Unexplained rapid bill-size changes in an urban bird during the COVID-19 anthropause

Diamant and Yeh surveyed urban (University of California, Los Angeles campus) and non-urban populations of dark-eyed juncos (Junco hyemalis) between 2018 and 2025 (1). Second-year urban birds had shorter beaks, except in 2022 and 2023 where no significant difference was found. The authors state that dispersal from non-urban to urban populations is likely low, thus 2022 and 2023 second-year birds correspond to birds hatched at UCLA in 2021 and 2022. The authors propose that the observed changes in beak length can be explained by adaptation, driven by a diet change caused by a reduction in food waste availability during COVID-19 lockdowns. Here we argue that this is an unlikely explanation for the observed changes, as such rapid changes in a population trait would only have been possible through an extraordinary level of evolutionary pressure, with conspicuous effects on population sizes.

We estimated the rate of evolutionary change in beak length between urban parental and offspring populations, assuming that all observed differences resulted from adaptation from standing variation, based on a previously published dataset of adult birds captured at UCLA (2). The resulting values (Table 1) are extraordinarily high. Indeed, among about 2,400 rates of phenotypic evolution due to human disturbances on natural populations reported in a meta-analysis (3; median = 0.19 haldane, max = 1.4 haldanes), the absolute values for UCLA’s juncos in 2021 (0.76) and 2023 (−0.86) would be, respectively, the third and second highest values reported.

Table 1.

Rate of evolution from putative parental to offspring populations based on annual changes in beak sizes

Year of reproduction	Parental population	Offspring population	Rate of evolutionary change $\mathit{H}=\frac{{\overline{\mathit{x}}}_{\mathit{o}}-{\overline{\mathit{x}}}_{\mathit{p}}}{{\sigma }_{\mathit{p}}}$ (haldanes)
2018	${\overline{x}}_p=7.22$; ${\sigma }_p=0.61;$ ${\mathrm{n}}_p=79$	${\overline{x}}_o=7.52$; ${\mathrm{n}}_o=24$	0.49
2019	${\overline{x}}_p=7.49$; ${\sigma }_p=0.49;$ ${\mathrm{n}}_p=55$	${\overline{x}}_o=7.63$; ${\mathrm{n}}_o=17$	0.28
2020	${\overline{x}}_p=7.41$; ${\sigma }_p=0.50;$ ${\mathrm{n}}_p=43$	${\overline{x}}_o=7.57$; ${\mathrm{n}}_o=18$	0.31
2021	${\overline{x}}_p=7.74$; ${\sigma }_p=0.50$; ${\mathrm{n}}_p=84$	${\overline{x}}_o=8.12$; ${\mathrm{n}}_o=23$	0.76
2022	${\overline{x}}_p=8.13$; ${\sigma }_p=0.46$; ${\mathrm{n}}_p=64$	${\overline{x}}_o=7.95$; ${\mathrm{n}}_o=39$	−0.40
2023	${\overline{x}}_p=7.91$; ${\sigma }_p=0.51$; ${\mathrm{n}}_p=27$	${\overline{x}}_o=7.47$; ${\mathrm{n}}_o=39$	−0.86

For comparison, a severe drought in 1976–1977 resulted in a change in the beak length of Darwin’s finches (Geospiza fortis) of 0.65 haldanes in 2 y (4). During this extreme event, 85% of the 1976 birds disappeared (died or emigrated) and no reproduction occurred in 1977 (4). Likewise, and even if the urban environment can be the source of strong selective pressures (5), changes in traits of the magnitude reported for the UCLA juncos’ beaks would have only been possible through major selective decreases in nesting success, in survival, or in both. Either way, this would have had dramatic and noticeable repercussions on population size, which have not been reported, making the hypothesis of a rapid adaptation from standing variation extremely unlikely.

The observed changes in UCLA junco beak size therefore remain to be explained. First, rapid changes in both directions across few generations may be the result of the beak developmental plasticity in response to environmental changes (e.g., ref. 6). Second, in the dataset comprising all adults, population beak length was already increasing in 2021, i.e., in birds born before the lockdowns (Table 1). This is consistent with either beak plasticity or increased immigration of non-urban birds into the campus, perhaps reflecting a breakdown in dispersion or establishment barriers with the release of human pressures. Furthermore, while it is appealing to link beak length to foraging behavior, it may affect other life history aspects [e.g., social interactions (7)], or be a by-product of changes in other traits sharing the same developmental pathway [e.g., stress responses, brain size (8, 9)].

Data, Materials, and Software Availability

Previously published data were used for this work (1, 2).