Figure 1.
Schematic diagram of the length of the Breeding Area Residence Time (BART) depending on the initial condition and both hypotheses for changes in migration dates.
A, Initial state: migration dates are unchanged; the length of the BART is stable. B, Hypothesis 1: both spring and autumn migration dates occur earlier; the length of the BART is stable but lagged. C, Hypothesis 2: spring migration date is earlier, autumn migration date is later; the length of the BART is extended.
Figure 2.
Location of the five sub-regions used in the study:
A, Sweden. B, Netherlands. C, French Alps. D, Massif Central. E, Pyrenees.
Illustrations of the seven species of raptor from left to right: red kite, marsh Harrier, hen harrier, merlin, sparrowhawk, common buzzard, and kestrel.
Figure 3.
Long-term changes in the timing of autumn migration in relation to autumn temperatures in Europe from 1980 to 2010.
A, B, PCA of the year-to-year changes in the mean passage date of seven species of raptors from 1980 to 2010. (A) Long-term changes in the first principal component (PC1), which represents 20.24% of the total variability. Abrupt shifts, identified by the Rodionov’s test, are superimposed (grey line). (B) Bar diagram of the first eigenvector normalised between -1 and 1. Negative values are in blue and positive values are in red. C, D, PCA of the year-to-year changes in the August-October temperatures from 1980 to 2010. (C) Long-term changes in the first principal component (PC1), which represents 44.25% of the total variability. Abrupt shifts, identified by the Rodionov’s test, are superimposed (grey line). (D) Mapping of the first normalised eigenvector, which shows the correlation between autumn temperatures and the first component. Negative values are in blue and positive values are in red.
Figure 4.
Effect of the abrupt shifts in autumn temperatures on the timing of migration.
A, mean anomalies of autumn temperatures and Mean Passage Date (MPD; in day) for all periods revealed by the Rodionov’s test (see Figure 2). Both autumn temperatures and MPD are expressed in anomalies with respect to the period 1980-2010. b, c, d, Repartition of the anomalies of MPD during (B) the first period 1980-1994 (T1=12.29°C), (C) the second period 1995-2006 (T2=12.91°C) and (D) the third period 2007-2010 (T3=12.66°C).