|
Résumé :
|
Global change affects the timing of bird migration. It presumed (and has been shown) that, (1) species arrived earlier in spring and that (2) this response to climate change is stronger in short-distance than in long-distance migratory species. To confirm this pattern we analysed long-term observational data from field ornithologists under the condition that (1) at least eight arrival dates for the periods 1970-1986 and 1987-2003 were available, (2) observers had not changed their study area during these years, and (3) at least three time series were available for a species. In total, we examined the first arrival dated of 17 migratory species at 13 sites in southwestern Germany. To analyse changes in arrival times we used two statistical procedures: linear regression for the period 1970-2003 and Mann-Whitney U-test to compare median arrival dates of the periods 1970-1986 and 1987-2003. Out of 103 time series 96 showed a trend towards earlier spring arrival, whereas series only for 6 even time series the reverse was true. Using linear regression for 58 species time series we found a significantly earlier arrival date. On average, arrival date changed 0.3 (max. 2.9) days per year. Only one species showed a significant later arrival date. The U-test revealed similar results. An analyses of variance (ANOVA) showed that this effect was stronger in short-distance than in long-distance migrants. We suggest that earlier spring arrival could be due to a shift of wintering areas towards the North - even in some long-distance migratory species.
|