Mechanical and aerobic energy costs of diving were measured simultaneo
usly by closed-circuit respirometry in six lesser scaup Aythya affinis
Eyton (body mass=591+/-30 g) during bouts of voluntary feeding dives.
Durations of dives (t(d)=13.5+/-1.4 s) and surface intervals (t(i)=16
.3+/-2.2 s) were within the normal range for ducks diving to 1.5 m dep
th. Mechanical power output (3.69+/-0.24 W kg(-1)) and aerobic power i
nput (29.32+/-2.47 W kg(-1)) were both higher than previous estimates.
Buoyancy was found to be the dominant factor determining dive costs,
contributing 62 % of the mechanical cost of descent and 87 % of the co
st of staying at the bottom while feeding. Drag forces, including the
contribution from the forward-moving hindlimbs during the recovery str
oke of the leg-beat cycle, contributed 27 % and 13 % of the mechanical
costs of descent and feeding, respectively. Inertial forces created b
y net acceleration during descent contributed approximately 11 % durin
g descent but not at all during the feeding phase. Buoyant force at th
e start of voluntary dives (6.2+/-0.35 N kg(-1)) was significantly gre
ater than that measured in restrained ducks (4.9+/-0.2 N kg(-1)). Loss
of air from the plumage layer and compression due to hydrostatic pres
sure decreased buoyancy by 32 %. Mechanical work and power output were
1.9 and 2.4 times greater during descent than during the feeding phas
e. Therefore, energetic costs are strongly affected by dive-phase dura
tions. Estimates by unsteady and steady biomechanical models differ si
gnificantly during descent but not during the feeding phase.