Figure 3

Relationship between departure fuel load and total fuel deposition rate of northern wheatears in Wales, Alaska. Departure fuel load and total fuel deposition rate did not correlate significantly (blue dots and blue trend line; n = 21; R _S = 0.3, 95% CI = -0.22 – 0.65). Black dashed line indicates departure fuel load to accomplish a 9 h flight, which is the maximum night length experienced during the study period. If birds had minimized the overall energy costs of transport, data points should have scattered around this line [1]. For comparison we give departure fuel loads and fuel deposition rates of wheatears departing from Heimaey, Iceland (light grey dots and light grey trend line; n = 10, R _S = 0.81, 95% CI = 0.16 – 1.0) [9], and from Helgoland, Germany (dark grey dots and dark grey trend line; n = 13, R _S = 0.61, 95% CI = 0.003 – 0.91) [8] during autumn. Correlation coefficients did not differ significantly from each other (compare 95% CI and Steiger’s Z-tests: z-score < 1.93, P > 0.054). Black curves i – iii are the predicted relationships for time-minimizers assuming global variation with search and settling time of one day (i), three days (ii), and five days (iii). Curve iv is the predicted relationship assuming local variation as previously described [64], with recapture on Helgoland without supplement food (black cross: total fuel deposition rate = 0.030 day^-1, departure fuel load = 0.304). There were no recaptures without supplement food in Wales.