FORAGING GROUP SIZE: MODELS AND A TEST WITH JAEGERS KLEPTOPARASITIZING TERNS

Abstract

The formation of foraging groups is likely to result from individual decisions constrained by spatial distribution of food and number of foragers. Foraging groups may form via different mechanisms depending on the absolute levels of at least two relative spatial scales. I present four group size models for environments composed of two nested patch scales. Second-order patches contain first-order patches that correspond to food patches. Individuals within a given second-order patch are able to monitor their conspecifics’ behavior, and the number of individuals attending a first-order patch defines group size. I then test the models’ predictions with Parasitic Jaegers (Stercorarius parasiticus) kleptoparasitizing fish from Common Terns (Sterna hirundo). In this context, tern fishing grounds and flying terns that carried fish correspond to the jaegers’ second and first-order patches, respectively. The Dispersion and Aggregation models assume that foragers have perfect knowledge of the spatial distribution of food and form groups in conformity to an ideal free distribution. The Producer–Scrounger (PS) and Information-Sharing (IS) models, however, assume that foragers are unaware of the food patches’ location and form groups as some foragers join and exploit the food discoveries of others. I used four fitness currencies: probability of obtaining a fish, PF; gross and net rates of energy intake; and energy efficiency. All currencies decreased with group size except the PF currency, which did not vary for group sizes of one and two. Because fitness never increased with group size, the Aggregation model was ruled out. Fish-securing rates of jaegers that initiated and joined chases did not differ. Assuming that initiating and joining a chase correspond to a producing and a scrounging event, respectively, both the PS and IS model then predict that group size should match jaeger abundance. Because group size leveled rapidly with jaeger abundance, both models were discarded. The Dispersion model predicts that group size should decrease with an increase in tern availability when the latter is limiting. Otherwise, jaegers should not form groups unless there is no cost to do so, such as with the PF currency in which solitary jaegers and pairs experience equal fitness. The facts that group size did not reach 1.0 at high tern availability and that it averaged 1.5 when it did not decrease further with tern availability suggest that jaegers may form groups according to the Dispersion model with the PF currency. This conclusion is discussed in relation to the other assumptions underlying the Dispersion model.