|
Résumé :
|
At the ringing station on the island of Helgoland (German Bight), migratory birds have been trapped since 1909 in three consecutive funnel traps. All-year-round trapping effort with up to seven trapping runs daily remained unchanged throughout the last four decades. Between 1960 and 2001, a total of about 490.000 individuals (153 species) were trapped and ringed. Sample sizes were sufficient to calculate spring migration means (HZMW) over 42 years for 12 short/medium-distance migrants (KMZ), 10 long-distance migrants (LZ) and for Blackcap and Common Chiffchaff, which conform to both migration categories. Autumn migration means could be calculated for 16 KMZ, 8 LZ and for Blackcap and Chiffchaff.In spring, 20 of the 24 species showed a trend towards earliness in HZMW throughout the study period, with a significant advancement in 14 species (5 KMZ, 8 LZ and Blackcap) of up to 12 days. This trend did not differ between KMZ with a mean of 5.8 days and LZ with an average of 7.3 days. In 22 species, the advancement of HZMW was associated with an increase in the local mean species-specific spring migration temperature, significant in 11 species (7 KMZ, 2 LZ, Blackcap and Chiffchaff). In 23 species, the trend towards earlier HZMW coincided with the trend towards increasingly positive winter-indices of a large-scale climatic phenomenon, the "North Atlantic Oscillation" (NAO), with significant correlations in 13 species (5 KMZ, 7 LZ and Chiffchaff). The HZMW (single species or pooled for each migration category) is probably influenced to a larger extent by local temperature in KMZ, whereas LZ seem to be more susceptible to the winter-NAO.The autumn migration means (WZMW) showed a delay in 14 of 26 species between 1960 and 2001, which was significant in 5 species (2 KMZ, 2 LZ and Chiffchaff) of up to 9 days. A non-significant trend towards earlier WZMW was observed in 6 species. Pooled LZ showed an average (not significant) delay in WZMW by 3 days whereas pooled KMZ showed no such trend, with no significant difference between the two migration categories. There were no significant correlations with climatic variables measured during summer breeding and autumn migration, which have changed less than variables measured in winter and spring.In 20 species, sample sizes were sufficient to calculate HZMW as well as WZMW The period between HZMW and WZMW increased in 16 species, significantly in 13 species (7 KMZ, 5 LZ and Chiffchaff) by up to 16 days, largely as a result of earlier spring migration. This may reflect a prolongation of the time spent on Scandinavian breeding grounds with the opportunity to increase breeding success.
|