Gone with the Wind? A Comment on Butler et al. (1997)

Abstract

Because wind speed and flight speed of migratory birds generally are of the same magnitude, wind direction has a significant effect on the ground speed of birds. Consequently, migrants should prefer days of favorable tailwinds for commencing migratory flights, or at least they should avoid strong headwinds. These predictions generally are the case as revealed by radar studies (Richardson 1990). Based on estimated flight costs of shorebirds, Butler et al. (1997) concluded that tailwinds are necessary for successful spring migration of Western Sandpipers (Calidris mauri). Butler et al. (1997) proposed that the migration strategy for Western Sandpipers consists of maintaining relatively large fuel reserves and departing on migratory flights when tailwinds are present. This strategy was put forth as an alternative to the strategy of time minimization of migration (sensu Alerstam and Lindstrom 1990). We think it is premature to reject time minimization as a migration strategy for Western Sandpiper on the basis of the data presented by Butler et al. (1997), and we outline our arguments below. First, we show that new information concerning the aerodynamic drag of bird bodies indicates that Western Sandpipers may complete their spring migration without wind assistance. Next, we provide a critical discussion of the assumptions, methods, and conclusions put forward by Butler et al. (1997). Flight costs: They can do it.-Butler et al. (1997) calculated that when using Pennycuick's (1989) model of bird flight performance, the arrival mass will be 17.5 g for a male Western Sandpiper in the absence of wind assistance. Two amendments to the theory of bird flight may change this conclusion. First, the default value for the body-drag coefficient (Cpar), which determines the magnitude of parasite drag (i.e. drag of the body), has been overestimated previously. The old default value (Cpar= 0.4) was based on measurements using frozen bird bodies in a wind tunnel (Pennycuick et al. 1988), although the experiments were suspected to overestimate Cpar. Based on new wind-tunnel experiments on live birds, Pennycuick et al. (1996) found that CP,, is much lower than previously measured on frozen birds, and they rec-