Social and habitat correlates of immigrant recruitment of yearling female Mallards to breeding locations

Conspecific avoidance and attraction hypotheses have been proposed to explain patterns of animal spacing behavior and have been frequently used to explain habitat selection by dispersers. In birds, tests of these hypotheses have been limited to fine spatial scales and have not considered dispersal distance. We used discriminant function analysis (DF) of feather δ34S, δD, δ15N and δ13C values to identify yearling female Mallards (Anas platyrhynchos) that dispersed long distances (>400 km relative latitude) from natal areas to settle at 22 breeding locations in the Canadian parklands, 1993–2000; these sites varied in social (breeding pair density) and breeding habitat (wetland density and percent nest cover) conditions. DF indicated that the proportion of yearling female recruits estimated to be long-distance immigrants in these breeding populations averaged 12% across sites (range 0–38%). Yearling immigration rates were positively correlated with conspecific breeding pair density and negatively correlated with percent wetland habitat. Settling patterns by yearling immigrants were most consistent with the prediction of social attraction, but effects were statistically weak possibly because inter-regional movements were less variable annually due to relatively stable wetland conditions in the southern prairies during the period of study. Our results suggest immigrants arriving from distant natal sites used conspecific abundance as a positive cue of habitat quality when choosing the new breeding sites. Fitness consequences of these choices should be evaluated.