Data quality problems undermine analyses of endotherm upper critical temperatures

Abstract

In an analysis of avian and mammalian thermal tolerances recently published in this journal, Khaliq et al. (2015) reported that endotherm thermal niches are phylogenetically conserved in tropical, but not temperate, regions. However, closer examination of the data upon which this analysis was based reveals that many of the upper critical temperature (UCT) data are not valid. Approximately 55% and 42% of avian and mammalian UCT data, respectively, originated from studies in which animals were not exposed to air temperatures high enough to elicit an increase in metabolic rate above minimum levels; the cited UCT values are merely the highest air temperatures at which measurements took place. An additional 18% and 25% of avian and mammalian UCT data, respectively, represent values based on just one individual per species and/or measurements at too few air temperatures above the thermoneutral zone (TNZ) to reliably estimate the UCT.