1.1. Background

Predation is a violent interaction in nature, where one part loses its life. Almost all animals are destined for a violent death, which is often caused by predation. This excludes top predators – although they are often the prey of man. Conventionally, predation was considered only to keep track prey population killing rather ‘doomed surplus’ (Erringhton 1956). Also prey individuals fallen as an offer were often assumed to be injured or ill and predators as ‘health officers’ in nature. Predation was, thus, considered mainly compensatory (Kenward 1986, Korpimäki & Krebs 1996). A classic example of predator-prey interaction, snowshoe hare - lynx time series from the 19th century, based on statistics from Hudson Bay company, suggested a predator driven cyclicity of prey but it can also be interpreted in a conventional way. The early, simple, Lotka-Volterra model suggests a fluctuating nature of predator-prey interaction (see Begon et al. 1990). In more realistic models, the stability of the system depends mainly on prey recruitment rate, the predator’s searching efficiency and a time-lag between predator and prey (Rosenzqeig & MacArthur 1963, May 1973, Maynard-Smith 1974, Hanski et al. 1991). A lot of work done in the ecology of voles and snowshoe hares produced several hypotheses attempting to explain their periodic fluctuations (Korpimäki & Krebs 1996, Krebs 1996). However, most population regulation theories put forward in the 1960s stressed social and even genetic feed-back mechanisms not to mention plant-herbivore interactions as a cause of population fluctuations (see review Krebs 1996). Starting in the beginning of the 1980’s, several field and theoretical studies on small mammals turned their interest towards the role of predation in regulating prey populations (Keith et al. 1977, Angelstam et al. 1984, Erlinge et al 1984, Hansson 1984, Kenward 1986, Henttonen 1987, Korpimäki et al. 1991, Hanski et al. 1991, 1993, Norrdahl & Korpimäki 1995, Korpimäki & Krebs 1996).

The ‘armsraces’ between prey and predator, co-evolution, is probably responsible for many specific adaptations found in animals. For example camouflage, mimicry and different anti-predator behaviours like flocking are cases where avoidance of predation is apparent. Most of these adaptations have a genetic basis but also, especially in higher vertebrates, some behaviours can be learned (Taylor 1984). According to the life-history theory, animals tend to increase the proportion of their genes in future generations. Thus the predator that can allocate the most energy for breeding also obviously increases its genes most in the next generation. The predator must therefore decide where to hunt and what to hunt. Optimising energy gain will, according to the optimal foraging theory, sometimes lead to specialisation in prey choice, but sometimes a more beneficial way is to attack every potential prey. (Charnov 1976, Pyke et al. 1977, Pyke 1984, Stephen & Krebs 1986, Reynolds et al. 1988). The responses of a predator to changing prey numbers can be classified into numerical and functional responses according to Solomon (1949). The shape of the response curve is considered to be important in the regulation of prey population (Holling 1959, Taylor 1984). A sigmoid shaped response curve hints to stabilising of prey population while a convex shaped curve to destabilising it.

The goshawk is a large, relatively common and widely distributed raptor species in the northern hemisphere (Fischer 1980). It captures bird and mammal species varying greatly in size, from small mammals weighing a few grams up to the size of hares and capercaillie cocks weighing 4kg (Höglund 1964, Sulkava 1964, Brull & Fischer 1981). Because many small game species, interesting also man, include goshawk’s diet, its foraging habits have been studied in Europe since the 1940’s (see Brull & Fischer 1981). In Central Europe, the goshawk’s food base depends greatly on pidgeons, corvids, thrushes and rabbits (e.g. Opdam et al. 1977, Cozszynski & Pilatowski 1986), while in the boreal forests of Northern Europe goshawks hunt mainly grouse species; black grouse Tetrao tetrix, capercaillie Tetrao urogallus, hazel grouse, Bonasa bonasia and willow grouse, Lagopus lagopus (Sulkava 1964, Höglund 1964, Linden & Wikman 1983, Widen 1987, I, II). In the boreal forests of North American mammals, primarily the snowshoe hare Lepus americanus constitutes the main prey base (Storer 1966, McGowan 1975). Because of a high proportion of game animals in the diet, the goshawk has caused a lot of controversy between protectionists and game managers. Recent studies on game animals adopting radio-telemetry as a research tool have revealed predation to be the their most marked proximate cause of death (Angelstam 1984, Willebrand 1988, Kastdalen & Wegge 1989, Wegge et al. 1990, Swenson 1991, Marjakangas 1992, Valkeajärvi & Ijäs 1994). The goshawk has turned out to be one of the most important predators of adult grouse. When goshawk predation was studied in Sweden on pheasant Phasianus colchicus, population decreases of up to 60-70 % were found, depending on the density of pheasants (Kenward et al. 1981). When a similar study was conducted in boreal forests for winter predation on grouse, very low predation was apparent (Widen 1987). Yet, the highest losses of grouse taking place in the early phase of breeding as depredated eggs and chicks are presumably not due to goshawk predation. Recently, changes in forest structure are considered to be the most important cause for the poor success of grouse (e.g. Henttonen 1989, Andren 1995, Wegge et al. 1990, Kurki et al. 1997). Forest fragmentation has probably increased voles and their predators; red foxes Vulpes vulpes, stoats Mustela erminea and martens Martes martes, which would also increase grouse kills.

Goshawk is considered an old forest species based on its nest site selection (e.g. Link 1986). Decrease of the amount of old forests has probably negatively affected on its living conditions via loss of nesting and hunting habitats (Widen 1997). However, goshawk is fairly well adapted to agricultural landscape of central Europe (Kenward 1982). Therefore, its adaptation to modern man-dominated boreal forest might mainly be a question of sufficient food supply, especially in winter.