(A) Calibration relationship between heart rate (f_H) and rate of oxygen consumption () in eight Australasian gannets. (B) Distribution of mean f_H (calculated every 3 min) measured in six free-ranging Australasian gannets. The range of f_H used in the calibration procedure encompasses 85% of f_H observed in the free-ranging gannets. Data were obtained while birds rested and walked on a treadmill. In A, each data point represents the mean values of f_H and calculated for bins of 10 beats min⁻¹ for each bird. The solid line shows the regression line through these data points, weighted by the square root of the number of points used to calculate each mean (=8.58e^0.0098f_H, r²=0.89, P<0.001). Short-dash lines are 95% prediction intervals; long-dash lines are 95% confidence intervals. Parameters required to calculate the confidence intervals are as follows: number of individuals (N)=8, number of data points=131, standard error of slope estimate (s.e.e. slope)=0.000334, mean f_H during calibration=146.1, variance component for individuals=0.02957, variance component for error=0.11287.

Diel variation in mean (+s.e.m./s.e.e.) f_H (A) and (B) of Australasian gannets while in the laboratory (light grey bars, N=8) or free-ranging in the field (dark grey bars, N=5) from 11 January to 7 February. Field estimates of (+s.e.e.) were made using a calibration relationship between f_H and (Fig. 1). The horizontal black bar represents the mean duration of daylight for the study period.

Estimated mean (±s.e.e.) daily field metabolic rate (FMR, A), net metabolic rate (NMR, B) and minimum metabolic rate (MMR, C) of free-ranging Australasian gannets (N=6) for each day of the 2004–2005 breeding season. Dashed lines show significant linear regression relationships between date and metabolic rate. The timing and duration of the incubation and chick-rearing phases of the study animals are indicated by the horizontal black and grey bars, respectively, in C.

Mean (±s.e.m.) time spent in flight (A), time spent submerged (B) and number of dives (C) by free-ranging Australasian gannets (N=4) for each day of the 2004–2005 breeding season. The timing and duration of the incubation and chick-rearing phases of the study animals are indicated by the horizontal black and grey bars, respectively, in C.

Calibration relationship (OCal) between f_H and in Australasian gannets (grey line) including an extrapolation beyond the range of f_H measured during calibration (dashed section) most likely associated with flight. Black lines show MinCal (lower) and MaxCal (upper) calibration lines including a similar extrapolation (dashed sections), which were derived to account for potential errors in the extrapolation of OCal and the use of it for all behaviours and modes of locomotion. Also shown (black circle) is the estimate of maximum aerobic power output during flight (with 95% confidence intervals) made using Method 2 of Bishop (Bishop, 1997) for the corresponding highest f_H values recorded in the free-ranging gannets, which are assumed to be during flight (Fig. 1).