is required to offset these costs, suggesting that bird survival depends on maintaining a balance between food and thermoregulation. To understand how food and thermoregulatory costs interact to influence birds, we analyzed the interacting effects of mast (both mean and annual anomalies), temperature, and solar radiation on winter bird abundances in a joint species distribution model of mast-consuming bird species. We included interaction terms between mast and both temperature and solar radiation in the model, and we interpret the fitted coefficients of these interactions to understand how the effect of food changes across these gradients. We expected that the full effect of mast will be larger when temperature and solar radiation are low, i.e. a negative interaction coefficient, due to the high energetic cost of thermoregulating in these conditions.
Results/Conclusions
Mean conifer and mean fruit have a larger effect in colder winters, and mean conifer also has a larger effect when there is less solar radiation. Conifer and fruit anomalies have weaker or no interactions with temperature and radiation. These results indicate that areas with high mean conifer and fruit mast support more birds in winters when thermoregulatory costs are high, whereas the impact of interannual variation in conifer and fruit mast does not change substantially depending on thermoregulatory costs. This may be because birds tend to return to the same wintering grounds each year, so they are less sensitive to fluctuations in food. Conversely, the effect of oak mast increases with temperature and radiation. Again, the interaction is stronger for mean oak mast than oak mast anomaly. This pattern suggests that oak consumers are limited by mast, thermoregulatory costs, or both: when one increases, the effect of the other also increases. These results have important implications for our fundamental understanding of the drivers of biodiversity patterns. We demonstrate that winter bird distributions are determined, in part, by a balance between thermoregulatory costs and food availability.