As heart rate (
f
H
) can be used to determine the energy expenditure of black-browed albatrosses (
Diomedea melanophrys
) (Bevan
et al.
1994), data loggers - recording
f
H
and abdominal temperature (
T
ab
) -were implanted into free-ranging black-browed albatrosses breeding at South Georgia. Five birds also had salt water switches (sws) attached to one leg to record when the birds were on the water, and two others had satellite transmitters attached to their back to determine the birds’ position at sea. The birds were released into their natural environment and recaptured, on average, 23 days later when the data loggers were removed. The
f
H
data were then converted into estimates of energy expenditure (ee) using a previously derived equation. The mean EE during incubation and brooding were 2.22 and 2.42 W kg
-1
, respectively. When the birds were foraging at sea, EE increased to between 4.63 and 5.80 W kg
-1
, depending on the phase of the reproductive cycle. As the birds spent approximately the same length of time at the nest and at sea during incubation and brooding, the overall mean ee during these phases were 3.63 and 3.54 W kg
-1
respectively. These rates are significantly lower than that during the chick-rearing phase when a high level of foraging EE is maintained almost continuously. By combining information from the sws with the
f
H
data, it was possible to determine the EE of the birds when on the water (5.77 W kg
-1
) and when flying (6.21 W kg
-1
). These values are approximately twice the estimated basal metabolic rate (BMR) for the species. The energy costs of flight are half previous values, estimated using the doubly labelled water technique, because of the previous assumption that birds on the water have an EE equivalent to BMR. When the birds were on the nest,
T
ab
was 39.3 + 0.4 °C and this changed very little with time. However, when they were at sea,
T
ab
showed large variations, depending on the behaviour of the bird. Information from the sws indicated that all large drops (> 0.5 °C) in Tab occurred when the birds were on water. The mean minimum value reached was 32.5 + 2.0 °C. It is likely that ingestion of prey or water are the major causes of this decrease. This is the first study to have used
f
H
extensively to determine the EE of a free-ranging marine bird. The advantages of using this technique are that data can be obtained over long durations with high resolution, permitting the EE of different activities to be estimated.
Validating accelerometry-derived proxies of energy expenditure using the doubly-labelled water method in the smallest penguin species