Is the European Honey-buzzard (Pemis apivorus) a feeding specialist? The influence of social hymenoptera on habitat selection and home range size
In the years 1984-88 and 1996-98 a long-term study of the food specialist European Honey-buzzard (Pernis apivorus) was conducted in southern 8urgenland, Austria. The study examined the influence of social Hymenoptera, in particular of wasps of the genus Vespidae, on habitat selection and home range size.
(1) 404 prey items were collected at 56 nest sites and the occurrence of Hymenoptera in the nests was compared with their abundance (Iine transects, wasp nest density) in the environment. A few hymenoptera species comprised 81,8 % of prey items found (76,4 % wasps, 5,4 % bumblebees). Less abundant were frogs at 7,5 %, birds at 6,3 %, lizards at 1,1 % and various invertebrates at 3,3 % (Fig. 2). In comparison to their abundance large colonies of Vespula-species (V. vulgaris and V. germanica) were definitely prefered, whereas hornets (Vespa crabro) and field wasps (Pofistes spp.) were rather avoided. The frequency of Dolichovespula-species found as prey was similar to their occurrence in the environment.
(2) In examining habitat utilization more than 2/3 of all observations (n := 157) occured in forests. Of 8 habitat types distinguished in the study area, mature and medium-aged forests, orchards, and small wetlands were preferred. Monocultures of arable fields, hay meadows and dense young wood stands were avoided by foraging European Honey-buzzards (Fig. 4). In habitats with the highest Hymenoptera density the hunting success (excavated nests) was highest (Fig. 6) and therefore they are the most important on es for the spedes.
(3) Individual birds could be identified according to plumage characteristics. Altogether 45 home ranges (18 females and 27 males, minimum convex polygon) were analysed. Three different phases of the breeding cycle were analysed: (A) arrival start of breeding, (8) incubation, and (C) rearing time -independence of young, according to which 3 partial home ranges were calculated (Tab. 1). Home ranges were smallest (females 2,6 km2, males 3,2 km2) in phase (A) and largest in Phase (C) (females 14,6 km2, males 15,4 km2). Only during (8) incubation did the home ranges of females (3,7 km2) and males (7,2 km2) differ to a large extent.
years when Hymenoptera were abundant, home range size varied between 7,9 and 16 km2 in poor Hymenoptera years between 16 and 25 km2• About 50 % of all observations of females (n 267) took place less than 1 km from the nest site; in contrast almost 50 % of all hunting males (n =622) were observed between 1 and 2 km (Fig. 8) from the nest. Most birds hunted within a radius of 3 km ot the nest site. Maximum distances from the nest for females were > 6 km, and for males> 7 km. When data from both sexes were pooled, significant differences in ranging behaviour in relation to Hymenoptera density were found. In years with abundant Hymenoptera, observations were concentrated within a radius of 1 km from the nest. In years with lower numbers of Hymenoptera birds flew for larger distances (observation peak 1-2 km, Fig. 9) when searching for food.