Avian colouration in a polluted world: a meta-analysis.

Brilliant, diverse colour ornaments of birds were one of the crucial cues that led Darwin to the idea of sexual selection. Although avian colouration plays many functions, including concealment, thermoregulation, or advertisement as a distasteful prey, a quality-signalling role in sexual selection has attracted most research attention. Sexually selected ornaments are thought to be more susceptible to external stressors than naturally selected traits, and as such, they might be used as a test for environmental quality. For this reason, the last two decades have seen numerous studies on the impact of anthropogenic pollution on the expression of various avian colour traits. Herein, we provide the first meta-analytical summary of these results and examine whether there is an interaction between the mechanism of colour production (carotenoid-based, melanin-based and structural) and the type of anthropogenic factor (categorised as heavy metals, persistent organic pollutants, urbanisation, or other). Following the assumption of heightened condition dependence of ornaments under sexual selection, we also expected the magnitude of effect sizes to be higher in males. The overall effect size was close to significance and negative, supporting a general detrimental impact of anthropogenic pollutants on avian colouration. In contrast to expectations, there was no interaction between pollution types and colour-producing mechanisms. Yet there were significant differences in sensitivity between colour-producing mechanisms, with carotenoid-based colouration being the most affected by anthropogenic environmental disturbances. Moreover, we observed no significant tendency towards heightened sensitivity in males. We identified a publication gap on structural colouration, which, compared to pigment-based colouration, remains markedly understudied and should thus be prioritised in future research. Finally, we call for the unification of methods used in colour quantification in ecological research to ensure comparability of results among studies.