Loading...
HomeMy WebLinkAboutInterior Moose Studies Vol 1 1973ALASKA D E P A R T M E N T 0 F F I S H J U N E A U, A L A S K A STATE OF ALASKA William A. Egan, Governor DEPARTMENT OF FISH AND GAME James W. Brooks, Commissioner DIVISION OF GAME Frank Jones, Director Donald McKnight, Research Chief AND INTERIOR MOOSE S T U D I E S by John W. Coady Volume I Project Progress Report Federal Aid in Wildlife Res toration Projects W-17-4 and W-17-5, Jobs 1.3R, 1.4R and 1.8R GAME Persons are free to use material in these reports for educational or informational purposes. However, s ince most reports treat only part of continuing studies, persons intending to use this material in scien- tific publications should obtain prior permission from the Department of Fish and Game. In all cases, tentative conclusions should be identifi ed as such in quotation, and due credit would be appreciated. (Printed July, 1973) State: Cooperator: Project Nos: Job No: Job No.: JOB PROGRESS REPORT (RESEARCH) Alaska John W. W-17-4 & W-17-5 1.3R Coady Project Title: Big Game Investigations Job Title: Evaluation of Moose Range and Habitat Utilization in Interior Alaska Job Title: Evaluation of Moose Browse and Rumen Fermentation in Interior Alaska Period Covered: July 1, 1971 to June 30, 1973 SUMMARY This report describes major activities undertaken between February 1971 and June 1973. A vegetation type map of the Tanana Flats was com- pleted in spring 1972. The vegetation, browse selection by moose and hares, and soil conditions in different habitats in the Tanana Flats and in the Elliott Creek burn on the Little Chena Drainage were described during summer 1972. Gross trends in moose -food habits were examined by identifying .the botanical composition of over 100 rumen content samples collected during four seasons near Fairbanks. Estimates of seasonal energy requirements and rates of rumen fermentation and food consumption by moose were obtained between May 1972 and April 1973 by measuring volatile fatty acid (VFA) production, total weight and water content, and chemical composition of rumen contents, and by extensively reviewlng the literature. Nutritional status of moose and quality of moose range may be reflected by rates of rumen fermentation. i Summary •• Objectives ••• Background. Procedures. Vegetation Type Map. • • . . CONTENTS Botanical Composition and Soil Conditions. • Botanical Composition of Rumen Contents .• Chemical Composition of Moose Rumen Contents, Energy Requirements and Rumen Fermentation in Moose • Findings and Conclusions •.••••..•..• Vegetation Type Map ••••••.•.••• Botanical Composition and Soil Composition • Moose Use. • • • • • . • • • • i 1 1 1 1 3 3 5 5 5 5 9 Botanical Analysis of Moose Rumen Contents • 14 Chemical Composition of Moose Rumen Contents and Energy Requirements and Rumen Fe~mentation in Moose • Acknowledgments • • Literature Cited. OBJECTIVES . 23 24 24 The objectives of this study are: 1) to construct a vegetation type map of the Tanana Flats; 2) to characterize botanical composition and soil conditions of major vegetation types in the Tanana Flats, and evaluate browse preference by both moose (AZces aZces) and hares (Lepus americanus) in each habitat type; 3) to analyze the botanical composi- tion of moose rumen contents collected near Fairbanks during different seasons; 4) to analyze the chemical composition of moose rumen contents collected during different seasons for components which reflect the relative digestability of the forage; and 5) to estimate the seasonal energy requirements and measure factors associated with rumen fermenta- tion rates in moose which reflect nutritional status of the animal and quality of the food resource. BACKGROUND See Job 1.8R of this report. PROCEDURES Vegetation Type Map A vegetation type map of the Tanana Flats was preoared from black and white aerial photographs (scale 1" :1320 1 ) taken during the sunnners of 1968, 1969 and 1970. Identification of taxonomic units was ba8ed on stereoscopic examination of photographs and on field observation. Map units delineated on photographs were transferred using a light table to USGS topographic maps printed on photographic drafting film. Blue-line 1 copies of the finished type map were printed from the film. Several qualifications regarding the type map should be noted. Shapes of map units may differ from their appearance in the field since distinct boundaries between vegetation types seldom exis.t in nature. Aerial photographs were taken during relatively wet (June) and dry (August) months over three years, and therefore water levels in ponds and sloughs may differ from their present condition. In particular, 1968 photographs may indicate above normal water levels due to previous flooding in 1967. Aberrations in photographs may slightly distort true shapes of particular map units, although positions relative to surround- ing units are correct. In a project of this magnitude, using black and white photographs, errors in photo interpretation are unavoidable. Some taxonomic· units have undoubtedly been incorrectly identified, especially when the inter- preter was required to classify certain mixed vegetation stqnds based on . dominant species. However, these errors are believed to represent only a small fraction of all map units and should not significantly affect the usefulness of the map. Five major taxonomic units were identified: Herbaceous Bog: designated by number "1" (blue), and consisting of marshes, shallow ponds and their aquatic margins, and shallow streams where movement of water is slow enough to support emergent vegetation. Heath Bog: designated by number "2". (white), and consisting primarily of low shrubs and widely scattered trees. Scattered patChes of shrub birch (Bet.ula glandulosa), paper birch (Betula papyrifera), willow, spruce, and tamarack (Larix laricina) also occur. Tall Shrub: designated by number "3" (yellow), and consisting primarily of willow and alder. Deciduous Tree: designated by number "4" (red), and consisting primarily of paper birch, quaking aspen (Populus tremuloides), and cottonwood ~ (Populus balsamifera). Conifer Tree: designated by number "5" (gre~n), and consisting primarily of black spruce (Picea mariana), white spruce (P. glauca) and tamarack. Two burns were outlined on the map. One burn of approximately 200 km2 on the east side of the Flats and south of Salchaket Slough occurred in 195 7. The secortd burn of approximately 25 km2 near the Blair Lakes occurred during summer 1969. A solid line indicates known boundaries, while a dotted line indicates approximate boundaries of the burns. Unclass.ified area is designated by "uc" and consists primarily of deep lakes and steep rocky slopes. 2 Botanical Composition and Soil Conditions Site characteristics and plant associations in six heath bog, six tall shrub, two deciduous tree and three conifer tree habitats on the Tanana Flats, and in two tall shrub and one heath bog habitats in the Elliott Creek burn in the upper Little.Chena River drainage were described using a sampling scheme developed by Ohmann and Ream (1971). The scheme includes determining the approximate age of the stand by counting annual growth rings of the largest trees, recording species, size, and density of trees and shrubs, and determining the species and percent coverage of low growing plants. Depth to permafrost, moisture content, and mineral composition of the soil in each stand was also recorded. Utilization of browse species by·moose and hares in three heath bog, five tall shrub, and one deciduous tree stand in the Tanana Flats was evaluated using range survey techniques similar to those described by Cole (1963). Sixty to 200 individual plants of each major browse species in a stand were sampled using closest plant sampling techniques, and data were recorded on Alaska Department of Fish and Game Browse Survey forms-(Fig. 1). A numerical index of one to three was used to indicate age class , where one indicates seedlings less than four and one-half feet (137 em) high, two indicates saplings greater than four and one-half feet (137 em) high but less than two inches (5 em) in diameter at breast height (dbh), and three indicates trees greater than two inches (5 em) dbh. Canopy range is the height range above ground from the first lateral branch to the top of the plant. Browse range for moose and for hares is the height range above ground of browsing by each species during the previous year. A numerical index of 0 to 4 was used to indicate browse class for moose and for hares, where 0 indicates no browsing during the previous year, .and-1, 2, and 3 indicate 1-33 percent, 34-66 percent, and 6 7-100 percent, respectively, of the lateral branches within the browse range for moose and hares browsed during the previous year. A numerical index of 0 to 3 was used to indicate barking by hares, where 0 indicates no barking, and 1, 2, and J indicate a subjective evaluation of light, moderate, and severe (or girdled) barking, respectively. Botanical Composition of Rumen Contents In this preliminary study, no attempt was made to standardize the quantity of material collected from individuaL moose stomachs. A one- quart subsample was randomly collected from each sample and analyzed. This was done without knowing whether the one-quart subsample was representative of the gallons of material normally in a moose stomach. The one-quart subsamples were emptied into a gang of sieves and washed with fresh water. Material retained by a 6. 35 mm mesh sieve was separated by plant parts (twigs, fruit, leaves, etc.) to species where possible. This mesh sieve was used without determining its effectiveness in retaining food items. Food items that could not be readily identified to species were grouped into general categories of foods (e.g. , leaves , twigs, fruits). All data were coded and recorded for direct automatic data processing, using summary programs developed by Cushwa et al. (19 73). Frequency of occurrence, the number of samples each food item occurred 3 l I i I I I I I ! : ' I I Figure 1. Alaska Department of Fish and Game browse survey form. (See · text for description of figure.) Stand Area Date Veg Type Observer Remarks Moose Hares Age Canopy Browse Browse Browse Browse Sp Cl Range Range .Class Range Class Barking i I I ! I I I ! I i I i I ~ l i ! I I i r i I ' ; • ' I i I ' I i I I I I I i ! i i I ! ! I ! i : I I I I I I I I I i I I ! l I I i i I i I I I I ; i I I I ! 4 in, was calculated and reported. Neither volume nor importance value were reported because of the large number of factors which influence volume calculations over which an investigator has no control. Chemical Composition of Moose Rumen Contents, Energy Requirements and Rumen Fermentation in Moose Procedures have been described in two previous publications (Coady and Gasaway, 1972; Gasaway and Coady, 1973), and will not be reported here. FINDINGS AND CONCLUSIONS Vegetation Type Map Approximately seven man-months of labor, $1,000 in materials, and 20 hours of air charter were devoted to this project. The type map was completed in April 1972 and is located in the Fairbanks roame Division office. Botanical Composition and Soil Composition The regional vegetation of northern boreal forests in Alaska, referred to by the Russian term "taiga," consists primarily of low, open-growing spruce forests, occasional stands of well developed spruce and hardwoods, and frequent tracts of treeless or sparsely timbered bogs. On south facing slopes and well drained sites the forest consists ___ of _white_ s12r1.1ce and hardwood stands of quaking_ aspe_n .;m,ci_pg.p~r l:>ir_ch_, __ while on cool, north facing slopes and poorly drained lowlands climax vegetation is generally black spruce, tamarack and bogs. Extensive lowlands, locally referred to as "flats" or "muskeg," cover broad alluvial plains throughout interior Alaska, from the south slope of the Brooks Range to the southern coastal forests, and from the Alaska-Canada border nearly to the Bering and Chukchi seas (Wahrhaftig, 1965; Johnson and Hartman, 1969). Approximately 30 percent of this area is forested, while the remaining land consists of bogs, shrub thickets, and alpine tundra (Viereck, in press). Surface vegetation patterns are closely related to topography, drainage, presence or absence of perma~ frost and past forest fire history. Surface features in lowland areas frequently include extensive flood plains with little relief, meander scars and oxbow lakes, terraces, and alluvial outwash deposits (Black, 1958; Wahrhaftig, 1965). Loess, sand, and outwash of Quaternary age, organic deposits formed in bogs, and recently deposited alluvium frequently overlay a micaceous schist bedrock (Dutro and Payne, 1957; Viereck, in press). Forest soils are generally shallow with poorly developed profiles. Piedmont st'reams, many of glacial origin, change from braided to tightly meandering tributaries as they enter lower elevations. Permafrost, or permantly frozen ground, is a widespread phenomenon 5 throughout much of the Alaskan taiga. Between the Alaska Range and the Brooks Range permafrost occurs in all areas except for south facing slopes and recently deposited alluvium, while $OUth of the Alaska Range permafrost is sporadic in occurrence, and is found only in bogs and on north facing slopes (Vier~ck, in press). An important influence of permafrost on vegetation patterns results from the frozen impervious substrate which prevents lateral movement and downward percolation of soil water (Benninghoff, 1952). Thus, permafrost results in saturated soils or standing water throughout much of the taiga. Regional distributions of vegetation types and permafrost in low- land areas are inextricably interrelated. Insulation provided by black spruce and bog vegetation prevents melting of permafrost during summer months' while permafrost governs climax vegetation during the course of succession (Benninghoff, 1952; Drury, 1956). Disjunct stands of shrubs and deciduous trees scattered throughout Alaskan lowlands ·are frequently the result of burning of climax vegetation, which results in lowering of permafrost tables, and formation of a substrate temporarily favorable to subclimax vegetation (Viereck, pers. comm.). Geologically and vegetatively the flats of interior Alaska consist of treeless or nearly treeless bogs and more or less forested areas surrounding or occurring tvithin the bogs. Drury (1956) and Viereck (1970a, 1970b) have discussed lowland forest succession and origin of bogs along braided or meandering streams in interior Alaska. Freshly deposited alluvium is first colonized by willow (Salix spp.) or alder (Alnus spp.), and later by balsam poplar (Populus balsamifera). An understory of low shrubs and horsetails (Equisetwn spp.) along with white spruce seedlings may develop beneath the poplars. As white spruce matures an organic ground layer of mosses, herbs, and low shrubs develops, resulting in permafrost formation and a substrate more favorable to black than to white spruce after 200 to 300 years. Local disturbance of the insulating organic layer in black spruce forests may result in shallow thawing of the permafrost, water accumula- tion, and bog formation (Benninghoff, 1952; Drury, 1956). · Development and expansion of bogs are frequently indicated by angular growth of trees due to soil instability. The resulting vegetation over extensive lowland areas becomes an intricate mosaic of black spruce forests, bogs,. shrub and hardwood sub-climax communities, a8 well as numerous intermediate stages. Floristics of northern lowlands have been studied by several workers including Ritchie (1959) and Larsen (1965) in subarctic Canada, Hanson (1951, 1953) in western Alaska, Drury (1956) in the upper Kuskokwim River region of Alaska, and Johnson and Vogel (1966) in the Yukon Flats of Alaska. In addition, Anderson (1959), Hulten (1968), and Viereck and Little (1972) have described the circumpolar distribution of trees, shrubs, and herbs found in Alaska. Recent alluvial deposits on lowland floodplains throughout interior Alaska are generally colonized first by horsetails (Equisetwn arvense), grasses (Calamagrostis canadensis), willows (Salix alaxensis~ S. 6 arbusculoides, S. bebbiana), and alders (lllnus tenuifolia). As balsam poplar and later, white spruce or mixed white spruce-paper birch become established, herbs such as wintergreen (J>yrola spp.) and fi reweed (Epilob1:wn angustifolium), and low shrubs such as roses (Ro::a acic:ulor•i::), currants (Ribes triste), and high bush cranberry (Wbur'Ylurn edulc) appear .. Accompanying the replacement of white spruce and paper birch by black spruce is a gradual increase in the sphagnum moss coVer (Sphagnum capiUaceum~ S. girgensobnii, S. fuscus, and S. rubeUum) and growth of a dense shrub layer of bog blueberry ( Vacciniwn uliginosum), Labrador tea (Ledwn groenlandicum, L. decumbens) and birch (B. glandulosa, B. nana). A dense ground cover of sphagnum and low shrubs, along ~vith willows (S. pulchra~ S. bebbiana), and widely spaced paper birch, black spruce, or tamarack may replace stands of black spruce. The recurring process of bog formation and subsequent reforestation has been described in detail by Drury (1956). All stages of· bog development, from open water to black spruce forest are found in Alaskan lowlands. Sphagnum mosses, sedges (Carex spp. and Eriophorum spp.) and pond lilies (Nuphar polysepalum and Nymphaea tetragona) are common in open water, while other sedges (C. aquati lis) and horsetails (E. fluviatile, E. palustre) are found along margins of ponds and shallow flowing water. Bog shrubs, such as bog rosemary (Andromeda polifolia), Labrador tea, bog blueberry (Vaccinium uliginosum), swamp cranberry (Oxycoccus microcarpus), leather-leaf (Chamaedaphne calyculata), cloudberry (Rubus chamaemorus) , dwarf birch (Betula nana) and shrub birch along with sphagnum mosses are common on moist ground. The vegetation of the Tanana Flats, a 130 0 square-mile alluvial lowland lying immediately south of Fairbanks, Alaska, has been recently studied (Coady and Simpson, in ~·). The area is bounded on the south by the rugged Alaska Range, on the north and east by the glacial Tanana River, anci on the west by the glacial Wood River, and is part of the much larger Tanana-Kuskokwim physiographic province described by Wahrhaftig (1965). Surface deposits from glacial streams flowing into the Tanana Flats on the south form a belt of broad coalescing fans that grade from coarse sand and gravel near the mountains to fine sand and silt at lower elevations. Haterial manteling the eastern and northern portion of the Flats has been deposited by the Tanana River (Andreasen, et al., 1964). Except for1 scattered low hills of granite, ultramafic rocks, and possibly Precambrian schist, the Flats are an area of little relief (Andreasen, et al., 1964). The entire region is underlain by permafrost (Black, 1958; Wahrhaftig, 1965), and drainage is poor, resulting in numerous small, shallow ponds, extensive bogs, and meander scars. Herbaceous bogs occur primarily in the northern portion of the Flats and cover approximately 7 percent of the area. Vegetation is dominated by emergent species, and live trees and shrubs are totally absent (Table 1). Stagnant or slowly flowing water depths vary season- ally, although they tange from several inches to several feet after spring run-off. Bog bottoms consist of a meter or more of dead and 7 Table 1. Floristics of herbaceous bog community; Tanana Flats, Interior Alaska. Trees Tall Shrubs Low Shrubs None None None Herbaceous Rotentilla palustris Gramina spp. Eriophorum. spp . Equisetum hiemale var. califor~icum Equisetum fluviatile Equisetum palustre Iris. setosa subsp. interior Ranllnculus Gmelini subsp. Purshii · Ranunculus sp • Sanguisorba officinalis Petasites frigidus Petjsites sagittatus Valeriana capitata Caltha palustris subsp. arctica Gardamine sp. Rum:ex sp. Carex spp. Potentilla spp. Utricularia vulgaris subsp. macrorhiza Typha latifolia Nymphae tetragona ChEysoplenium tetrandrum Hippuris vulgaris utricularia minor decaying vegetation, and permafrost depths are presumably well below the upper surface of organic material. Heath bogs are widespread throughout the Flats, occupying approxi- mately 40 percent of the land. ·Dominant vegetation consists of mosses and shrubs, although scat"tered trees and various herbs. rooting on precipituous sedge hummocks are common (Table 2). Both mineral soil and permafrost tables occur within a meter of the surface, although seasonal thaw may extend to greater depths in some areas. Soil moisture is high and shallow standing water is common in many areas. Tall shrub communities occur throughout approximately 10 percent of the Flats but are most frequent along rivers, streams, and sloughs and along margins of ponds and meander scars. Recent wildfire burns may also support tall shrub communities. Vegetation ranges from pure to mixed stands of willow and alder with a dense understory of mosses, herbs, and low shrubs in poorly drained sites (Table 3). Exposed miner.al soil, low moisture content, and absence of permafrost are common on recent alluvial deposits, while a thick organic layer, impeded drainage, and high permafrost tables ~re found in other areas. Discontinuous pure or mixed stands of paper birch, quaking aspen, or balsam poplar occur throughou~ approximately 8 percent of the Flats, particularly on slightly elevated land and on coarse river alluvium. Understory vegetation ranges from a dense herbaceous cover in cqttonwood stands to mixed herbs and low shrubs in aspen and birch stands. Scattered willows and alders are common among cottonwood communi ties (Table 4) . Well drained mineral soil lies close to the surface and permafrost tables are deep or nonexistent. Scattered conifer stands in the western portion and extensive low, dense black spruce and occasional tamarack forests in the southern area cover approximately 35 percent of the Flats. Mature white spruce forests with a ground vegetation of low shrubs and mosses are common near streams, while black spruce forests underlain by a dense mat of moss, herbs, and low shrubs grow in poorly drained areas (Table 5). Soil organic layer, moisture content, and permafrost tables range from low in young white spruce stands to high in black spruce and tamarack stands. Moose Use The seasonal importance to moose of local flatlands south of the Brooks Range is diverse. Lowland areas, by virtue of their abundant herbaceous vegetation, are generally important summer ranges for moose of all sex and age groups. Furthermore, seclusion provided by numerous dense stands of tall shrubs and trees frequently creates favorable calving areas for large numbers of moose (Bishop, 1969). On the Tanana Flats moose commonly feed in herbaceous bogs from spring thaw to late summer. However, greatest use of this habitat appears to be during early to midsummer. During late summer moose may feed more frequently on herbaceous arid woody browse in heath bog and tall shrub communities. 9 ...... 0 Table 2. Floristics of heath bog community; Tanana Flats, Interior Alaska. Trees Betula papyrifera subsp. humilis Picea glauca Picea mariana Larix laricina var. alas kens is Tall Shrubs Salix bebbiana Salix glauca Salix arbusculoides Salix planifolia subsp. pulchra Betula papyrifera x glandulosa Low Shrubs Salix brachyocarpa subsp. niphoclada Salix myrtillifolia Betula nana :Ledum paltiStre subsp. ·decumbens . Ledum palustre subsp. groenlandictnn yaccirtium uliginosum subsp. alpinum Vaccinium vitis-idaea subsp. minus Andromeda polifolia :Myrica gale var. tomentosa Potentilla fruticosa Chamaedaphne calyculata Linnaea borealis Arctostaphylos uva-ursi Arctostaphylos rubra · Rosa. acicularis Oxycoccus microcarpus Herbaceous Rubus chamaemorus Potentilla palustris Eriophorum spp. Gramina spp. Rubus arcticus Stellaria spp. .. Epilobium angustifolium subsp . angus tif o li um Cornus canadensis Pyrola spp. Equisetum pratense Equisetum variegatum ·Equise tum arvense Equisetum scirpoides · Equisetum fluviatile Pedicularis labradorica Saus.surea sp . Mertertsia paniculata var. paniculata Mertensia sp. Solidago sp. Lupinus arcticus Iris setosa subsp. interior s;nsuisorba officinalis Petasites frigidus Trientalis europea subsp. arctica Ranunculus sp. Spiranthes Romanzoffiana Table 3. Floristics of tall shrub community; Tanana Flats, Interior Alaska. Trees Tall Shrubs Populus balsamifera subsp. Salix arbusculoides balsamifera Salix bebbiana Picea glauca Salix monticola Picea mariana Salix glauca Betula papyrifera subsp. Salix lanata subsp. humilis richardsonii Salix lasiandra Salix hastata Salix planifolia subsp. pulchra Salix novae-angliae Alnus crispa subsp. crispa Alnus incana subsp. tenuifolia Betula papyrifera x glandulosa Low Shrubs Rosa acicularis Potentilla fruiticosa Ledum palustre subsp. groenlandicum Vaccinium uliginosum subsp. alpinum Vaccinium vitis-idaea subsp. minus Arctostaphylos rubra Arctostaphylos uva-ursi Ribes hudsonianum Ribes tris te Rubus idaeus subsp. melanolasius Viburnum edule Linnaea borealis Myrica gale var. tomentosa Oxycoccus microcarpus Herbaceous Gramina spp. Carex spp. Equisetum arvense Equisetum palustre Equisetum pratens.e Equisetum fluviatile Equisetum scirpoides Potentilla spp. Achillea siberica Rubus arcticus Stellaria spp. Pyrola secunda Pyrola spp. Epilobium angustifolium subsp. arigustifolium Cornus canadensis Erigeron sp. Trientalis europaea subsp. arctica Petasites saggitatus Petasites frigidus Moneses uniflora Aster sp. Galium boreale Galium trifidum subsp. trifidum Fragaria virginiana subsp. glauca Mertensia paniculata var. paniculata Rubus chamaemorus Parnassia palustris subsp. neogaea Table 4. Floristics of deciduous tree community; Tanana Flats, Interior Alaska. Trees Populus balsamifera subsp. balsamifera Populus tremuloides Betula papyrifera subsp. humilis Picea glauca Picea.mariana Tall Shrubs Alnus crispa subsp. crispa Alnus incana subsp. tenuifolia Salix alexensis Salix interior Salix lanata subsp. richardsonii Salix novae-angliae Salix planifolia subsp. pulchra Low Shrubs Rosa acicularis Ribes triste Ribes sp. Rub.us idaeus subsp. melanolasius Herbaceous Stellaria ~· . Pyrola . secunda Epilobium angustifolium Gramina spp . Equisetum silvaticum Equisetum palustre Equisetum arvense Equisetum pratense Poiygonum alpinum Mertensia paniculata var. Trientalis europea subsp. Thalictrum sparsiflorum Galium boreale Artemsia sp. paniculata arctica Table 5. Floristics of conifer tree community; Tanana Flats, Interior Alaska. Trees Tall Shrubs Picea glauca Alnus crispa subsp. crispa Picea mariana Alnus incana subsp. Larix laricina var. tenuifolia alaskensis Salix alexensis Betula papyrifera subsp. Salix bebbiana humilis Salix glauca Populus balsamifera subsp. Salix arbusculoides balsamifera Low Shrubs Viburnum edule Rosa acicularis Ribes hudsonianum Ribes sp. Rubus idaeus subsp. melanolasius Empetrum hermaphroditum subsp. nigrum Arctostaphylos uva-ursi Vaccinium vitis-idaea subsp. minus Linnaea borealis Herbaceous Equisetum arvense Equisetum pratense Equisetum fluviatile Equisetum palustris Equisetum scirpoides Cornus canadensis Cornus stolonifera Gra.niina spp . Trientalis europea subsp. arctica Stellaria spp. Mertensia paniculata Rubus arcticus Pyrola spp. Geocaulon lividum Potentilla ~· Epilobium angustifolium subsp. angustifolium Goodyera repens Galium sp. Thalictrum sparsiflorum Galium trifidum subsp. trifidum Cypripedium guttatum subsp. gutta tum Platanthera obtusata Listera borealis Use of different lowland areas by moose during fall and winter depends largely upon the availability of adequate quality winter browse. Following the 600 square-¢le 194 7 burn on the Kenai Peninsula lowlands, white birch regrowth created major winter rarige ·for large numbers of moose.· Spencer and Chatelain (1953) discussed other lowland areas in Alaska which have become important winter moose range following fires. Generally, the Tanana Flats do not seem to be good winter moose range, and large scale emigration of animals during fall and early winter supports this conclusion. While several species of willows are widely scattered throughout portions of the Flats, they are often old and extremely decadent. Apparently, changes in local edaphic factors, such as soil temperature, moisture, and organic content, due to the dynamic nature of bog formation, frequently create substrates unfavorable to continued growth of tall shrub and other sub-climax communities. However, vigorous growth of shrubs occurs on·well drained recent deposits along some streams, and these areas may supnort large numbers of moose during winter. The effects of fire in creating winter moose brows·e on the Tanana Flats'have been variable. Although much of the Flats has burned within the past 50-100 years, early seral stages have frequently not developed or have been of short duration and consequently of minor importance to moose. However, a 75 square-mile burn in 1957 on the east side of the Flats has resulted in exposure of mineral soil and considerable reduction in permafrost levels in many areas; Consequently, much of the area is well drained and supports moderately dense, mixed stands of willow (Salix glauca, S. arbusculoides, S. bebbiana), poplar (Populus trerrruloides, P. balsamifera), and paper birch. For the past several vears the area has supnorted a sizable number of moose during winter. Other recent burns of considerably smaller size have provided limited winter browse to a few animals. A more complete description of the vegetation of the Tanana l"lats will be presented in a future publication (Coady and Simpson, in prep.). Data from the moose and hare browse study have not been fully summarized, and therefore, no quantitative results are available. However, observations leave little doubt that deciduous shrubs and · trees preferred by moose are also used extensively by· hares. These species include willows (Salix spp.), paper birch, quaking aspen and cottonwood. The portion of the plant hrowsed by the two animals differs, of course, depending largely on snow depths. The only conifer trees occurring in interior Alaska, tamarack, black S.Pruce and white spruce, are used extensively by hares and rarely by moose. Botanical Analysis of Moose Rumen Contents During the winter of 1970-71 (September through }fuv) the Fairbanks weather station recorded the heaviest total snowfall (307 em) since it began keeping records during the winter of 1933-34. Maximum snow depths on the ground exceeded 101 em in January. Examination of 44 samples of moose stomach contents collected during this period revealed a. diet 14 primarily of deciduous woody plants (Table 6). Of the identified food items, twigs of willow, birch, aspen, and alder, respectively, were most frequently eaten. Other foods, ranked by frequency, included fruit (catkins, seeds, or flower parts) , dry aspen leaves, spruce twigs, willow fruit, and dry leaves. A small quantity of unidentified herbaceous material was found in one sample. During the winter of 1971-72, total snowfall was 229 em in January. Examination of 10 samples collected during this time indicated that the diet again consisted primarily of deciduous woody materials (Table 7). Of the identified food items, willow twigs, birch twigs, birch fruit, alder, aspen, and cottonwood twigs, alder fruit, followed by dry birch leaves, green spruce needles, willow fruit and dry leaves, were the most frequently eaten foods. One sample contained a small quantity of unidentified sedges. The record snowfall during the winter of 1970-71, when 140 em more snow than fell during 1971~72, did not influence the frequency of occurrence of major food items found in stomach contents of moose. It is also interesting to note that during periods of greater than normal snowfall, when moose died from malnutrition, only one of the 44 samples contained spruce. It therefore appears that spruce does not constitute a major portion of the moose's winter diet near Fairbanks, Alaska. It also appears that spruce is not used as an emergency food by moose. During the spring of 1971, 15 samples of moose stomach contents were collected near Fairbanks. Analysis of these samples revealed that hardened twigs of willow and birch were most frequently eaten (Table 8). Other foods, in descending order, were fruit and dry willmv leaves, hardened aspen twigs, green willow leaves, hardened alder twigs, sedges, hardened Vaccinium twigs, and succulent willow twigs. · Six samples collected during spring 19 72 were analyzed. Hardened willow and birch twigs were found in all samples, and birch fruit and hardened alder twigs occurred in half the samples (Table 9). Seven samples were collected during the fall of 1971. Hardened twigs of willow and birch were followed in descending order by birch fruit, dry and green willow leaves, hardened aspen twigs, green birch leaves, hardened alder twigs, and Equisetum (Table 10). During the fall of 1972, 15 samples from the Fairbanks area con- tained hardened willow and birch twigs, dry leaves of willow and birch, hardened twigs of cottonwood and alder, and Equisetwn (Table 11). Ten samples were coll.ected during the summer of 19 72. All samples contained green willow leaves, while succulent willow twigs and Equisetum were found in nine samples. Six samples contained green shrub birch leaves, five grass, and four hardened birch twigs (Table 12). A more complete analysis and discussion of the data has been recently completed (Cushwa and Coady, 1973), and will not be nresented ·here. 15 Table 6. Moose stom'ach contents analysis from Game Management Unit 20A and 20B during winter 1970-71. Sample size = 44. Food Item Woody Identified Trees Aspen Birch Spruce sp. Tall Shrubs Willow sp. Alder Unidentified Herbaceous Unidentified Plant Part Pry leaves Hardened twigs Hardened twigs Hardened twigs Fruit Dry leaves Hardened twigs Fruit Hardened twigs Hardened stems Dry leaves Green leaves Dead stems I bark Dry sterns anq leaves 1 Relative frequency of occurrence. 16 Frequertcy1. 2.3 56.8 70.5 2.3 2.3 2.3 86.4 4.6 25.0 93.2 31.8 2.3 34.1 2.3 Table 7. Moose stomach contents analysis from Game Management Unit 20A and 20B during winter 1970-71. Sample size = 10. Food Item Woody Identified Trees Aspen Birch Cottonwood Spruce sp. Tall Shrubs Willow sp. Unidentified Herbaceous Unidentified Plant Part Hardened twigs Fruit Dry leaves Hardened twigs Hardened twigs Green needles Fruit Dry leaves Hardened twigs Hardened twigs Dry deciduous leaves Green deciduous leaves Dead twigs I bark Sedge 1 Relative frequency of 9ccurrence. 17 Frequencyl 20.0 70.0 10.0 90.0 20.0 10.0 10.0 10.0 10.0 100.0 40.0 5.0 10.0 10.0 Table 8· Moose stomach contents anaJ,ysis from Game Hanagement Unit 20A during spring 1971. Sample size = 15 .. Food Item Woody Identified Trees Aspen Birch Spruce sp. Tall Shrubs Willow sp. Alder Shrub birch Low Shrubs Vaccinium sp. Unidentified Herbaceous Unidentified Plant Part Succulent twigs Hardened twigs Hardened twigs Hardened twigs Fruit Green leaves Dry leaves Succulent twigs Hardened twigs Fruit Hardened twigs Hardened twigs Green leaves Hardened twigs Hardened stems Dry deciduous leaves Dead stems I bark . Grass Mushrooms Sedges 1 Relative frequency of occurrence. 18 Frequency1 6". 7 20.0 80.0 6.7 26.7 20.0_ 26.7 13.3 86.7 . 6. 7 13.3 6.7 6.7 13.3 100.0 13.3 33.3 6.7 6.7 13.3 Table 9. Moose stomach contents analysis from Game Management Unit 20A during spring 1972. Sample size = 6. Food Item Woody Identified Trees Aspen Birch Cottonwood Tall Shrubs Willow sp. Alder Unidentified Plant Part Hardened twigs Fruit Hardened twigs Hardened twigs Hardened twigs Fruit Dry leaves Hardened twigs Hardened twigs Green deciduous 1 Relative frequency of occurrence. 19 Frequency1 16.7 50.0 100.0 16.7 100.0 16.7 16.7 50.0 100.0 leaves 16.7 Table 10. Moose stomach contents analysis from Game Management Unit 20A and 20B during fall 1971. Sample size= 7. Food Item Woody Identified Trees Aspen Birch Tall Shrubs Willow sp. Alder Shrub birch Low Shrubs Vaccinium sp. Unidentified Herbaceous Idetiti fied Equiseturn Unidentified Plant Part Hardened twigs Fruit Green leaves Dry leaves Hardened twigs Green leaves Dry leaves Hardened twigs Hardened twigs Dry leaves Hardened twigs Green leaves Hardened stems Dry leaves Green leaves Dry stems and leaves Sedges Fruit Grass 1 Relative frequency of-occurrence. 20 Frequency1 28.6 42.9 28.6 14.3 71.4 4.2. 9 42.9 85.7 28.6 14.3 14.3 14.3 100.0 14.3 28.6 28.6 14.3 14.3 14.3 14.3 Table il. Moose stomach contents analysis from Game Management Unit 20A and 20B during fall 1972. Sample size = 15. Food Item Woody Identified Trees Aspen Birch Cottonwood Spruce sp. Larch Tall Shrubs Willow sp. Shrub birch Alder Low Shrubs Vaccinium vitis-idaea Ledum Arctostaphylos Unidentified Herbaceous Identified Equisetum Unidentified Plant Part Dry leaves Hardened twigs Green leaves Dry leaves Hardened twigs Green leaves Hardened twigs Dry needles Hardened twigs Green leaves Dry leaves Hardened twigs Dry leaves Fruit Green leaves Dry leaves Hardened twigs Green leaves Green leaves Dry leaves Hardened twigs Dry deciduous leaves Green deciduous leaves Grass Hush rooms 1 Relative frequency of occurrence. 21 Frequencyl 13.3 6.7 6.7 20 .o 66.7 6.7 20.0 6.7 6.7 6.7 40.0 86.7 13.3 6.7 6.7 6.7 20.0 6.7 6.7 6.7 86.7 20.0 53.3 20.0 6.7 13.3 Table 12. Moose stomach contents analysis from Game Management Unit 20A during summer 1972. Sample size = 10. Food Item Woody Identified Trees Aspen Birch Tall Shrubs Willow sp. Shrub birch Alder . Salix pulchra Unidentified Herbaceous Identified Equisetum Unidentified Plant Part Green leaves Succulent twigs Hardened twigs Green leaves Succulent twigs Green leaves Dry leaves Succulent twigs Green leaves Dry leaves Succulent twigs -Green leaves Succulent· twigs Green deciduous leaves Dead twigs I bark Grass 1 Relative frequency of occurrence. 22 Frequency1 10.0 10.0 40.0 100.0 90.0 60.0 10.0 10.0 10.0 10.0 10.0 20.0 20.0 10.0 90.0 50.0 Chemical Composition of Moose Rumen Contents and Energy Requirements and. Rumen Fermentation in Moose Findings have been reported in two previous publications (Coady and (;asaway, 1972; Gasaway and Coady, 1973), and will not be nresent.ed again at this time. However, an abstract of the most recent publication (Gasaway and Coady, 1973) is given below: A review of seasonal energy requirements and utilization of food by moose, with reference to other wild and domestic species, is presented. Energy requirements are difficult to estimate because no metabolic studies have been conducted with moose and comparative data from other wild and domestic species differ widely. It is assurn:ed that basal metabolic rate (BMR) conforms to the empirical relationship of wei~h t to metabolic rate, where BMR = 70 • 75 , and maintenance demands approximate 1. 7 times BMR. Energy requirements of female moose begin to increase significantly in Marcl). due to pregnancy and reach a peak of three to four times BMR in June, due to lactation and linogenesis. Major seasonal differences in rumen contents and estimates of food consumption by moose are described. Rumen fill in cow moose was greatest during early winter, lowest during late spring, and intermediate during summer. Percent dry matter was lowest during sunrrner and highest during winter. Washed rumen contents were higher in crude protein and lower in acid detergent fiber and lignin during sunrrner than during winter, reflect- ing the superior quality of summer forage. Estimates of food intake by moose vary greatly in the literature, although there is considerable evidence indicating that a greater quantitv of food is consumed during summer than during winter. Dry matter consumed by adult females was estimated to be three to four times greater during sunrrner than during winter. Increased rumen fill and decreased food intake during winter apparently result from slow passage of low quality food which restricts. additional food intake, and from voluntary reduction of forage consumption. Volatile fatty acids (VFA) produced by rumen microbes by fermenta- tion of dietary carbohydrates and prote,ins constitute approximately 57 percent of the digestible energy of rurrdnants. VFA production, which is directly related to food quality, was determined seasonally on free- ranging moose in interior Alaska using the "zero time rate" method. Production rates varied from a mean low of 18 eq VFA/ml rumen liquor/ hr durinp: winter to 60 eq VFA/ml rumen liquor/hr during summer. Moose undergo a large seasonal change in body ~•eight which corre- sponds closely to seasonal rates of VFA energy production. Metabolizable energy (ME), calculated from estimated VFA production, increased from 7,300 kcal/day in females during winter to 20,900 kcal/dav in lactating moose during summer. It was estimated that approximately 6,000 kcal/day of ME was required for BMR. During winter an estimated average of 3,900 kcal/day was obtained from catabolism of fat and protein reserves to meet the energy requirements not provided by forage, while during summer 7,600 kcal/day of fat and protein were deposited. 23 A review of effects of malnutrition on rumen function shows that a decrease in food quantity or quality depresses microbial populations and rates of fermentation. The value and practical application of using various parameters of rumen fi.mction to evaluate nutritional status of ruminants and quality of the habitat are discussed. ACKNOWLEDGMENTS Photographs and technical equipment for constructing the vegetation type map were generously supplied by the State of Alaska Lands Division. SpeCial thanks are due Larry L. Johnson, temporary, for exclusive prepa- ration of the type map. Dorothy Simpson, botanist and secretary, ADF&~, and Ronald Severns, temporary, provided expertise, initiative, and excellent company during long, sometimes uncomfortable days conducting vegetation surveys. Institute of Northern Forestry personnel, especially Charles Cushwa, Nonan Noate; and Les Viereck, provided necessary techni- cal and analytical assistance associated with vegetation sampling._ Les Viereck and Dorothy Simpson identified plant material from moose rumen contents. Numerous ADF&G employees, including Carol Ericson, Wayne Heimer, Edward Kootuk, Georganna Ranglack, John Trent, and Mike Vierthaler, assisted in collecting material at various times during energy require- ment and rumen fermentation studies. ADF&G biologists Albert Franzmann and Paul Arneson permitted use of the Kenai Moose Research Center facility and provided assistance irt testing radio collars and conducting energy requirements studies. . The above individuals have made this study both possible and enjoyable, and their valuable contributions are gratefully acknowledged. LITERATURE CITED Anderson, J. P. 1959. Flora of Alaska and adjacent parts of Canada. Iowa State Uni v. Press , Ames. 543 pp. Andreasen, G. E., C. Wahrhaftig, and I. Zietz. 1964. Aeromagnetic . reconnaissance of the east-central Tanana lowland, Alaska. US~S, Washington. 3 pp. Benninghoff, W. S. 1952. Interaction of vegetation and soil frost phenomena. Arctic 5:34-44. Bishop, R. H. 1969. Preliminary review of changes in sex and age ratios of moose and their relation to snow c9nditions on the Tanana Flats, Alaska. Paper presented at 6th Annu. N. Am. Moose Committee Meeting, Feb. 3-5, Kamloops, B. C. 14 pp. Xerox. Black, R. F. 1958. Lowlands and plains of interior and western Alaska in H. Williams, ed. Landscapes of Alaska. Univ. Calif. Press, Berkeley. pp. 76:...81. 24 Coady, J. W. and W. C. Gasaway. 1972. Rumen function and energy pro- duction of moose in interior Alaska. In R. B. Addison ed. Proceed- ings of the 8th North American Moose Conference and Workshop. Queen's Printer for Ontario. pp. 80-104. Cole, G. F. 1963. Range survey guide. Grand Teton Natl. Hist. Assoc. Moose, Wy. 22 pp. Cushwa, C. T. and J. W. Coady. 1973. Moose food habits in Alaska, a preliminary study using stomach contents analysis; MS 21 pp. Xerox. ____________ ., S. A. Liscinsky, R. F. Harlow and M. J. Puglisi. 1973. Rumen analysis in white-tailed deer management -development, use, and evaluation of a procedure. MS 40 pp. Drury, W. H!, Jr. 1956. Bog flats and physiographic processes in the upper Kuskokwim River region, Alaska. Contrib. Gray Herb. No. 178, Harvard Univ. 130 pp. Dutro, J. T. and T. G. Payne. 1957. Geological map of Alaska. USGS, Scale 1:2,500,000. Gasaway, W. C. and J. W. Coady. 1973. Review of energy requirements and rumen fermentation in moose and other ruminants. Int. Symp. on Moose Ecology, Quebec City. March 1973. 37 pp. Hanson, H. C. 1951. Characteristics of some grassland, marsh, and other plant communities in western Alaska. Ecol. Monogr. 21(4): 317-378. 1953. Vegetation types in northwestern Alaska and comparisons with communities in other arctic regions. Ecology 34(1) :111-140. Hulten, E. 1968. Flora of Alaska and neighboring territories. Stanford Univ. Press. 1,008 pp. Johnson, P. L. and T. C. Vogel. 1966. Vegetation of the Yukon Flats region, Alaska. CRREL Research Report 209. Hanover, N. H. 53 pp. Johnson, P. R. and C. W. Har.tman. 1969. Environmental atlas of Alaska. Inst. of Arctic Env. Eng. and Inst. of Water Res. Univ. Alaska, College. 111 pp. Larsen, J. A. 1965. The vegetation of the Ennadai Lake area, N.W.T.: Studies in subarctic and arctic bioclimatology. Ecol. Monogr. 35(1) :37-59. Ohmann, L. F. and R. R. Ream. 1971. Wilderness ecology: virgin plant communities of the Boundary Waters Canoe area. USDA Forest Serv. Res. Paper NC-63. 25 Ritchie, J. C. Paper 3. 1959. 56 PP· The vegetation of northern Manitoba. AINA Tech. Spencer, D. H. and E. F. Chatelain. 1953. of the moose in southcentral Alaska. Conf. pp. 539-552. Progress in the management Trans. 18th N. Am. Wildl. Viereck, L~ A. 1970a. Forest succession and soil development adjacent to the Chena River in interior Alaska. Arctic and Alpine Res. 2"(1) :1-26. 1970b. Soil temperatures in river bottom stands in interior Alaska. In Ecology of the subarctic regions, Proc. Helsinki Symp. (Ecol. and Conser. No.1). Paris, UNESCO. 364 pp. (in press). Wildfire in the taiga of Alaska. J. Quant. Res. 75 pp. Viereck, L. A; and E. L. Little, Jr. 1972. Alaska trees and shrubs. U.S. Dept. Agr. Handbook 410. 265 pp. Wahrhaftig, C. 1965. Prof. Paper 482. PREPARED BY : John W. Coady Game Biologist SUBMITTED BY : Richard H. Bishop Physiographic divisions of Alaska. Geol. Surv. U. S. Gov. Printing Office, Wa5h. 52 pp. APPROVED BY: ~~·smz~;:t ·Research Chief, Di visio of Game Regional Research Coordinator 26 State: Cooperator: Project Nos: Job No: JOB PROGRESS REPORT (RESEARCH) Alaska John W. W-17-4 W-17-5 1.8R Coady & Project Title: Big Game Investigations Job Title: Snow Characteristics in Relation to Moose Distribution in Interior Alaska Period Covered: July 1, 1971 to June 30, 1973 Sill1MARY This report describes major activities undertaken between February 1971 and June 1973. This research project was initiated in February 1971 during a winter of record snowfall in interior Alaska. Substantial winter mortality of moose permitted collecting numerous data on sex and age, body weights and measurements, reproductive status of females, and bone marrow fat content of moose in the Interior. Collection of these data has continued to the present as opportunity has permitted. Both NRC inscruments and the Swiss Rammsonde Penetrometer have been used since November 1972 to monitor physical characteristics of the snow cover at various sites in Game Management Units 20A and 20B. Frequent aerial surveys and ground track counts hav~ been conducted near snow study sites to relate moose movement and distribution to snow conditions. Although rarely used in biological studies, the ease, accuracy, and properties of the snow measured by the Rammsonde appear to make the instrument useful for assessing severity of snow conditions for large ungulates. A limited radio telemetry program initiated in November 1972 provided additional information regarding moose movement in response to snow conditions and habitat use, and indicated that telemetry may be a useful tool for studying certain aspects of moose behavior in interior Alaska. i Summary .. Background. Objectives .. Procedures. CONTENTS Production, Survival and Physical Condition of Moose . Snow Measurements. • •. Radio Telemetry. . • . • . . . . . • . . . Findings and Conclusions. . .....••.•. Production, Survival and Physical Condition of Moose • Moose-Snow Relationships • Radio Telemetry. • . . Acknowledgments . Literature Cited ••• BACKGROUND i 1 2 3 3 4 8 10 10 30 44 45 46 Moose (AZces aZces) are Holarctic in distribution (Rausch, 1963), occurring in several Eurasian countries, Canada, the conterminous United States, and Alaska. Four Eurasian races (Heptner et al., 1961) and four North American races (Hall and Kelson, 1959) of moose are recognized. The Alaskan moose (A. a. gigas) is of pre-Wisconsin age, remains having been found in Illinoian beds in the "Cripple Creek sump" near Fairbanks (Pewe and Hopkins, 1967). Moose are expanding their range throughout North America (Kelsall, 1972; Kelsall and Telfer, 1973). In Alaska, moose are widespread within the boreal forest zone (Alaska Dept. of Fish and Game, 1973), although many reports suggest they are more common beyond tree line in several areas than in previous years (Rausch, 1951; Bee and Hall, 1956; Lutz, 1960; Pruitt, 1966; Chesemore, 1968; and LeResche et al., 1973). North- ward dispersal of moose may be due in part to an increase in riparian climax vegetation resulting from gradual Holarctic warming trends during the last half century (Leopold and Darling, 1953). This apparent expan- sion may also result from recent increased human activity in northern regions, the-reby increasing opportunity for sighting animals (Kelsall, 1972). Moose play an important economic and cultural role in Alaska. Between 1963 and 1972 an average of 7500 moose per year were reported killed by hunters. In 1971 alone, over 8800 moose were reported killed, 1070 of which were from the Fairbanks region (Game Management Unit 20) . Thousands of recreationists photograph and observe moose each year. The extensive distribution, large numbers, and relatively visible nature of the species has made moose a symbol of Alaskan wildlife. Since statehood in 1959 the Alaska Department of Fish and Game has actively engaged in collecting information in interior Alaska relating to moose biology. Major emphasis has been pl-aced on extensive aerial surveys during fall to document changes in productivity, to assess the effect of hunting on bull-cow ratios, to determine survival of moose to 1 1.5 years of age, and to document ~hanges in relative abundance and distribution of moose. Annual aerial surveys have been conducted in selected areas during spring to assess survival of moose to 1 year of age, and to monitor changes in relative abundance and distribution of moose. Harvest statistics and basic life history information have also been collected. In spring 1966 a moose calf tagging program was initiated on the Tanana Flats south of Fairbanks to help determine distribution, movement, and population identity of moose calving in the area. Between 1966 and 1969 over 800 moose calves were tagged, and locations of tagged moose continue to be recorded as sightings are made. Major contributions to moose studies in interior Alaska have been made by Robert A. Rausch (Rausch, 1967, 1971; Rausch and Bishop, 1968) and Richard Bishop (Bishop, 1969, 1970; Bishop and Rausch, 1973). Together these individuals conducted and directed many early studies of moose in interior Alaska. Theirwork has provided the background for and continues to provide ·guidance in developing current moose studies. \ In February 1971, a winter of record snowfall in interior Alaska, a moose research program was initiated at Fairbanks to further id~ntify and quantify major control factors affecting moose populations. Basic emphasis of the program centers around effects of certain relationships between snow and range conditions on the production, survival, and dis- tribution of moose in interior Alaska. Information ·obtained from this project will hopefully contribute to efficient management of. moose in interior Alaska, where growth of human populations and increased demands on wildlife and land resources are expected. OBJECTIVES The objectives of this study are 1) to obtain certain data pertain- ing to production, survival, and general condition of moose in interior Alaska; 2) to evaluate the morphological adaptation to snow of various sex and age class moose; 3) to monitor distribution, survival, and productivity of moose in relation to char.acteristics of the snow cover in interior Alaska; 4) to test different instruments and techniques for evaluating the severity of winter conditions to moose; and 5) to determine the influence of snow and habitat on the seasonal movement and distribu- tion patterns of radio-collared moose breeding in the Little Chena drainage, and to test the feasibility of using biotelemetry to study movements of moose in interior Alaska. · The following data were collected: 1) Body weights and measurements of pre-and post-natal moose; 2) Chest height and foot load of post-natal moose; 3) Reproductive status of females over 1 year of age; 4) Percent femur marrow fat; 2 5) Sex and age of moose dying from both malnutrir.ion and unnatural causes; 6) Snow depth, hardness, density, temperature, and Rammsonde hard- ness at various sites in relation to distribution and abundance· of moose; 7) Visual s~arch time, location, habitat selection and activity of radio-collared moose. PROCEDURES Production, Survival and Physical Condition of Moose Moose dying from natural (malnutrition, predation) and unnatural (highway and railroad collisions, shooting) causes between February 1, 1971 and June 1, 1973 within 100 miles of Fairbanks were examined. Moose killed on highways were retrieved with a truck, while specimens from animals killed on the railroad were collected with a rail-car. Snow machines, fixed-wing aircraft, and helicopters were used to collect animals or specimens in remote areas. Private individuals also collected numerous specimens. Post-natal body weights were measured with a commercial truck in Fairbanks or with a spring scale_suspended under a helicopter. body weights were measured with a hand-held spring scale. ·The following measurements were obtained when possible: 1) Heart girth: measured immediately behind front legs; 2) Total length: measured along contour of body from hairless portion of nose to base of tail; scale Fetal 3) Total height: measured along contour of body from tip of hoof to top of shoulder or scapula; 4) Chest height: measured from tip of hoof to brisket or sternum, with leg in "normal" position relative to body; 5) Hind leg: measured from tip of hind hoof to hock or tuber calcis; 6) Knee: measured from tip of front hoof to middle of knee or carpus; 7) Foot load: determined by measuring area of four feet and divid- ing total area by body weight, according to the procedure described by Kelsall and Telfer (1971). Only heart girth and body con tour length were measured on pre-natal animals. 3 Reproductive tracts of females over 1 year old were examined for pregnancy. Ovaries were collected and prese:r:ved in formalin. Incisors were collected from animals over 1 year old, and the age determined by examining cementum growth layers. Percent fat in femur or tibia marrow was determined by drying marrow from fresh or frozen bones at 65°C to constant weight according to a method developed by Neiland (1970). Although marrow is composed of water, fat, and non-fat residue, the non-fat residue is small (2-6 per- cent of total weight), and the dried marrow is .. assumed to consist entirely of fat. Unless otherwise stated, 0.05 was the probability level at which. the null hypothesis was rejected. Snow Measurements Methods of measuring properties of a snow cover have been described by several authors (Klein et al., 1950; Benson, 1962; Keeler, 1969; Test Lab, 1970; and others). Two similar sets of instruments, a National Research Cbuncil of Canada (NRC) kit and a USA Cold Regions Research and Engineering Laboratory (USACRREL) kit have been used to measure tempera- ture, density, and hardness of a snow cover. However, modified snow study kits have also been used (Richens and Madden, 1973). These measurements are generally obtained in "pit" studies, in which a trench is dug in the snow to ground level. Thickness, temperature, density, hardness, and sometimes crystal type and size of major strata are recorded; Snow samples or measurements are generally taken iri a horizon- tal plane, although vertical as well as horizontal hardness measurements should be made (Pruitt, 197la). The International Workshop on Rangifer Winter Ecology (Pruitt, 197la) has recommended measuring the thinnest distinct strata which a given instrument size will allow. Depth of snow cover is probably the most common and important measurement obtained in studying snow ecology of moose. In addition to ease of measurement, snow depth provides a measure of the thickness of the medium through which an animal must move if not supported by the snow .. Depth of snow or thickness of strata can be measured with any conveniently calibrated probe. In addition, in interior Alaska 5 em diameter stakes, clearly calibrated at 30.5 em (12 in) intervals, have been permanently located in remote or inaccessible areas to measure snow depth from fixed wing aircraft. Depth can consistently be estimated within 3 to 5 em of the actual snow depth, although this method has the disadvantages of limiting measurements to the number of stakes at a study site, and limiting study sites to relatively open areas where the aircraft may be flown near ground level. Snow depths relative to anatomical features of moose also provide a useful and reasonably accurate estimate of snow depth from the air. Snow temperatures of each substrate or at intervals on the pit wall, beginning at ground level or in the subnivian space and ending within 2 em of the surface, may be measured. However, the International 4 Workshop on Rangifer Winter Ecology (Pruitt, 197la) has questioned the significance of measuring snow temperature in Rangifer research, and has recommended limiting measurements to ground level and air temperatures. While a relationship between snow temperature and moose behavior has been alluded to by Des Meules (1964), the ease and speed of temp-erature measurement in pit studies and its possible signtficance to moose behavior probably merit its continued measurement. Measurements may be made with either an alcohol or bimetallic thermometer. Density, determined by weighing a known volume of snow, is probably the most widely used index of snow type, and under certain snow condi- tions it may be correlated with hardness (Keeler and Weeks, 1967; Bilello et al., 1970). Increased density presumably causes increased drag on legs or the body of ungulates during movement, and thereby inhibits locomotion. Several snow cutters for obtaining snow density samples are available. The NRC kit employs two 250 ml snow cutters; one for soft and one for hard snow, while the USACRREL kit uses 500 ml sampling tubes. Swedish workers use a Swedish Army density "box" which reportedly gives reliable results because of its large 1,000 cm3 volume (Fruit t, 19 7la) . In interior Alaska a 650 cm3 plastic cylinder has been used with · satisfactory results. Snow hardness reflects the degree of bonding between crystals and in most types of snow increases as density increases and/or snow tempera- ture decreases (Gold, 1956). In addition to snow depth, hardness is probably the most critical parameter of a snow cover to ungulates since it reflects the force which must be exerted to move legs or body through the snow, and the ability of the snow to partially or fully support the animal. NRC and USACRREL snow hardness gauges are similar, and consist of a spring-loaded push rod with provision for mounting discs of different areas on one end and a calibrated gauge on the other end. The disc is pressed against a snow surface and the maximum stress associated with the initial collapse of the snow structure is noted on the gauge. By using a high and low range gauge and different sized discs, hardnesses of 0 to 100,000 g/cm2 can be measured. Useful modifications to the instrument developed in Sweden and.now used by Pruitt (per. comm.) include a ratchet which retains the calibrated gauge at the extended position reached at the instant of snow collapse, thereby providing more accurate measurements. Snow hardness may ·also be measured with a cone penetrometer, commonly referred to as a Swiss Rammsonde. The instrument and its use have been described by several workers (Bader et al. , 19 39; Benson> 1962; Keeler, 1969; Test Lab., 1970; and others). Basically, it consists of a hollow steel shaft with a 60° conical tip 4 em in diameter and 3. 5 em high. A solid rod mounted on top of the steel shaft guides a hammer . which is dropped from a measured height. The height of the drop, weight ·of the entire instrument, .and depth of penetration may be related to the resistance of snow to penetration by·the cone using the following formula:. R = Whn + W + Q, where R = ram hardness number or resistance --x to penetration, W = weight (kg) of penetrometer. The equation ignores friction between the cone and snow arid elasticity in the penetrometer. 5 However,.error is small, especially for snow of relatively low hardness (Keeler, 1969; Benson, per. comm.). The ram hardness number, R, indicates the resistance (kg) of a layer of snow to penetration by the cone of the penetrometer. In prac- tice it may be useful to determine the ram hardness value for the total snow depth. To do this an integrated ram hardness number, Ri (kg-em), is calculated by multiplying each depth increment (em) times its ram hardness m.nnber,. R (kg), and summing the values from the snow surface to ground level or to any given depth. The irttegrat·ion indicates the work done by the penetrometer as it moves through the snow to a given depth. The Rammsonde can be used in several ways. It can rapidly distin- guish different strata and provide a ram hardness profile or an integrated ram hardness value of the total snow cover without digging a pit. Rammsonde measurements are less subject to operator error than are NRC or USACRREL snow hardness values (Benson, 1962). Limitations to this procedure include lack of resolution at low snow hardness. However, a modified 120° cone, 10 em in diameter, has been satisfactorily used by Abele (1968) in soft snow.. The hardness number obtained with· the large cone is divided by_ a factor of 10 to obtain the approximate ram· hardness value of the standard cone (Test Lab, 1970). Rammsonde hardness numbers can be correlated with other derived snow measurements under most temperature and snow conditions, and with the support capacity of snow. Bull (1956) and Keeler (1969) have correlated ram hardness values with snow densities, while Abele (1963) has related ram values to compression strength or hardness. Abele et al. (1965, ·1968), Wuori (19o2, 1963) and others have related ram hardness to vehicle support capacity of snow roads and runways. While the Rammsonde penetrometer has been used extensively to study mechanical properties of a snow pack, it has been used rarely in biologi- cal applications. Other than Lent and Knutson (1971) working with musk- oxen (Ovibos moschatus), Coady (this report) working with moose, and some investigators in Canada and Scandinavia working with reindeer (Rangifer sp.) (c.f. Pruitt, 197lb), the instrument has not been widely used in studies of ungulate snow ecology. Additional measurements and experience are certainly required to evaluate the usefulness of the Rammsonde penetrometer in biological studies. Ease, speed, and consis- tency of identifying stratigraphic horizons and measuring hardness of a snow cover may make the Rammsonde a potentially valuable instrument for assessing the resistance to movement through or the support capacity of snow. Both NRC and Rammsonde snow measurements were recorded at each station on a data form (Fig. 1). Site characteristics, weather, general snow conditions, presence or absence of moose in the vicinity, and if moose are present, the depth of foot penetration into the snow were recorded. NRC measurements consisted ofweight of snow, hardness, temperature, and crystal type (optional). in each snow layer on the pit wall. Rammsonde measurements were obtained by noting weight of hammer, 6 Fig. 1. Alaska Department of Fish and Game data form used to record National Research Council (NRC) of Cailada and Swiss Rammsonde Penetrometer snow measurements. ALASKA DEPARTMENT OF ·FISH AND GAME SNOW SURVEY Station Altitude Exposure Vegetation .Date Time Observer Air Temp Weather Snow Depth Snow Surface Moose Present Track Depth NRC Measurements Depth Gross Wt Density_ Hardness Temp Crystals Remarks Rammsonde ·Measurements Penetration Resistance Wt Ht (1) (2) (3) (4) (5) (1) (2) (3) (4) ·(5) 7 height of hammer drip, and total depth of Rammsonde penetration after each hammer drop (recorded in a vertical coltnnn). Ramms.onde measurements were repeated five times at each station. The resistances for each increment of Rammsonde penetration and for each trial were calculated at a later date. Trends in snow severity for moose in interior Alaska were examined by plotting weather bureau records of snow depth on the ground in Fairbanks versus month throughout the winter, and measurtng the area under the curve with a planimeter. A similar procedure could be used to evaluate temperature severity. While weather bureau measurements may not indicate actual conditions on winter moose range, they probably reflect relative differences between years and long:..term trends. Depth, temperature, density, .hardness (NRC hardness gauge, and Rammsonde hardness of the snow cover) and relative abundance of moose were monitored at weekly to monthly intervals in different habitats in the Chatanika drainage, the Chena drainage, the Tanana Fl.its, and the north side of the Alaska Range foothills. The major study area is located near 40 mile Steese Highway (Chatanika drainage), where rela- tionships between altitudinal moose movements and snow conditions were monitored during the winters of 1971-'72 and 1972-73. The seven day. total (counted daily) of fresh moose tracks crossing a one-half mile long transect in the Chatanika River valley was recorded between November and May. The valley is located at 245 m elevation, and represents typical winter ·riparian moose habitat. The number of moose counted during frequent intensive aerial surveys in a 75 km 2 drainage above the_ valley transect was also recorded. The upland site ranges from 550 to 670 m elevation and consists of mixed conifer and deciduous trees and shrubs which characteristically support modest numbers of moose during summer and fall in interior Alaska. Snow measurements were obtained in shrub communities in the upland site, and :in shrub, deciduous tree, conifer tree, and grass communities in the valley site. Radio Telemetry Six a:dult female moose were radio collared in the 15-year-old Elliott Creek burn along the upper Little Chena River (elevation 300- 700 m) during November 1972. Collaring procedures and transmitter specifications are on file at the Alaska Department of Fish and Game office in Fairbanks, and will not be reported here. Collared animals were located from the air at two-to ten-day intervals, and the location of individual moose was plotted on one inch to one mile USGS maps or on four inch to one mile vegetation type maps. Habitat selection (herbaceous bog, heath bog, shrub, deciduous tree, conifer tree), activity (bedded, standing), and time spent visually searching for the animal after her location was determined electronically were recorded on data forms (Fig. 2). Snow depth in the Elliott Creek burn was monitored from the air using permanently located calibrated stakes. Monthly aerial moose surveys were conducted in the Elliott Creek burn ·between November and March. 8 Fig. 2. Alaska Department of Fish and Game d·ata form used to record observations of radio collared ti:10ose in GMU 20A and 20B. RADIO~COLLARED MOOSE OBSERVATIONS Area Snow Date Pilot/Observer Temp Weather Wind Animal 1 2 3 4 5 6· Detect Signal? Locate? Visually Time of Day Visual Search Time Habitat Activity Grouping 9 FINDINGS AND .CONCLUSIONS Production, Survival and Physical Condition of Moose Weights and. Measurements: Weights and measurements were obtained from 22 moose fetuses collected in interior Alaska (Table 1). Date of death is recorded to the nearest day or month, when known. Five fetuses, ranging in weight from 13.6 to 17.2 aDd averaging 15.2 kg, were collected on the Tanana Flats on May 24, 1971. Calving was underway. at this time, and therefore the above weights are considered to represent the approxi- mate birth weight of moose calves in the area. Growth rates for moose fetuses have been recorded by Edwards and Ritcey (1958) in British Columbia, by Rausch (1959) in Alaska, and by Markgren (1969) in Sweden. Growth rate of fetuses reported in Table 1, based on total contour length and date of death,when known, agree closely with that reported by Rausch. Birth weights of 6 to 16 kg have been reported for moose in the USSR (Knorre, 1961); of 11.2 kg for moose in Michigan (Verme, 1970)·, and of 11.3 to 15.9 kg for moose in Alaska (Rausch, 1959). Markgren (1966) measured weights of 11.5 and 13.5 kg for two 4-day-old moose in Sweden. The average weight of 15.2 kg reported in this study for near-term moose fetuses agrees most closely with the findings of Rausch; and is somewhat greater than the weights recorded by other investigators. Knorre (1961) found an average birth weight ot 12.8 kg for single and 10.3 kg for twin calves. Individual weights ·of the one set of twins near term in this study were similar to weights of the three single calves (Table 1). Weights and measurements were obtained from .approximately 25 moose calves under 1 year of age (Table 2). Based on limited data presented in this report, birth weight (see Table 1) increased over 2. 5 times by four weeks, over 4.5 times by six weeks, and by approximately 10 times by 6 to 12 months of age. The extremely rapid growth rate of moose calves has been noted by many workers. Rausch (1959) also reported that moose calves in Alaska increase their birth weight by a factor of 10 after six months. Verme (1970) found that calves in Michigan tripled their birth weight by.four weeks and weighed 159 kg by six months of age. In Alberta Blood et al. (1967) found average weights of 197 and 174 kg for male and female . calves, respectively, after six months. Weights of moose calves have also been reported by Denniston (1956), Dodds (1959), Knorre (1961) , Peek (1962), and Markgren (1966). 'Earlier studies have been summarized by Peterson (1955). Weights and measurements from approximately 55 moose older than 1 year were obtained during winter and spring (Table 3), sununer (Table 4), and fall (Table 5). During each period maximum body weight and skeletal measurements were reached after approximately three years of life (Figs. 3-7) • These age specific data suggest that moose, at least females, in 10 Table 1. Weights and measurements of moose fetuses from interior Alaska (GMU 20A and 20B). Total Contour Length Heart Girth Date of Age of Sex Wt (k~} (em) (em) Death Cow F 0.64 33.0 20.3 12/70 4 M 3.86 66.0 38.1 3/12/71 11 F 2.28 52.7 31.1 2 F 1.81 45.7 14 M 2.94 63.5 34.3 3/71 14 F 0.79 29.8 18.4 10 M 1.02 49.5 30.5 11 F 4.08 62.2 33.7 3/71 13 ~ (twin) 1.36 41.9 26.7 (twin) 1.13 40.6 27.2 r; (twin) 0.57 33.7 18.4 13 I ·F t~ (twin) 0.91 33.7 19.1 M 3.40 62.2 34.3 3/20/71 11 M 4.98 69.9 40.6 4/8/71 15 F 0.24 26.6 10 M 3.44 61.1 33.0 16 F 4.54 69.9 38.7 16 F 17.23 106.7 55.9 5/24/71 11 F 13.61 100.3 53.3 5/24/71 M 14.51 96.5 53.3 5/24/71 2 G (twin) 15.87 96.5 53.3 5/24/71 8 F (twin) 14.97 95.3 52.1 5/24/71 8 11 Table 2. Weights (kg) and measurements (em) of moose calves from interior Alaska (GMU 20A and 20B). Age and Measurement No. Mean s. D. S. E. Range 4 weeks Weight 2 41.5 . 28.6-54.4 6 weeks Weight 1 70.0 Heart girth 1 88 Total length 1 128 Shoulder height 1 86 6-12 months Weight 23 157 25 5 113-238 Heart girth 17 143 15 4 127-186 Total length 22 204 11 2 173-213 Shoulder height 21 148 16 4 122-194 Chest height 21 84 8 2 66-95 Hind foot length 20 63 5 1 51-74 12 Table 3. Weights (kg) and measurements (em) of male and female moose over 1 year of age during winter and spring in interior Alaska (Gt-ru 20A and 20B) . ·Age and Measurement No. Mean S. D. S. E. Range 1 ear Weight 1 254 Heart girth 1 203 Total length 1 254 Chest height 1 94 Hind foot length 1 69 2 ears Weight 2 356 295-417 Heart girth 3 195 19 .11 173-211 Total length 4 275 18 9 262-302 Total height 5 183 13 6 163-:-198 Chest height 5 101 4 2 97-105 Hind foot length 6 71 4 2 66-76 3 ears Weight 1 279 Total length 1 254 Shoulder height 1 175 Chest height 2 97 97-97 Hind foot length 2 71 71-71 4 ears Weight 2 368 361-374 Heart girth 2 196 194-198 Total length 2 271 270-272 Total height . ·2 181 17 5-187 Chest height 2 103 101-104 Hind foot length 2 82 79-85 6 ears Weight 1 385 Heart girth 1 188 Total length 1 254 Shoulder height 1 183 Chest height 2 93 91-94 Hind foot length 2 65 61-69 13 Table 3. Continued. Age and Measurement No. . Mean S. D • S. E. Range 7-8 ears Weight 2 368 361-374 Heart girth 2 192 192-192 Total length 2 277 274-280 Total height 1 177 Chest height 1 108 Hind foot length 1 77 10-11 years Weight 5 351 69 34 266-435 Heart girth 3 177 4 2 173-180 Total length 2 281 269-292 Shoulder height 6 182 12 6 163-196 Chest height 5 102 5 2 97-109 Hind foot length 5 63 5 3 61-71 13+ ears Weight 2 325 313-336 Heart girth 3 179 8 4 175-188 Total length 4 274 13 7 259-290 Shoulder height 5 179 11 5 165-193 Chest height 5 100 2 1 97-102 Hind foot length 12 68 4 1 61-74 14 Table 4. Weights. (kg) and measurements (em) of cow moose over 1 year of age during late June in interior Alaska (GHU 20A). Age and Measurement 1 ear· Weight Heart girth Total length Shoulder height Chest height 2 ears Weight Heart girth Total length Shoulder height . Chest height 3 ears Weight Heart girth Total length Shoulder height Chest height 4 ears Weight Heart girth Total length Shoulder height Chest height 6 ears Weight Heart girth Total length Shoulder height Chest height 10-11 years Weight Heart girth Total length Shoulder height Chest height No. 1 1 1 1 1 3 3 3 3· 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 Mean 15 227 156 203 151 92 395 177 265 180 112 463 204 272 188 110 390 172 273 184 113 367 186 284 188 105 413 191 269 182 107 ·Range 381-408 174-182 242-278 177-183 106-118 376-449 188-194 268-269 174-190 106-107 Table 5. Weights (kg) and-measurements (em) of male and female moose over 1 year of age during October in interior Alaska (GMU 20A and 20B). Age and Measurement No. Mean Range 3 ears Weight 1 490 Heart girth 2 216 202-230 Total length 2 295 276-314 Shoulder height 1 209 Chest height 1 107 Hind foot length 1 88 4 ears Weight 3 524 501-546 Heart girth 3 213 210-214 Total length 3 . 289 270~306 Shoulder height 3 188 183-192 Chest height 3 111 107-118 5 ears Weight 1 512 Heart girth 1 186 Total length 1 304 Shoulder height 1 192 Chest height 1 106 Hind foot length 1 79 7 ears Weight 1 458 Heart girth 2 209 204-214 Total length 2 275 260-289 Shoulder height 1 172 Chest height 1 108 Hind foot length 1 82 10-11 years Weight 1 517 Heart girth 3 208 204-212 Total length 3 306 297-311 Shoulder height 1 180 Chest height 1 106 Hind foot length 2 86 83-89 16 Fig. 3. Hind foot and total body length of male and female moose obtained in GMU 20A and 20B during winter and spring, 1971 to 1973. 300~~--~----4 CD CD CD 200 E (.) _.., ..c +- Ol c (J) _J 100 0 +- ~ +- 0 ~ '"0 c I CD ® ® (I I l. 3 &-8 Age (Year.s} 17 - Fig. 4. 400 Heart girth and total body weight of male and female moose obtained in GMU 20A and 20B during winter and spring, 1971 to 1973. CD CD CD 0 .... _ _,_ ___ _,_ ___ ..&... ___ ....__, I I I I I ,_I_.___. I i I I I 2 3 ,_8 I0-1/ I 3+ Age (Yea. r s) I 18 19 Fig. 6. Total body weight and body length of male and female moose obtained in GMU 20A and 20B during fall, 1971 and 1972. f f (j) ® 300 ¥ ,...... s (J '-' p:: ~ z 1>-4 ...:! 200 ~ E-< 0 E-< ' 100 ~l '1-_j I I I ~ CD (j) l ([) CD J 400 I ,...... i blJ ..,: I E-< I l t5 ' H ~ ~ >-< I I t=l ! 0 ~ i I ~ 1 200 l I ! i I I I{ I 1/ I 3 4 5 7 . /J 10-11 AGE (YEARS) 20 s tJ p:: ~ z r.-:1 .... N f-' !;;;1 t) E-< I i I I I 1 2 3 5 AGE (YEARS) 6 10-11 13+ N 0 > ~ p.. N 0 bd the study area may reach maximum body mass and skeletal dimensions dur- ing their third year. Body mass or weight is a widely used measure of growth rate and indicator of physical condition (Klein, 1970). Nevertheless, skeletal measurements better reflect growth rate since they are not subject to short-term fluctuations associated with fullness of the digestive tract, seasonal accumulation .and utilization of fat reserves, etc. (Klein, 1964, 1965). In addition, skeletal measurements are generally easier to obtain than body weight from large animals. Total length, and to a lesser extent hind foot length, and shoulder height best reflected a pattern of age specific growth. Heart girth involves measurements of the skeleton as well as overlying muscle and fat, and therefore reflects the mass of the animal more closely than skeletal growth (Brandy et al., 1956). Chest heights were recorded to evaluate the potential performance of the animal in deep snow for this study. Moose have been reported to reach maximum body size within two to four years in other areas. Blood et al. (1967) noted that moose in Alberta increased in weight until three and one-half years of age, or perhaps longer, while Verme (19 70) found that females in Hichigan reached maximum size in two and. one-half years, and males increased in size until three and one-half years old. Rausch (1959) found that female moose in southcentral Alaska grew very little beyond wear age class III (two to four years). However, other workers have reported that moose grow in both weight and skeletal size for several years. Skunke (1949) noted the moose in Sweden increase in weight until 7 to 10 years of age, and LeResche and Davis (1971) indicated that growth of moose on the Kenai Peninsula in Alaska may continue throughout the life of the animal. After reviewing the European and American studies, Jordan et al. (1970) concluded that body growth continued until year six in females and year 11 in males. In interior Alaska maximum body weights occurred during fall. The average weight for two bulls and five cows collected during mid-October, well into the rut, was 507 kg. The heaviest animal was a 4~year-old bull weighing 546 kg, while the next heaviest was a 4-year-old cow weighing 524 kg (Table 5). Body weights of moose reported from other areas are generally less than those cited here. The heaviest bull and cow weighed by Blood et al. (1967) in Alberta were 478 and 429 kg, respectively, vJhile Verme (1970) found a maximum weight of 506 and 385 kg for a bull and cow in Hichigan, respectively. Rausch (1959) noted that few bulls exceed 545 to 635 kg in Alaska, and reported a weight of only 517 kg and a body length of 287 em for the largest bull he examined. LeResche and Davis (1971) reported a weight of 445 kg for the largest cow handled at the Kenai Hoose Research Center, Alaska. Moose in interior Alaska apparently experience considerable seasonal 22 .fluctuation in body weight. Average body weights for moose older than three years (see Tables 3, 4, 5) weue 338 kg in December through May, 409 kg in June, and 507 kg in October. These values indicate a 33 per- cent weight loss in animals between fall and winter and spring. Although sample sizes were not large, and some of the weights obtained during winter and spring were from "winter..:..killed" moose, it is apparent that substantial seasonal fluctuations in weight of moose in interior Alaska normally do occur. Jordan et al. (1970) concluded that seasonal weight fluctuations in moose amounted to 6.6 percent for females and 10.3 percent for males. Ho~ever, Rausch (1959) and LeResche and Davis (1971) reported seasonal weight fluctuations of 20 percent and '15-30 percent, respectively, for moose in southcentral Alaska. Verme (1970) found that a "winter...;killed" bull in Michigan had lost 33 percent of his pre-winter weight. Energetic demands of lactation resulted in a substantially lower body weight during the summer than for non-lactating females (Table 6). Average body weight of four non-lactating females was approximately 12 percent greater than that of lactating females. Each group included 2-year-old animals. Energy requirements for lactation are discussed in greater detail in a recent publication (Gasaway and Coady, 1973). Sequential weights of cows at the Kenai Moose Research Center indicated a weight loss of 8 to 18 percent associated with calving and rearing young (LeResche and Davis, 1971). Rapid growth rates and large body size of moose reported in this study suggest favorable range conditions. Growth of ungulates in both body mass and skeletal dimensions is closely related to plane of nutrition (Brandy et al., 1956). Maximum size and growth rates have been positively correlated with quantitative and qualitative aspects of the food supply for numerous ungulates, including reindeer (Klein, 1964, 1968), apd black-tailed deer (Odo~oiZeus hemionus coZumbianus) (Cowan and Wood, 1955). LeResche and Davis (1971) indicated that low growth rates and small body size of moose on the Kenai' Peninsula may be due to relatively poor range conditions. Klein (1968) further suggested that quantitative aspects of the food resource was important in determining growth rate andbody size. Thus, range "carrying capacity" must consider winter periods when physiological requirements of animals are low and food supply is critical, and summer periods when physiological demands for reproduction, fat deposition, and body growth are high and high quality food is important. Klein (1965) has discussed the annual nutritional cycle of deer. Winter nutritional requirements, though reduced, are generally in excess of available forage quantity, resulting in a gradual weight loss through- out the winter. During summer nutritional requirements are relatively great although high quality forage may be relatively abundant. This physiological cycle is apparently controlled largely through genetic mechanisms, since Wood et al. (1962) reported that captive black-tailed deer fawns stopped growing and adults lost weight during fall and winter, even though a high quality ration was available throughout the year. 23 Table 6. Weight and total length of lactating and non-lactating cow moose over 2 years of age during June in interior Alaska (GMU 20A) . Non-lactating Weight Total length Lactating Weight Total length No. 4 4 No. 4 4 24 Mean 429 264 Mean 379 276 Range 395-463 242-275 Range 367-390 269-284 Large fluctuations in body weight of adult moose reported in this study may also reflect adequate nutritional status and favorable range conditions. Nordan et al. (1968) reported that captive black-tailed deer maintained on a low plane of nutrition exhibited smaLler annual fluctuations ln body weight than deer fed a high quality ration. The magnitude of the a.nnual cycle was apparently related to the amount of fat deposition during summer months which determines the ability of the animal to effectively .participate in reproductive activities and its success in surviving the winter. Reproduction: Thirty female reproductive tracts collected between February and May 1971 from moose two years old or older were examined. Twenty-one (70%) of the animals were pregnant. Of 13 females obtained from the inhabited Fairbanks area, only seven (54%) were pregnant. Fourteen (82%) of the remaining animals, obtained from the uninhabited Tanana Flats were pregnant. Three of five (60%) adult females examined during April 19 73 on the Tanana Flats were pregnant. No yearlings examined at any time were pregnant. Ovarian analyses from approximately 70 female moose older than calves have not been completed at this time. Pregnancy rates observed in this study during the winter of 1970- 1971 and during April 1973 were somewhat lower than those reported by other workers. Edwards and Ritcey (1958), Pimlot t (1959), Rausch (1959), Simkin (1965), Rausch and Bratlie (1965), and Houston (1968) reported average pregnancy rates of 75, 81, 94, 87, 90, and 90 percent for females older than yearlings in British Coltnnbia, Newfoundland, Alaska, Ontario, Alaska, and Wyoming, respectively. Edwards and Ritcey (1958) included animals collected during October in their calculations, and therefore may not have detected some pregnancies or pregnancies resulting from later oestri. Pimlott (1959) reported variation in pregnancy rates of 74 to 87 percent among moose from different areas of Newfoundland. A similar range of pregnancy rates has been reported by Cheatum and Severinghaus (1950) for white-tailed deer (OdocoiZeus virginia) in New York. Low pregnancy rates observed in this study are difficult to explain. There is little evidence in the literature for resorption, abortion, or stillbirth among cervids. Only two out of 283 reproductive tracts examined by Pimlott (1959) indicated resorption, while Rausch (1959) found no definite indication of mortality among embryos or fetuses in Alaskan moose. However, resorption in moose has been recorded by several workers in the USSR, particularly following severe winters (in Markgren, 1969). Thus, the possibility of intrauterine death contribut- ing to the low pregnancy rates in the Fairbanks area must be considered. Another potential cause of low pregnancy rates may be due to non- breeding of females. While this has never been implicated as a factor contributing to low calf production in Alaska, the possibility of females not being bred under certain situations, either because of low bull:cow ratios or because of continuous anoestrus during the normal breeding season, cannot be discounted. 25 Of 22 fetuses examined between February and May 1971, 15 were female and 7 were male. Sex ratios at birth for most species of mammals are generally near 1:1, or slightly in favor of males (Robinette et al. , 1957). However, among some species under favorable environmental condi- tions, males tend to predominate. Rausch (1959) listed several factors which may influence sex ratios at birth, including age structure of the population, reproductive history of the female, nutritional status before and after birth, and severity of the winter. Generally under adverse conditions males tend to suffer the greatest intrauterine mortality, and this factor, considering the severity of the 1970-1971 winter, may have been responsible for skewed fetal sex ratios observed here (assuming intrauterine mortality occurs in moose). On June 27, 1972 a female moose fetus was collected from a 3-year- old cow on the Tanana Flats (Coady, 1973). Based on fetal growth rates reported by Rausch (1959), measurements (weight, 6. 71 kg; girth, 42 em; total length, 70 em) indicated that the fetus was approximately six to eight weeks from term. Since the gestation period for moose in North America is about 240-246 days (Peterson, 1955; Rausch, 1959) , conception of the above fetus may have occurred during December. Normally, the peak of conception occurs around the first of October for moose in Alaska (Rausch, 1959). Markgren (1969) reported that although births during April and early May are rare in Sweden, calving during midsummer is quite common, and births during August, September, and even October have been recorded. Although occasionally reported verbally, the only written account known to this author of a late pregnancy in North American moose was of a pregnant cow shot in Alaska on the lower Tanana River in September (Davis, 1952). Late pregnancies indicate conceptions decidedly later than during the main periods of oestrus. Recurrence of oestrus in moose has been reported by Edwards and Ritcey (1958), Rausch (1959), and Markgren (1969) • Data presented by Edwards and Ritcey (1958) indicate an interval of approximately one month between successive oestri. Males are apparently capable of breeding for a considerable period during the fall since Rausch (1959) found spermatozoa in the epididymis from mid-August to December. Thus, the late pregnancy observed in this study suggests conception during a third or fourth oestrus. Several factors contributing to conception after an initial oestrus have been suggested. Russian data (in Markgren, 1969) indicate that the "heat" during which the cow will allow copulation lasts less than 24 hours, while Altmann (1959) reports for American moose that the bull stays with a cow for only 7 to 12 days. Therefore, a bull conceivably may not be available for mating during an initial oestrus. Other Russian information cited in Markgren (1969) indicates that young (primarily yearlings) and nutritionally deficient cows may have delayed oestri. Although exceptions to and irregularities in most physiological processes can be found, "normality" of nearly all individuals is the 26 rule. Among moose populations the synchrony and uniformity of oestrus cycles and subsequent calving among the overwhelming majority of the population under widely varying environmental conditions remains the most remarkable phenomenon. Femur Marrow Fat: Percent fat in femur bone marrow was analyzed from 65 adult and 61 calf moose examined between February and May 1971 (Fig. 8). A significant difference (P< 0.01) in marrow fat between adults dying from malnutrition and from unnatural causes (road and rail- road accidents, shooting, etc.) was found, while a similar comparison between calves revealed no significant difference (P> 0.50). Duncan's new multiple range test indicated no significant difference in marrow fat between moose dying from unnatural causes in remote areas and from apparent wolf (Canis lupus) predation, and between moose shot in May and from malnutrition (Fig. 9). Marrow fat of animals dying from apparent wolf predation, unnatural. causes near Fairbanks, and shooting in May were intermediate in value and not significantly different. Moose which appeared to be in the poorest physical condition were selected for shooting. Quantative analysis of depot fat has frequently been used as a criterion of physical condition of animals, and various procedures for determination of total body fat, subcutaneous fat, abdominal fat, perinephric fat, and bone marrow fat have been reviewed (Cheatum, 1949; Riney, 1955; Smith, 1970). Riney (19.55) noted that deposition or mobilization of fat, in response to favorable or unfavorable. conditions, occurs first in subcutaneous depots, followed by abdominal, perinephric, and bone marrow depots. Thus, fat mobilization from bone marrow tissue probably occurs only after other fat reserves have been largely depleted. Marrow fat values have been used to evaluate physical condition of white-tailed deer (Ransom, 1965), elk (Cervus canadensis) (Greer, 1968, 1969), moose (LeResche, 1970), pronghorn (AntiZocapra americana) (Bear, 1971), and caribou (Rangifer tarandus) (Miller, 1972). Greer (1968) noted that elk in excellent physical condition have marrow fat values of 82 to 99 percent, while marrows from "winter-killed" elk contain less than 1 percent fat. Mech (1970) found that only 15 percent of wolf-killed adult moose examined on Isle Royal had low marrow fat content. Relatively low mean marrow fat content among all groups of adult moose dying from unnatural as well as natural causes (Fig. 9) reflects the severity of the 1970-19 71 winter and the poor physical condition of most animals. However, it is interesting to note that moose dying from unnatural causes in remote areas had both the highest marrow fat and the highest pregnancy rate of any group of animals. While marrow fat from adult moose collected in this study clearly reflects a difference in nutritional status of the animal, samples from calves were uniformly low (Fig. 8). Among young animals Abrams (1968) reported that the sequence of tissue development is nervous system, skeletal system, musculature, and finally fat. Thus, fat is apparently a "luxury" which moose calves in interior Alaska, if anywhere, cannot afford. 27 Fig. 8. Average.percent fat (circles), standard deviation (triangles), and range (bars) in femur bone marrow of male and female moose dying from malnutrition and unnatural causes in GMU 20A and ~ 0 lL ~ ~ ~ ~ 20B between February and May, 1971. Number in circles indicate sample size. lOOp· --------------------~ AJu. I u t 80 :·~o--~ r ~ ~- ' I I i. f ~ 60 lr-[ ' -, ·~1 r H ¥J ~ t1.1 ! ~ :J ~ 1J f 'l 40 ~j f >\'-~t H d ! ., ~ l I I ·'·l I !i ! 20 :: ra ·-~- t: <..:::> I· ii i· i. ol I' fl ; I< I ,, Calve I' [I 1: s ,r I t] I; c il 60 .-0 l' +-\\i '-b +- ::J -:$ c -~ 40 J-0 1 ~ :1 c c :s 20 -~· 0 28 Fig. 9. Average percent fat (circles), standard deviation (triangles), and range (bars) in femur bone marrow of male and female adult moose dying from various causes in GMU 20A and 20B between February and May, 1971. Number in circles indicate sample size. Solid bar indicates no significant difference between sample means as measured by Duncan's new multiple range test. I ~) I ~ I (ltb I I (oo) I ~ ~~~~~~~~~ 0 CX) 0 ~ ~ 0 .::J o/o 29 0 ~ :.,= ·-... ~ s:: -IS ~ ~ -tal: ~ VlC ·- _, ~ ~ ~ ~ L C t~~ 'dGJ-0 ~~.t -.J ::5 C\} LL ~ ~ V"\ _"<d a " s.. '--:1~ ~v c-1-= c 0 :5E QJ ac _.c + di QJ Q ~ 0 QJ \1) ':5 <d v Winter Mortality of Moose: Due to unusually early and heavy snow during the winter of 1970-1971, winter mortality throughout a large por- tion of interior Alaska was relatively high. In addition to deaths resulting from malnutrition, numerous animals attracted to plowed roads and railroads were killed by vehicles. MOose deaths resulting from natural and accidental causes in the Fairbanks area occurred primarily between December 1970 and May 1971. During this period the Alaska Department of Fish and Game office in Fairbanks received over 400 reports of dead, dying, and nuisance moose near Fairbanks. Approximately 250 of those reports were received during February and March. Between December 1970 and May 1971, 154 moose dying from natural and accidental causes were examined. Calves were the most common age group in the sample. Male and female calves died in approximately equal numbers, while among adults females tended to predominate (Fig. 10). The sex ratio of both dead calves and adults may, in part, reflect the sex ratio in the population of the two age groups. In the Susitna River Valley in southcentral Alaska, Pitcher (per. comm.) also found that approximately equal numbers of male and female calves predominated among winter-killed moose during 1970-71. Amortg adults, however, males died in higher proportion than they were found in fall sex and age composition counts. The relationship between winter weather conditions and winter mortality was first reported by Severinghaus (1947) for deer, and recently discussed by Bishop and Rausch (1973) for moose. Severity of winter snow conditions in the Fairbanks area (Fig. 11) is reflected by survival of moose calves in the Fairbanks area to 1 year of age (Fig. 12). Calf survival over the two winters of the most severe snow conditions (1965-66 and 1970-71) was exceptionally low, while a gradual increase in the ratio of yearlings per one hundred females occurred following winters of less severe snow conditions. In Fig. 11, the dotted line indicates the average chest height of calves, while the solid line indicates the ' average chest height of adult moose. Moose-Snow Relationships The following has been taken, in part, from a recent publication by Coady (1973). The structural, physical, and mechanical properties of a snow cover vary greatly, depending on conditions of deposition and subsequent meta- morphism. Seasonal snow covers behave in an extremely dynamic fashion, and the only completely predictable phenomenon is change itself. Meta- morphic processes which take place within a seasonal snow pack have been described by numerous workers (Bader et al., 1939; Formozov, 1946; Gold, 1958; Kingery, 1960; Benson, 1967, 1969; Trabant, 1970). While wind action is a major factor affecting snow as it precipitates (Sommerfeld, 1969), diagenetic processes resulting from temperature, time, and settling (Keeler, 1969) begin immediately after deposition. All features of a snow pack reflect post-depositional changes as much or more than they reflect the character of snow at the time of deposition. High temperature relative to the melting point of snow and steep temperature 30 ~ MALNUTRITION UNNATURAL DEATHS .... 0'1 28 1-' 0 . 24 !loll> ~0'1 20 Males Males li (D .... t:l Ill OQ'd (D 16 ( n ........ t:l H> (/) rtl-'- 12 !D n 1'1 -v Ill 0 (D 1-':< E 8 \0 "'-11'1 011> I rt c 4 '-I t-'• 1-'0 . <! 0 0 H> ., s ~ 0 '+-0 0 4 Ill (D ?I" ~ 8 t-'• 1-' <l) 1-' CD ...0 p. E 12 t-'• t:l :::::J 16 Females Females ~ z N 20 0 > 24 ~ I I I I I I I I I I I I I p. I I I I I I I I I I I I I I I I I I I I I I I I N 28 0 <I 2 4 6 8 10 12 14 16 18 <I 2 4 6 8 10 12 14 16 18 ? t:1:l Fig. 11. Relative yearly snow severity near Fairbanks, Alaska, calculated by plotting snow depth on ground on the last day of the month, between October and April, and determining the relative area under the curve with a planimeter. Data are taken·from National Weather Service Records from Fairbanks . . ,-.: .. • ; RtLflrwl! · SnoUJ StvUtTY·······IlS . ..; . -··-·-··· :ual. ._. . l'f( lfl bl. '\l. It>() f----+-~--:--:-----A duff Chest C hes+ Hei.9J,.,t_ l-le·tc.; ht----Cal-f 3 So. 0 <: lf1 .r\ \ I I oL---+---~~----~----+4~---r~----+l~(-----r\ 1'1'1'1-SO, I'ISI-SJ. lbO \31 10 ----·--,·,---.- '. -----r ·,~·------ ! ' ! "'b "'S S..l. 1'\0 .1')5" l\0 C: \---' . r 1\0 .10 HoO / \ I \;\ . ' \ /'<....... i ·~ j \ , "\\ dl ,~ r/ , -~ --+-+----+.----+----·-· I CJ 71-?:l. /9?).-?3 'l') ·~s4 -ss Fig. 12. Survival of moose to one year of age as indicated by spring sex and age composition counts in GMU 20A. Data are taken in part from Rausch (1971). 50 - 45 - 40 - Cf.l 35 -~ (..) 0 30 -0 r-i .. Cf.l c.!> 25 -z 1-1 ~ 20 -~ >-< 15 - 10 - 5 - 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 YEAR ., gradients are effective in promoting vapor transport, a major cause of metamorphism. Constant fluctuations in air temperature and accumulation rate of snow cause continual changes in both thennal and gravitational forces which are perceptible within days or even hours. The properties of snow can be divided into fundamental and derived characteristics. A discussion of tundamentai properties, which include size, shape, orientation, and packing of particles, is beyond the scope of this review. While fundamental properties of a snow cover do bear a relationship to.the mechanical properties and are important for a com- plete understanding of snow metamorphism, they are probably of minor significance in ungulate snow ecology (Pruitt, 197la). Derived properties of snow are those which depend on fundamental properties for their magnitude and rate of change (Keeler, 1969) •. Generally, they are more easily measured than fundamental properties, and are useful indices to the nature of snow. Although derived properties are numerous and diverse, relatively few are generally measured to characterize a snow cover. The most commonly measured derived properties used in biological work are depth, temperature; density, and hardness. Physical characteristics of moose affecting their mobility in snow have been related to height of the animal and weight-load-on-track of the feet. When moose sink into snow to depths approaching chest heigh.t and are forced to "plow" or bound through the snow, energy required for movement is greatly increased. However, snow depths below chest height may hinder movement by increasing resistance to movement of che legs. Weight-load-on-track is a measure of the weight per unit area on the feet, and reflects the extent to which a moose may be supported by a s·ubs trate. Under situations in which snow will partially or fully. support a moose, resistance to and energy required for movement may be reduced. Procedures for measuring chest heights and track loads have recently been described in detail. Kelsall (1969) measured chest heights from the tip of the thoracic limb, with leg extended perpendicular to the body, diagonally to the sternum. The measurement may slightly over- estimate actual chest height since the leg may be less extended on a standing than on a decumbent animal. In many studies hoof load, or weight on hooves alone, has been used co measure the bearing surface of the legs. However, Kelsall and Telfer (1971) and. Telfer and Kelsall (19 71) indicated that the entire foot from the tip of the hoof to the dew claws supports an ungulate in soft snow, and described a procedure for measuring foot areas and calculating foot loads. Foot loads measured by the previous technique are probably minimal since the measurement does not allow for spreading of hooves and angular placement of legs in snow. However, the actual foot load is probably somewhat greater since only when standing is weight supported on only twci feet. In addition, Kelsall (1969) thought that as much as three-fifths of the weight· of a standing ungulate is distributed on .its forefeet. The significance of chest heights and track loads of moose reflect- ing adaptation to snow was first shown by studies in Russia and later in North America. Nasimovich · (1955) reported that chesc heights of adults 34 averaged 105 em or greater. Kelsall (1969) found that average chest heights of male moose in Eastern Canada increased from 81.9 em for calves, to 98.9 em for yearlings, and 104.7 em for ariimals older than four years. Kelsall and Telfer (1971) found an average chest height of 106 em for male moose older than four years in Western Canada. Similar chest heights of 84, 96, and 104 em for 60 calves, yearlings; and moose older than two years, respectively, have been found in interior Alaska. Nasimovich (1955) noted that moose on the Kola Peninsula in Russia were unaffected by snow depths of 40 to 50 em, while movement was definitely impeded by depths of 60-70 em. At 60-70 em calves frequently followed in the trail of adults. Nasimovich (1955) concluded from the Russian.literature that snow depths of 90 to 100 em may be considered critical to moose, since at that depth winter mortality substantially increased. Kelsall (1969) reported similar observations from. Eastern Canada, noting .that movement was unrestricted by depths of 44 err and severely restricted by depths of 70-99 em. Snow depths greater tha:n 90-100 em were critical for moose unless of very short duration. Ritcey (1967) and Prescott (1968, in Telfer 1970) found that depths of 60-70 em restricted mobility of moose in British Columbia and Nova Scotia, respectively. In Alaska substantial winter mortality has occurred in several areas of the state when snow depths exceeded 90 em for several months. The above data suggest that snow depths up to 40 em, or depths approximately equal to the carpus or tarsus height, cause little or no hindrance to movement. From 40 to 60-70 em, or depths approaching two- thirds of the chest height, movement is only slightly rest: ric ted. At depths greater than 70 em movement is definitely impeded, while at depths greater than 90 em, or approximately equal to or slightly less than chest height of standing moose, movement is greatly restricted to the extent that adequate food intake may become impossible. Calves, because of their shorter legs, may be restricted by snow depths somewhat less than those affecting adults, while large males may be·least affected by deep snow. Differential movement of sex and age groups during winter reported by several workers (LeResche, 1973; Pulliainen, 1973) may reflect, in part, the relative ability of the different groups to move in deep snovJ. Heights of moose from interior Alaska and the Kenai Peninsula have been compared to illustrate that leg length in moose may be adaptive to snow conditions. In interior Alaska in winter moose habitat snow depths of 70 em or more, that persist for several months are the rule, and depths in excess of 90 em are not unusual. On the Kenai Peninsula, however, depths in winter moose habitat range near 40 em for short: periods, and seldom reach 60 em. Since chest heights of moose from the Kenai Peninsula were not available, a ratio of shoulder height to total length was used to reflect relative differences in leg length and presumably in chest height. of moose from the two areas.. Average ratio for 31 fully grown moose from the Inte;rior was ~68, while that of 64 similar animals from the Kenai Peninsula was .59. Simi~ar ratio for 18 calves 6 to 12 months of age from the Interior, and 50 calves from the Kenai Peninsula were . 72 and . 70, respectively. Average shoulder heights of moose from the Interior and from the Kenai Peninsula were 182 em and 35 172 em for adults, and 148 em and 141 em for calves, respectively. both relative and absolute height of moose is lower, particularly adults, from the Kenai Peninsula than from interior Alaska. Thus, among Variations in body size of animals may be due to genetic or nutri-' tional differences. Since data from the Kenai Peninsula were obtained from moose on relatively poor range (LeResche and Davis, 1971), and data from the Interior were from moose on relatively good range (Coady, 1973), differences in skeletal dimensions, of animals from the two areas may reflect nutritional differences. While nutritionally related differences in skeletal growth do occur (c.f. Klein, 1964), there is little reason to expect that poor range on the Kenai Peninsula is responsible for preferential growth of body length over foreleg length. Thus, while nutritional deficiencies may account for smaller absolute shoulder height of Kenai Peninsula moose, it is probably not responsible for reduced height of animals relative to length. \fuile long legs per se are not necessarily an adaptation to deep snow (e. g., height facilitates the browsing habit), the selective advantage to moose of increased leg length in regions of deep snow is obvious. Therefore, observed differences in relative height of moose between the Kenai Peninsula and the Interior in Alaska are probably of genetic origin and may be related to differences in snmv conditions between the two areas. Nasimovich (1955) noted that reindeer-from the taiga zone where deep, soft snow is common have longer legs than those animals from tundra areas. Nevertheless, additional studies would be useful. Nasimovich (1955) reported that the average track load of "several moose" in Russia was 420 g/cm2 .. This value presumably represents the total foot load, and not just hoof load, of the animal. Kelsall and Telfer (1971) measured average foot loads of approximately 710 g/cm2 for male moose four years and older during December in Western Canada. Higher average hoof loads of 789 to 922 g/cm2 were found by Kelsall (1969) for similar aged male moose from two areas of Eastern Canada. However, the higher values were measured prior to rut, while the_ lower hoof loads were obtained following the rut and presumably after consider- able weight loss. Foot loads of moose in interior Alaska are not uniform btit vary with age of animal and with season. Average foot load decreased from 593 g/cm2 for eight adult cows in October to 432 g/cm2 for 19 adult cows between April and June. Thus, as winter progresses and snow depth and hardness . increase, foot loads of adults decrease due to seasonal loss of body weight. Foot loads of nine calves between April and June averaged 317 g/cm2 , over 100 g/cm2 less than adults during the same season. Thus, the shorter legs of calves may in part be compensated for by lower foot loads. No seasonal increase in hoof size was noted for Alaskan moose, as Pruitt (1959) reported for caribou. · Snow density has been related to track depth of moose. Kelsall and Prescott (1971) concluded from their extensive observations that snmv- densities of 0.10 to 0.19 g/cm3 do not support moose, densities of 0.20 36 to 0.29 g/cm3 provide some support, and densities of 0.30 to 0.39 g/cm3 limit foot penetration to approximately 50 percent of the snm..r depth. Nasimovich (1955) reported that snow densities of 0.20 to 0.22 g/cm3 provide little support to a running moose, while densities of 0.24 to 0.26 g/cm3 limit foot penetration to two-thirds of the total snow depth. However, under these conditions, moose experience difficulty lifting legs from holes in the snow. Kelsall (1969) and Kelsall and Prescott (1971) were unsatisfied wi.th attempts to relate ungulate support to snow hardness. Theoretically, a standing moose should 9e support~d by a vertical snow hardness equal to or greater than its track load (Kelsall, 196 9; Kelsall and Prescott, 1971). However, Kelsall and Prescott (1971) found the support capacity of snow to be extremely variable, depending on the presence or absence of surface crusts and the hardness of underlying snow layers. Both white-tailed deer and moose frequently broke through crusts that should easily have supported the animal. Moose, with maximum track loads of 1,000 g/cm2 , broke through crusts of 8,000 g/cm2 at the surface, 10,000 g/cm2 at 15 em, and 90,000 g/cm2 at 34 em. Moose also broke through crusts of 20,000, 10,000, 30,000, 40,000 and 25,000 g/cm2 to ground level at a depth of 73 em. On other o~casions moose were supported by surface crusts of 2,000 to 30,000 g/cm . Peek (197la) noted that surface crusts of 7,500 g/cm2 supported moose in Minnesota. In interior Alaska extremely hard crusts are unusual, although an adult moose walking on a trail penetrated 20 em in 30 em deep snow when the hardness was 2,000 to 4,000 .g/ cm2. On another occasion both a cow and calf walking on a trail penetrated 39 em in 90 em deep snow when the hardness was 1,000-2,000 g/cm2 . Preliminary attempts to use the Rammsonde penetrometer to quantify the support capacity of snow for moose have been attempted. The inte- grated ram hardness (Ri) was calculated for the total ram hardness of the snow to foot penetration depth (Table 1). Average Ri ranged from 188 to 570 kg/em for penetration depths of 22 and 42 em, respectively. Although noi: immediately evident from the limited data above, further study may reveal a predictable relationship between Ri and depth of foot penetration in or resistance to movement through snow. Throughout most of the circumboreal range of moose and within favor- able habitat, snow conditions that significantly benefit moose by provid- ing support apparently are seldom extensive or persistent. Even on the tundra of Alaska where snow density and hardness are very great, the snow cover in winter riparian habitat provides little or no support to moose. In other areas supporting crusts are usually extremely localized, and are apparently rarely consistent enough to facilitate travel. Murie (1944), Nasimovich (1955), Kelsall and Prescott (1971), Peek (1971a), and others indicated that snow conditions which only partially support moose may make movement more difficult and hazardous because of the resistance to movement of legs provided by the dense snow and/or the danger of abrasion from hard crusts. Snow conditions in which depth of penetration is variable may require an animal to expend more energy recovering from breaking through crusts and climbing onto crust·s than would be required to move through deep snow offering no support. However, 37 dense, hard snow offering uniform support, such as ski or snow machine trails, may be extensively used. Comparison of foot loads of moose from different regions may indi- cate adaptation to varying snow conditions. Iri Western Canada Kelsall and Tel fer (19 71) measured average foot loads of 710 g/ cm2 for adult moose durin2 December, whereas in interior Alaska I report foot loads of 593 g/cm for adults during October. Normal snow conditions at the collection site in Western Canada were not given, but presumably snow depth and hardness were at least as great as those in interior Alaska. Although I followed procedures described by Kelsall and Telfer, p0ssible differences in measuring foot area must be considered. The influence of snow on seasonal movements of moose has been reported by several workers, although quantitative observations are relatively limited. The greatest effort to document and review relation- ships between snow and moose migrations has occurred in the USSR (Formozov, 1946; Nasimovich, 1955; Knorre, 1959, 1961; Egorov, 1965; Heptner and Nasimovich, 1967 in Van Ballenberghe and Peek, 1971). However, signifi- cant contributions have also been made in Europe (c.f. Pulliainen, 1973) and in North America (Edwards and Rit cey, 1956; Ritcey, 196 7; Knowlton, 1960; Houston, 1968; Kelsall and Prescott, 197~). Nasimovich (1955) drew several conclusions from the copious Russian literature regarding the influence of snow on moose migrations in both mountains and flatlands of the USSR. , Basically, in regions where maximum snow depth averages less than 50 em and deep snov1s are of short duration, extensive seasonal migrations are uncommon, although local movements may occur. However, in areas where maximum snow depths in excess of 70 em persist for long periods, seasonal movements occur from areas of deep to less deep snow. The longest migrations, ranging from 150 to 300 km, occur among animals living on flat terrain, although migrations of 100 to 150 km are common across divides or to lower elevations in mountainous regions. Gradual movements generally occur between October and January, and may coincide either with the first lasting snow cover or with snow depths of 25 to 45 em. However, some animals migrate before snowfall while others remain in summer habitat until snow depths reach 60 to 70 em. Pulliainen (1973) reviewed the literature describing relationships between snow.and moose migrations in Scandinavia. Movements in most areas are closely correlated with prevailing snow conditions. Gradual movements from high to low elevations usually begin in November or December, although they may be 'de1ayed or may not occur during years of little snow. Some animals, particularly cows with calves, begin migrat- ing at first snowfall, while others, particularly bulls and cows without calves, remain at high elevations until snow depths reach 60 to 70 em. However, formation of icy crusts may initiate downward migration of almost all animals (Krafft, 1964). Return to summer rarige is usually abrupt, and occurs during May after snow melt has exposed patches of ground. In North America several early workers, including Murie (1934) on 38 Isle Royal, Hosley (1949) in Maine, Bauman (1941, in Hosley) in Yellowstone Park, Murie (1944) in Alaska, and Hatter (1946, in Hosley) have copl!llented on snow depths and moose movements. However, Edwards and Ritcey (1956) in British Columbia were the first to present detailed observationson the relationships between moose migrations and snow conditions. They found that a gradual altitudinal movement from 1525- 2135 m to 760-1220 m during fall and winter coincided with a gradual increase in snow depths on summer range. A rapid return to higher elevations during spring coincided with a rapid snow melt. Segregation of winter.and summer ranges was not complete, although most animals had departed higher elevations by the time snow depths J;eached 75 em. Upward movement to summer range was initiated when melting had reduced snow depths t:o 30 to 45 em. cold t~mperatures appeared to moderate the effect of snow by speeding movement downward in the fall and retarding movement upward in the spring. Ritcey (1967), also in British Columbia, noted that deep snows .at high elevations were responsible for the fall and winter movement to elevations below 1050 m. Arrival on winter range generally began in November when snow depths were less than 15 em, and continued throughout the winter. Departure from winter range began in late February or March, while snow depths were as great as 125 em but declining. Kelsall and Prescott (1971) and Telfer (1967a, b) in the Canadian Maritime Provinces, studied winter segregation of white-tailed deer and moose in relation to moose sickness induced by Pa:rela:phostrongylus tenuis. Although segregation was not complete, deer generally wintered at elevations below 200 m, while moose remained at elevations above 200 m. Snow depths of 85-90 -em above 200 m did not initiate downward migra- tion of moose, even though snow depths were more favorable and browse more abundant at lower elevations. However, relatively high density and hard crusts due to winter thaws and rains may provide some support to moose, thereby reducing the effective snow depth (Telfer, per. comm.). In Montana Knowlton (1960) and Stevens (1970) reported that deep snows in summer range above 1830-2135 m initiated movements to lower elevations. Movements were gradual, and frequently lasted from December to Harch. Harry (1957) in Wyoming reported that increasing snow depths at high elevations resulted in a gradual downward movement and concentra- tion of moose in mountain valleys by December. Snow condit:Lons associated with these studies in Montana and Wyoming were not reported. However, Houston (1968) in Jackson Hole, Wyoming, found that downward movements from 2190 m to winter range at lower elevations began in late December in response to snow depths of about 80 em on the summer range. Movement to spring and summer range began in late March in response to a snow crust formation capable of supporting mooose ~d in response to disappear- ance of snow from south and east slopes. Moose densities on winter range were 10 moose/km2 in Mont~na (Stevens, 1967:7 1 in Stevens, 1970) and 19 moose/km2 in Jackson Hole (Houston, 1968). 1 • Seasonal movements noted by some workers. 1956 at high elevations of moose in response ~;o snow in Alaska have been Rausch (1958) reported\ that an early snowfall in in southcentral Alaskal caused an early migration 39 in November to lowland areas. Rausch concluded from his extensive observations that snow influences but does not cause seasonal movements of moose. Fall migrations in hills and·mountains of interior Alaska generally occur as a gradual downward movement between December and February or March. The extent, time, and composition of the migrating animals appear to be closely related to snow conditions. In late November and early December, 1970, snow depths of 90 em (55 em above average) at elevations of 600 to 915 m in hills near Fairbanks apparently caused an abrupt down- ward migration of moose to elevations below 300 m. Over 1200 animals were seen on or moving toward low elevation riparian habitat during 23 hours of Alaska Department of Fish and Game aerial surveys in early December. Almost no animals were found on the usual upland fall and early winter range. Snow depths of 110 to 120 em persisted until early April and moose remained along rivers until mid~March when they apparently dispersed into adjacent timbered areas. Similar observations during 1970 were reported by Bishop (1971) for western interior Alaska, where snow depths of 60 em during the end of November apparently precipitated an early movement of moose from upland areas to lower elevation riparian habitat, where they remained until late March. The response of moose to winter weather factors has been studied in interior Alaska since 1971. For example, relationships between moose movement and snow conditions during 1971-1972 were examined on a study area near Fairbanks (Fig. 13). "Tracks-Valley" indicates the seven-day total of fresh moose tracks crossing a one-half mile long transect in a valley. The valley is located at 245 m elevation, and represents typical winter riparian moose habitat. "Moose-Hills" indicates the number of moose counted during frequent intensive aerial surveys in a 75 krn 2 drainage above the valley transect. The upland site ranges from 550 to 670 m elevation, and consists of mixed conifer and deciduous trees and shrubs which characteristically support modest numbers of moose during Stmlrner and fall in interior Alaska. While neither "Track-Valley" nor "Moose-Hills" indicates actual number of animals, they are thought to reflect the trend of animal abundance on each site. Snow depths and integrated Rannnsonde resistance of the total snow cover (Ri) in shrub communities on the two sites are also noted. A decrease in moose observed at high elevations and an increase in fresh tracks at low elevations occurred during late December and early January. The decrease in fresh tracks in the valley during mid-January may have been related to reduced activity during extremely cold tempera- tures (-40° to -50°C) during that period, while the decrease in tracks after late February may have resulted from a dispersal of animals away from the riparian habitat where the transect was located. After January the number of moose in the hills remained low throughout the winter. Snow depths at high elevations gradually increased to about 80 ern at the time of movement in late December. While lowland snow depths throughout January and early February ranged from 15 to 25 em below those in the hills, depths at the two sites remained nearly identical during the rest of winter. The R. of the snow cover sharply increased 1 40 Fig. 13. Relationship between moose movement between and snow depth and Rammsonde hardness at high and low elevations in the Chatanika River drainage (GMU 20B) during winter 1971-72. See text for explanation of figures. 0 9 I I ~-,..-r -~ I : I: I ·\ ~ J4 ~ I ),? I J, < ? I I >. I >. <l> I W(f) I -(f) --I -I - -· o I >o ·-I > J: (._"( ~ C~ I :r: I l \ <" I 0 0 i . 0 0 .. .-:'' ('··<> '·.' ~ 0 4 J ', ' /.,) \ ~ I \ ~ ~-' -."-.......... -,......, ..._ ' ·',) ~ '""-' .,,J \ ;. I J ---~ 0 -.... I I \ ....._ I 1 ! I 0 ' I I I . -I , I ) I .Q t) 6 :q __ /· I : \ }., / l -:~ : -'' "\ (------< G (. ', \ ~ -J " ', \ "' 0\'' \ ' ( 9 \ n ? <} . - l, : \q l \ -J ~j ~ 0 0 \ I I \ I \ . I )., I I y ~ •6 () <> -~)0 { 1> \ 0 b . z I 1-L__ , 0 0 0 0 0 0, oo 0 0 0 0 0 L() 0 l{) 0 l[):. (\j (\J ---( w~. 0 ~) .'C:J (W:J} l/ldtJO /110U$ S//,IH-CJSOO}Y /.:Jjj0/1-s.y:JD.I.J. 41 during December preceding movement of animals. The increase resulted from both an increase in total snow depth and an increase in ram hardness of given depth increments. The dispersal of moose from lowland riparian habitat to adjacent areas during March may have been influenced by the declining Ri making travel less difficult. Dispersal from riparian habitat in March may also have been related to the lower snow depth, density and hardness in deciduous and conifer tree communities during that time. The range for snow density of settled snow at the upland shrub site increased from 0.16 -0.24 g/cm3 in late November to 0.20 - 0.31 g/cm3 in late December, while the range of.snowhardness for settled snow increased from 10-50 g/ cm2 to 50 -500 g/ cm2 during the same period. Conclusions regarding the significance between integrated Rammsonde resistance and movement of moose would be highly premature at this time. However, based upon the above data and upon similar correlations between Ri and moose behavior in other study areas during both 1971-1972 and 1972- 1973, further studies using the Rammsonde penetrometer appear justified. Movement of moose onto non-riparian lowland habitat during 1971- 1972 corresponded closely with that onto riparian areas (Fig. 14). Sites 1 and 2, located 90 km from the above study area, are each approximately 1. 5 kni2 in size. Land clearing 10 to 15 years ago has resulted in a dense regrowth of shrubs and low trees. Moose movement onto both sites 1 and 2 began in late December, while by early March nearly all animals had dispersed from the sites, apparently into more densely vegetated areas. Snow depths on the sites averaged about 70 em in late December when animals began to appear, and remained near 80 to 90 em until late April. The maximum number of animals was 11 (density 7/km2 ) on site 1 in mid-January. During the preceding winter snow depths on the sites averaged 115 em in January and the maximum number of animals during. that time was 36 (density 24/km2). Seasonal migrations of moose in interior Alaska are not always influenced by snow conditions. Movement of some animals from lowland summer range to either upland or riparian shrub habitat may begin in August, well before snowfall occurs, and continue throughout the winter. However, while initial movements may be related to factors other than snow, the speed and extent. of migration are apparently influenced by snow. During winters of early or deep -snow, movement of most animals from lowland summer ranges may occur sooner and to a greater extent than during winters of late or little snow. Availability of browse as affected by snow depth may be particularly important in influencing movements over flat terrain where local differences in snow conditions are not great. For example, late snowfall may have accounted for exceptionally heavy and extensive use of some low (40 to 60 em high) willow (Salix puZchra) communities on the Tanana Flats near Fairbanks during fall and early winter, 1972. Movement to summer range in interior Alaska apparently occurs dur- ing a relatively short period after snow melt has exposed patches of bare ground. Observations of animals and tracks suggest that when snow cover persists into mid-to late May, substantial movement of animals does not occur until that time. However, an early thaw results in an 42 Q) (j) 0 0 00::::::::: c::;:_ ~ C) .D c c.. - --0.. () 0 5 0 c (/) Fig. 14. Movement of moose and snow depth at two small lowland wintering. sites near Fairbanks, Alaska (GMU 20B), during winter 1971-1972. Number of moose were determined by aerial count. 1·0 IO ,-./ ,-.}--o---'-...) I //~ 50i-;O-d 10-~ . Site 1 Site 2 ~ /....,.------o--~ \ 'o O l---~--~---~~--~~----~--~-. I I ~OV 1 DEC 1 JAN 1 FEB 1 MAR 1 APR 1 MAR 1 Month 43 early movement to summer range during late April and early May. Advanced snow melt during spring 1973 resulted in a 70 kin migration between April 10 and April 20 of a radio-collared moose from winter to summer range. Although the above data are preliminary and highly limited in scope, they illustrate an approach to studying moose-snow relationships which may prove useful in other areas. Detailed observations of moose distri- bution and movements, snow parameters, and temperature and wind conditions in several areas of interior Alaska over three years wili be reported in a future publication. Habitat selection and movement on winter range in relation to snow conditions have been reported in several excellent studies (Nasimovich, 1955; Des Meules, 1964; Telfer, 1970; Berg, 1971; Peek, 197la, b; Van Ballenberghe and Peek, 1971), and will therefore not be further con- sidered here. However, most studies suggest an increased use of dense covE7r with an increase in snow depth, density, and/or hardness, and a relatively small winter home range, although actual snow conditions causing a change in habitat selection or home range size are variable. Since February 1971, 195 moose observations, 34 aerial moose surveys, and 56 weekly track counts have been completed in the Fairbanks area. Data analyses are incomplete, and will be presented in a future report. All records are on file in the Alaska Department of Fish and Game office, Fairbanks. Radio Telemetry The six radio transmitters were tested on horses at Fairbanks for one day and at Kenai on Moose Research Center penned moose for two days. All transmitters and collars functioned satisfactorily during testing. The operational life of the radio collars was as follows: one unit, seven weeks; one unit, eleven weeks; two units, seven months; two units, seven months plus (still operating). However, the signal of one .currently operating transmitter was not received for seven months after it was placed on the moose. On June 18, 1973 the signal was received in the area where the moose was instrumented. It is not known whether movement of the moose out of the search area and beyond the range of the receiver or or intermittent operation of the transmitter was responsible for not receiving the signal. During the operation of the transmitters, the moose were located a total of 158 times between November 21, 1972 and June 18, 1973. Three of the radio-collared moose remained in the Elliott Creek burn until late December, when they migrated out of the burn within 10 days of each other. One of the animals moved into a mature deciduous tree stand (elevation 600m), midway between the burn and the Chena River), and remained in that area until late April. During a period of ten days the moose then moved 55 km to a black spruce (Picea mariana)-bog habitat in the Tanana Flats (elevation 130m). The other two moose moved 30 to 40 km directly from the burn to shrub communities along the Chena River and 44 the Chena Hot Springs Road (elevation 170 rn) where they remained until late April. One of the animals then occupied adjacent black spruce~ deciduous tree-bog habitat. The transmitter failed on the second moose. A fourth moose remained in the burn· until early April, when she moved directly into.black spruce-bog habitat along the Chena River. Mopthly aerial surveys in the 90 krn 2 Elliott Creek burn also indi- cated a gradual movement of moose out of the area. Between the end of November and the end of December, snow depths in the burn increased from approximately 30 to 55 ern. Movement of radio- collared moose out of the burn during late December may have been influenced by rapid accumulation of snow during that time. Snow depths in the burn between mid-January and mid-March ranged between 50 and 60 ern. Therefore, the sharp decline in total number of moose in the burn between January and March was apparently not in response to increasing snow depths, but was influenced by other factors. Movement of radio- collared moose into black spruce-bog habitat during late April corre- sponded with the appearance of patches of bare ground at that time. Between November and mid-April, while the ground was 100 percent snow covered, 93 percent of the collared moose located electronically were then.visually spotted. However, after mid-April, with the appear- ance of bare ground, only 58 percent of the collared moose located electronically were visually sighted. The decrease in visibility was probably due to a number of factors, such as greater difficulty sighting moose on snow-free ground, change in habitat selection to more dense cover, and a more secretive nature with the approach of parturition. Analysis of moose telemetry data is incomplete, and will be pre- sented in a future publication. However, the use of radio telemetry in combination with other methods of monitoring seasonal movement patterns and habitat selection in response to various factors appears justified. However, a larger number of transmitters (20 to 25) of longer operational life (one or more years) would be of value. ACKNOWLEDGMENTS The ADF&G laboratory staff, especially David Harkness and EdWard Kootuk, along with ADF&G technician Mike Vierthaler, collected and obtained weights and measurements on many moose . Mike Vierthaler ably made detailed snow measurements on numerous occasions. H. Don Draper, resident of 40 mile Steese Highway and sourdough extraordinaire, ·pro- vided extensive moose observations and daily counts of moose tracks during two winters. ADF&G biologists Richard Bishop and Robert LeResche offered valuable suggestions and ne·eded assistance in radio-collaring moose. ADF&G biologist Tony Smith located radio~collared moose when the author was unavailable, and flew monthly aerial surveys during winter 1972-73 in the Elliott Creek burn. Among pilots, William Lentsch stands out for his ability, persistence, and interest in locating radio-collared moose. The assistance of the above individuals and numerous others is greatly appreciated. A special acknowledgment is· extended to John J. Burns, 45 ADF&G biologist, fat,;. timely advice and. guidance in defining res·earch · needs and designing many aspects of this and other studies during their initial stages~ <ti LITERATURE CITED ·Abele~ G. 1963. A correlation of unconfined compressive strength and ram hardness of processed snow. USA CRREL Tech. Rep. 85. · 14 . pp. ____ ._. 1968. An experimental snow runway pavement in Antarctica. USA GRREL Tech. Rep. 211. ----'-., R. 0. Ramseier and A. F. Wuori. tr,ansportation methods in deep snow. 1965. · A study of sub-surface USA CRREL Tech. Rep. 160. 196~. Design criteria for snow runways. USA CRREL Tech. Rep. 212. Abrams, J. T. 1968. Fundamental approach to the nutrition of the captive wild herbivore. In M. A. Crawford, ed. Comparative nutrition of wild animals. Symp. Zool. Soc. London 21:41-62. Alaska Department of Fish and Game. 1973. Alaska's wildlife and habitat. ADF&G. 590 pp. Altmann, M. 1959. Group dynamics in Wyoming moose during the rutting season. J. Mammal. 40:420-424. Bader, H. , R. Haefeli, E. Bucher, J. Neher, 0. Eckel and C. Thams. 19 39 • Snow and its metamorphism. Beitrage zur Geologie der Schweiz, Geotechniche Serie, Hydrologie, Liefereung 3, Bern. U. S. Army Snow, Ice and Permafrost Res. Establish. Transl. 14, 1954. 313 pp. Bear, G. D. 1971. Seasonal trends in fat levels of pronghorns, AntiZocapra americana, in Colorado. J. Mammal. 52(3) :583-589. Bee, J. W. and E. R. Hall. 1956. Mammals of northern Alaska on ·the Arctic Slope. Mus; Nat. Hist. Misc. Publ. No. 8. Univ. Kansas,· Lawrence. 309 pp. Benson, C. S. 1962. Stratigraphic studies in the snow and firn of the Greenland ice sheet. USA SIPRE Res. Rep. 70. 91 pp. 1967. A reconnaissance snow survey of interior Alaska. Rep. USGR-190. Geophysical Inst., Univ. Alaska, College. 71 pp. 1969. The seasonal snow cover of Arctic Alaska. Arctic Inst. N. Am. Res. Paper No. 51. 47 pp. Berg, W. E. 1971. Habitat use, movements, and activity patterns of moose in northwestern Minnesota. M.S. Thesis. Univ; Hinn., St. Paul. 98 pp. 46 Bilello, M. A., R. E. Bates and J. Riley. tics of the snow cover, Fort Greely, Rep. 230. 1970. Physical characteris- Alaska, 1966-67. CRREL f'ech. Bishop, R. 1969. Moose Report, Vol. 10. Alaska Dept. Fish and Game Annu. Proj. Seg. Rep.,' Proj. W-15-R-3, Work PlanK. 153 pp. 19 70. Preliminary review of changes in sex and age ratios of moose and their relation to snow conditions on the Tanaria Flats, Alaska. Paper presented at 6th Annu. N. Am. Moose Com. Meet., Kamloops, B. C. Feb. 1970. 14 pp. Xerox. 1971. Annual report of survey-inventory activities, Part 1, moose, deer, and elk. Fed .. Aid in Wildl. Rest. Proj. Alaska, Proj. W-17-3, Vol. II. 119 pp . -----• and R. A. Rausch. 1973. Moose population fluctuations in Alaska, 1950-72. Paper presented at the Int. Symp. on Hoose Ecol., Quebec City. March 1973. 37 pp. Blood, D. A., J. R. McGillis and A. L. Lovaas. 1967. Weights and measurements of moose in Elk Island National Park, Alberta. Can. Field Nat. 81:263-269. Brandy, P. J., L MeT. Cowan, W. D. Kitts and A. J, Wood. 1956. A method for the assessment of the nutritional status of wild ungulates. Can. J. Zool. 34:48-52. Bull, c. 1956. The use of the Rammsonde as an instrument in determin- ing the density of firn. J. Glaciology 2:713-725. Cheatum, E. L. 1949. Bone marrow as an index of malnutrition in deer. N.Y. Conserv. 3:19-22 . -------. and C. W. Severinghaus. 1950. Variations in fertility of white-tailed deer related to range conditions. Trans. N. Am. Wildl. Conf. 15:170-190. Chesemore, D. L. 1968. Occurrence of moose near Barrow, Alaska. J. Mammal. 49(3) :528-529. Coady, J. W. 1973. Influence of snow on behavior of moose. · Paper presented at the Int. Symp. on Moose Ecology, Quebec City, March 1973; 35 pp. Cowan, I. MeT. and A. J. Wood. 1955. The growth rate of the black- tailed deer (Odocoileus hemionus columbianus). J. Wildl. Manage. 19 (3) :331-336. Davis, H. 1952. Recollections. In Herbert L. Heller, ed. Sourdough Sagas. World Publ. Co. , Cleveland. 47 Denniston, R. H. 1956. Ecology, behavior, and population dynamics of the Wyoming or Rocky Mountain moose, Alces alces shirasi. Zoologica 41(3) :105-118. Des Meules, P. 1964. The influence of snow on the behaviour of moose. Trans. N.E. Wildl. Conf. 21. 17'pp. Dodds, D. G. 1959. Feeding and growth of a captive moose calf. J. Wildl. Manage. 23:231-232. Edwards, R. Y. and R. W. Ritcey. 1956. The migrations of a moose herd. J. Mammal. 37(4):486-494. 1958. Reproduction in a moose popula- tion. J. Wildl. Manage. 22(3) :261-268. Egorov, 0. V. 1965. Wild Ungulates of Yakutia. Israel Program for Scientific Translations. 204 pp. Formozov, A. N. 1946. Snow cover as an integral factor of the environ- ment and its importance. Moscow· Society of Naturalists, Moscow, USSR. Boreal Institute, University of Alberta, Canada, Occasional Paper No. 1. 176 pp. Gasaway, W. C. and J. W. Coady. 1973. Review of energy requirements and rumen fermentation in moose and other ruminants. Paper presented at the Int. Symp. on Moose Ecology, Quebec City. March, 1973. 37 pp. Gold, L. W. 1956. The strength of snow in compression. J, Glaciology 2 (20) :719-725. 1958. Changes in a shallow snow cover subject to a temperature climate. J. Glaciology 3(23):218-222. Greer, K. R. 1968. A compression method indicates fat content of elk (wapit·i) femur marrows. J. Wildl. Manage. 32(4) :747-751. 1969. Femur marrow reveals the condition of game animals. Montana Fish and Game Department, Bozeman. Hall, E. R. and K. R. Kelson. 1959. The mammals of North America. V. 2, Ronald Press, New York. Harry, G. B. Wyoming .. 1957. Winter food habits of moose in Jackson Hole, J. Wildl. Manage. 21 (1): 53-57. Heptner, V. G., A. A. Nasimovich and A. G. Bannikov. 1961. Mlekopitaiu- shchie Sovetskogo Soiuza. V. 1, Parnokopytnye i neparnokopytnye. Moscow, Vysshaia Shkola. Hosl'ey, N. W. 1949. The moose and its ecology. U. S. Fish and Wildl. Serv. L~afl. 217. ·51 pp. 48 Houston, D. B. 1968. The Shiras moose in Jackson Hole, Wyoming. Grand Teton Natural History Association Tech. Bull. No. 1. 110 pp. Jordan, P. A., D. B. Botkin and M. L. Wolfe. 1970. Biomass dynamics in a moose population. Ecology 52(1) :147~152. Keeler, C. M~ 1969. Some physical properties of alpine snow. USA CRREL Res. Rep. 271. 70 pp . -------. and W. ·F. Weeks. 196 7. Some mechanical properties of alpine snow, Montana 1964-66. USA CRREL Res Rep. 227. 35 pp. Kelsall, J. P. 1969. Structural adaptations of moose and deer for snow. J. Mammal. 50(2) :302-210. 1972. The northern limits of moose (A lees alces) in western Canada. J. Mammal. 53(1) :129-138 . . and W. Prescott. 1971. Moose and deer behaviour in snow -------in Fundy National Park, New Brunswick. Can. Wildl. Serv. Rep. Ser. 15,; 25 pp. --------.· and E. S. Telfer. tion of big game for snow. Proceedings of the Snow and tion Symposium. Iowa Coop. 1971. Studies of the physical adapta- Pages 134-146 in A. 0. Haugen, ed. Ice in Relat.ion to Wildlife and Recrea- Wildl, Res. Unit. Ames. 280 pp. 1973. Biogeography of moose with particular reference to western Canada. Paper presented at the International Symposium on Moose Ecology, Quebec City, March 1973~ 26 PP· Kingery, W. D. 1960. On the metamorphism of snow. Mass. Inst, Technol. Ice Res. Lab. Rep. No. 61-1. 9 pp, Klein, D. R. 1964. · Range-related differences in growth of deer reflected in skeletal ratios. J. Mammal. 45 (2) :226-235. 1965. Ecology of deer range in Alaska. Ecol. Monogr. 35: 259-284. 1968. The introduction, increase, and crash of reindeer on St. Matthew Island. J. Wildl. Manage. 32(2) :350-367. 1970. Food selection by North American deer and their response to overutilization of preferred plant species. Pages 25- 46 in A. Watson, ed. Animal populations in relation to their food resources. Blackwell, Oxford . ------• , C.-C. Pearce and L. W. Gold. 1950. Method of measuring the significant characteristics of a snow cover. Nat. Res. Counc. Can. Tech. Mem. 56 pp. Mimeogr. 49 Knorre, E. P. 1959. ·Ecology of the Moose. Proceedings of the Pechora- Ilych StateNature Preserve. 7:167 pp. (in Russian). 1961. The results and perspectives of domesticat-ion of moose. Papers of the Pechora-Ilych State Reservation. 9:396 pp. Knowlton, F. F. 1960. Food habits, movements and populations of moose in the Gravelly Mountains, Montana. J. Wildl. Manage. 24(2) :162-170. Krafft, A. 1964. Management of moose in a Norwegian Forest. The Norwegian State Game Research Insititute, 2(16) :61 pp. Lent, P. C. and D. Knutson. 1971. Muskox and snow cover on Nunivak Island, Alaska. Pages 50-62 in A. 0. Haugen, ed. Proceedings of the Snow and Ice in Relation to Wildlife and Recreation Sumposium. Iowa Coop. Wildl. Res. Unit, Ames. 280 pp. Leopold, A. S. and F. F. Darling. 1953. Wildlife in Alaska. Ronald Press Co. , New York. 129 pp. LeResche, R. E. 1970. Moose Report. Annu. Proj. Seg. Rep. Fed. Aid Wildl. Rest., Alaska Dept. Fish and Game, Juneau. 1973. Moose migrations in North America. Paper pre- sented at the Int. Symp. on Moose-Ecology, Quebec City, March 1973. 41 pp. --------., R. H. Bishop and J. W. Coady. 1973. Distribution and habitats of moose in Alaska. Paper presented at the Int. Sy~p. on Moose Ecology, Quebec City, March 1973. 43 pp. ---------~-· and J, L. Davis. 1971. Moose research report. Annu. Proj . Seg. Rep. Fed. Aid Wildl. Rest. , Alaska Dept. Fish and Game, Juneau. 88 PP• Lutz, H. J. 1960. Early occurrence of moose on the Kenai Peninsula and in other sections of Alaska. Misc. Publ. Alaska Forest Res. Center, 1:1-25. Markgren, G. 1966. A study of hand-reared moose calves. Viltrevy 4 (1)·: 1-42. 1969. Reproduction of moose in Sweden. Viltrevy 6(3) :127- 299. Mech, L. D. 1970. The Wolf: The ecology and behavior of an endangered species. The Natural History Pre.ss, Garden City, N. Y. 384 pp. Miller, D. R. 1972. Observations of wolf predation on barren-ground caribou in winter. Abstr. First Int. Reindeer/caribou Symp., Univ. Alaska, College. 50 Murie, A. 1934. The moose of Isle Royale. Univ. Michigan Mus. ZooL Misc. Publ. No. 25, Ann Arbor. 44 pp. 1944. The wolves of Mount McKinley. U. S. Gov. Print. Off., Washington, D. C. 238 pp. Nasimovich, A. A. 1955. The role of the regime of snow cover in the life of ungulates in the USSR. Moskva, Akad. Nauk SSSR, 403 'PP. Translated from Russian by Can. Wildl. Serv. Ottawa. Neiland, K. A. 1970. caribou femurs. Weight of dried marrow as indicator of fat in J. Wildl. Manage. 34(4):904-907. Nordan, H. C., I. MeT. Cowan and A. J. Wood. 1968. Nutritional require- ments and growth of black-tailed deer, Odocoi Zeus hemionus coZumbianus, in captivity. Pages 89-96 in M. A. Crawford, ed. Comparative Nutrition of Wild Ariiinals. Symp. Zool. Soc. London, No. 21. · Peek, J. M. 1962. Studies of moose in the Gravelly and Snowcrest }fountains, Montana. J. Wildl. Manage. 26:360-365. 197la. Moose habitat selection and relationships to forest management in northeastern Minnesota. Unpubl. Ph.D. Thesis. Univ. Minn., St. Paul. 250 pp. 197lb. Moose-snow relationships in northeastern Minnesota. Pages 39-49 in A. 0. Haugen, ed. Proceedings of the Snow and Ice in Relation to Wildlife and Recreation Symposium, Iowa State Univ., Ames. 280 pp. Peterson, R. L• 1955. North American Moose. Univ. Toronto Press. 280 pp. Pewe, T. L. and D. M. Hopkins. 196 7. Mammal remains of Pre-Wisconsin Age in Alaska. In D. H. Hopkins, ed. The Bering Land Bridge. Stanford Univ. Press, Stanford. 495 pp. Pimlott, D. H. 1959. Reproduction and productivity of Newfoundland moose. J. Wildl. Manage. 23(4) :381-401. Pruitt, W. 0., Jr. 1959. Snow as a factor in the winter ecology of the barren ground caribou (Rangifer arcticus). Arctic 12(3) :159-179. 1966. Ecology of terrestrial mammals. InN. J. Wilimovsky and J. N. Wolfe, eds. Environment of the Cape Thompson Region, Alaska. USAEC Oak Ridge, Tenn. 1250 pp. 19 7la. Report on International Workshop on Rangifer Winter Ecology, Vittangi, Sweden, March 20-28. 5 pp. Xerox. 197lb. Scandinavian-Canadian Field Workshop on Rangifer Snow Ecology. Arctic 24(2) :145. 51 Pulliainen, E. 1973. Seasonal movements of moose in Europe. Paper presented at the Int. Symp. on Moose Ecology, Quebec City, March 1973. 43 pp. Ransom, A. B. 1965. deer condition. Kidney and marrow fat as indicators of white-tailed J. Wi1dl. Manage. 29(2):397~398. Rausch, R .. A. 1958. Moose management studies. Job Completion Rep. Fed. Aid Wildl. Rest., 12:138 pp. Proj. W-3-R-12, Alaska Game Connniss ion, Juneau. 1959. Some aspects of population dynamics of the rail- belt moose populations of Alaska. MS Thesis, Univ. Alaska. 81 pp. 196 7. Report on 1965-66 moose studies, Vol. VII, ADF&G Annu. Proj. Seg. Rep., Proj. W-15-R-1, Work PlanK. Juneau. 129 pp. 19 71. Moose Report, Vol. 11, ADF&G, Prog. Rep. , Proj . W-17-1, Work Plan K. Juneau. 118 pp. ------=--· and R. Bishop. 1968. Report on 1966-67 moose studies, Vol. 8 and 9, ADF&G Annu. Proj. Seg. Rep. Proj. W-15-R-2 and -3, Work Plan K. 263 pp. ______ . and A. Bratlie. 1965. Annual assessments of moose calf production and mortality in southcentral Alaska. Annu. Conf. W. Assoc. State Fish and Game Connnissioners, 45. 11 pp. Himeo. Rausch, R. L. 1951. Notes on the Nunamiut Eskimo and mannnals of the Anaktuvuk Pass region, Brooks Range, Alaska. Arctic 4:14 7-'195. 1963. A review of the distribution of holarctic recent mannnals. Pages 29-43 in J. L. Gressitt, ed. Pacific Basin Biogeography. Bishop Museum Press, Honolulu. Richens, v. B. and C. G. Madden. 1973. An improved snow study kit. J. Wildl. Manage. 37(1):109-113. Riney, T. 1955. Evaluating condition of free ranging red deer (Cervus . elaphus), with specia} references to New Zealand. New Zealand J. Sci. and Tech., Sec. B. 36(4) :429-464. Ritcey, R. W. 1967. Ecology of moose winter range in Wells Gray Park British Columbia. Proc. 4th Workshop on Moose Research and. Manage- ment, Edmonton. 15 pp. Xerox. Robinette, L. W., J. S. Gashwiler, J. B. Low and D. A. Jones. 1957. Differential mortality by sex and age among mule deer. J. Wild!. Manage. 21(1) :1-16. Severinghaus, C. W. 194 7. Relationship of weather to winter mortality and population levels among deer in the adirondack region of New York. 12th North American Wildlife Conference, pages 212-223. 52 Simkin, D. W. 1965. Reproduction and productivity of moose in north- western Ontario. J. Wildl. Manage. 29(4) :740.,.-750. Skunke, F. 1949. Algen. Stokholm, Norstedt. 400 pp. Smith, N. S. 1970. Appraisal of condition estimation methods for east African ungulates, E. Afr. Wildl. J. 8:123-129. Sonnnerfeld, R. A. 1969. Classification outline for snow on the ground. USDA Forest Serv. Res. Paper RM-48. 24 pp. Stevens, D. R; Montana; 1970. Winter ecology of moose in the Gallatin Mountains, J. Wildl. Manage. 34(1) :37-46. Trabant, D. C. 1970. Diagenesis of the seasonal snow cover of interior Alaska. Unpubl. M.S. Thesis. Univ. Alaska, College. 48 pp. Telfer, E. S. 1967a. Comparison of a deer yard and a moose yard in Nova Scotia. Can. J. Zool. 45:485-490. 1967b. Comparison of moose and deer winter range in Nova Scotia. J. Wildl. Manage. 31(3) :418-425. 19 70. Winter habitat selection by moose and white-tailed deer. J. Wildl. Manage. 34(3) :553-559. _______ . and J. P. Kelsall. 1971. Morphological parameters for mammal locomotion in snow. Paper pr.esented at 51st Annual Meeting of the Am. Soc. Mammal., Univ. B. C., Vancouver, June 20-24. 10 pp. Xerox. Testlab. 1970. The Swiss Rammsonde. Special Technical Paper No. 1. 15 PP· Van Ballenberghe, V. and J. M. Peek. moose in northeastern Minnesota. 1971. Radiotelemetry studies of J. Wildl. Manage. 35(1):63-71. Verme, L. J. 1970. Some characteristics of captive Michigan moose. J. Mammal. 51(2):403~405. Wood, A. J. , I. MeT. Cowan and H. C. Norday. 1962. Periodicity of growth in ungulates as shown by deer of the genus OdocoiZeus. Can. J. Zool. 40:593-603. Wuori, A. F. 1962. Supporting capacity. of processed snow runways. USA CRREL Tech. Rep. 82. ·, __ 53 1963. Snow stabilization for roads and runways. USA CRREL Tech. Rep. 83. PREPARED BY: John W. Coady Game Biologist SUBMITTED BY: Richard H. Bishop Regional· Research Coordinator 54 APPROVED BY : ARLIS Alaska Resources Library & Information Service: Ailchc~2:!e. 1-Jask;:J