|
Résumé :
|
Data from 5,134 Skylarks ringed at Castricum, NW Netherlands, during the autumns of 1977-1992 were analysed to describe migration patterns and sex composition. Birds on active migration were attracted to tape-lures and clap-netted. Weekly estimates of the proportions of males and females, and their mean wing lengths, were made from the wing length distributions. Females passed earlier in the autumn than males. A comparison with samples of Skylarks which had collided with lighthouses along the Dutch coast strongly suggested that males were more likely to be attracted by the tape-lures than females. The strength of the male bias decreased in the course of the autumn. The male bias resulted in a shift in the seasonal pattern based on catches, since relatively more of the later migrants (which were males) were caught. However, this effect could not fully explain the difference of eight days between the migration patterns at Castricum, derived from ringing, and at Hoorn, 35km away, derived from counts. A tenfold increase in the capture probability for both sexes over the course of the autumn had to be assumed to account for this. Such a seasonal bias might have been caused by variation in capture effort at the ringing station, and/or by an increase in the proportion of first-year birds among migrants. When both effects are accounted for, females appear to pass Castricum eight days earlier than males; this difference was three days when based on uncorrected data. The earlier passage of females is in accordance with the preponderance of males among birds that try to overwinter in continental W Europe, and suggests that females winter further to the south or west than males. In analysing ringing data, it should be kept in mind that capture bias may affect not only the composition of the sample, but also the observed seasonal patterns and distribution of recoveries.
|