Metabolic Adaptation and Climatic Constraints on Winter Bird Distribution

I used two methods to estimate limiting environmental temperatures for 28 North American passerine species studied in the physiological literature: indirect estimation from body mass and allometric equations, and direct estimation from physiological measurements of basal metabolic rate (BMR), lower critical temperature, and thermal conductance. I compared these predicted limiting temperatures to mean minimum temperatures at northern winter range boundaries determined from Christmas Bird Count data. Estimates derived from physiological parameters predicted minimum temperatures at range boundaries more accurately than did allometric equations. Phylogenetic analyses of independent contrasts under two branch length hypotheses corroborated that physiological parameters were superior to allometric equations as predictors of range boundary temperature. Thus using body mass as a surrogate for physiological data underestimates the association between physiological cold tolerance and northern range limits. However, this association could be explained by phenotypic acclimatization to different environments as well as by genetic differences among species; thus low values of the physiological temperature limit may be a result rather than a cause of northerly ranges. The distribution of predicted metabolic rates at northern range boundaries relative to BMR did not cluster at a clear upper boundary, so it is uncertain whether a constraint on maximum metabolic rate limits distribution of some species. Metabolic rate at the northern range boundary was a smaller multiple of BMR for insectivores than for non-insectivores, suggesting that food availability may limit winter ranges of insectivores, but analysis of independent contrasts did not verify this effect. The data do not indicate clearly whether the distribution of non-insectivorous bird species is limited by abiotic or biotic factors.