Studies of long-term trends in arrival times of birds are mostly based on dates of first sightings, as it is these dates that are most frequently recorded, in some European localities for periods exceeding 300 years (Lehikoinen et al. 2004). The problem with first arrival dates is that many refer to only single individuals, which may not be typical. They also depend on the number of observers, which in many regions has increased in recent decades, raising the likelihood of extreme early birds being detected. The median or mean arrival dates of populations of individuals in their breeding areas are more representative, but have been recorded much less often, and chiefly in recent decades. They depend on detailed studies in particular areas, where the arrival of each individual (or occupant of each territory) was recorded as it occurred. Bird observatories provide a third source of migration dates, where observations or trapping were maintained throughout the migration seasons each year, enabling median or mean passage dates (and standard deviations) to be calculated. Another approach has been to combine the records from different bird observatories in the same region and calculate mean values (see Table 21.2 for examples). Whereas arrival (or departure) dates refer to birds from a single population breeding at a particular locality, passage dates usually refer to birds from a much wider area, counted at a point on their migration. Some studies have compared first and median or mean passage dates from the same site over a period of years, and found the various dates to be correlated (Huppop & Huppop 2003, Jenni & Keri 2003, Vahatalo et al. 2004, Sparks et al. 2005). In years that were early, the total arrival period became more prolonged. But whatever the recording process, long-term counts from all these sources reveal similar general long-term trends in migration timing.
Was this article helpful?