Repeatability in spring arrival dates in Pied Flycatchers varies among years and sexes

Timing of arrival in long-distance migration could have fitness consequences: arrival too early impairs survival chances, whereas arrival too late reduces current reproductive success. Evolution thus may have favoured a phenotype that arrived at the optimal time. However, individuals within populations of long-distance migrant species arrive over a considerable time span, and often show consistency in whether they are early or late. This repeatability in arrival varies between studies, and we hypothesise it to be affected by conditions encountered en route or in winter. Here we report on the spring arrival dates of Pied Flycatchers Ficedula hypoleuca to their Dutch breeding sites during eight consecutive years. Our field estimates of arrival were highly accurate, as validated by geolocator data on 13 individuals. Years differed in mean arrival dates. Within years and sexes, arrival date generally spanned more than two weeks. First-year individuals arrived on average 4-5 days later than older individuals. Using repeated arrival dates of more than 500 individuals we show that (1) the overall arrival repeatabilities were similar for females and males, (2) arrival repeatabilities varied temporally, with individuals in consecutive years having sometimes moderate (R = 0.2) and sometimes rather high (>0.40) repeatabilities, and (3) individual females arrived later in their first than in their second year. In females, repeatabilities of arrival and laying dates were similar. We hypothesize that individual flycatchers have a high individual consistency in their spring migration departure date from the wintering grounds. However, previous studies suggest the expression of this individual schedule to be affected by environmental circumstances at the wintering grounds or by what is encountered en route, determining whether this variation is still present at arrival on the breeding grounds. Sexes seemed to differ in this respect, as year-to-year variation in repeatabilities of timing was explained by individual consistency in females, but not in males. We discuss the relevance of the observed variation for the potential for an evolutionary response when environments change.