Winter Habits of the Lynx (Lynx lynx L.) and Wolverine (Gulo gulo L.) as Revealed by Tracking in the Snow

1. What they did

In the 1960s, large predators in Sweden were few (some ten or twenty wolves, a few hundred wolverines, three to four hundred bears and a few more lynx), but reindeer herders and hunters complained of heavy losses. Opinions about how much damage the predators did differed widely, and losses are greatest in winter. To build a sound basis for judging this, the author organised a five-year study of the winter habits of the four big predators by following their tracks in snow. This summary covers the lynx and the wolverine.

• Tracking (1960 to 1964). Patrols of two men on skis searched for the tracks of a predator and followed them as far as possible without frightening the animal. If a track was very fresh, they first followed it backwards for a day or more. The first track they found was usually the one followed. Each day they filled in a standard form, with identical instructions for every tracker.
• Track measurements. When the snow allowed, tracks were measured to estimate the animal’s size and, sometimes, to tell individuals apart.
• Collecting remains. From large prey the trackers collected the lower jaw, a front leg, and where possible the heart, head, pelvis and part of the backbone. About 15 whole reindeer and roe deer carcasses were brought out. Scats from tracked animals were collected and analysed for parasites and food, and prey remains were examined at the National Veterinary Institute in Stockholm.
• Other material. Lynx stomachs from museums, roe deer hunting records from Hällefors in southern Sweden, compensation records for reindeer killed by predators, and reindeer killed by trains (as a “normal” comparison) were also studied.

2. What they found

The tracking effort

• Only 14% of the reports had no information on predators. Of the distance covered on active days, 37% was actual tracking; the rest was searching and travel.
• Most wolverine tracking was done in March and April; the early winter in the mountains was too dark and cold.

Lynx: movement and territory

• Adult males averaged 17.9 kg (13 animals) and females 16.8 kg (10 animals). Yearlings in December and January averaged 9.4 kg.
• A lynx usually spends the day in one long-time bed, leaves about an hour before dark, rests briefly in the darkest part of the night, and hunts again before taking its next day bed.
• Lynx travelled an average of 19.2 km a day along the track (2371 km over 123 day beds). The straight-line distance between two day beds averaged 7.6 km (78 cases).
• One old male in Hällefors was tracked for more than 500 km and used an inner territory of about 300 km². A female with two young seemed to share his route, and no other lynx killed roe deer there.
• In Lapland, lynx travelled further, followed river valleys, and went up to 1100 m in Sarek.

Lynx: hunting

• Lynx locate prey by hearing and sight, stalk with great patience, then attack explosively in a few leaps. In 70% of successful hunts the prey was caught within 20 m.
• Females with young spread out in a line, the young on the flanks. Two cases of real cooperation on hares were seen.
• Of 64 attacks on reindeer, 45 ended in a kill; of 35 attacks on roe deer, 23 did. Roe deer more often spotted the lynx first (nine times, against once for reindeer). Hares were caught in 35% of attacks, capercaillie in 29%, and hazel grouse and willow grouse in 19%.
• A thin crust of snow that would not carry the lynx made hunting impossible; lynx then walked through roe deer areas without trying.
• In reindeer country, reindeer were 66% of kills and the lynx killed more than it needed. Prey shifted from small game in October and November to mostly deer by February, as hares and grouse became scarce.
• Usually a lynx killed no more than one prey a day. Most double kills happened when a second animal “exposed” itself, or, among reindeer, when the lynx could see the next animal from the first kill.
• The lynx jumps on the back of large prey and bites the throat or neck; many examined carcasses showed signs of suffocation. In 23 roe deer and 45 reindeer kills, no head had been taken: the author concludes the common belief that lynx decapitate roe deer is wrong, and suggests foxes may carry off the heads.
• A normal meal is about 2 kg. Lynx often returned to a roe deer kill, but usually ate only once from a reindeer; three times reindeer were killed and not eaten. Hares were eaten almost completely.
• Tracked lynx never ate old carcasses, which the author links to frozen meat being too hard for their teeth. Lynx attacks disturbed prey animals less than those of the other big predators.
• Red foxes were scarce and appeared to fear lynx: some made long jumps over lynx tracks or turned away. Cut-off fox tails left in a track seem to be a sign of a lynx hunt.
• Signs of the rut were seen in mid-March. Kittens stayed with their mother until the March rut, still suckled in December, and shared her bed. Lone kittens near houses were often starving.
• In Hällefors, lynx returned in 1953 after about 30 years. From 1959 roe deer bag records fell to about 60% of the 1958 and 1959 peak (about 40% in areas with lynx), while they rose in areas without lynx.

Wolverine

• 957 km of wolverine tracks were followed, about the same as for lynx in reindeer areas.
• Wolverines were often active in daylight and rested irregularly; a normal day’s travel seemed to be at least 30 km. They repeatedly used the same paths (“wolverine trails”) to visit caches and old kills.
• Reindeer were the main winter food. Wolverines visited three times as many carcasses as they killed; 85% of these carcasses were reindeer. Of about 50 old kills visited, 15 were known wolverine kills and seven were lynx kills. In 32 of 38 scats there were reindeer remains; lemmings, voles, shrews, plant fragments and bone splinters up to 20 mm long were also found.
• Wolverines are not skilful hunters. They chased rather than stalked, mostly in deep snow, and gave up if they did not catch up; the longest chase was 1 km. In about 1000 km of tracking there were nine successful reindeer hunts and seven failed ones. One large male killed seven reindeer, and later ten, in very deep snow. All hunts for small game failed, including six hare chases of up to 5 km or more.
• No wolverine attacked a moose in the study areas. Just outside, one wolverine attacked a cow moose, which died later.
• In 14 cases wolverines followed lynx tracks (77 km in all), 90% of it in early winter when food was short. Carcasses became more frequent through the winter: one every 23.9 km of tracking before March, 8.7 km in March.
• Wolverines stored food in 23 observed caches: dug into snow, in rock crevices and rock heaps, and once in water, never in trees. Meat was carried up to 10 km or more; one female carried meat 8 km to her den, resting many times. Signs of mucous vomit near females with young suggest they may carry food in their stomachs.
• Wolverines jump on a reindeer’s back and bite the back of the neck or the withers. Some bitten reindeer were left alive, perhaps because the bite caused temporary paralysis.
• Dens were usually dug in hard snowdrifts by rocks above the tree line. Several cases of a male and female travelling or meeting together were recorded. Litters averaged 2.4 young (73 litters from the literature).
• Wolverines marked with urine and scat, especially at caches and in breeding areas, and bit pine, spruce and juniper (once birch).

Selection of reindeer

• Using bone marrow fat, more than 50% of reindeer killed by predators were in normal condition, but poor condition and the parasite disease cerebrospinal nematodiasis (12.5% of predator kills against 1.5% of reindeer killed by trains or slaughtered) were far more common than in the comparison reindeer.
• Lynx killed mainly younger reindeer and wolverines middle-aged or older ones. Both killed about 75% females, likely matching the herds.

3. What they concluded

• Snow conditions largely decide whether lynx and wolverines can hunt: soft snow helps a stalking lynx, a weak crust makes hunting hopeless, and deep snow or a strong crust lets wolverines catch reindeer.
• The lynx is basically a small-game (hare) hunter whose fixed hunting pattern spreads its killing over a wide area. Its heavy predation on deer is probably recent, and in reindeer herds it kills more than it needs. Its loss to herders is mainly animals killed, not disturbance.
• The lynx does not normally decapitate its prey or eat carcasses. The wolverine very rarely kills moose in Sweden and should not be counted among moose predators.
• The wolverine depends heavily on carcasses, many left by lynx, which have taken over the wolves’ role of supplying wolverines with meat. It stores large amounts of food for the late winter and for its young.
• Both predators feed to a considerable extent on prey or carcasses of low quality, and wolverines seem to hunt more selectively than lynx. The study could not say whether this is selection in the hunt, or because predators hunt where reindeer are in poor condition.
• The tracking days represent active days; lynx are inactive in bad weather (perhaps 20 to 30% of winter), so the data may overestimate lynx activity.

4. How tracking was or was not used

Tracking in snow was the main method. Patrols followed lynx and wolverine trails on skis, often backwards first and then forwards, and recorded beds, hunts (attack distances, success and failure), kills, how carcasses were eaten, caches, scats, urine and scent marks, tree biting, dens and contacts between animals. Track size was measured to judge size and sex and sometimes to tell individuals apart. The author found the hind foot length the best bear measurement, wolverine tracks measurable with good accuracy, and lynx tracks about 30% wider in loose snow than on hard ground. The paper also lists what made tracking hard: wind and drifting snow erasing tracks, hard crust after thaw, melted snow in spring, and reindeer or moose trampling the trail.

5. Tracker or observer certification

Reported but not validated. The trackers were game wardens, forestry staff, professional hunters, Lapps (Sámi) and other local people, described as carefully chosen and all experienced hunters. Each patrol usually paired a local person with someone in an official position, and all had the same written instructions. Their skill was not tested. Some sex and age judgements came from track size, and the author notes that sex is hard to tell in the field.

6. Other technologies, skills or innovations

• A standard daily tracking form, with identical instructions for every tracker.
• Some tracking in District I was done from a helicopter.
• Laboratory analysis of scats (parasites and food), prey remains and bone marrow fat at the National Veterinary Institute and the University of Stockholm.

7. Implications for conservation (from the paper)

• Reindeer losses to lynx are actual kills, not disturbance of the herds.
• Lynx spread their predation widely, which should help keep prey populations dense enough within their territories.
• Wolverines should not be counted among moose predators in Sweden.
• In Hällefors the roe deer bag records point to pressure from lynx, though many factors affect bag records and more research is needed.