Changing distribution of flocking sea ducks as non-regenerating food resources are depleted

JO Bustnes, GH Systad, RC Ydenberg
2013 Marine Ecology Progress Series  
Sutherland & Anderson (1993; J Theor Biol 160:223-230) considered how the distribution of consumers should change as prey resources fixed at the start of a period of exploitation are depleted by foraging activities. Their model predicts that consumers initially aggregate on a single or a few patches offering the highest feeding rate. As the prey density in those patches is reduced, the feeding rate declines to the level attainable at the next best patches, which are then also exploited.
more » ... e, an increase in the number of flocks and a decline in flock size should be observed as individuals spread out over the available feeding opportunities. Further, once occupied, a patch is exploited for the remainder of the winter. We tested these predictions with winter survey data on the number, size and location of flocks of common eiders Somateria mollissima and king eiders S. spectabilis wintering in Kvalsundet, a sound in northern Norway. Both species are benthivorous, and there is little or no growth or recruitment of their prey in winter at high-latitude sites. The green sea urchin Strongylocentrotus droebachiensis was the dominant prey of both species in Kvalsundet, and the density of this prey species declined over the winter in kelp beds, the preferred winter habitat of common eiders. Our data showed that both common and king eiders began their winter periods in Kvalsundet aggregated into a few large flocks. As the winter progressed, the distribution changed and birds gathered into more and smaller flocks. In the case of common eiders, flock locations remained fixed once a foraging location began to be exploited. King eiders formed a few large flocks early in the season which drifted in heavy tidal currents, but when these split up into smaller flocks in February, they started to exploit fixed locations. We interpret these results as consistent with the Sutherland & Anderson model.
doi:10.3354/meps10339 fatcat:2asbyaptv5gyjnbcbtr52a5ola