Trait Correlations Shift in Sign and Magnitude in Response to Urbanization

The correlation between traits can affect how populations evolve in novel and changing environments. Moreover, as environments change, trait correlations can be altered in both sign and magnitude. However, how urban environments affect the expression and evolution of trait correlations has not been explored. In this study, we conduct a systematic review of the literature and meta-analysis to test if and how phenotypic trait correlations shift between non-urban and urban habitats. Cities make an excellent model to test shifts of trait correlations as multiple co-occurring novel stressors challenge populations with strong environmental changes from non-urban habitats. In exploring the literature, we find evidence that trait correlations can switch signs between urban and non-urban populations, but that no change in sign is much more common. We also find that the distribution of trait correlations differs between urban and non-urban populations for field measurements but not under common-garden conditions. Likewise, the magnitude of trait correlations is greater for non-urban populations compared to urban populations for field measurements but not under common-garden conditions, suggesting a strong role of phenotypic plasticity in driving shifts in phenotypic trait correlations. Together, these results offer a view into how populations respond to the multivariate nature of urbanization and suggest potential hypotheses for future exploration.