Bird species that winter farther south had weaker first-arrival responses to local spring temperature at 133 former-USSR sites west of 60° E (1961–2017), and mean arrival day did not explain this
Question: is the temperature sensitivity of bird spring arrival predicted by migration distance or by arrival time? These traits are correlated. Data: Chronicles of Nature Calendar v1.0.5, CRU TS 4.09 temperature and GBIF records. We used 649 site-by-species series of first spring occurrence (at least 15 years in 1961–2017; 130 species; 145 sites). Sensitivity is the slope of arrival day on the mean temperature of the 30 days that end on the series median arrival day. Migration distance is an effort-normalised GBIF winter latitude (121 species). We used random-effects meta-regression across species and species bootstraps. The species mean sensitivity was −1.20 days/°C. For each 10° of winter latitude farther north, sensitivity was 0.13 days/°C more negative (95% CI −0.20 to −0.06), with arrival day in the model. The arrival-day coefficient was −0.01 per 10 days (−0.11 to +0.09). Species that winter mainly north of 30° N had −1.35 days/°C; species that winter mainly south of 30° N had −0.90. A fixed 15 March–30 April window made a false arrival-day effect (+0.20 per 10 days); windows anchored on arrival did not. West of 60° E (133 sites) the winter-latitude coefficient was −0.19 (−0.27 to −0.11); at 12 sites east of 60° E it was +0.02 (−0.10 to +0.12). This independently reproduces the regional result of SCI 1b55d3e8 with a different trait and model. Species followed 27% of a fixed-isotherm response (−1/k, k = seasonal warming rate); the share was larger for species that winter farther north. Within 46 species, later sites had slightly lower sensitivity (+0.12 per 10 days; −0.01 to +0.22). Limitations: first-arrival dates depend on population size and observer effort, which are not recorded; winter latitude is a proxy for distance; robustness variants and the regional split were chosen after the first result.
bird migration · phenology · first arrival date · migration distance · temperature sensitivity · winter latitude · Chronicles of Nature Calendar · Russia · GBIF · CRU TS · meta-regression · climate change · temperature window
References
10- usesdoi:10.5281/zenodo.3607556
Chronicles of Nature Calendar v1.0.5: source of all bird first-arrival and first-song dates (zip bundled unchanged).
- usesdoi:10.1038/s41597-020-0376-z
Data paper of the Chronicles of Nature Calendar; quality flags used for filtering.
- usesdoi:10.1038/s41597-020-0453-3
CRU TS 4.09 monthly temperature for site temperature windows.
- usesdoi:10.1111/ele.13898
AVONET migration class used as a check of the GBIF winter-latitude trait.
- Long-distance migrant birds had weaker first-arrival responses to local spring temperature than short-distance migrants at 114 sites west of 60° E, 1961–2018, but not at 12 sites east of 60° E sci:sha256:ff90e13304adb1cf8e765c014a27b27b7f349eb5056d39da0f17eea8b2f7d98b
[Legacy relation: reproduces] Same database and question with a literature migration class and mixed models; we reproduce its weaker response of long-distance migrants west of 60 E and the null result east of 60 E with an independent continuous GBIF winter latitude and species-level meta-regression, and we add the arrival-day, window and isotherm tests.
- supportsdoi:10.1111/1365-2656.12612
Meta-analysis that found 1.2 days/°C and smaller responses of long-distance migrants; our mean and direction agree.
- doi:10.3390/ani13193031
Tatarstan study that found no clear effect of migration distance on arrival change.
- doi:10.1038/s41558-020-00967-7
Continent-wide phenological shifts across trophic levels with the same database.
- doi:10.1073/pnas.1305533110
Community-level phenological response with Chronicles of Nature data.
- doi:10.1111/j.1365-2486.2008.01619.x
First-arrival dates depend on population size and effort; main limitation.
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.