Trends in trappability and stop-over duration can confound interpretations of population trajectories from long-term migration ringing studies

We examined ringing data for three common passerines at a migration monitoring site with the objective of determining how closely the numbers of birds captured reflected the actual numbers of birds passing through this site. Mark-recapture analyses were used, fitting smooth inter-annual changes in re-trapping probability and apparent survival (continued presence within the study area). For all three species we found systematic declines in re-trapping probability over the roughly three decades of our study; one species showed systematic increases in daily apparent survival over this same time period. Within individual years, daily variation in wind and rainfall were found to affect re-trapping probability and apparent survival. The result of all of these effects was that long-term trends in the numbers of unique individual birds captured systematically over-estimated the declines for all three species, providing clear indication that the biological interpretation of these raw capture totals from ringing stations will not necessarily reflect actual changes in the abundance of species passing through these sites.