Make your own free website on Tripod.com
Wolves
Wolf History, Conservation, Ecology and Behavior
[www.wolfology.com]

Ecology of an Exploited Wolf Population in South-Central Alaska
Warren B. Ballard, Jackson S. Whitman & Craig L. Gardner
1987
Abstract
During April 1975 through April 1982, 151 gray wolves (Canis lupus) in 30 packs were captured and radio-collared in a 61,600 km2 area of south-central Alaska. Area traversed by wolf packs was correlated with number of radiolocations, pack size, and density of ungulate prey. Territory sizes of packs located >60 pack days averaged 1,645 km2....Wolves did not follow migratory movements of moose (Alces alces) or caribou (Rangifer tarandus granti) outside of their pack areas but did follow elevational movements of moose within their areas. Twenty-eight percent of 135 wolves dispersed from their original area. Dispersal occurred mainly during April through June and October through November. Seventy-four percent of dispersers were males. Twenty-two percent of all wolves that dispersed were accepted into existing packs. Parturition in south-central Alaska occurred during May. Natal dens were usually abandoned between late June and late July....Average distance between dens of neighboring packs was 45 km.
Wolf densities ranged from 10.3/1,000 km2 in autumn 1975 to 2.6/1,000 km2 in spring 1982. Mean annual litter sizes in November varied from 3.7 to 7.3 pups....Intraspecific strife and other natural factors caused 20% of the annual wolf mortalities; the remaining mortalities resulted from legal and illegal human harvest. Harvest by humans >40% or total mortality >50% of autumn wolf numbers resulted in a population decline. During 1976 through summer 1978, 60 wolves were killed in the 7,262 km2 Susitna River Study Area to determine the effects of reduced wolf densities on moose survival. Spring wolf densities were reduced from precontrol levels by 42% to 58% annually. Annual rates of increase of wolves during years of control ranged from 0.57 to 0.81. When reduction efforts were terminated, wolf densities quickly increased with annual rates of increase ranging from 1.04 to 2.40.
Wolf packs were observed at 439 kills; 70% were moose. Wolves preyed upon moose calves slightly less or in proportion to their occurrence in the moose population during late May through October. Proportionately more calves were killed by wolves during winter, but adult moose were overall the most common prey (38% of the kills.) Caribou were the second most common prey....Wolf predation rates during summer...ranged from 1 kill/7 to 16 days/pack, whereas winter rates ranged from 1 kill/5 to 11 days/pack....Wolf-killed ungulates usually were heavily utilized by wolves....
Spring wolf densities were negatively correlated with subsequent autumn moose calf:cow ratios, suggesting that reduced wolf densities improved survival rates of calf moose. Calf moose mortality studies indicated that predation by brown bears was the most important calf mortality factor. Estimates of moose recruitment compared with estimates of annual mortality from wolf predation during 1980-81 and 1983-84 within the Susitna River Study Area suggested that wolves were not preventing moose population growth at that time.
INTRODUCTION
Game Management Unit 13 (GMU 13), which encompasses the Nelchina and upper Copper and upper Susitna River basins, has been one of the most important moose hunting areas in Alaska for 3 decades. During 1963-74, approximately 18% of the reported statewide moose harvest occurred in the area. The area also has been popular for nonconsumptive uses of wildlife.
Prior to 1953, few data on wildlife populations in GMU 13 were available. The moose population probably increased through the late 1940's and 1950's and peaked about 1960. Severe winters in 1961-62, 1965-66, and 1971-72 resulted in high moose mortality. Moose population indices indicated a general decline in recruitment (calves:100 cows) during intervening years of mild winters, and the population continued to decline.
....[E]fforts to identify the causes of low recruitment were initiated in 1975. Bishop and Rausch (1974) and McIlroy (1974) reviewed available data and concluded that snow depth, range quality, and hunting may have contributed to the population decline but were probably not the primary causes of low recruitment.
An extensive program of wolf poisoning and aerial hunting, conducted by the federal government during 1948-53, apparently resulted in low wolf densities and a period of moose population growth. In 1953 only 12 wolves were believed to remain in the area. Following termination of the program, wolf numbers increased to approximately 125 by 1961 when the moose population reached its peak. Wolf numbers may have increased to a peak of 350-450 by 1965 when the moose population was declining. These moose and wolf population trends and the high occurrence of calf moose hair in wolf scats suggested that predation by wolves was a major factor contributing to the decline of the moose herd.
An understanding of the moose population decline was complicated by the presence of other potential predfators, especially brown (Ursus arctos) and black bears (Ursus americanus), and such alternative prey species as barren-ground caribou, Dali sheep (Ovus dalli), beaver (Castor canadensis), and snowshoe hare (Lepus americanus)....Caribou numbers declined from about 71,000 in 1962 to approximately 10,000 in 1973. The contribution of predation to the caribou decline was unknown, and contrasting views as to its importance exist.
In 1975 a wolf study was initiated to determine food habits and territory size, to examine population dynamics, and to determine the impacts of wolves on moose. This monograph focuses primarily on results of wolf studies conducted during 1975 through April 1982, but also includes pertinent data from concurrent related studies on moose and brown bear during 1975 through 1984....
Study Area
The study was conducted in GMU 13, an area of 61,600 km2. Thirty-one percent (18,800 km2) of the area lies above 1,220m elevation, which is considered poor year-round habitat for ungulates and wolves. Wolves were killed by Alaska Department of Fish and Game (ADF&G) personnel during 1976 through 1978 in an area of 7,262 km2 within GMU 13, referred to as the Susitna River Study Area (SRSA)....
....[B]oth temperature and precipitation are quite variable within GMU 13. Physiography and vegetation types...are diverse....Willows, the most important moose winter forage, occurred in nearly all vegetation types but were most abundant in riparian areas....Although much of GMU 13 is roadless and sparsely populated, it is traversed by four major road systems: the Richardson, Glenn, Parks, and Denali Highways. Much of GMU 13 is accessible by all-terrain vehicle, snowmobile, or light fixed-wing aircraft....
METHODS
Wolf Studies
Wolves were captured for radio-collaring by darting from a helicopter....Three additional live wolves were purchased from trappers.
....Radio-collared individuals were located and, when possible, visually observed from fixed-wing aircraft....Monitoring intensity varied from pack to pack, but locations were usually obtained at least twice each month during winter....
Ages of captured and released wolves were determined from tooth eruption and wear. Ages of wolves observed but not handled were estimated from relative size and other criteria....Age and sex compositions of some packs were not ascertained until the animals had been killed by hunters or trappers. Hunters and trappers were encouraged to provide carcasses of wolves taken in GMU 13 by an offer of $10/carcass....
Areas occupied by individual wolf packs were computed for 4 time periods that coincided with changes in area boundaries, number of wolves, establishment and study of new packs, and differences in study objectives. The periods were (1) April 1975 through June 1976, a period that included the loss of several packs killed as part of an experimental wolf reduction program, (2) July 1976 through June 1978, while wolf reduction efforts continued, (3) July 1978 through June 1980, when substantial shifts in pack areas and boundaries occurred following termination of wolf reduction efforts, and (4) July 1980 through April 1982, a period followed by high rates of harvest by the public....
Spring and autumn wolf densities in GMU 13 were estimated by extrapolating known numbers of wolves within radio-marked packs to other portions of GMU 13 with similar habitat where observations were more limited. Active wolf dens...were inspected on the ground after the wolves vacated. The vicinity of each den was searched, scat collected, and food remains identified....
During wolf radio-tracking flights, prey carcasses associated with wolves were assumed to have been killed by wolves. If other predators such as bears were observed with the wolves and the species that made the kill could not be determined...wolf predation was not assumed to be the cause of death. Cause of death was classified as winter-kill (starvation) when there was no evidence of predation or accidental death. These moose were usually lying on their sternums with numerous pellet groups nearby, suggesting that such animals had been there several days prior to death. Sex of adult moose and caribou killed by wolves...often was judged from fixed-wing aircraft by presence and shape of antlers. Wolf-killed prey were classified as calves, yearlings, or adults using combinations of body size, pelage, antler morphology, and presence of vulval patch....
....During winters 1978-79, 1979-80, and 1980-81 and summers 1977, 1978, and 1981, an attempt was made to locate and examine all kills made by selected radio-collared wolf packs during a 2-3 month period....In an effort to maintain several packs at stable numbers for predation rate studies, portions of GMU 13 were periodically closed to hunting and trapping of wolves.
An attempt was made to kill all wolves inhabiting the SRSA during 1976 through 1978 by shooting from either helicopter or fixed-wing aircraft. Our objective was to determine the effect of wolf removal on moose survival. Carcasses of wolves removed from the experimental area were necropsied, aged, and assessed for nutritional condition and reproductive status. Repopulation of the removal area was documented by radio marking and monitoring activities of colonizing wolves, and by periodic wolf censuses....
Moose Studies
During 1976 through 1984, radiotelemetry was used to study moose movements and population dynamics in several parts of GMU 13....Moose home ranges and survival rates were determined using methods described earlier for wolves.
Sex and age composition were surveyed each autumn beginning in the early 1950's within specific areas. These low-intensity surveys were the basis for minimum estimates of moose density in count areas....
....CAPTURE AND TELEMTRY RESULTS
One hundred fifty-one wolves were captured and radio-collared in GMU 13 during April 1975 through April 1982....Four wolves died during capture or handling: 1 drowned, 1 was fatally injured by the dart, and 2 died from drugs.
Sex and age composition of the captured wolves was 45 adult males, 29 adult females, 10 yearling males, 8 yearling females, 28 male pups, and 30 female pups. One wolf was not sexed or aged....In an effort to select a pack member that would not disperse, a deliberate effort was made to capture a large male first....
During April 1975 through April 1982, 151 radio-collared members of 30 individual packs (a pack is defined as [two or more] wolves) were individually located from a fixed-wing aircraft on a combined total of 4,810 occasions. These flights resulted in 8,970 wolf sightings (collared + uncollared) and a total of 2,274 pack days (a pack day is defined as 1 day on which [one or more] pack members were located on [one or more] occasions)....The average number of locations per pack was 78 (range 5-240)....
BEHAVIOR
Pack Areas
Areas traversed by wolf packs varied greatly in size, ranging from 60 km2 for a pair of wolves to 3,077 km2 for a pack of 3 nondenning males. Total minimum area traversed by individual wolf packs was correlated with both number of days packs were monitored and autumn pack size. Areas were stratified into 2 groups based on number of days monitored. Size of areas for packs located >60 pack days were not correlated with either number of days monitored or autumn pack size, suggesting that the size of these areas was fully defined. These latter areas are referred to as territories. Eleven territories were considered defined, ranging in size from 943 to 2,541 km2. These territories were larger than 15 pack areas traversed by packs located <60 days....Territory sizes (located [60 pack days or more]) reported in this study were relatively large compared with those reported elsewhere in North America....
Territory size was also influenced by prey availability. Deep Lake, Ewan Lake, and Tolsona Ridge Packs had the largest territories of the denning packs, averaging 2,308 km2. These packs were responsible for 51% of the caribou kills observed during 1975 through 1980. Moose densities in these packs' territories were relatively low (0.09 moose/kmw), whereas caribou numbers were variable each year depending on the wintering location of the Nelchina caribou herd. These territories were characterized by flat terrain and a relatively homogenous spruce-bog forest that is considered poor moose habitat. We suspect that relatively large territories were the result of low densities of moose, the preferred ungulate prey during this study.
Distribution and Stability. -- Areas occupied by wolves were distributed continuously throughout GMU 13 in areas [1,220 m or less] in elevation. Although several pack areas contained habitats >1,220 m elevation, wolves were rarely observed there even during summer. It appears that the absence of moose and caribou during winter, and vulnerability to aircraft-assisted hunting, renders such areas in GMU 13 uninhabitable to wolves....Previous wolf location and distribution data were useful in predicting where colonizing pairs or packs might be found that had not yet been detected. For example, the Susitna Lake, Tolsona Ridge, Fish Lake, and Jay Creek Packs were initially located by searching likely but apparently unoccupied areas between known pack locations.
The general distribution of wolf packs was similar before and after wolf reduction. During winter 1975-76, most members of 5 packs were killed in the SRSA. Over the next 2-year period, the area was colonized by singles and pairs, most of which were also killed. No wolves were killed by ADF&G after 1978, and singles and pairs repopulated the area. During 1980 through spring 1982, recolonization continued through immigration and reproduction....
Distribution and territory boundaries of wolf packs seem to be strongly affected by interaction with other packs and by distribution and density of prey. For example, the Mendeltna Creek Pack was either eliminated or greatly reduced by aircraft-assisted hunting early in 1978. During the subsequent 2 years, members of the adjacent Tolsona Ridge Pack, which may have included 1 or 2 descendants of the Mendeltna Creek Pack, expanded their range westward into the old Mendeltna Creek Pack territory. Ultimately, the Tolsona Ridge Pack abandoned the southeast portion of its original territory, and by 1982 this pack occupied the previous Mendeltna Creek Pack territory. The Tolsona Ridge wolves had moved to an area of higher moose density (0.08 vs. 0.04 moose/km2)....During 1982-83, a pair of wolves colonized the original Tolsona Ridge Pack territory.
The sex-age composition of remaining pack members following heavy harvests also influenced whether an area continued to be occupied. If a single yearling or adult female survived, she generally remained in the same, but smaller area, attracted a new mate, and attempted to reproduce. Single males, however, usually dispersed and attempted to colonize vacant habitat or joined existing packs. If remaining pack members were only pups, they made extensive forays away from their original territory and eventually dispersed as yearlings. Other wolf sex and age combinations normally remained in the area.
Overlap. -- Even though obvious dispersals and forays were excluded from area analyses, considerable overlap of pack areas and territories occurred. Many overlaps resulted from the method used to depict territory and pack areas. Overlap did occur along territory or pack edges, but that overlap was usually separated by season of use. Also, many of the overlaps occurred along peripheral areas that were frequented less than core areas.
Annual variations in pack area or territory boundaries also were evident. Wolf packs appeared to change their use of peripheral areas, whereas the use of core areas was similar each year....Haber (1977) in nearby Mt. McKinley National Park (now Denali National Park and Preserve) believed that territory boundaries of unexploited packs remained stable...over relatively long periods. In northwestern Minnesota, Fritts and Mech (1981) recorded shifting boundaries resulting from establishment and abandonment of territories in an increasing wolf population. Territorial boundaries could be relatively stable from year to year in unhunted, saturated populations where prey abundance remains relatively stable. However, in heavily exploited populations like those in GMU 13 or increasing populations as in northwestern Minnesota, shifts in territory boundaries are probably common, due to adjustments with other packs, changes in pack members, and changes in prey density.
In other parts of North America, territories are maintained through howling, scent marking, aggression, and avoidance. Territory sizes reported in those studies...were considerably smaller than those reported here, and rates of harvest by humans was much lower. Regardless of territory size, those same mechanisms undoubtedly help to maintain territories in GMU 13.
Pack Areas in Relation to Movements of Ungulate Prey. -- During October 1976 through April 1982, movements, productivity, and mortality of moose were investigated in the areas where wolves were studied. Over 3,500 aerial relocations were obtained on 207 radio-collared moose ranging in age from 6 months to 19 years.
Moose typically migrated from summer range to winter range in November. In some years these migrations occurred as early as October or as late as January and appeared to be initiated by the first heavy snowfall. Not all moose migrated each year. During winters of low snowfall, many remained on summer range....Once spring migration began, movement to summer range usually took 4-6 weeks. Distances between summer and winter home ranges of migratory moose ranged from 16 to 93 km.
Seasonal and total home range sizes of GMU 13 moose were considerably larger than those reported elsewhere in North America. Total area occupied by individual radio-collared adult moose averaged 224 km2, ranging from 4 to 2,911 km2....
In addition to resident moose, migratory moose traveled through wolf pack territories during both spring and autumn. Wolf packs did not follow these migratory moose beyond their established pack territories. Thus, there were large seasonal variations in moose densities within some pack areas. However, all wolf pack areas were occupied year-round by some nonmigratory moose.
The seasonal and elevational distribution of wolves closely followed those of their principal prey, i.e., moose....Lowest mean elevation for both wolves and moose occurred in February, with both species inhabiting progressively higher elevations through October. Although wolves did not follow the movements of migratory moose, they apparently responded to movement of both resident and migratory moose within their territory.
Availability of caribou to wolves also fluctuated seasonally during the study....Calving occurred in late May, primarily from Kosina Creek to the Oshetna River, after which most females inhabited the northern and eastern slopes of the Talkeetna Mountains between 1,000 and 1,800 m altitude. Because most of the herd was concentrated on the calving grounds, the bulk of the herd was unavailable to most wolf packs, which were widely distributed throughout the study area....
In areas of North America where migratory caribou compose most of the wolf diet, wolves are known to follow caribou as they migrate to both summer and winter range....GMU 13 wolves did not follow caribou during this study, probably because moose were sufficiently abundant in most areas to provide the bulk of their prey. Also, during the early years of this study, caribou numbers were relatively low. Because of the migratory habits of the area's caribou, wolves would have [had] to become less territorial if they were to depend heavily upon caribou as prey.
Dispersal
Excluding 4 wolves killed during tagging and 12 killed during experimental removal within GMU 13, 38 (28%) of 135 were known to have dispersed from the pack area in which they were originally captured. More male (28) than female (10) wolves dispersed. Average estimated age of dispersing males (30 months) vs. females (33 months) was not significantly different. Forty percent of female and 50% of male dispersers were [24 months of age or younger].
Wolves dispersed from their territories during all months of the year, but most (57%) dispersed during April through June and October through November. Straight-line distances moved by dispersing wolves ranged from 23 to 732 km. The longest movement was made by 2 radio-collared males from the St. Anne Lake Pack between April 1978 and March 1979. Although this dispersal distance was less than the distance reported for a single wolf that moved 917 km, from Minnesota to Saskatchewan (Fritts 1983), it represents the longest dispersal distance reported for a group of wolves. Excluding this record movement, females dispersed farther than males.
Twenty-three (61%) of the 38 dispersing wolves were accompanied by 1 or more wolves following dispersal. At least 8 appeared to be pair-bonded in formerly vacant areas. Another 15 were accompanied by 2-6 wolves when last observed, and at least 8 of these were apparently accepted into existing packs. There were several instances when wolves left a pack, apparently dispersing, but returned. Forays of this type have been reported elsewhere.
Temporary associations among members from different packs were observed on 20 occasions. Such associations often preceded permanent dispersal from the original territory. Temporary associations sometimes resulted in wolf mortality, particularly when the associations included > 2 members from 1 or both packs. At least 7 wolves (5 males, 2 females) were killed by other wolves. Both females that were killed were nondispersing members of small packs {[4 or less]} within their own territories when the conflicts occurred. Possibly the residing pack was unable to defend against intruding wolves because of its small size. In comparison, all 5 male deaths occurred during trespass into neighboring territories of packs comprised of at least 7 or more members. Two of these deaths occurred at moose kills within an overlap area between 2 packs, suggesting competition for food may have been a factor. Although the reasons for friendly meetings and temporary acceptance or rejection of foreign wolves are not known, fatalities more often occurred when pack sizes were relatively large ([seven or more wolves]). Large packs would probably consist of an alpha pair and possibly several other adults that could be less tolerant of foreign individuals than smaller packs that may contain no adults.
....The average size of packs from which wolves emigrated was larger than those to which they immigrated. These observations suggest that pack size influenced dispersal and acceptance of dispersing wolves.
....Fritts and Mech (1981) suggested that acceptance of lone wolves by established packs helped maintain pack structure and resulted in a stable social system within an increasing wolf population. This may be important for wolves in GMU 13 where packs consist of <3 adults and 5-6 pups. Exploitation of these packs during and following the March breeding season could result in the loss of 1 or 2 adult members. This loss of adults could result in a reproductive failure that year and the following year, even though exploitation levels were below maximum based upon pack size alone. Such packs probably would have an unstable or less rigid social system than found in an unexploited populating, creating a social environment favoring acceptance of foreign individuals. Rapid acceptance of dispersers in these situations would be advantageous and could minimize inbreeding.
Activities of Wolves at Den and Rendezvous Sites
....Average distance among 38 neighboring dens was 45 km, which is similar to the 40 km reported in north-central Alaska (Stephenson and Johnson, 1973)....
....Pups were first observed from fixed-wing aircraft at dens or rendezvous sites from 1 June to 15 September....Mech (1970) indicated that wolf pups were commonly observed outside dens at 3 weeks of age. Clark (1971) observed 10-day-old pups outside dens on Baffin Island....
Aerial observations of pup emergence and wolf-den usage patterns suggested that parturition occurred from mid- to late May in south-central Alaska. Ground observations at 2 den sites suggested that parturition occurred by 1 and 11 May in 1980 and 1981, respectively....
Natal dens usually were abandoned during July, with dates ranging from 4 June to 1 August. The earliest movements resulted from human disturbance. No pup mortality resulted from human disturbance. Distance between natal den and first rendezvous site ranged from a few hundred m to 14 km. In at least 2 instances packs moved to a second rendezvous site in August or early September. These latter sites were occupied for periods of several days. In general, pups began traveling regularly with adults between late August and mid-September.
....Away from the den site, wolves were more solitary during summer than during fall and winter; 55% of 143 and 6% of 142 relocations of radio-collared wolves were observed alone during summer and the remainder of the year, respectively. Even though wolves were more solitary away from the den site during summer than winter, the presence of adult males (presumed alpha males) was a contributing factor to the type and age of prey killed during summer. Adult males were involved in 92% of the fresh kills involving calf moose or larger prey. Although yearlings hunted alone, they were largely unsuccessful and scavenged old kills or killed small prey. This suggests that in studies of summer predation, all adult members of each study pack should be radio-collared to aid in locating all ungulate kills.
POPULATION DYNAMICS
Wolf Densities
Historical. -- The status of wolves in GMU 13 prior to 1948 is largely unknown. Reports of early explorers indicate that wolves and their ungulate prey were scarce in the late 1800's. By the 1930's, wolves were more common according to early residents....
....Rausch (1967, 1969) summarized estimates prior to 1969 and concluded that extensive predator control, begun in 1948 by the U.S. Fish and Wildlife Service, reduced wolf numbers to a low of about 12 in 1953. After wolf control ceased, wolves increased steadily to a peak of 350-450 by 1965. By 1967, wolf numbers had declined to a maximum of 300 or less, apparently due to hunting. The next population estimate was made by C. McIlroy, who estimated that 207 wolves were present in March 1974....
Present Study (1975-82) -- ....Wolf densities in GMU 13 (excluding the SRSA and areas > 1,220 m elevation) varied from 10.3 wolves/1,000 km2 in autumn 1975 to 2.6/1,000 km2 in spring 1982. Lower spring densities reflected losses due to trapping, hunting, dispersal, and natural mortality, whereas autumn densities reflected pup production and immigration. From spring 1975 to spring 1982, wolf densities in radio-marked packs declined by 58%. Autumn densities exhibited a similar trend, declining by 54%. From autumn to spring each year during 1975 through 1982, wolf numbers declined an average of 47% (range 31-60%). Annual increases from late spring (following hunting but prior to denning) to autumn (prior to hunting and when best estimates of pup production are available) averaged 72% and ranged from 11% to 106%.
Extrapolation of wolf density estimates to available wolf habitat...in GMU 13 (including the SRSA) yielded total population estimates ranging from 426 wolves in autumn 1975 to 109 wolves in spring 1982. Wolf population levels were highest in 1975-76 and declined each year thereafter due to aircraft-assisted ground shooting and Department wolf control....
Productivity and Survival
Litter sizes as indicated by counts of placental scars in 16 adults [three years old or older] averaged 6.1....
Minimum estimates of mean litter size also were available from observations of selected packs in early summer. Pack litter sizes ranged from 2 to 9 pups. Litters were larger in early June and July than in November suggesting that some pup mortality occurred during summer. Average pup survival rate during the first 6 months of life based upon average observed litter sizes could have been as high as 0.97. However, if an initial litter size of 6.5 was assumed, pup survival to 1 November in GMU 13 was estimated at 0.82....The lowest survival (1975-76) occurred during the highest spring densities; this suggests that recruitment to autumn populations may be partially density dependent. However, both moose and caribou numbers were lowest that year; therefore nutrition may have influenced wolf productivity or pup survival.
Wolf packs generally produce 1 litter of pups/year. During our study of 41 pack litters with well-documented histories, at least 3 packs (7%) produced 2 litters by different females at separate dens in 1 year....
Van Ballenberghe (1983) concluded that conditions necessary for multiple litter production in wolves (large territories with several potential den sites, adequate numbers of prey, and low human exploitation) occur rarely. Our data do not support his hypothesis. Territory sizes for 2 of 3 twin-litter packs...were average for GMU 13, whereas the third area...was relatively small (1,000 km2)....All pack areas in GMU 13 contain numerous old and potential wolf den sites. There were no significant differences in numbers of wolves per pack between those which produced 1 or 2 litters. Also, there was no relationship between numbers of adults per pack in spring and numbers of pups surviving to November each year. This latter finding was similar to that of Peterson et al. (1984) on the Kenai Peninsula, Alaska, but different from that of Harrington et al. (1983) in Minnesota....Reasons for the differences in correlation between the Alaska and Minnesota studies are not known. Although reductions in human-caused mortality may enhance conditions (more adult females per pack) for production of multiple litters at relatively low densities, the opposite appears true at higher wolf densities when prey may be relatively less abundant.
Harrington et al. (1982) concluded that in North America the frequency of multiple litters in the wild was greater than previously thought, ranging from 22% to 41%. In the present study, 7-10% of the packs produced multiple litters. Multiple litters would be advantageous in allowing a wolf population to quickly increase following high mortality, but such litters have not been documented in saturated or increasing unexploited populations. Therefore, multiple littering may represent a form of compensatory natality in heavily exploited wolf populations. Although the frequency of multiple littering is probably greater than previously thought, our data suggest that the estimate proposed by Harrington et al. (1982) may be too high.
....Annual survival of radio-collared adult ([2 years or older]) wolves from 1975-76 through 1981-82 averaged 0.59 (range 0.45-0.77). Adult survival was similar to that of yearlings (0.68) and both exceeded pup survival rates, which averaged 0.35 (range 0.13-0.60). The average annual mortality rate of all radio-collared wolves was 0.45....Comparison of observed finite rates of increase in other North American wolf populations...with mortality rates and finite rates of increase observed in this study...suggests that mortality was sufficiently high to cause the population to decline.
Mortality
....Human Caused Mortality. -- Trapping was the most common harvest method from 1972-73 through 1975-76, accounting for 59% of the total harvest. Beginning with the 1976-77 hunting season, ground shooting became more common, accounting for 55% of the total harvest from 1976 through 1978-79.
The 1977-78 wolf harvest was the highest recorded for GMU 13 since 1968 when the hunting and trapping season was opened. Only 6% of the 1977-78 harvest was due to ADF&G wolf reduction efforts in the SRSA. Most of the increase could be attributed to the increased efficiency of 2 or 3 aircraft-assisted hunters (Alaska regulations allow trappers to land and shoot wolves from the ground on the same day as airborne); during 1976 through 1980, their harvest ranged from 12% to 61% of the total GMU 13 annual harvest. Beginning in 1979-80, trapping again became the most common method of harvest. During 1976 through 1980, successful trappers took an average of 1.5 wolves/season, whereas aircraft-assisted hunters averaged 3.4 wolves/season.
Sex ratios of harvested wolves were not significantly different from 50:50 annually or in total. According to harvest records, pups composed 42-74% of the annual harvest, averaging 57% from 1975 through 1982....
....An overall mortality rate of 50% from all causes would allow wolf packs to remain stable. Excluding natural mortality, which accounted for 20% of the wolf mortality (10% of population, see next section), human exploitation in excess of 40% would cause a population decline. For individual packs the exact number would vary considerably depending on the sex and age composition of each pack and the timing of the harvest. Heavily exploited packs are composed of only a few adults and several pups; therefore any additional adult mortality could greatly affect reproductive success.
Keith (1983) concluded that wolves could sustain annual harvests of 23-38% and that stationary populations, whether exploited or not, were composed of 40% pups by autumn or winter. Increasing and decreasing exploited wolf populations, however, would have larger percentages of pups. Results reported here are in basic agreement with Keith's (1983) conclusions. The average maximum sustainable harvest indicated by GMU 13 data (40%) and those examined by Keith (1983) and Peterson et al. (1984) are somewhat lower than the 50% or more mortality suggested by Mech (1970) as required to reduce wolf densities....
Mortality of Radio-Collared Wolves. -- ....Harvest by humans (excluding experimental removal and collaring mortality)  was the largest cause of death among radio-collared wolves. Legal (46%) and illegal (34%) human harvest made up 80% of the total mortality. Remaining losses were attributable to intraspecific strife (11%) and such miscellaneous causes as drownings, starvation, vehicle collisions, and injuries by moose (9%). In spite of the comparatively high level of human-caused mortality, the natural mortality rate observed in this study was relatively high in relation to observed wolf population densities and to other North American studies....
Experimental Wolf Reduction and Repopulation. -- During January 1976 through July 1978, 60 wolves were removed from the SRSA by ADF&G personnel as part of a wolf control program. During this period, 9 additional wolves were reportedly removed from the area by public hunting and trapping....
Immediately prior to initiation of wolf removal in 1976, wolf densities in the SRSA (7,262 km2) ranged from 4 wolves/1,000 km2 in spring 1975 to 7 wolves/1,000 km2 in autumn 1976. During the experiment, spring densities (before whelping) were reduced by 42-58% from precontrol levels....While control was in effect, spring densities were an average of 57% lower than in the remainder of GMU 13....
....Prior to termination of wolf control, annual finite rates of increase (based on autumn population estimates) ranged from 0.57 to 0.81. Once the experiment terminated in July 1978, the wolf population quickly increased due to immigration and reproduction. The finite rate of increase from autumn 1978 to autumn 1979, after control efforts terminated, was 2.40, an apparent record rate of increase for a wolf population in North America. Thereafter, the finite rate of increase in 1979-80 and 1980-81 was 1.23 and 1.04, respectively. The area was open to public hunting and trapping throughout the period of study, and several wolves were killed. By 1981-82 the wolf population was limited by public hunting and trapping with 55% of the autumn 1981 population being harvested.
FEEDING ECOLOGY AND BEHAVIOR
Food Habits
Aerial Observations. -- During April 1975 through April 1982, radio-collared wolves were observed at 439 kills; 91% were ungulates. Smaller prey species (e.g., snowshoe hares) were undoubtedly underestimated during aerial observations.
Moose composed 70% of the observed kills. During May and June calves composed 30% of the classified moose kills. During these months, newborn calves made up 38-48% of the population. Thus wolves appeared to be taking calves less or in proportion to their occurrence in the moose population. This conclusion was supported by scat analyses.
Wolves continued to take calves in proportion to availability through October. During July-October (1975-82), 14% of the moose killed were calves, and they composed 12-20% of the moose counted in November surveys during 1975-82. Most neonatal mortality had occurred prior to July, so the reduced use of calves reflected, in part, a decrease in their availability.
During November-April, calf moose were selected disproportionately to their occurrence in the moose population. Calves made up 40% of the classified wolf kills during November through April but represented only 12-20% of the moose counted in November. This result suggests increased vulnerability of calves due to several winter-related conditions such as snow cover.
....Adult moose were the most common prey, accounting for 38% of the 439 observed kills....[A]dult moose apparently were preyed upon in approximate proportion to their occurrence in the population during May through October. Thereafter, they were killed less frequently with proportionately more calves being killed. Adults were not as vulnerable as calves during periods of snow cover.
Caribou were the second most important prey, composing 21% of all kills....
Dall sheep were not important prey during the study. No identifiable sheep kills were observed although sheep remains were occasionally found in summer wolf scats....[A]t least 9 pack areas contained habitat where sheep periodically or regularly occurred. For example, during 1976 through 1982, 130-209 sheep were counted annually within the Watana, Jay, and Coal Creek Pack areas, but no sheep kills by these packs were identified. In other sheep habitats, such as the Talkeetna Mountains, wolves were uncommon. Wolves that periodically attempted to colonize portions of these areas usually did so where small bands of caribou over-wintered....
In nearby Denali National Park and Preserve, Dall skeep were preyed upon by wolves, but to a much lesser degree than caribou and moose. Murie (1944) suggested that predation on sheep was probably proportionately greater when wolf populations were relatively large....
Scat Analyses. --