Plumage and timing of migration in bar‐tailed godwits: a comment on Drent et al. (2003)

Abstract

Rudi Drent and co-workers presented data on the plumage and apparent quality of bar-tailed godwits (Limosa lapponica ) migrating northwards through the Netherlands en route from Africa to Siberia (Drent et al. 2003). They identified two groups of males in their samples based on the extent of the prebreeding moult birds with extensive red breeding plumage (‘dark’ birds) and birds with little red plumage (‘pale’ birds). The pale birds were considered to be low quality individuals, which showed a higher incidence of cestode infestation (32.4% of individuals c.f. 12.2%) and, while the dataset was very small, lower subsequent survival (no pale birds were recaptured in later years whereas 11 red males were). They radio-tracked birds to determine the migration date of pale and dark birds towards the breeding grounds and were surprised to find that, contrary to expectation, the pale birds left on migration earlier than the dark birds. Noting that there was no evidence for any age-related changes in the breeding plumage score of male godwits (though their Fig. 7 indicates they never recaptured any pale birds so could not test this), Drent et al. erected a hypothesis to explain the unexpected finding. The pale birds were suggested to be ‘desperates’, birds of low quality whose only chance of successful breeding was to arrive early on the breeding grounds and achieve the benefits of prior occupancy on good territories. Given that it is generally high-quality individuals that can afford the risks of early arrival on the breeding grounds (Moller 1994), the lower apparent survival of these pale godwits reflected the costs associated with this strategy. Appealing as this idea may be, data collected on the non-breeding grounds of a more accessible population of godwits show that most pale birds in Drent et al.’s study are almost certainly immature birds, potentially on their first migration. Together with colleagues from the Ornithological Society of New Zealand, I have individually colour-banded godwits (subspecies baueri ) in New Zealand since 2004 and aged these birds at capture. Godwits show delayed maturity and young birds remain on the non-breeding grounds for up to 3 4 years before migrating northward (McCaffery and Gill 2001). Where possible birds were aged as 1 (in the 1st year of life), 2 or 3 /, though at some times of year birds could simply be designated as immature (age 2/3). Ageing was by plumage (juveniles have distinctive upperpart contour feathers that are retained until birds reach the non-breeding grounds), timing of wing moult (adults start wing moult only after arrival on the non-breeding grounds in September-October, whereas immatures that did not migrate to the breeding grounds start earlier; juveniles may undergo partial wing moult in the southern autumn) and feather condition (juvenile primaries become highly abraded by the time of the southern winter; adult breeding plumage contour feathers and tertials become very worn during the breeding season). Checks were then made at high-tide roosts to determine the presence of colour-banded birds and, when practical, the breeding plumage of males was scored on the same scale as used by Drent et al. (where 1 /complete nonbreeding plumage, 2 /a trace of breeding plumage, 3 /25% breeding plumage, 4 /50% breeding plumage, 5 /75% breeding plumage, 6 /only a trace of non-breeding plumage remains, and 7 /complete breeding plumage). Resightings of marked birds in New Zealand show that the breeding plumage grown by male godwits is strongly age-related (Fig. 1). Across all birds, most age 1 and 2 godwits had very little breeding plumage, whereas many 3rd-year and most ‘adult’ godwits had substantial breeding plumage scores (Fig. 1, upper). The pattern of