Short communication Landscape and season effects on the diet of the Goshawk RISTO TORNBERG, 1 * MIKKO MO ¨ NKKO ¨ NEN 2 & SAMI M. KIVELA ¨ 1 1 Department of Biology, PO Box 3000, FIN-90014 University of Oulu, Oulu, Finland 2 Department of Biological and Environmental Science, PO Box35,FIN-40014 University of Jyva¨skyla¨, Jyva¨s- kyla¨,Finland Keywords: corvids, grouse, predation, prey avail- ability, vulnerability. There are two general effects of habitat loss and frag- mentation of mature boreal forests (Schmiegelow & Mönkkönen 2002). First, fragmentation by farmland creates stable structures such as permanent edge zones with enrichment of species diversity and density (Andrén 1992, Berg & Pärt 1994). Secondly, modern forestry with clear-cuts creates sharp, unstable bound- aries between forest and open areas, usually with less pronounced edge effects (Helle 1983, Schmiegelow & Mönkkönen 2002). Considering the vast array of stud- ies on the effects of habitat loss and fragmentation on bird populations, relatively little attention has been paid to the role of predators, other than nest preda- tors, across different landscapes (Lampila et al. 2005). Predators’ searching efficiency may improve due to a diminished area where prey live (Storaas et al. 1999). By killing smaller predators and nest predators, top predators may contribute positively to prey species populations (Petty et al. 2003, Mönkkönen et al. 2007). Increased availability of alternative prey as a result of landscape change may deflect predation from the main prey species (Angelstam et al. 1984). The final outcome of these landscape-related predator–prey interactions is likely to depend on direct functional and numerical responses of predators to the variation in the abundance and vulnerability of the main and alternative prey, as well as on the indirect controlling effect of top predators on smaller predators and nest predators. In northern latitudes the Northern Goshawk Accipi- ter gentilis relies mainly on grouse as a staple food dur- ing most of the year (Tornberg 1997, 2001, Tornberg & Colpaert 2001). Breeding season diet, however, con- tains a large spectrum of alternative prey species, mainly birds (Tornberg 1997). The proportion of grouse in the diet is at the lowest during late nestling phase when fledglings of alternative prey such as larger passerines and waterfowl are readily available (Lindén & Wikman 1983, Tornberg 1997). Goshawks mainly use mature forests for nesting (Penteriani 2002), but they are more flexible in their choice of hunting habi- tats (Kenward & Widén 1989, Tornberg & Colpaert 2001). Even though the diet and habitat associations of the Goshawk are relatively well known, we do not have a clear picture of how these vary with landscape structure. In this study, we examined Goshawk predation on grouse (Willow Grouse Lagopus lagopus, Black Grouse Tetrao tetrix, Capercaillie Tetrao urogallus , Hazel Grouse Bonasa bonasia) along a landscape gradient. We also examined whether predation on alternative prey was dependent on the same landscape gradient. METHODS Study area The study area comprised roughly 1700 km 2 of coastal lowland situated near the city of Oulu in northern Fin- land (25°30¢E, 65°00¢N). Almost one third of the area is covered by peat-lands, natural and drained bogs. There are few lakes, but many rivers in the area forested mainly by Scots Pine Pinus sylvestris and Norwegian Spruce Pi- cea abies mixed with Birch Betula pendula and Aspen Populus tremula. Collection of food remains Food remains were collected in active Goshawk territo- ries between 1989 and 2003, during the nest building and incubation period from the beginning of April to the end of May (hereafter spring), and the nestling period from the beginning of June until mid-July (hereafter summer); at least once during both periods in each year the territory was occupied by a breeding pair. In spring, collection was done searching the surroundings of the nest where feeding takes place. In summer, collection was made from the nest, because food remains accumu- late there, especially during the last week before fledg- ing. Identification of food remains to species level was carried out with reference to collections of the Zoologi- cal Museum of University of Oulu. Collection years per territory varied from 1 to 10. Prey weights were taken from Wikman and Tarsa (1980) and specimens from the Zoological Museum of University of Oulu. *Corresponding author. Email: risto.tornberg@oulu.fi ª 2009 The Authors Journal compilation ª 2009 British Ornithologists’ Union Ibis (2009), 151, 396–400