Christoph Rohner

Population changes of the vertebrate community during a snowshoe hare cycle in Canada’s boreal forest

We measured the density changes of 22 species of vertebrates during a snowshoe cycle in northern Canada. Hares were the dominant herbivore in the system and changes in their numbers were correlated with changes in numbers of arctic ground squirrel, spruce grouse, ptarmigan, lynx, coyote, great horned owl, goshawk, raven and hawk owl. Hare numbers were not correlated with numbers of red-backed vole which showed peaks during the low, increase, and early decline phases of the hare cycle. Hawk owls were the only predator whose numbers correlated with changes in red-backed voles while boreal owls and weasels were correlated with densities of Microtus. Red squirrel, American kestrel, red-tailed hawk, northern harrier, wolverine, magpie, and gray jay showed no correlation with hare or vole numbers. We conclude that species in the boreal forests of Canada do not exhibit the strong synchrony found between voles and other members of the vertebrate community in northern Fennoscandia. We discuss some of the possible reasons for these differences.

Owl predation on snowshoe hares: consequences of antipredator behaviour

We show evidence of differential predation on snowshoe hares (Lepus americanus) by great horned owls (Bubo virginianus) and ask whether predation mortality is related to antipredator behaviour in prey. We predicted higher predation on (1) young and inexperienced hares, (2) hares in open habitats lacking cover to protect from owl predation, and (3) hares in above average condition assuming that rich food patches are under highest risk of predation. Information on killed hares was obtained at nest sites of owls and by monitoring hares using radio-telemetry. The availability of age classes within the hare population was established from live-trapping and field data on reproduction and survival. Great horned owls preferred juvenile over adult hares. Juveniles were more vulnerable to owl predation before rather than after dispersal, suggesting that displacement or increased mobility were not causes for this increased mortality. Owls killed ratio-collared hares more often in open than in closed forest types, and they avoided or had less hunting success in habitats with dense shrub cover. Also, owls took hares in above average condition, although it is unclear whether samples from early spring are representative for other seasons. In conclusion, these results are consistent with the hypothesis that variation in antipredator behaviours of snowshoe hares leads to differential predation by great horned owls.

MORTALITY IN FLEDGLING GREAT HORNED OWLS FROM BLACK FLY HEMATOPHAGA AND LEUCOCYTOZOONOSIS

Black fly feeding alone and in concert with Leucocytozoon spp. infection caused mortality in fledgling great horned owls (Bubo virginianus) in the Yukon, Canada 1990 to 1991. These mortalities occurred during a year of food shortage corresponding with a decline in the population of snowshoe hare (Lepus americanus), the main prey for great horned owls. We hypothesize an interaction between food availability and the consequences of host-parasite interactions.

Northern Hawk-Owls in the Nearctic Boreal Forest: Prey Selection and Population Consequences of Multiple Prey Cycles

We studied hawk-owls in the southwestern Yukon, Canada, from 1987-1993. Most information on hawk-owls originates from studies in Europe, and very little is known about the subspecies Surnia ulula caparoch in North America. The boreal forest communities in the two continents differ remarkably in the composition of cyclic herbivore populations. Fennoscandia is dominated by 3-4 year microtine cycles, whereas northern Canada and Alaska experience a 10-year cycle in snowshoe hare numbers, with voles fluctuating at lower levels. We studied the diets of nine nesting pairs by pellet analysis, and we observed prey deliveries at five nests. The proportion of voles in the diets was lower than reported from Fennoscandia, and snowshoe hares made up 40-50% during the peak of the hare cycle. Estimates of prey densities by live-trapping revealed that hawk-owls strongly prefer voles over snowshoe hares and squirrels. Among voles, Microtus were preferred and Clethrionomys were avoided. Hawk-owls showed, however, a functional response not only to voles but also to juvenile hares, and they may be critically dependent on larger prey during certain nesting stages when vole abundance is moderate or low. Breeding densities and winter observations changed concurrently over years of different prey abundance. Prey selection translated into population consequences: hawk-owls did not respond numerically to Clethrionomys outbreaks, but to the combined densities of Microtus and snowshoe hares. We conclude that the Northern Hawk-Owl is less of a vole specialist and more affected by the prey composition in specific systems than commonly assumed, and we discuss this pattern from an evolutionary perspective