This is a study of the relationship between predators and their prey. We shall focus on the Arctic fox Alopex lagopus, the Arctic skua Stercorarius parasiticus and the long-tailed skua S. Longicaudus, all of which are dependent on lemmings as a food resource. One aspect of the study is how the population dynamics of these predators are related to their prey species. This includes both the dietary response (the functional response) and the changes in population densities (the numerical response). Another aspect deals with predator–predator interaction, especially their spatial organisation and how different species compete for the same food resources.

Large fluctuations in lemming numbers are perhaps the most fascinating trait of Arctic and sub-Arctic ecosystems. Lemming populations often reach peak numbers every three to five years. Most of the hypotheses put forward for these population cycles fall in to one of three categories: food, intrinsic factors or predation. The predation hypothesis suggests that predators cause the cyclic behaviour of rodent populations by periodically reducing rodent numbers. This theory has received support from many studies. The tundra ecosystem harbours several mammalian and avian predators that are dependent on lemmings for successful reproduction. Although the tundra might seem to be a simple ecosystem, the relationship between lemmings and their various predators is complex and has baffled scientists throughout the 20th century. A better understanding of the tundra ecosystem requires studies that include both lemmings and their predators. In particular, it is necessary to study the factors which limit the population size of each species and those which permit co-occurrence of species with similar niches.

The Arctic fox is a circumpolar carnivore with opportunistic feeding habits, hut in most inland areas it is dependent on lemming population peaks for successful reproduction. During peak years there can be up to 18 cubs in a single litter, while there usually is no reproduction at all during lemming lows. Arctic foxes build dens which, if not destroyed by permafrost or landslides, can last for centuries and become very large, with 200 openings covering an area of 400 m². They are territorial in the breeding season, hut spend the winter ranging over large areas on the tundra or on sea ice. They are capable of migrations of up to 2300 km. Lemmings are their staple diet during the summer season. Other important types of food in summer are reindeer carcasses, voles, hares, ptarmigan, willow grouse and other birds. In winter, the most important food resources are carcasses of large mammals, ptarmigan and grouse.

Skuas appear to show great fidelity to breeding territories, often returning to the same territory in consecutive years. Non-breeders and failed breeders leave the tundra as early as July. In both long-tailed and Arctic skuas, successful pair formation and breeding on the tundra require large numbers of lemmings. The summer diet of the long-tailed skua is 50-99% lemmings. Aggression among skua species is common in the breeding season. They lay a clutch of two eggs in a scratched hollow in the ground. Both species are long-distance migrants, spending the winter pelagically in the southern oceans.

Together with the rough-legged buzzard and the snowy owl, these species form a foraging guild, i.e. they exploit the same food resource. Also, their production of offspring is related to the abundance of lemmings on the tundra. Theoretical studies have suggested that competition between species, predation and recruitment processes determine guild structure. However, there is no strong evidence for this. In contrast, recent studies suggest that predation could affect the distribution and assemblages of species. Furthermore, the importance of super-predation, when predator kills predator, has attracted increased attention.

Another interesting field of study is the relationship between the predators and the two lemming genera Lemmus (brown lemmings) and Dicrostonyx (collared lemmings). The population dynamics of these rodents differ and their interaction presumably depends on their relative effect on predators. In Siberia, Lemmus is the dominant prey in most areas. Data suggest that Arctic foxes have the capacity to regulate Dicrostonyx where Lemmus is most common. However, at most Canadian sites, Dicrostonyx are the dominant or even the only lemming species. This presents us with a question to explore: Whether, for example, Arctic foxes are truly specialized on Lemmus or whether they simply show preference for, and are regulated by, the most common prey.

Methods

We censused mammals and birds at each study site. The study was carried out during the breeding periods of all species. On foot, we surveyed between 8 and 90 km² at each of the 17 sites, the area depending mainly on weather and visibility conditions. We used binoculars to spot mammals, birds and nests/dens. For a number of mammal and bird species, we counted all individuals and for the predator species, we also investigated breeding attempts, counted young and noted nest/den positions. Fox dens were studied in detail.

We also collected prey remains, bird pellets and fox scats for diet analyses, and noted lemming winter nests as a gauge of previous densities. In total, we surveyed 758 km² on foot. At each site, rodent populations were surveyed by other researchers (Angerbjörn et al.). Furthermore, we carried out an aerial count of birds and large mammals by means of helicopter transects which also included a vegetation survey. The se transects were 40 to 118 km long and were performed at each of 15 sites.

Results and discussion

Together with the data collected in Siberia on the Tundra Ecology 1994 expedition, the present study provides an opportunity to combine detailed data from the entire Holarctic populations of Arctic foxes, skuas and their prey. In total, 13 breeding Arctic fox dens were found during surveys on foot. This is a low number for such a large total area. We also observed 12 adult Arctic foxes and 5 red foxes, but no red fox breeding attempts. Also avian lemming predators were scarce at most sites. Stationary long tailed skuas were found at 8 sites and Arctic skuas at 6 sites (table 1). The reason for meagre predator numbers was the low number of rodents, as described by Angerbjörn et al.. This absence of peak numbers in lemmings and their predators impaired data collection, as there were very few fresh faeces and prey remains to be found. The presence of wolves was indicated by tracks and faeces at two sites, and that of weasels by a few winter nests. Musk-oxen, caribou and geese were very common at same sites (table 1). The aerial surveys covered larger areas and showed that lemming trapping and surveys on foot were generally performed in the most productive habitat at each site. From the vegetation mapping, we will be able to compare these survey types and perform comparative analyses on different spatial scales.

We have collected a large amount of faecal droppings and bird pellets. From the analysis of these, we will discuss functional response and other aspects of the relationship between Arctic foxes and their prey populations. This is an important part of the analysis, since the functional response of a predator, together with the numerical response, enables us to analyse the regulation of prey populations.

Arctic fox dens were in most cases quite small, with fewer than 50 openings. This is most likely du e to soil movements caused by permafrost and may cause the splitting up of litters, making litter size estimates difficult. Den structure was dependent on the geomorphology at each site. The data collected on fox den structure will be used in comparisons with those of other Arctic areas and with the dens of swift and kit foxes, two close relatives of the Arctic fox.

Our aim is to reveal the structure of the lemming predator guild by comparing the relative densities of different predators, their spatial distribution within each site and their diet. An interesting question is the extent to which the predators eat each other.  Some other questions of interest are: To what extent do these species compete? How do they avoid competition when resources are scarce? What is the potential effect of combined predator pressure on lemming populations? Our results will be analysed in co-operation with the other projects connected with lemming cyclicity.