Intraspecific consistency and geographic variability in temporal trends of spring migration phenology among European bird species

Abstract

In the course of the 20th century, migratory birds have shown rapid phenological changes in response to climate change. However, the spatial variability of phenological changes, as well as their intraspecific consistency, remains largely unexplored. Here we analysed 672 estimates of change in first arrival dates of migratory birds and 289 estimates of mean/median arrival dates, based on time series with a minimum duration of 15 yr, collected across Europe from 1960 to 2006. There were highly significant advances in arrival date, significantly more so for first than mean arrival date. Change in arrival dates significantly varied among species, implying that response to climate change is a species-specific feature, and showed substantial phylogenetic effects, since ca. 50% of the variation in the observed trends was attributable to differences among species. The advance in first arrival date was weaker at extreme latitudes and stronger at intermediate latitudes, while geographic variation in mean arrival dates was less pronounced. Both first and mean arrival dates advanced the most for short-compared to long-distance migrants. These findings emphasize the reliability of estimates of phenological trends of avian species, which are therefore suitable to be included in comparative analyses aimed at identifying species-specific traits that favour adaptation to climatic changes. In addition, our results suggest that analyses of factors that have affected phenological responses to climate change should take into account spatial variation in the response, which could be due to spatial differences in the strength of climate change.

Populations of migratory bird species that did not show a phenological response to climate change are declining

“…While controlling for the effect of population trend during 1990-2000, which strongly predicted change in migration timing, we found that long-distance migrants advanced spring migration date the least, and that species laying a greater number of clutches showed a stronger advance in timing of spring migration. A stronger advance in timing of spring migration in short-distance versus long-distance migrants is in accordance with most previous studies [e.g., (8,9), but see (35)]. This pattern is indeed to be expected, because short-distance migrants spend the winter closer to the breeding areas compared with longdistance migrants, which may allow such species to better tune the timing of spring migration to concomitant weather conditions (8,9), and the timing of migration of short-distance migrants may have a stronger environmental component compared with long-distance migrants (8,36).…”

“…Considerable heterogeneity in advancement in timing of spring migration among species (8,9) may be due in part to spatial heterogeneity in climate change across Europe, with spring temperatures increasing in some regions but not in others (10).…”

“…However, despite such heterogeneity, spring phenological responses to climate change are species-specific and consistent in different populations, making an investigation of predictors of species-specific variation in change in migration timing, considered as an indicator of the phenological response to climate change, meaningful (9).…”

“…We predicted that migratory bird species that did not anticipate timing of spring migration (i.e., that did not show a phenological response to climate change) should demonstrate declining populations, while taking into account several variables previously shown to be associated with population trends (see above and 12-19) or response to climate change. Among variables potentially affecting the phenological response to climate change, we considered migration distance, intensity of sexual selection, and total population size; a small population size, and thus lack of genetic variation, potentially could explain lack of response to climate change (8,9,21,22). All analyses were performed while accounting for phylogenetic relatedness among species by using phylogenetically independent linear contrasts (23).…”

“…We deliberately avoided using change in first arrival dates as a cue to phenolog-ical response to climate change, because this phenological datum is strongly dependent on sampling effort and may be affected by changes in population size (8,9).…”