On-line Book

Book Cover






Study Area

Elk Population

Population Dynamics


Elk Habits

Effects on Habitats

Elk Management






Research Report GRTE-N-1
The Elk of Grand Teton and Southern Yellowstone National Parks
National Park Service Arrowhead


Habitat Use

Patterns of elk habitat use within Grand Teton and refuge valley areas and mountain areas extending into southern Yellowstone Park were determined by recording the numbers of animals observed on the different vegetation types while covering routes. Multiple records were made when undisturbed animals moved from one type to another. Animals were about equally observable when they were using all but the forest types. Numbers observed in the forest types during the August through October period of greatest use were weighted by assigning the difference from maximum July counts to these types.

Valley Areas

General patterns of habitat use within valley areas were shown by 82,223 recorded elk observations between April and December (Figure 10). The sample included 42,237 May through October observations from 1963 and 1964 that were made by Martinka (1965). Additional observations were obtained within and on either side of this period from 1963 through 1966. The average sample size for monthly periods was 9,136 (5,525 to 19,398). April, November and December averages were from selected samples when large numbers of elk were freeranging off refuge feed grounds.

Fig. 10. Relative use of different vegetation types in valley areas as indicated from observations of 82,223 elk between April and December, 1962-1966. (click on image for an enlargement in a new window)

The recorded observations from late April illustrate the animals' maximum use of the bunchgrass-shrub type on upland slope sites after they dispersed from feed grounds and ranged over the northern portions of the refuge. Extensive areas of the type on an alluvial fan and hayfields were used to a greater extent when the animals first moved off feed grounds. May observations illustrate the animals' progressive shift to the sagebrush type as they migrated onto and used valley floor areas in Grand Teton Park. The bunchgrass-shrub type on south slopes remained important for its more advanced stages of vegetation growth. The predominant use of the sagebrush type for foraging is shown by the June through August observations.

Shifts to greater use of forest types occurred in August and increased during September. The October observations showed the animals increased their use of the valley meadow type and hayfields. This mainly occurred in October of 1962 and 1963 when large numbers of elk migrated to the refuge. Restrictions of human disturbances in Grand Teton areas closed to hunting and special hunting seasons reduced these movements in subsequent years and led to the greater use of the sagebrush type within the park.

November and December observations illustrate the animals' use of extensive bunchgrass-shrub and valley meadow types. Use at this time of year appeared to be almost exclusively on high producing bottomland sites or within swales on uplands. Variable additional use occurred on hayfields in some years. This appeared to depend upon whether fall rains caused regrowth or they had been left uncut. The start of artificial feeding in January caused the majority of the animals on the refuge to abruptly leave natural food sources and move onto feed grounds.

Mountain Areas

Patterns of use on different vegetation types in mountain areas were shown by 20,017 recorded elk observations between June and October of 1962 through 1966 (Figure 11). Sample sizes for periods ranged from 641 during the first half of June to 6,543 during the last half of July.

Figure 11 shows that the majority of the animals arriving in mountain areas in early June used the relatively low elevation valley meadow (willow-sedge stage) and forest park types. These were principally female-calf groups. Late June through July observations illustrate progressive shifts in elk use from the low elevation types to the intermediate elevation burn type and finally, the herbland type. Maximum use of the highest elevation subalpine meadow type usually occurred during late July. August through October observations illustrate the elks' dispersals into and predominant use of forest types.

Fig. 11. Relative use of different vegetation types in mountain areas as indicated from observations of 20,017 elk between June and October, 1962-66. (click on image for an enlargement in a new window)

In 1962, a late emergence of molesting insects appeared to hold substantial numbers of elk on the herbland and subalpine meadow types into the first half of August (Table 12). In late July and early August of 1964, molesting insects were scarce. Elk occurred in scattered small groups and used forest and low elevation forest park types to the greatest extent observed during the study. In 1965, snow accumulations along migratory routes appeared to preclude the usual use of lower elevation mountain areas for calving. Large female-calf groups moved directly onto Yellowstone's high elevation herbland types in late July. In 1966, an early July emergence of molesting insects caused concentrations on high elevation herblands comparable to late July and early August highs observed in other years. Subsequent late July and early August use of high elevation ranges was the lowest observed during the study.

Food Habits

Information on food habits was obtained by recording 262,602 instances of plant use at 473 elk feeding sites. The unit for recording one instance of use from April 16 through November 15 was a rooted stem for forbs and single-stemmed grasses, a leader for shrubs, or a distinct clump of stems for bunchgrasses. A unit recorded during this period could be considered roughly equivalent to a "bite."

During the November 16 through April 15 winter period, elk frequently grazed bunchgrasses closely. On sites where shrubs were also used the above system was modified. Recorded instances of use on bluebunch wheatgrass, or other grasses with a similar life form that were grazed to average heights of 1, 2, 3, or 4 inches, were multiplied by weight values of 25, 20, 15 or 10, respectively. Instances of use on needle-and-thread, or other comparable grasses that were grazed to 0.5, 1, or 2 inch average heights, were multiplied by weight values of 11, 8, or 5, respectively. These weight values made units of use on bunchgrass plants more equivalent in weight to units for shrubs and the few forbs used during this period.

The base weight values of 25 and 11 were obtained by dividing the average air dry weight of 100 2-inch shrub leaders (N = 400) into the air dry weight of 100 bluebunch wheatgrass plants clipped to 1 inch and 100 needle-and-thread plants clipped to 0.5 inch. Leader samples were from Douglas rabbitbrush and winterfat and weights were approximately equal. Weight values for other bunchgrass stubble heights were obtained from proportion calculations that utilized the "utilization gauge" developed by Lomason and Jenson (1938).

The only departure from the described method was that the side branches of tall larkspur and individual stems of giant wild rye plants were considered special units for recording one unit of use. Most feeding sites were examined while fresh use could be detected, but were allowed to accumulate use over known periods when elk were only animals involved. Examinations were purposely restricted to sites that had not been used to the extent that animal preference obscured.

Elk use was mainly on cured grass during the November 16 through April 15 winter season. New growth of grasses usually became increasingly availabile as a food source through the first half of spring; forbs, during the last half of the period (April 16 to May 16 or June 30). Variation occurred with late and early springs. Forbs were readily available on most habitat types through the summer (up to September 15), though some cured to the extent that they ceased to be used. The freezing or curing of forbs in fall (September 16 to November 15) resulted in elk shifting their use to other forage sources.

Plant terminology generally follows Davis (1952) and Booth and Wright (1959), except that grasslike sedge and rush species are included in the grass forage class. Grazed and vegetative forms of fleabane or aster plants could not be readily identified and are collectively called asters. Instances of elk use on plant species (or related groups) were calculated as percentages of the total recorded use at each feeding site. Percentages were averaged for the different vegetation types by seasonal periods. Results by forage classes are shown in Table 16. Plants that received 5 percent or more of the recorded use and/or were used at 25 percent or more of the sites are shown in Appendix Tables V through IX. An alphabetical listing of common and scientific plant names is given in Appendix I. Discussions of food habits by seasons follow.


Grasses were the predominant class of forage used by elk feeding on the upland bunchgrass-shrub and valley meadow types (Table 16). Bluebunch wheatgrass and Douglas rabbitbrush were the two most important plant species used on the upland type as a whole (Appendix V). Needle-and-thread or bluegrasses were additionally important food items within swales during the first half of winter and again on a variety of other upland sites as new growth started during late winter. Combined use on these four plant species averaged 79 percent of the recorded instances of use. Sedge, bluegrass, and willow were the most important species used on the valley meadow type (Appendix VI). Combined use averaged 76 percent.

Shrubs were the predominant class of forage used by elk feeding within forest types during winter. Willow, narrowleaf cottonwood, aspen, and silverberry in combination with bluegrasses were the most important items used in aspen or cottonwood stands (Appendix VII). Combined use was 70 percent. A variety of shrubs and sedge served as the main food source for animals feeding within coniferous forest stands (Appendix VII). Combined use was 72 percent.

Table 16.—Instances of plant use by forage class at 473 elk feeding sites on different vegetation types during seasonal periods.1

Seasons2 Vegetation types No.
of use
Grass3 Forbs Shrubs

Winter Bunchgrass-Shrub51163,83772  226
Valley-Meadow86,15271  227
Forest (Deciduous)43,01643  651
Forest (Coniferous)99,65737  954
Spring Bunchgrass-Shrub244,60675 1411
Valley-Meadow197,83660 36 4
Sagebrush438,25754 44 2
Forest Park378,10733 66 1
Burn51,42257 42 1
Herbland51,10248 52 0
Forest (Deciduous)164,12750 49 1
Forest (Coniferous)183,05340 3921
Summer Valley-Meadow91,484 23 770
Sagebrush597.637  492 4
Forest Park335,369 1482 4
Burn152,546 1383 4
Herbland305,488 1585 0
Subalpine Meadow143,351 5050 0
Forest (Deciduous)173,256 255025
Forest (Coniferous)263.960 104743
Fall Valley-Meadow3 2,63293 5 2
Sagebrush10 1,3487425 1
Forest Park6 1,265562915
Herbland3 1654555 0
Forest (Coniferous)9
52  840


1 Data for valley areas involving 5,361 instances of use included from 44 sites examined by Martinka (1965).

2 Winter, November 16-April 15; spring, April 16-June 15 in valley areas, May 16-June 30 in mountain areas; summer up to September 15; fall up to November 15.

3 May include grasslike sedge and juncus spp.


Grass continued to be the predominant class of forage used on most types over the spring season as a whole. Forb use about equaled or exceeded that on grass on half the types. The bunchgrass-shrub type was mainly used before forbs became readily available. Bluebunch wheatgrass and bluegrasses were the most important food items (Appendix V). Combined use averaged 58 percent.

Bluegrass and sedge species were the most important plants used on the valley meadow type (Appendix VI). Combined use was 55 percent. Bluegrass, Idaho fescue and balsamroot were the most important species used by elk moving onto or migrating over the outwash plain, sagebrush type during spring (Appendix VIII). Combined use on these species was 42 percent.

Sedge, mountain and nodding brome, geranium, dandelion and other forbs served as the main food items for elk feeding within the forest park type during spring (Appendix IX). Forb and grass use was 66 and 33 percent respectively. Mountain brome, sedge and geranium were the most important plants used on the burn type during spring (Appendix IX). Combined use was 58 percent. Mountain brome, slender wheatgrass, violet and agoseris were the most important items on the high elevation herbland type. Combined use averaged 65 percent.

Bluegrass and a variety of forbs served as the most important food for elk feeding in deciduous forest stands (Appendix VII). Sedge and pinegrass were the most important individual items in coniferous stands. Aggregate forb and grass use was about equal in both deciduous and coniferous stands, but shrubs continued to be used to a greater extent within the latter.


Forbs were the predominant class of forage used by elk on all but the valley meadow and the highest elevation subalpine meadow types. Willows were the most important food item on the former type, averaging 70 percent (Appendix VI). Sedge and aster species and lupine were the most important food items on the subalpine meadow type, aggregating 62 percent (Appendix IX).

Lupine was the most important single food item on the sagebrush type over the summer as a whole, averaging 50 percent (Appendix VIII). It was used to the greatest extent after other vegetation cured during late summer or was killed by frost. Agoseris, buckwheat, little sun flower, and balsamroot were more important during the early summer. Combined use on these and lupine was 79 percent.

Agoseris, potentilla, dandelion, and aster were the most important forb species used by elk feeding on the forest park, burn and herb land types during the summer (Appendix IX). Combined use averaged 55, 42, and 60 percent for each type, respectively. These were obviously highly preferred forage plants which were selectively used even on sites where they were a minor component of the vegetation. The species had widespread distribution. However, they were relatively abundant on sites that had been disturbed by pocket gophers (Thomomys talpoides) on immature soils, and on ridgetop sites that elk may have maintained in a disclimax stage.

Aster, dandelion, and agoseris were the most important forbs used in both deciduous and coniferous forest stands during the summer (Appendix VII). These three species, nodding brome, wheatgrasses, and aspen received 51 percent of the elk use in deciduous stands. They received 56 percent, in combination with lupine, spirea, and huckleberry, in coniferous stands.


Grass was the predominant class of forage used on most types during fall. Bluegrass was the most important single item on sampled valley meadow types, averaging 91 percent of the recorded use (Appendix VI). Junegrass, bluebunch wheatgrass, bluegrass, and lupine were the most important on the sagebrush type. Combined use was 70 percent (Appendix VIII). The increased use of grass on these types appeared to be in response to most forbs losing their succulence and fall rains causing new leaf growth on bluegrasses and junegrass.

Mountain brome and lupine appeared to be the main plants used by the relatively few elk that continued to forage on the herbland type during fall (Appendix IX). Combined use was 59 percent. Sedges and the aster-fleabane-goldenrod group of plants were most important food items within the forest park type. Combined use was 76 percent. Geyer's sedge, pachystima, spirea, and huckleberry were the most important plants used by elk feeding within coniferous forest stands (Appendix VII). Combined use averaged 79 percent.

Seasonal and Yearlong Averages

The two forest types used during winter were estimated to provide at least 10 percent of the November 15 through April 15 diet for free-ranging elk as a whole. Proportionately weighted grass, forb, and shrub averages for the two forest types and the nonforest types which provided the main source of winter food were 69, 2, and 29 percent, respectively.

Respective grass, forb and shrub averages were 50, 45, and 5 percent for the different habitat types used during spring; 19, 62, and 19 percent for summer. The unadjusted values would be expected to vary with late springs resulting in an increased use of grass and shrubs; early springs, an increased use of forbs. The unweighted summer values probably would not vary greatly during years when animals used certain habitat types to a greater extent than others. The general use of different habitat types through spring and summer was considered to preclude any meaningful weighting of averages.

Recorded observations of elk indicated that their relative use of forest and nonforest types may have been in a ratio of about 75 to 25 percent during the fall (Figure 11). By this relationship, fall use of grass, forbs, and shrubs averaged 56, 13, and 31 percent, respectively.

The winter, spring, summer, and fall forage class averages extended over about 5, 2, 3, and 2-month periods, respectively. Proportionate weighting of seasonal forage class figures gave an average yearlong food habit of about 51 percent grass, 26 percent forbs, and 23 percent shrubs for free-ranging elk. These values could be expected to vary with elk groups using particular vegetation types extensively, during years with extreme weather, or with animals that also used artificial food on feed grounds. What is perhaps best shown by the food habits study is that elk are extremely versatile and generalized feeders on all classes of forage and on a variety of plant species. Important winter food plants were major components in either seral stages or climax stands. This obviously enables the elk to contend with a broad spectrum of environmental change and persist as a faunal dominant in the Grand Teton and Yellowstone environments.

top of page Top

Last Modified: Tues, Jan 20 2004 10:00:00 pm PDT

National Park Service's ParkNet Home