Predator size, prey size and threshold food densities of diving ducks: does a common prey base support fewer large...

1. Allometry predicts that a given habitat area or common prey biomass supports fewer numbers of larger than smaller predators; however, birds from related taxa or the same feeding guild often deviate from this pattern. In particular, foraging costs of birds may differ among locomotor modes, while intake rates vary with accessibility, handling times and energy content of different-sized prey. Such mechanisms might affect threshold prey densities needed for energy balance, and thus relative numbers of different-sized predators in habitats with varying prey patches. 2. We compared the foraging profitability (energy gain minus cost) of two diving ducks: smaller lesser scaup (Aythya affinis, 450-1090 g) and larger white-winged scoters (Melanitta fusca, 950-1800 g). Calculations were based on past measurements of dive costs with respirometry, and of intake rates of a common bivalve prey ranging in size, energy content and burial depth in sediments. 3. For scaup feeding on small prey <12 mm long, all clams buried deeper than 5 cm were unprofitable at realistic prey densities. For clams buried in the top 5 cm, the profitability threshold decreased from 216 to 34 clams m(-2) as energy content increased from 50 to 300 J clam(-1). 4. For larger scoters feeding on larger prey 18-24 mm long, foraging was profitable for clams buried deeper than 5 cm, with a threshold density of 147 m(-2) for clams containing 380 J clam(-1). For clams <5 cm deep, the threshold density decreased from 86 to 36 clams m(-2) as energy content increased from 380 to 850 J clam(-1). If scoters decreased dive costs by swimming with wings as well as feet (not an option for scaup), threshold prey densities were 11-12% lower. 5. Our results show that threshold densities of total prey numbers for different-sized ducks depend on prey size structure and depth in the sediments. Thus, heterogeneity in disturbance regimes and prey population dynamics can create a mosaic of patches favouring large or small predators. Whether a given area or total prey biomass will support greater numbers of larger or smaller predators will vary with these effects.