[}:{J£[ffi~£c~rn3&\@@@ Susitna Joint Venture Document Number
Please Return To DOCUMENT CONTROL
INFLUENCE OF SNOW ON BEHAVIOR OF MOOSE
_j .'r:·q-r, II' c
f:atura.liste- can .. 101: 417-436 (19741]
INFLUENCE OF SNOW ON BEHAVIOR OF MOOSE J. W. COADY Alaska Department of Fis/1 and Game, Fairbanks, Alaska, United States
Resume L'auteur passe en revue les <:1daptations morphologiques et ethologiques de l'orignal a Ia neige, les proprietes pertinentes de Ia neige a cet eg3rd (epais seur, densite et durete) et Jes met.hodes de mesure appropriees. L'epaisseur refere a Ia couche non-portante que !'animal doit ecarter en se depla<;ant; Ia densite, determinee en pesant un volume donne de neige, permet d'evaluer Ia friction exercee sur les pattes ou sur le corps: Ia durete, sans doute Ia propriete Ia plus importante, determine Ia force que !'animal doit appliquer pour se deplacer et Ia capacite portante de Ia neige. La durete peut etre evaluee au moyen d'un gabarit a ressort ou d'une sonde de type ccRammsonde». Chez. f'orignal, on peut considerer que Ia longueur des pattes et .,e mode de repartition du poids refiE~tent en quelque sorte un ajustement morphologique a Ia neige. Si Ia neige au sol depasse le tarse, les mouvements ne sont pas genes. Mais si Ia neige atteint les deux-tiers de Ia hauteur comprise entre le sabot et le poitrail, les mouvements sont partiellement restreints; its le sont presque com pletement, quand Ia neige atteint ou depasse Je poitrail. L'auteur interprete certaines differences dans Ia longueur des paties comrne une adap1ation aux differences regionales observees qua11t a l'epaisseur de Ia neige. La repartition, entre les quatre membres, du poids par unite de surface permet cf'evaluer Ia durete requise pour assurer le support de !'animal. Les mouvements graduels de !'habitat d'ete a !'habitat d'hiver prennent place entre octobre et mars, en reponse a diverses conditions de neige. Cependant on assiste !n§quemment a une migration acceleree vers les quartiers d'hiver, a Ia suite d'une accumulation hative de neige. En Alaska, Ia durete de Ia neige dans les quartiers d*ete peut egalament influencer les deplacements de l'orignal. Ces deplacements sont ordinairement rapides et coincident avec Ia fonte de Ia neige et Ia reapparition du soL Par contre, les deplacements durant l'hiver sont generalement limi~es, particulierement dans les periodes marquees par une forte accumulation de neige au sol.
' ,, . ~ La neige, sans etre Ia cau~· e des migrations, en influence Ia date et I' am ~·~a t, : f ' ~~: ~ plitude. Elle peut intervenir dans le oifan energetique, en augmentant le~ besoins metaboliques et en restreignant l'acces aux ressources alimentaires. Elle peut egale ment jouer un role majeur en suscitant des deplacements vers des quartiers d'hiver nage dans lesquels Ia qualite, Ia quantite et Ia disponibilite de Ia nourriture sont meilleures.
l~bstract
Properties of ~now influencing moose, methods of measuring those pro perties and morphological and behavioral adapta·:_,ns of moose to snow are re viewed. To moose, snow depth, density and hardness are probably the most important c:l1aracteristics. Depth indicates the thickness of the medium tnrough which an animal must move if not supported by the snow; density, measur-:-d by weighing a known vo'ume of snow, inhibits locomotion by increasing drag on legs or body; hardness is pE:rhaps thE' most important property of snow to moose . slnce it determines the force which must be exerted to move through the snow ' •j,, ..... : ...... and the capacity of snow to support the anlm~t. Hardness can be measured with .i '1 eitr.-.r a spring loaoed hardness gauge or with a Rammsonde ponetrometer• Q '
:·! ,: ·~, ·.t ~
I I' • 1)' 1 . ' 418 .LE NATURALISTE CANADIEN, VOL 101, 1974
Morphological adaptations of moose to snow may be related to leg length and to foot load. Snow depths below tarsus height provide little or no hindrance to moose: depths approaching two-thirds of the· ches"t height partially restrict movement. and depthS equal to or greater than the chest height severely restrict movement. Differences in leg length of moose between two areas in Alaska suggest a poss:ble adaptation of leg length to regional differences in snow depth. Foot loads reflect the weight per unit areas of all four feet. and theoretically indicate the hardness of snow requir~d to support a moose. Gradual movement of moose from summer to winter range generally occurs between Gctober and March in response to a wide range of snow conditions. However, abrupt migrations to winter range in response to eariy and deep snow frequently occur. Hardness of snow on summer rang~ {Rammsonde values) may be related to altitudinal movements of moose in Alasi
Introduction cal properties of a snow cover vary The circumboreal distribution of greatly, depending on conditions of de moose (Aices alces) throughout regions position and subsequent metamor . phism. Seasonal snow covers behave in ; ' characterized by a seasonal snow f?Over an extremely dynamic fashion, and the . dt illustrates the successful adaptation of only completely predictable phenome ' i the species to the nival environment .. non is change itself. Metamorphic pro . ··~·· Adaptation of moose to snow can be ex s ~ cesses which take place within a sea '. f'. plained, in part, on the basis of physical L sonal snow pack have been described ' ;'lt• and behavioral characteristics. Long legs, which increase tolerance to deep by numerous workers (Bader et a/., i :~ ;;;: ' snow, and movement to the most favo 1939; Formozov, 1946; Geld, 1958; Kin ~· ~: ~' : .. rable habitat when snow conditions gery, 1960; Benson, 1967, 1969; Trabant, make travel laborious and food difficult 1970). While wind action is a major factor affecting snow as it precipitates 1 to obtain are important factors facilita I ting winter survival. (Sommerfeld, 1969), diagenetic proces ses resulting from temperature, time In the following review, properties of and settling (Keeler, 1969) begin imme snow important to moose and methods diately after deposition. All .features of of measuring those properties, and phy a snow pack reflect post-depositional sical and behavioral adaptations of changes as much or more than they moose to snow will be discussed. Needs reflect the character of snow at the time for future research to better understand of deposition. High temperature relative and predict the influence of snow on to the melting point of snow and steep mcose will be indicated. temperature grad-ients are effective in promoting vapor transport, a major Properties of snow and methods of cause of metamorphism. Constant fluc measurement tuations iii air temperature and accumu The structural, physical, and rnechani- lation rate of snow cause continual . "
COADY: SNOW AND MOOSE BEHAVIOR 419
changes In both thermal and gravita are recorded for each of the major stra tional forces which are perceptible ta. Snow samples or measurements are within days or, sometimes, even hours. generally taken in a horizontal plane, although vertical as well as horizontal The properties of snow can be di hardness. measurements should be vided into fundamental and derived made (Pruitt, 1971 a). The International characteristics. A discussion of funda Workshop on Rangifer Winter Ecology mental properties, which include size, (Pruitt, 1971 a) has recommended mea shape, orientation, and packing of par suring the thinnest distinct strata which ticles, is beyond the scope ·of this re a given instrument size will allow. view. While fundamental properties of a snow cover bear a relationship to Depth of snow cover is probably the mec~anical properties and are impor most common and important measure tant for a complete understanding of ment obtained in studying snow ecology snow metamorphism, they are probably of moose. Snow depth data are easily of minor significance in ungulate snow obtained and provide a measure of the ecology (Pruitt, 1971 a). thickness of the medium through Derived properties of snow are those which an animal must move if not sup which depend on fundamental proper ported by the snow. Depth of snow or ties for their magnitude and rate of thickness of strata can be measured change (Keeler, 1969). Generally, they with any conveniently calibrated probe. are more easily measured than funda In addition, in interior Alaska, 5 em dia mental properties, and are useful indi meter stakes, clearly calibrated at 30.5 ces to the nature of snow. Although em (12 in) intervals, have been perma derived properties are numerous and nently located by the author in remote diverse, relatively few are generally or inaccessible araas to measure snow measured to characterize a snow cover. depth from fixed wing aircraft. Depth The most commonly measured ·'erived can consistently be estimated within 3 ' ,. , . properties used in biological work are to 5 em of the actual snow depth, J ,j ,:,; depth, temperature, density and hard- although this method has the disadvan . : · ness. tages of limiting measurements to the number of stakes at a study site, and ·· .; >' Methods of measuring derived pro- limiting study sites to relatively open ,. perties have been described by several ' i-' areas where the aircraft may be flown '"'-l authors (Klein et at., 1950; Benson, near ground level. Snow depths relative :i !.~: 1962; Keeler, 1969; Test Lab, 1970a, . :~1 to anatomical features of moose also ' "~ 'tl, 1970b; and others). ·Two similar sets of provide a useful and rea~onably accu instruments, a National Research Coun rate estimate of snow depth from the air. cil of Canada (NRC) kit and a USA Cold Regions Research and. Engineering Snow temperatures of each strata or
.~ ,, Laboratory {USACRREL) kit have been at intervals on the pit wall, beginning :. •, <~ ;+!' used to measure temperature, density, at ground level or in the subnivian space . : ', and hardness of a ~mow cover. However, and ending within 2 em of the surface, . · {:.. modified snow stuc'y kits have also been may be measured. The imponance of used (Richens and Madden, 1973). temperature to diagenetic processes in ·.(' •. These measurements are generally ob a snow cover was noted above. The In .;r ,:, tained in .. pit" studies, in which a trench s; • ternational VVorkshop on Rangifer Win i' ( : is dug in the snow to ground level. ter Ecology (Pruitt, 1971 a) has question Thickness, temperature, density, hard·· ed the significance of measuring snow n&ss, and usually crystal type and size temperature in Rangifer research, and . . l I
420 LE NATURALISTE CANADIEN. VOL 101, 1974
has recommended limiting measure rod with provision for mounting discs ments to ground level and air temper of different areas on one end and a atures. However, a relationship be calibrated gauge on the other end. The tween snow temperature and moose disc is rr6ssed against a snow surface behavior has been alluded to by Des and the maximum stress associated with Meules (1964). The ease and speed of the initial collapse of the snow struc temperature measurement in pit stu ture is· noted on the gauge. By using dies and its possible significance to moo a high and low range gauge and dif se behavior probably merit its conti ferent sized discs, hardnesses of 0 to .· nued measurem~nt. Measurements may 100,000 g/cm 2 can be measured. Useful be made with either an alcohol or bi modifications to the- instrument devel metallic thermometer. oped by 0. Eriksson in Sweden and • now used by Pruitt (pers. comm.) in Density, determined by weighing a clude a ratchet which retains the cali known volume of snow, is probably the brated gauge at the extended position most widely used index of snow type, reached at the instant of snow collapse, and under certain snow conditions it thereby providing more accurate measu may be correlated with hardness (Kee rements. ler and Weaks, 1967; Bilello eta!., 1970). Snow hardness may also be measured Increased . density presumably causes with a cone penetrometer, commonly increased drag on legs or body of un referred to as a Swiss Rammsonde. The gulates during movement, and thereby instrument and its use have been de inhibits locomotion. Several snow CLAt scribed by $everal workers (Bader et a/., ters for obtaining sno-vv density samples 1939; Benson, 1962, Keeler,1969; Test are available. The NRC kit employs two Lab~ 1970a, 1970b; and others). Basical 250 ml snow cutters, one for soft and ly, it consists of a hollow steel shaft one for hard snow, while the USACRREL with a 60" conical tip 4 em in diameter kit uses 500 ml sampling tubes. Swe and 3.5 em high. A solid rod mounted dish workers use a Swedish Army den on top of the steel shaft guides a ham sity "box" which reportedly gives re mer which is dropped from a measured liable results because of its large 1,000 height. The height of the drop, weight of em 3 volume (Pruitt, 1971 a}, In Interior the entire instrument, and depth of pe Alaska, a 650 em 3 plastic cylinder has netration may be related to the resis been used by the writer with satisfactory tance of snow to penetration ~"'Y the results. cone using the following formula: Snow hardness reflects the degree of R:::: Whn + w +a. bonding between crystals and in most X types of snow increases as density in where R = ram hardness number or resistance creases and/or snow temperature de to penetration, W = weight (kg) of the dropped creases (Gold, 1956). In addition to hammer, h = height (em) of dropped hammer, snow depth, hardness is probably the n = number of hammer blows, x "'"'" penetration rnost critical parameter of a snow cover (em) after n blows, and Q ·.:: weight (kg) of penetrometer. to ungulates since it reflects the force which must be exerted to move legs This equation ignores friction be Jr body through the snow, and the abi tween the cone and snow and elasticity lity of the snow to pnrtially or fully in the penetrcmeter. However, the error support the animal. NRC and USACR is small, especialiy for snow of relatively REL snow hardness gauges are similar, low hardness (Kealer, 1969; Benson, pers. and consist of a spring loaded push comrn.) ..
I ' j
COADY: St,~OW AN~' MOCSE BEHAVIOR 421
The fam hard11ess number, R, mdic··f'S the with moose, and some investigators in resistanre (! ,...... - ...... ,..~-""-"""~-¥...... ,,.~.-.,l,-....,"'~~ ...... :J<"-""'"'"''''""'"'"""""'·~-- .. ~·"'..... ¥ ....ni:~··~~~ .-~- j 422 LE NATURALISTE CANADIEN, VOL. 101, 19r4 : ' ,__.,. ~ med to compute a severity index. Picton the thoracic limb, with leg extended pin and Knight (1971) computed an index pendicular to the body, diagonaily to to winter weather conditions for big the sternum. The measurement may I game based on weather bureau mea slightly overestimate actual chest height surements. Basically, daily snow depths since the leg may be less extended on I ~ on ground are multiplied by the maxi a standing ihan a decumbent animal. { mum daily degrees below ooc, and the In some studies (Kelsall, 1 969) hoof load, fi daily points are summed to provide a or weig!1t on hooves alone, has been cumulative winter index. · . used to measure the bearing surface of the legs. However, Kelsall and Tel Coady (1973) has examined trends in I fer (1 971) and Telfer and Kelsall {1971) i ' snow severity for moose in Alaska by indicated that the entire foot from the plotting U.S. National Weather Service . tip of tile hoof to the dew claws sup records of snow depth on the ground J ports an ungulate in ~oft Si10W, and de versu~~ month throughout the winter scribed a procedure for measuring .)ot I and measuring the area under the curve areas and calculating foot loads. Foot with a planimeter (cf. Bishop and loads measured by the previous techni Rausch, 1 974). A similar procedure que are certainly minimal since the could be used to evaluate temperature I measurement does not allow for spread-· severity. While such weather records ing of hooves and angular placement of may not indicate actual conditions on legs in snow. However, the actual foot winter moose range, they probably re load is probably somewhat greater since flect relative differences between years only when standing is weight supported and long term trends. on all four feet. In addition, Kelsall (1969) thought that as much as three Adaptation of moose to snow fifths of the weight of a standing ungu Major physical characteristics of late is distributed on its forefeet. moose ir!fluencing their mobility in snow are chest height of the animal The significance of chest heights and and weight load on the feet.' When foot loads of moose reflecting adapta .. moose sink into snow to depths ap tion to snow were first studied in Rus '•, I ~ '; • proaching chest height and are forced sia and later, in North Ame, ica. Nasimo ' .. to "plow'' or bound through the snow, vitch (1 955) reported that chest height energy required for movement is great of adults averaged 105 em or greater. ly increased. However, snow depths be Kelsall (1 969) found that average chest low chest height may hinder movement height of male moose in eastern Canada increased from 81 .9 em for calves to by increasing resistance to movement ' of the legs. Weight load on the feet 98.2 em for yearlings, and 104.7 em for is a measure of the weight per unit area animals older than four years. Kelsall on the feet and reflects the extent to and Telfer (1 971) measured average which a moose may be supported by chest heights of 106 em for male moose a substrate. Under situations in which older than four years in western Canada. snow will partially or fully support a Similar chest heights of 84, 96, and moose, resistance to and energy re 104 em for 60 calves, yearlings, and quired for movement may be reduced. moose older tllan two years, respective ly, have been found by Coady (unpubL) Procedures tot r:neasuring chest in interiorAiaska. heights and foot loads have recently been described in detail. ~(elsall (1969) Nasimovitch (1 u55) noted that rnoose measured chest heights from the tip of on the Kola Peninsula in Russia were - COADY: SNOW AND MOOSE BEHAVIOR 423 .unaffected by snow depths of 40 to 50 of the different groups to move in deep em, while movement was definitely im snow. peded by depths of 60-70 em. At 60-70 Height of moose from intedor Alaska em calves frequently followed in the trail and the Kenai· Peninsula have been of adults.· Nasimovitch (1955) concluded compared to illustrate that leg length from the Russian Werature that snow in moose may be adaptive to snow con depths of 90 to 100 ern may be consi ditions. In interior Alaska snow depths · dered critical to moose, since at that of 70 em or more, persisting for several depth winter mortality substantially in- months in winter moose habitat are the . creased. Kelsall (1969) reported similar rule, and depths in exc8ss of 90 em are observations from eastern Canada no not unusual. On the Kenai Peninsula, ting that movement was unrestricted ho\Never, .depths in winter moose habi by depths of 44 em and severely restrict·· tat range near 40 em for short periods, ed by depths of 70-99 em. Snow depths and seldom reach 60 em. Since chest greater than 90-100 em were critical for heights of moose from the Kenai Pe moose unless of very short duration. ninsula were not available, a ratio of Ritcey (1967) and Prescott (1968, cited shoulder height to total length was used in Telfer, 1970) found that depths of to reflect relative differences in leg 60-70 em restricted mobility of moose in .·. length and presumably in chest height British Columbia and Nova Scotia, res of moose from the two area~>. The av pectively. In Ala5ka, substantial. winter erage ratio from 31 fully grown moose rtwrtalit~· has v ... curred in several areas from the interior was .68, while that of the state when snow depths exceed from 64 similar animals from the Kenai ed 90 em for several months (Coady, Peninsula was .59 (t = 6.13. P < 0.001 ). 1'973). Similar ratios of 18 calves 6 to 12 months of age from the Interior, and The above data suggest that snow 50 calves from the Kenai Peninsula depths up to 40 em, or depths approx yielded averages of .72 and .70, respec imately equal to the carpus or tarsus tively (t = 1.82, P<0.1). Absolute differ , height, cause little or no hindrance ences in average shoulder height of I 1, to movement. From 40 to 60-70 em, or moose from the Interior and from the ·}! depths approaching two-thirds of the Kenai Peninsula were 182 em and 172 ' chest height, movement is only slightly em for adults, and 148 em and 141 em , .; . restricted. At depths greater than 70 em for calves, respectively. Thus, both rela ·, ! movement is definitely impeded, while tive and absolute height of moose is ', ~; h depths greater than 90 em, or approx- lower, particularly among adults, from '1 :, imately equal to or slightly less than the Kenai Peninsula than from interior ,.;,.,· ·... · the chest height of standing moose, Alaska. •: ; :: t movement is greatly restricted to the ex- :;; tent that adequate food intake may Va.riations in body size of animals :-;!: become impossib~e. Calves, because of may be due to genetic or nutritional ·.; ~ their shorter legs, may be restricted by differences. Since data from the Kenai : ' ~; 1 snow depths somewhat less than those Peninsula were obtained from moose on : . ' affecting adults, while large males may relatively poor range (LeResche and .:·:; be least affected by deep snow. Diffe Davis, 1971) and data from the Interior .. : rential movement of sex and age groups from moose on relatively good range ·;< :during winter reported by several work (Coady, 1973), possible nutritional d if ·:r: ers {LeResche, 1974; Pulliainen, 1974) ferences in skeletal dimensions of ani ; c: may reflect, in part, the relative ability mals from the two areas is possible. 424 LE NATURALISTE CANADIEN, VOL 101, 1974 vVhiie differences in skeletal growth re and by using the area of the entire lated to nutrition do occur (Klein, 1964), foot. there is little reason to expect that poor range on the Kenai Peninsuia is re Foot loads of moose in interior Alaska sponsible for preferential growth of body (Coady, unpubl.) are not uniform but length over foreleg length. Thus, while vary with age of animal and with sea nutritional deficiencies may account for soh. Average toot load decreased from smaller absolute shoulder height of st 9b3 g/tcm2 + 94 for ~ ad u1t fcows in doc-t If Kenai Peninsula moose, it is likely o er o 432 g 1em + 63 or 1·9 a u1 not responsiqle for reduced height of cows between April and June. Thus, as animals relative to length. winter progresses and snow depth and hardness increase, foot loads of adults decrease due to seasonal loss of bo While long legs per se are not neces dy weight. Foot loads of 9 calves be sarily an adaptation to deep snow (e.g., tween April and June averaged 317 height facilitates the browsing habit), g/cm2 +28, over 100 g/cm2 _1ess than the selective advantage to moose of adults during the same season. Thus, increased leg length in regions of deep the shorter legs of calves may in part snow is obvious. Therefore, observed be compensated for by lower foot loads. differences in relative height of moose No seasonal change in hoof size was between the Kenai Peninsula and the noted for Alaskan moose, as Pruitt Interior in Alaska are probably of gene (1959) reported for caribou. tic odgin and may be related to differ ences in snow conditions between the Snow density has been related to two areas. Nasimovitch {1955) noted track depth of moose. Kelsall (1971) that reindeer from the taiga zone where concluded from his observations that deep, soft snow is common have longer snow densities of 0.10 to 0.19 g/cm 3 legs than those animal's from tundra do not support moose, densities of 0.20 areas. Nevertheless, additional studies to 0.29 g/cm 3 provide some support, would be useful. and densities of 0.30 to 0.39 g/cm 3 limit foot penetration to approximately 50 Nasimovitch (1955} reported that the percent of the snow depth. Nasimovitch average "track load" of "several moose" (1955) reported that snow densities of in Russia was 420 g/cm 2 • This value 0.20 to 0.22 g/cm3 provide little support presumably represents the total foot to 3 running moose, while densities of load, and not just hoof load, of the ani 0.24 to 0.26 g/cm 3 limit foot penetra mal. Kelsall and Telfer (1971) measured tion to two-thirds of the ·total snow average foot loads of approximately depth. However, under thase conditions, 710 g/cm 2 for male moose four years ·moose experience difficulty lifting legs and older during December in western from holes in the snow. Canada. Higher average hoof loads of 789 to 922 g/cm 2 were f.ound by Kel Kelsall (1969) and Kelsall and Prescott sai, (1969) for male moose of similar (1971) were unsatisfied with attempts age from two areas of eastern Canada. to relate ungulate sup[)ort to snow hard The higher hoof loads from eastern ness. Theoretically, a standing moose Canada were measured before the rut should be supported by a vertical snow using the area of the hoof alone. How hardness equal to or greater than its ever, the lower foot loads from western foot load (Kelsall, t 969; Kelsall and · Canada were determined following the Prescott, 1971). However, the support rut, presumably after some weigtlt loss, capacity of snow is extremely variable, COADY: SNOW AND MOOSE BEHAVIOR 425 depending on presence or -absence of TABLE I surface crusts and the llardness of Integrated Rammsonde resistance of snow (Ri) · underlying snow layers. Both white-tai- to depth penetrated by adult moose in interior .led deer and moose frequently broke Alaska through crusts that should easily have supported the animal. Moose, with max Snow depth Track depth XRt RangeR; (em) (em) (kg-em) (kg-em) imum track loads of 1,000 g/cm 2, broke thro.ugh cr~sts of 8,000. g/cm2 50 . 22 188 167-211 at the surface, 10,000 g/cm2 at 15 em, 85 19 340 140-540 and 90,000 g/cm2 at 34 em. Moose al 80 12 409 326-525 90 42 570 535-632 so broke through crusts of 10,000, 20, tOO, 30,000, 40,000 and 25,000 g/cm2 t<.' ground level at a depth of 73 em. dom extensive or persistent enough Or' other occasions moose 'vYem sup to significantly benefit moose. Even ported by surface crusts of 2,000 to on the tundra of Alaska where snow 30,000 g/cm2• Peek (1971 b) noted density and hardness are very great, the that surface crusts of 7,500 g/cm2 snow cover in winter riparian habitat supported moose in Minnesota. In pr:>vides little or no support to moose. interior Alaska such extremely hard In other areas supporting crusts are ·crusts are unusual, although I obser - usually extremely localized, and are ed that an adult moose walking on a apparently rarely consistent enough to trail penetrated 20 em in 30 em deep facilitate travel. Murie (1944), Nasimo snow when the hardness was 2,000 to vitch (1955), Kelsall and Prescott (1971), 4,000 g/cm2• On another occasion both Peek (1971 a), and others indicated that a cow and calf walking on a trail pen snow conditions which only partially etrated 39 em in 90 em deep snow support moose may make movement when the hardness was, 1 ,000-2,000 more difficult and hazardous because g/cm2• of the resistance to movement of legs caused by the dense snow and/or Preliminary attempts to use the Ram the danger of abrasion from hard msonde penetrometer to quantify the crusts. Snow conditions in which depth support capacity of snow for moose of penetration is variable may require an have been attempted by Coady (un animal to expend more energy recov . publ.). The integrated ram hardness 1 ~; ~ ering from breaking through crusts and ~ > ' {Ri) was calculated for the total ram climbing onto crusts than would be re : I l hardness of the snow to foot penetra- quired to move through deep snow :~·! . ,, . tion depth (Tahle 1). Average R i ran- offering no support. However, dense, ged from 188 to 570 kg/em for pen hard snow offering uniform support, etration depths of 22 and 42 em, such as ski or snow machine trails, respectively. Although .t is not imme may be extensively used. diately evident from the limited data · ,c above, further study may reveal a pre Comparison of foot loads of moose ,: dictable relationship between Ri and from different regions may indicate depth of foot penetration in or resis adaptation to varying snow conditions. . tance to movement through snow. Kelsall and Telfer (1971) in western " ( Canada measured average foot loads of 2 r 1 Throughout most of the circum bo 710 g/cm for adult moose during se real range of moose and within favor December, whereas I reported foot ;.~: ( able habitat, snow conditions that loads of 593 for adults during Octo provide support are apparently sel- ber (see above). Normal snow condi" •' v •'· 0 ' - :) \-. '• I 426 LE NATURALISTE CANADIEN, VOL 101,1974 tions at the collection site in western tainous regions. Gradual movements Canada were not given, but presuma generally occur between October and bly snow depth and hardness are at January, and may coincide either with least as great as those in Interior Alas the first lasting snow cover, or with ka. Although I followed procedures snow depths of 25 to 45 em. How described by Kelsall and Telfer (1971) ever, some animals migrate before possible differences in measuring foot snowfall while others r-emain in summer area must be considered. habitat until snow depth reaehes 60 to 70 em. Movement of moose in response to Pulliainen (1974) reviewed the iiter • snow ature describing relationships between The influence of snow on seasonal snow and moose migrations in Scan movements of moose has been reported dinavia. Movements in most areas are by several 'harkers, although quanti closely correlated with prevailing snow tative observations are relatively conditions. Gradual movements from limited. The greatest effort to docu high to low elevations usually begin in ment and review relationships between Novemb1~r or December, although they snow and moose migrations has occur may be delayed or rna~' not occur dur red in the USSR (Formozov, 1946; Nasi ing years of little snow. Some· animals, movitch, 1955; Knorre, 1959, 1961; particularly cows with calves, begin Egorov, 1965; Heptner and Nasimovitch, migrating at first snowfall, while others, 1967 in van Ballenberghe and Peek, particularly bulls and cows without 1971). However, significant contribu calves, remain at high elevations until tions have also been made in Europe snow depth reaches 60 to 70 em. How (Pulliainen, 1974). and in North America ever, formation of icy crusts may ini .- tiate downward migration of almost ·"';,~' {Edwards and Ritcey, 1956; Ritcey, 1967; ·, . ~ ... ;;; " '-', Knowlton, 1960; Houston, 1968; Kelsall all animals (Krafft, 1964). Return to sum and Prescott, 1971). mer range is usually abrupt, and occurs during May after snow melt has exposed Nasimovitch (1955) drew several patches of ground. conclusions from the copious Russian literature regarding the influence of fn North Arr.erica several early work snow on moose migrations in both ers, including Murie (1934) on Isle ;I. mountains and flatlands of the USSR. Royale, Hosley (1949) in Maine, Bau '"": i"• Basically, in regions where maximum man (1941, cited in Hosley, 1949) in ~ ~ ~ '! snow depth averages less than 50 em Yellowstone Park, Murie (1944) in Alas .;,.,, . ··:·,;.;.;.,.~;·-.···· and deep snows are· of short duration, ka, and Hatter (1946, cited in Hosley1 ; ..,, extensive seasonal migrations are 1949) have commented on snow depth ... uncommon, although local movements and moose movements. However, Ed may occur. However, 1n areas where wards and Rltcey (1956) in British Co maximum snow depths in excess of lumbia were the first to present detail 70 em persist for long perjods, season ed observations on the relationships al movements occur from areas of between moose migrations and snow deep to less deep snow. The longest conditions. They found that a gradual migrations, ranging from 150 to 300 altitudinal movement from 1525-2135 km, occur among animals living m to 760-1220 m during fall and winter on flat terrain, although migrations of ~oincided with a gradual increase in 100 to 150 krn are common across snow depth on summer range. A rapid divides or to lower elevations in moun~ return to hi~Jhcr elevations during 1.::. l -- 1 1 I h.. ' ·spring coincided with a rapid snow initiated movements to lower eleva melt Separation of winter and tions. Movements were gradual, and summer range was not complete, al frequently lasted from December to though most animals departed higher March. Harry {1957) reported that in elevations by the time snow depth VVyoming, increasing snow depth at ·reached 75 em. Upward movement to high elevations resulted in a gradual summer range was initiated when melt downward movement and concentra ing reduced snow deptl1s to 30 to 45 tion of moose in mountain valleys by em. Cold temperatures appeared to alter December. Snow conditions associated the e-ffect of snow by speeding· move with these studies in Montana and Wyo ment downward in the fall and retarding ming were not reported. However, movement upward in the spring. Houston (1968) in Jackson Hole, Wyo ming, found that downward move Ritcey (1967), also in British. Colum ments from 2190 m to winter range at bia, noted that deep snows at high lower elevations began in late December elevations were responsible for the fall in response to snow depth of about 80 ahd winter movement to elevations em on the summer range. Movement to beiow i 050 m. Arrival on winter range spring and summer range began in generally began in November when late March in response to a snow crust snow depth was less than 15 em and formation capable of supporting moose continued throughout the winter. and in response to disappearance Departur~ from winter range began in of snow from nouth and east facing late February or early March, while slopes. Moose densities on winter range snow depth was as great· as 125 em were 10 moose/km 2 in Montana but declining. (Stevens, 1967:7 cited after Stevens, 1970) and 19 moose/km2 in Jack Kelsall and Prescott (1971) and Tel son Hole (Houston, i 968). fer (1967a, b) in the Canadian Mariti me Provinces,. studied winter segrega Seasonal movements of moose in tion of white-tailed deer and moose lh response to snow in Alaska have been relation to moose sickness induced by noted by some workers. Rausch (1958) . ~. the meningeal worm, Parelaphostrongy- reported that an early snowfall in lus tenuis. Although segregation was 1956 at high elevations in southcentral : not complete, deer generally wintered Ala~ka caused an early migration in 1 at elevations belovv 200 m, while moose November to lowland areas; he con ~. remained at elevations above. 200 m. cluded from his extensive obser . : Snow depths of 85-90 em above 200 m vations that snow influences but dces ;:! did not initiate downward migration of not cause seasonal movements of ~ :' moose even though snow depths were moose. :~ more favorable and browse more abun . ;: dant at lower elevations. However, Fall migrations in hills and moutains .:• relatively high snow density and hard of interior Alaska generally occur as a :. crusts due to winter thaws and rains gradual downward movement be ·: may have provided some support to tween December and March (LeRes 'moose, thereby reducing the effective che, 1974). The extent, time, and com · sngw depth (Telfer, pers. comm.). position of the migrating animals appear to be closely related to snow condi I ·:( Knowlton (1960) and Stevens (i 970) tions. In late November and early :reported that in Montana, deep snow December, 1970, snow depth of 90 em ;on summer range above 1830-2135 m {55 em above average) at elevations of I I I \ c:-1 I',, -rte-. . I -~-· -~ .. --·------L''~.~-~~~~--~~l· I , \%-.-;...,~.,..;,,,,"..,_....,,_.;.,,,._,. _ ... li.•·~"•"'l:: . .;:~...... _- ...l...... ,.v_,~ .. ~~c .... 4v..~ ...... ~~~~~4.~~~-S4 .. ~.1'.~~·•.. ~~.:L-. / .l··~ l. 428 LE NATUAALISTE CANADIEN, VOL 101, 1974 ...... _...,., ...,._., I 600 to 915 m in hills near Fairbanks ed Rammson(de) r:sisthancbe on the ~ot.tal -l apparently caused an abrupt down snow cover Ri m s ru -- commun1 1e1 ward migration of moose to elevations on the two sites are also noted. ( below 300 m. Over 1,200 animals were seen on, or moving toward low elevation A decrease in the number of moose riparian habitat during 23 hours of observed at high elevations and an Alaska Department of Fish and Game increase in the number of fresh tracks I aerial surveys in early December. Al at low elevations occurred during !ate j' most no animals were found on the December and early January {Fig. 1}, usual upland fall and early winter The decrease in fresh tracks in the range. Snow depths of 110 to 120 em valley during mid-January may have I persisted until early April and moose been related to reduced activity during!l remained along rivers until mid-March extremely cold temperatures (-40° to when they apparently dispersed into -50° C) during that period, while the adjacent timbered areas. Similar ob decrease in tracks after late February servations during 1970 were reported may have resulted from a dispersal of j by Bishop (1971) for western Interior animals away from the riparian habitat ! Alaska, where snow depth of 60 m dur where the transect was located. After ing the end of November apparently January the number of moose in the precipitated an early movement of hills remained low throughout the win- moose from upland areas to lower ter. elevation riparian habitat, where they remained until late March. I have stu died the response of moose to winter Snow depth at high elevations grad weather factors in Interior Alaska since ually increased to about 80 em at the 1971. For example, relationships be time of movement in late December. tween moose movement and snow While lowland snow depths through~ out January and early February ranged conditions duri~g 1971-1972 were ..:xam ined on a study area near Fairbanks. from 15 to 25 em below that in the In Figure 1, "Tracks-Valley" indicates hills, depth at the two sites remained the 7-day total of fresh moose tracks nearly identical during the rest of crossing a one-half mile long transect winter. The Ri of the snow cover sharply in a valley. The valley is located at increased during December preceding 245 m elevation, and represents ty movement of animals. The increase pical winter riparian moose habitat. resulted from both an increase in total "Moose-Hills" indicates the number of snow depth and an increase in ram moose counted during frequent intensive hardness of given depth increments. aerial surveys in a 75 km 2 drainage The dispersal of moose from lowland above the valley transect. The upland riparian habitat to adjacent areas during site ranges from 550 to 670 m elevationj Mar-h may have been influenced by and consists of mixed conifer and deci the declining Ri making travel less duous trees and shrubs which charac difficult. Dispersal from riparian habita~ teristically support modest numbers of in March may also nave been related moose during summer and fall in inte to. the lower snow depth, density and rior Alaska. While neither nTracks-Val hardness in deciduous and coniferous ley" nor ''Moose-Hills'' indicates actual tree communities during that time (Coady, unpubl.). The range for snow number of animals, thE;y are throught to density of settled snow at the upland .... , reflect the trend of animal abundunce shrub site increased from 0.16 ·· 0.24 j 3 ·.. · · on each site. Snow depths and integrat- g/cm in late November to 0.20 -- I .. · :·. .::.~~:·~ .· · ·-f~~~~~~~' - '-'~""t"f~~-::r~:l',·T.r•f~il'i"r~'i~'l"t\':<'1;'?"'~~,.~"!""'\"''~;.:F... : . . . . ------.,.;,;:~-··""!"fl~'?'l·-- . . -- "'· , ...... _ ""'· · · . I! r COADY: SNOW AND MOOSE BEHAVIOJ-l 429 0.31 g/cm3 in 1:-.te December, while ted in a dense regrowt:1 of shrubs and the range of NRC snow hardness for low trees in both the non-riparian (site settled snow increased from 10-50 no. 1) and riparian (site no. 2) hab g/cm2 to 50 ·- 500 g/cm2 during the itats. Moose movement onto both si same oeriod. . tes began in late. December and paral ' leled each other very closely until early Conclusions regarding the signifi March when nearly all animals had ·oance between integrated Rammsonde dispersed from the sites, apparently 1esistance and movement of moose into more .densely. vegetated areas. would be highly premature at this time. Snow depths on both sites averaged However, based upon the above data about 70 em in late December when and upon similar correle.tions between animals began to appear, and remain Ri and moose behavior in other study ed near 80 to 90 em until l~te April. areas during both 1971-1972 and 1972- The maximum number of animals was 1973 (Coady, unpul)l.), further studies 11 (density 7 /km2) on site 1 in mid using the Rammsonde penetrometer ~January. During the preceding winter, appear justified. snow depth on the sites averaged Movement of moose onto lowland 115 em in January and the maximum habitat during the 1971-1972 winter number of animals during that time was studied on two sites located 90 was 36 (density 24/km2) (Fig. 2). km from the above study area, and each approximately 1,5 km2 in size. Land Seasonal migrations of moose in In clearing 10 to 15 years ago has resui- terior Alaska are not always influenced 200r------·------~------•------~ E ·-·--..__.~.o-- __ . -- ';'0" 150 ,....,. -o- __ ..:.c 1 /. 0.:..---o. ,.... --- ··-o / r ,...,..._ ,. -• ,/· -"'0" o Volley ... 100 .-0 .... 0: , 0..----o---o',. • Hills • c .• . ;,., I '; '. NOV 1 DEC l JAN I FEB i MARl APR I MAY I Figure 1. Movement of rnoose, snow depth, and integrated Ram hardness (Rt) of the snow .:cover at high and low elevations during winter 1971-72 near Fairbanks, Alaska. Ordinate values \ for the two lower curves represent the number of animals (Moose Hillr,j observed cturing aerial surveys c;1nd the seven day total of fresh tracks (Tracks Valley) crossing a one·half mile transect in a ~alley. \ \ ! I - -· ..- ... ' ~ 430 LE NATURALISTE CANADIEN, VOL 101, 1974 melt has exposed patche~. of bare Site l ground. Observations of ar,imals and j ~l tracks suggest that when snow cover persists into mid - to late - May, ~D olo-1. substantial movement of animals does 0 not occur until that time. However, ~ IO't Site2 an early thaw results in an early nove ?--.. / ...... ment to summer range during April 0 _a.~ 1.-fil I - --.L-...I..e--'-- and early May. Advanced sno\1\ melt. ! 100 [ • ;:>- - -- -o- - -<\ during spring 1973 resulted in a 70 - p~-4 \ km migration between April 10 and g. / b f 50:<>'~-d April 20 of a radio-collared moosH from i winter to summer ranqo. (Coady, 1973), (ll ol I I I NCNI DEC I JAN I FEB I MAR I APR I MAR I Although the above data are preli Month minary and highly limited in sc'lpa, they t Figure 2. Moveme:.it of moose and snow depth illustrate an approach to studying on two sites characterized by dense deciduous L regrowth during winter 1971-72 near Fairbanks, moose-snow relationships wh:c 1 may Alaska. Snow depths were similar on both prove useful in other areas. Cetailed sitr.s. observations of moose distributi )n and by snow conditions. Movement of some movements, snow parameters ar d tem animals from lowland summer range perature and wind conditions in several to either upland or riparian shrub areas of Interior Alaska over three years habitat may begin in August, well will be reported in a future ~~ublica before snowfall occ.urs, and continue tion. throughout the winter. However, Habitat selection and moverr:ent on while initial movements may be related winter range in refation to sno N con to factors other than snow, the speed ditions have been reported in several and extent of migration is apparently excellent studies (Nasimovitch, 1955; influenced by snow. During winters DesMeules, 1964; Telfer, 1970; Berg, of early or deep snow; movement of 1971; Peek, 197'Ja, b; vanBallen most animals from lowland summer berg and Peek, 1971) and nviews range may occur sooner and to a in this volume (Berg and P 1illips, greater extent than during winters of 1974; Brassard et a!., 1974; Peek, late or little snow. Availability of 1974; Peterson and Allen, 1974) and will browse as affected by snow depth may therefore not be further considen·d he be particularly important in influen re. Most of these studies suggest an cing movements over flat terrain where increased use of dense cover wi~h an local differences in snow conditions increase in snow depth, densit {, o .. are not great. For example, late snow" l1ardness, and a relatively small v'intet fall may have accounted for exceptional horne range, although actual mow ly heavy and extensive use of some low conditions causing a change in he: bitat (40 to 60 ~m high) willow (Salix pulchrd) selection or home range size are communiiies on the Tanana Fluts near varinble. Fairbanks during fnll and early win ter, 1972. Activity of rnoose in response to snow Movement to summer mnge in into~ Rostriclad movement and snail rior Alaska apparently occurs during horne rnnuo of moose cJu ring whter a relatively short period aftor snow have boon reported by numerous w::lrk· .... /(-:··;~~:£ ~ • . '' .. "''1:i.··~~~IJIMI!IIIm· ~~"""'cy;~~~'IP"',""1J1,.~~~'::J!7!"'""'"'~ """~;:.,~~:'7'"""'''""'"'''"' ., : .. :: :s~-~l~0'7};p·~:~5.~5;.t,f;~:· :. ;v;,:j~; r ::::> .: : r': .: . ·... ·. ..·. . . . .) 'I j .• COADY: SNOW AND MOOSE BEHAVIOR 431 ers. Studies indicate that movements interior Alaska occurred within 25 to arc most limited during periods of deep 75 km 2 areas in upland shrub habitat snow. Knorre (1959) in .Russia found during December when snow ·depths that the average winter home range were 35 em. However, movements were of moose decreased from 225 ha during limited to 1.3 km 2 areas in upland shrub mi!d winters to 97 ha during average and l6wlana riparian habitat for three winters, to 5 ha during severe win to five weeks in January and February ters (Table II). A decrease in home range when snow depths had increased to size with increasing snow depth has 65 em in both areas. During late winter also been reported in Russia by Nasimo and particularly in spring when snow vitch (1965). Knorre (1959) found that was crusted, Timofeeva (1967) found a single moose occupied only 0.1 that moose in Russia remained in very ha during a 48-hour period of heavy limited areas for several days. However, snowfall, while van Ballenberghe and Timofeeva (1965) also noted that the Peek (1971) noted that a cow in Min distance traveled daily by some moose nesota occupied a 2.4 ha balsam fir may be greater during mid- to late (Abies balsamea) stand during 25 days winter than during early winter. Confi of rapid snow accumulation. nement during periods of deep snow has also been reported to occur in TABLE II riparian habitat by Nasimovitch (1955), Denniston (1956), Harry (1957), Knowl Size of winter range in relation to snow ton (1960), and Stevens (1970), and in conditions in Russia (from Knorre, 1959) non-riparian habitat by Peterson (1955), Average DesMeules (1962), Pivovarova (1965), depth of Size of range snow in Number (hectare) and Telfer (1967a, 1970). Type of March of While activity patterns of moose winter (em) Crust moose x Range have been recorded by several workers Mild 62 No 23 225 40-700 (Murie, 1934; Peterson, 1965; Denniston, Average 98 Yes 68 97 2-500 1956; Geist, 1963; Berg, 1971), the in Severe 127 Yes 96 5 0.1-20 fluence of snow conditions on daily ac~ tivity has not been widely studied. Both Berg (1971} found that winter home Timofeeva (1967) and Nasimovitch (1965) range of radio-collared mose in Minne noted that daily activity of moose in sota averaged 233 ha (compared to Russia decreased with increasing snow 1,400 ha during summer), although from depth. Knorre (1959} found that moose early March to mid-April movement was were active for a total of 11 hours per generally confined to areas less than day in November and December when 41 ha (Phillips and Berg, ~971). Berg snow depth was low, and for 10 hours {1971) also determined that daily move per day in February and March when ments during winter averaged less than snow depth was high. Animals were ac 0.4 km per day, while during summer tive for 14 hours per day during the they averaged 1 km per day. LeResche summer. Most of the active time during · and Davis (1971) reported that distances both winter and summer was spent feed moved by penned radio-collared moose ing. Timofeeva (1967} observed an adult on the Kenai Peninsula, Alaska, moose to average four steps and con · were smaller during ,january and Fe sume six shoots of browse per minute . bruary than during November and De when the snow was 45 em deep. The I cember. Coady (1973) found that move moose averaged 0.9 steps but still l ments of four radio-collared moose in consumed six shoots of browse per mi- I \ I 432 LE NATURALISTE CANADIEN, VOL 101, 1974. nute when the snow was 57 em deep. areas may occur. Relatively large track She noted that during deep snow, loads of moose tend to decrease moose consume all available browse throughout the winter due to seasonal within reach before moving. However, weight 'loss, thereby decreasing track Timofeeva (1967) concluded that the loads. rate of movement in deep snow depends Behavioral response of moose to largely upon the availability of forage. snow are represented by movement In areas with scattered browse, moose to areas of greatest food accessibility may be forced to move further to and by reduced activity. Increasing feed than in areas of more available snow depth during fall and early win food. ter generally results in a gradual mo The duration and frequency of active vement from su'mmer to winter range. and resting periods is apparently in The timing and magnitude of move fluenced by season and snow condi ment is closely related to the timing tions. Timofeeva (1965, 1967) found and rate of accumulation of snow. that during early winter when snow Although there are exceptions, · early depths were 50-60 em, moose rest three and deep snow may result in an to eight times and average five rest abrupt migration of most animals periods per day. Between January from summer to winter range, while and March when snow depths reach winters of little snow may result in a 70 em or greater, moose rest six to more gradual and delayed movement fifteen times and average eight rest of fewer animals. Spring movements periods per day. However, Geist (1963) from winter to summer range generally in British Columbia found that the occur after thawing has exposed pat duration of both active and resting pe ches of bare ground. Local movements riods was greater in winter than in and duration of activity are usually summer, indicating that fewer rest and restricted during periods of deep snow active periods per day may occur and during late winter. during winter than during summer. Factors influencing moose migra Timofeeva (1967) calculated for moose tions have been reviewed by LeResche that approximately 3.9 rest periods are (1974) and Pulliainen (1974). 'vVhile few, associated with a 1 km movement if any workers conclude that snow through 50 em deep snow, 4.7 rest pe is the ultimate cause of seasonal riods through 60-65 em deep snow, and movements, most reports indicate that 6.7 rest periods through snow depths it is an important factor influencing greater than 70 em. movements. Perhaps the major effect of snow is to alter thG energy balarce Conclusions of moose by either increasing metabolic requirements for movement or decrea- The most important and commonly sing access to energy sources by limi~ measured properties of a snow cover ting food intake. Deep or hard snow influencing the distribution and bohav- may restrict movement to the extent ' ·' ior of moose are d0pth, density, and that considerably more energy is ex- hardness. The integt::.tted effect of panded in moving to feed than i:;; as- j . thesP properties is to inlirec'!se the ef- similatod from ingostod food. Deep or j ·itt fort required for movement. Long legs 11ard .snow may also cover low growing . ~ .. ·. :r.~.~.· are perhaps the greatest pttysicai bn wse species, thereby reducing \ ... · _"';:;. adaptation to movement in snow, and their availnbility and requiring more \ ..:~ selection for long legs in deep snow extensive aGtivity to feed. l .' ~;, . l] . "•),_ . ' . ·~,:~:~~~-~ . • '! 1 '~1--"'.· . .-.:•·•·' ', .~1'1!W"YJ! ~~~t'~l!Mltw:JI\Itlflit'MYM>""!~~~'t"f'q""~~~ ,... . ,....., . ~, .---.r ·"".·.' ·\ ····- .. -.H'*'~-~- ··~:_~ ~: "'t•.;,'P,.: ·+._ ':., ·~~~/·.,.~~~~ 41';;-~:•*·r '*~~~ ..... d·;;..-~,..t1t·~'")} ::~\··r':·-:;~':'~:·---JIIO~--"".-~.-- -.:~t"'#_-,;ii,~ -~~~JY::'~~~, .. :l'.~~'f:'~I;rr~~r·~~':'"~ ~ -.~- ...... ·····-··...... - :'. .>S~!~ftlli';%;,;· ·'~::i;;,?;:> ,. " . ,.: ' .:: .' : ·, .· . '" .' . ' .. .. ·---~ .\ . 1 f r • ' COADY: SNOW AND MOOSE BEHAVIOR 433 Energy requirements for moose are BENSON, C. S., 1967. A reconnaissance snow .. somewhat greater than energy metabo survey of Interior Alaska. Rep. UAGR-190. Geophys. lnst., Univ. Alaska College, 71 p. lized frqm food during wir.ter (Gasaway and Coady, 1974), and therefore BENSON, C. S., 1969. The seasonal snow cover factors which add to energy require of Arctic Alaska. Arctic lnst. N. Am., Res. PRp.; No. 51. 47 p. ments or decrease energy availability further magnify the negative energy BERG, W. E., 1971. Habitat use, movements, balance and weight loss of the ani and activity patterns of moose in northwes tern Minnesot&. M. Sc. Thesis, Univ. of Minne mal. Movements of moose to areas sota, St. Paul, 98 p. (Unpubl.) where food quantity, quality, and acces sibility are greatest, depending on snow BERG, N. E. and R. L. PHILLIPS, 1974. Habitat conditions, minimize metabolic re use by moose in northwestern Minnesota wi!h reference to other heavily willowed quirements and enhance energy intake areas~ Naturalistc cnn., 101 ;101-116. for winter existence. BILELLO, M.A., R. E. BATES and J. RILEY, 1970. Physical characteristics of the snow cover Fort Greely, Alaska, 1966-67. U.S.A. Cold Regions References Research and Engineering Laboratory, Tech. Rep. No. 230, 33 p. ABELE, G., 1963. A correlation of unconfined compressive strength and ram hardness BISHOP, R. H., 1971. Annual report of sur of processed snow. U.S.A. Cold Regions vey-inventory activities. Part 1, moose, Research and Engineering Laboratory, Tech. deer, and elk, Fedl Aid Wildt. Restor. Proj. Rep .• No. 85, 14 p. Alaska, W-17-3, Vol. II, 119 p. : '~- ABELE, G., 1968. An experimental snow runway BISHOP, R. H. and R. A. RAUSCH. 1974. Moose pavement in Antarctica. U.S.A. Cold Re population fluctuations' in Alaska, 1950-1972. .. - gions Research and Engineering Laboratoy, Naturaliste can., 101; (in press). Tech. Rep. No. 211. BRASSARD, J. M .• E, M. AUDY, M. L. CRETE ABELE. G., R. 0. RAMSEIER and A. F. WUORI. and P. A. GRENIER, 1974. Distribution and 1965. A study of subsurface transportation winter habitat of moose in Quebec. Natura/is methods in deep snow. U.S.A. Cold Regions le can., 101 :67-80. Research and Engineering Laboratory, Tech. Rep. No. 160. BULL, C., 1956. The use of the Rammsonde as an instrument in determining the density ABELE, G., R. 0. RAMSEIER and A. F. WUORI, of firn. J. Glacio!., 2: 713-725. 1968. Design crite:ria for snow runways. U.S.A. Cold Regions Research Rnd Engi COADY, J. W., 1973. Moose Research Report. neering Laboratory, Tech. Rep. No. 212. Fedl Aid Wildt. Restor. Proj. Seg. Rep., Alaska Dep. Fish and Game., Juneau (in press). BADER. H.• R. HAEFELI, E. BUCHER, J. NEHER, 0. ECKEL and C. THAMS, 1939. Snow and its DENNISTON, R.H., 1956. Ecology, behavior metamorphism. Beitrage zur Geologie der and population dynamics of the Wyoming Schweiz. Geotechnische Serie, Hydrologie, or Rocky Mountain Moose, Alces alces Lieferung 3, Bern. U.S. Army Snow, Ice and shirasi. Zoologica, 41: 105-11"8. Permafrost Research Establishment Trans. No. OesMEULES, P., 1962, Intensive study of an 14, 1954, 313 p. early spring habitat of moose (A/ces alces van BALLENBEAGHE, V. and J. M. PEEK, americana Cl.) in Laurentides Park, Quebec. 1971. Radiotelemetry studies of moose in NE. Wildl. Cont., Monticello, N.Y. May 1962, northeastern Minnesota. J. Wild/. Mgmt, 12 p. (Mimeogr.) 35 (1): 63-71. DesMEULES, P., 1964. The influence of snow BAUMAN, J., 1941. The migration of moose. on the behavior of moose. Trans. NE. Yellowstone Nature Notes, 18: 33. Wildt. Cont., 21, 17 p. BENSON, C. S., 1962. Stratigraphic studies EDWARDS, R. Y. and R. W. RITCEY, 1956. The in the snow and firn of the Greenland Ice migrations of a moose herd, J. Mammal., \ Sheet. U.S.A. SIPRE Res. Rep. No. 70, 91 p. 37 (4):486·494. I \ I \ \ --- c} l i ~ COADY: SNOW liND Monm BEH!IVIOR 435 MURIE, /\., 1944~ The wolves of Mount Mcl NASIMOVITCH, A. A., 1965. The territory used PRUITT, W. 0 .. Jr.. 1971a. Report on interna by the moose, p. 10-22. In: R. 'Khozizdat (ed.) lional workshop on Rangifer winter ecology, The biology and commercial hunting of the V1ttangi, Sweden, 20-28 March. 5 p. {Mimeogr.) moose, Symp. 2, Moscow. PRUITT. W. 0., Jr., 1971b. Scandinavian-Cana OZOGA, J. J., 1968. Variations in microclimate dian field workshop on Rangifer-Snow ecolo in a conifer swamp deeryard in northern Mi gy. Arctic, 24 (2): 145 p. chigan. J. Wild/. Mgmt, 32{3):574-585: PULLIAINEN, E., 1974. Seasonal movement& of moose in Europe. Naturaliste can., PEEK, J. M., 1971 a. Moose habitat selection 101: 379-392. and relationships to forest management in northeastern Minnesota. Ph. D. Thesis, Univ. RAUSCH, R. A .• 1958. Moose management stu of Minnesota, St. Paul, 250 p. (Unpubl.) dies. Fedl Aid Wild!. Restor., Job Completion Rep., Vol. 12, Proj. W-3-R-12. Alaska PEEK, J. M., 1971b. Moose-snow relationships Game Commission, Juneau, "138 p. in northeastern Minnesota, p. 39-49. In: A. 0. Haugen (ed.}, Pr,.,c. snow and ice in rela RICHENS, V. B. and C. G. MADDEN, 1973. An • tion to wildlife and 'ecreation symposium . improved snow study kit. J. Wild/. Mgmt, Iowa Coop. Wildl. Res. Unit, Iowa State Univ. 37(1): 109-113. Press, Ames., 280 p. 1 • RITCEY, R. W .• 1967. Ecology of moose winter PEEK, J. M., 1974. On the nature of winter range in Wells Gray Park. British Columbia. habitats of Shiras moose. Naturaliste can., Proc. 4th Works., on Moose Res. Mgmt, 101 :131-141. Edmonton, 15 p. {Mimeogr.) PETERSON, R. L., 1955. North American ·moose. SOMMERFELD, R. A., 1969. Classification Univ. Toronto Press, Toronto, 280 p. outline for snow on the ground. U.S.D.A. Forest Serv. Res. Pap., No. RM-48, 24 p. PETERSON, R. 0. and D. L. ALLEN, 1974. Snow conditions as a parameter in moose-wolf STEVENS, D.R.. 1.967. Ecology of moose in coact10ns. Natura/iste can., 101 : (in press). southwestern Montana. Montana Fish Game Dep., Job Completion Rep. P-R. Proj. W- P.HILLIPS, R. L. and W. E. BERG, 1971. Home 98-R7, 28 p. range and habitat use patterns of moose STEVENS, D. 1970. Winter ecology of moose in northwestern Minnesota. Paper presented R.. in the Gallatin Mountains, Montana. J. at 7th N. Am. Moose Cor:~f., Saskatoon. (Abs tract). Wild!. Mgmt, 34(1): 37-46. TELFER, E. S., 1967a. Comparison of a deer PICTON, H. D. and R. R. KNIGHT, 1971. A nume yard and a moose yard in Nova Scotia. rical index of winter conditions of use Can. J. Zoo/., 45: 485-490. in big game management, p. 29-38, In: A. 0. Haugen (ed.), Proc. snow and ice in relation TELFER, E. S., 1967b. Comparison of moose to wildlife and recreation symposium. Iowa and deer winter range in Nova Scotia Coop. Wildl. Res. Unit, Iowa State Univ., J. WJ/dl. Mgmt, p. 418-425. Ames., 280 p. TELfER, E. S., 1970. Winter habitat selection PIVOVAROVA, E. P., 1965. The winter distribution by moose and white-tailed deer. J. Wild/. of moose in the hunting establishments of Mgmt, 34: 553·559. the Vladimir and Kaluga Provinces, p. 173- 180, In: Khozizdat (ed.), The biology and TELFER. E. S. and J.P. KELSALL, 1971. Morpho· commercial hunting of the moose, Symp. 2. tog1cat parameters for mammal locomotion Moscow. 10 snow. 51 st A. Mtg Am. Soc. Mammal. l7ill~~?l~·~·l".'i:~fi'::'Jl';:'~1--.:<:.~~"''""'""""'-~-···,. ·~··'''"' ...,.... ' ' .~ .; ""· ,. ... . I lr ...... -