Skip to main content
ARS Home » Plains Area » Miles City, Montana » Livestock and Range Research Laboratory » Docs » Range History

Range History
headline bar

Rangeland Management Research at Fort Keogh:
A Historical Perspective

M.R. Haferkamp, Rangeland Scientist


Range management research was initiated at Fort Keogh by the U. S. Forest Service in 1932. The first research focused on quantifying the effects of various stocking rates on the rangeland vegetation and cow-calf production. Since then, a wide array of studies has been conducted; all of which were designed to assist management personnel in their attempt to garner a living from the land while maintaining the highest level of stewardship possible. The objective of this paper is to broadly review the findings from these earlier studies.

Grazing Management

The first grazing study was begun in 1932. Two sets of pastures were used in this 24-year study with one set grazed for 6 months during winter and the other for 6 months during summer. Stocking rates averaged 1.6, 2.6, and 3.25 ac/AUM (animal unit month) for the heavy, moderate, and light grazing treatments, respectively. Response parameters monitored included basal area cover of vegetation, vegetation height-weight relationships, efficacy of sagebrush and cactus, changes in soil bulk density, organic matter content, water infiltration rates, and livestock performance.

Study results were published in several publications during the period from 1935 to 1970. Researchers concluded that moderate stocking at 2.6 ac/AUM (30.5 ac/cow yearlong) was advantageous to vegetation, soils, and livestock and that at this rate about 55% of the western wheatgrass plants were grazed on hills and uplands. The 30.5 ac recommendation was used extensively by producers and government agencies to establish appropriate stocking rates on Northern Great Plains rangeland. Use of this standard improved range management practices in the region by reducing rates of soil loss, increasing plant growth, and ultimately increasing production of both domestic livestock and wildlife.

Because amount and distribution of precipitation varied widely during this study, considerable information was derived concerning the impact of drought on the grassland vegetation. Primary effects of the extended drought of the 1930's were to reduce the total basal cover of vegetation, heights of grasses, production of palatable herbage, and livestock performance. Current year basal areas were found to primarily reflect moisture conditions during the previous growing season; whereas, mature height of grasses and production of herbage were more closely associated with current growing season precipitation. Of the major plant species, basal areas of Sandberg bluegrass increased during and decreased after the drought; whereas, those of bluegrama and western wheatgrass decreased and then increased slowly. Results showed also that reductions in the basal areas of threadleaf and needleleaf sedges, prairie junegrass, and buffalograss during the drought were less dramatic than for bluegrama and western wheatgrass and that recovery was more rapid. Prickly pear cactus and dominant shrubs were also sensitive to the drought conditions.

Prickly pear increased rapidly on all ranges immediately following the drought but then decreased again after 1940. This decrease coincided with a period of damp-cool weather during which other vegetation was approaching predrought status. Silver sagebrush decreased 50 to 100% with drought. Reestablishment was rapid on bottomlands after 1938, and by 1943, plant densities were 7 times greater than predrought densities. Big sagebrush also decreased with drought, and by 1938, 64 to 85% of the plants were dead. Post-drought recovery on upland sites was not as rapid as for silver sagebrush on bottomlands. Researchers reported also that heavy grazing appeared to favor proliferation of silver sagebrush but that it was detrimental to big sagebrush.

It was from these early findings at Fort Keogh that researchers first begin to realize that changes in perennial grass cover often follow very complex patterns that are highly variable. Weather, plant species composition, kind of grazing animal, and possibly other factors were found to modify temporal patterns of degradation and recovery. They found that responses in amount and composition of vegetation may be relatively insensitive for judging adequacy of management on these ranges over periods as short as 3 to 5 years. On the other hand, they found that the size and growth characteristics of many grasses appeared to be sensitive and reliable indicators of changes in range condition, with dwarfing or decrease in height occurring similarly with unfavorable weather or overgrazing. Researchers also concluded that natural increases and decreases of prickly pear were caused by weather and that fluctuations in density of sagebrush were more suited to evaluating long-term rather than short-term trends. They cautioned against the indiscriminate use of fringed sagewort as an indicator of range condition because severe range deterioration may occur long before increases in fringed sagewort are apparent. Researchers also reported that reductions in surface litter, soil porosity, organic matter content, rate of water infiltration, and depth of rainfall penetration occur during early stages of range deterioration long before changes in vegetation parameters are markedly apparent. They hypothesized these effects may carry over for at least 1 year following complete rest from grazing.

Researchers found also that grasshoppers as well as cattle selectively prefer current year's green vegetation to carry over herbage. They found grasshoppers often ate the new high quality herbage as rapidly as it grew, leaving only low quality old herbage for livestock. Grasshopper's affinity for grasses during drought was further explained by cooperating scientists in the 1970's. They found that grasshoppers detect and preferentially feed on grasses treated with the amino acids proline and valine, which commonly increase in plants under drought stress. Results suggest this ability might lead to insect concentrations on drought-stressed and nitrogen-enriched plants.

Research at Fort Keogh during the 30's and 40's also showed that drought induced losses in forage production severely impacted livestock performance. Drought not only resulted in depleted ranges and light weight cattle but also greatly increased costs of production by increasing the need for supplemental feeding and reducing returns from cattle sales. Reductions in livestock performance (e.g., reduction in calf birth weight, rate of gain, weaning weight, weaning grade, cow fertility, and spring and fall weights of dry cows) were more dramatic on heavily stocked compared to moderately or lightly stocked ranges. Research showed calves produced on heavily grazed pasture failed to attain weights equal to contemporaries from moderately and lightly grazed pastures even when fed hay free choice their first winter. Animals still suffered setback at 18 months of age after being fed free choice hay in winter and grazed on lightly stocked range the second summer.

In another early grazing study (1936-1941), researchers found that a minimum of 5.4 ac were required for an 8-month period beginning in March for normal development and maintenance of a dry yearling ewe. Resulting utilization (percent of plants grazed, stubble height) was 45%, 0.6" for blue grama; 35%, 2.5" for western wheatgrass; and 35%, 0.8" for threadleaf sedge.

Few grazing studies were conducted at Fort Keogh during the 1960's and 1970's. However, in the 1980's, researchers again initiated grazing studies designed to evaluate the impact of various range improvement practices on yearling steer gains during summer. Study results revealed that gains often peaked in late July to early August and then decreased thereafter. Results revealed also that some increase in gains could be expected with certain improvement practices.

In summary, Ellison and Woolfolk concluded that "proper" grazing intensities for the Northern Great Plains are those that insure a forage reserve is accumulated on an annual basis during "normal" years. They concluded that such an accumulation would provide protection during future droughts and encourage rapid post-drought recovery. This concept is as important today as it was in the 1930's as drought continues to be an important factor affecting the structure and function of grassland ecosystems including Northern Great Plains rangelands.

Range Improvement Practices

Range seeding. Increasing forage production on rangelands has always been an important concern in rangeland environments. Research at this and other stations has been designed to determine relationships between environment and both desirable and undesirable plant species and to develop methods of range seeding, soil tillage, fertilization, and weed control that will enhance production.

Adaptation trials comparing plant species and varieties have been conducted for many years at the Station. Of the many species evaluated at Fort Keogh, crested wheatgrass and Russian wildrye have proven to be the 2 most productive introduced perennial grasses evaluated in that both provide large amounts of high quality, cool-season forage, and as such, fit well into complimentary grazing schemes with native range. Both of these species and the dominant native grass, western wheatgrass, were found to survive the drought in the early 1980's better than many of the grasses that were currently being recommended for rangeland seeding. However, during the drought of 1988, stands of crested wheatgrass and Russian wildrye were found to decline. In other studies, both of these species, when seeded alone or with alfalfa and used as spring range, were found to be highly productive which resulted in increased livestock weight gains and in some cases improved reproductive efficiency.

During the 1980's, researchers working cooperatively with ARS plant breeders in Logan, Utah, evaluated early generations of `Newhy' a bluebunch wheatgrass x quackgrass hybrid. Results show this hybrid is well adapted to growing on saline soils that have been seeded historically to tall wheatgrass. But, in contrast to tall wheatgrass, hybrid plants are leafier, more palatable, and maintain a higher forage nutritional value into fall. Researchers at this location identified lines that were selected most frequently by steers and the characteristics of the plants associated with this preference. Variables such as basal area, height, phenology, leafiness, silica content, and digestibility were related to preference in some but not all trials. Carbohydrate content was also related to preference in some trials. Work was continued with this hybrid into the late 1980's to evaluate response of plants to fertilizer, defoliation by mowing, and irrigation. Researchers concluded that irrigated, nitrogen fertilized pastures of RS-2 harvested in midsummer, rested, and utilized again in fall or early winter should provide large amounts of high quality forage for grazing livestock in the Northern Great Plains.

Seedling establishment is a critical phase in establishing a productive seeding. Cool-season perennial grasses are often planted in late fall before soils freeze. In these dormant plantings, seeds germinate as the soil thaws and seedlings emerge thereafter. Research in the early 1980's showed that winter damage and the subsequent number of viable leaves present when growth began in spring were highly correlated with amount of spring and fall seedling growth in crested wheatgrass, Russian wildrye, and pubescent wheatgrass. As number of leaves increased from 1 to 4 during spring there was a corresponding increase in subsequent growth. Results suggested seeding in late summer in the Northern Great Plains would hasten stand establishment and reduce the length of grazing deferment necessary on newly seeded stands. However, the key to success with this seeding strategy is availability of adequate subsurface moisture that will allow seedlings to reach their optimal growth stage after germinating in late summer or fall. The amount of subsurface moisture required has not been quantified. As for optimal stage of development, no seedlings with more than 2 leaves died during winter and no seedlings with more than 3 leaves showed heavy or severe damage. Decrease in winter damage leveled as plants attained 3 leaves and as heights reached 2.6". This study was conducted during a winter with above average snowpack, and an open winter may provide different results.

Seeding techniques that improve stand establishment are constantly in demand. The Range Improvement Machine (RIM) developed by Currie and Erickson in the early 1980's showed potential for enhancing establishment of seeded stands on Northern Great Plains rangelands. Seedling establishment was significantly greater with the RIM compared to a power-till drill. The RIM created a relatively competition free seedbed furrow. Measurements from other studies using the RIM showed that treated areas enhanced soil water infiltration and retention.

Soil tillage and furrowing. Several methods of modifying rangeland soil surfaces have been evaluated at Fort Keogh. Range pitting was shown to reduce soil moisture stress and subsequently increase forage production on overflow and clayey range sites during a 6-year study in 1957-62. Contour furrowing was apparently first utilized to improve forage production at Fort Keogh during the drought of the 1930's, but the effects of such were not quantified until it was used in studies in the 1970's and 1980's. In one 5-year study, from 1974 through 1978, forage production was increased on an upland, medium textured range site from 538 lbs/ac on nontreated native range to 1206 lbs/ac with contour furrowing and seeding alfalfa. Furrowing applied with a contour furrower or the RIM effectively increased forage production during an 8-year period from 1983 to 1990. A series of treatments applied with the contour furrower or RIM machine produced June standing crops which averaged 286 lbs/ac greater than the 518 lbs/ac produced on the control pasture. Soil tillage and furrowing techniques were most effective on the heavier textured soil. Results showed the techniques increase production by holding moisture on site. Researchers also hypothesized that the increase was related to a releasing of nitrogen from destroyed vegetation. In some cases, soil nitrogen has been further enhanced by fertilization, seeding legumes, or both. These increases in forage production have resulted in some increases in livestock production when compared to untreated native range. In related studies on these pastures, researchers found also that sagebrush removal and interseeding of legumes did not markedly affect total grasshopper population trends but did influence species composition and provided additional food plants for some species.

Water enhancement. Beginning in 1936, water spreading systems were developed by building diversion dams and contour dikes. These studies were among the first in the U.S. to demonstrate that water normally lost to run-off could be used effectively to increase growth of native and introduced grasses. Research by cooperating agencies was conducted on the Station in the early 1980's to determine the response of vegetation to water availability at levels (0.24", 0.48", or 0.98"/wk) that might be applied through irrigation or cloud seeding. In a study with western wheatgrass, they found plant culm length was increased 6 to 7% with light showers (0.24 - 0.39") and 33-107% when soils were kept moist, but aboveground yields were significantly increased only by the wet treatments. Others reported in associated studies that water deposited from natural summer rainfall on these grasslands is likely to be lost in less than 2 days because 92% of the rain showers are <.4".

Fertilizer applications. Fertilization with nitrogen has been an effective method to obtain relatively rapid increases in forage production, particularly on overflow range sites. Research at Fort Keogh has shown that applying 30 lbs N/ac to native range may be economically profitable in some years. Nitrogen has also been applied in combination with other treatments such as contour furrowing. In one 5-year study, forage production on an upland medium textured range site ranged from 538 lbs/ac on nontreated native range to 1206 lbs/ac with contour furrowing and seeding alfalfa. Additions of 100 lbs/ac of nitrogen and 13.4 lbs/ac of phosphorous on interseeded range increased herbage production to 1480 lbs/ac. Average beef production over the 5-yr period was 20 lbs/ac on the control, 39 lbs/ac on the contour furrowed range, and 47 lbs/ac on the contour furrowed-fertilized range.

Burning. Burning has been used effectively in other regions to manage undesirable plants and increase herbage production. Burning in spring or fall was studied in the early 1980's at Fort Keogh to determine its effect on increasing production of grasses and decreasing stands of silver sagebrush. Results showed that spring burning with good soil moisture resulted in low mortality of silver sagebrush; whereas, fall burning with low soil moisture resulted in considerable mortality and reduced shrub growth. As fire intensity increased, mortality increased and regrowth decreased. Data from a study conducted in 1983 suggested spring burning of western wheatgrass and blue grama stimulated herbage production by mid- and late June; whereas, fall burning also stimulated herbage production but to a lesser degree. However, earlier season herbage production was not enhanced by this burning treatment. Researchers at Fort Keogh have shown that greatest reductions of downy and Japanese brome occur with fall burning, but good reductions are obtained with spring burning. When bromes were reduced, the remaining blue grama outproduced that growing in control plots. In general, the results from these studies showed that fire appears to be an effective tool for managing vegetation in this area of the Northern Great Plains.

Herbicides. Very few herbicide studies have been conducted at the Station. This reflects a lack of large amounts of undesirable plants that are actively competing with native vegetation. Houston and Woodward wrote: "These ranges appear resistant to extreme reduction in range condition due to heavy stocking. It would seem that the composition and vigor of the native vegetation under prolonged heavy grazing and below-normal precipitation are considerably reduced, but then they tend to stabilize at a low but fairly constant level. At this level, competition is still sufficient to prevent large-scale replacement by invader species. The increase of sod species under heavy stocking is probably the primary reason for this."

However, annual bromes have invaded ranges in this region, and some of the more aggressive perennials are present in varying densities. Thus, the use of soil-active herbicides has been studied at Fort Keogh as a method of controlling annual bromegrasses in seeded stands of crested wheatgrass, pubescent wheatgrass, Russian wildrye, and native rangeland dominated by western wheatgrass and blue grama. Results showed yields of annual bromegrasses averaged 91 and 47% less than controls the first and second year post treatment. Averaged over 4 study sites, bromegrass yields in 1983 and 1984 were 88 and 50, 88 and 8, and 94 and 84% less than the control in the atrazine, propham, and pronamide treatments. Yields of perennial grasses were increased from treatment the first year post treatment.

Most efforts at the Station and other locations to control brome have been shown to provide only short-term relief. Thus, current research is underway to gain a better understanding of the impact of Japanese brome on Northern Great Plains rangelands. Ongoing studies are examining: 1) the environmental effects on annual brome seed germination; 2) the impact of brome competition on western wheatgrass production; 3) the effect of brome on quantity and quality of forage produced and on livestock production; and 4) the impact of defoliation and nitrogen on growth and development of Japanese brome.

Autecology. Plains silver sagebrush is an important shrub in the Northern Great Plains. It is found on deep, lowland soils of floodplains and is generally considered to increase in response to cattle grazing. There are about 12.8 million ac of silver sagebrush in Montana. In the early 1980's, studies were implemented to examine the reproductive characteristics and mechanisms which are important in the success and maintenance of plains silver sagebrush. They determined that wind appears to be the most influential factor in the dispersal of achenes from plants. Researchers also found seeds germinate under a variety of environmental conditions, but greater germination occurs: 1) from seeds collected late in the growing season; 2) in complete darkness compared to light when plants are under some water stress; and 3) at a temperature of 68oF. All rates of germination declined with increasing water stress. Most plants, even small seedlings, showed some degree of rhizomatous (vegetative) growth. Phenological development, plant water potential, and soil water status were monitored on established plants. They determined phenological development could be predicted by using both plant water potential and calendar date.

Research at Fort Keogh has shown that weather has a far greater effect on abundance and vigor of plains prickly pear than any other influence investigated. Influences of weather tend to be cumulative up to a point, but weather may also have opposing influences. Precipitation is an important influence on the infestation of the prickly pear clumps by several insects, with some being favored by high and some by low precipitation. Thus, insect infestation is almost continuous. Data indicated that soil moisture and soil texture were most important soil characteristics, but topography or microenvironmental differences were also important. In one study there was an almost non-existent overall effect of stocking rate, at least in the range of 1.8 to 3.2 ac/AUM, on abundance and vigor of plains prickly pear. These findings were in contrast with findings of several other investigators. Stocking rates, however, interacted strongly with weather and soils in that a decrease in pads occurred on upland soils; possibly through trampling, soil compaction, and perhaps utilization. This reaction was, however, probably modified by weather.

Methods and Techniques

As in most phases of range research, development of techniques and equipment to assist in monitoring changes in vegetation and livestock performance is an ongoing process. For example, it was shown in the early 1950's that plant vigor, as expressed by the height of western wheatgrass plants, varied with range condition, amount of protection afforded by prickly pear, and annual precipitation regimen. The study established the utility of this system of measure of vigor as a criterion for the appraisal or estimation of range condition.

During the 1980's, several trials were conducted comparing different capacitance meters for estimating standing crops on native and seeded ranges on the Station. This work helped define the shortcomings of various meters and how plot size could be adjusted to obtain better relationships between clipped samples and meter readings.


The range research program at Fort Keogh is actively investigating vegetation-environment-animal relationships. Specific ongoing studies include studies designed to: 1) elucidate the impacts of annual bromes on the quantity and quality of forage produced and consumed by beef cattle; 2) quantify the impact of annual bromes on livestock production; and 3) quantify the interaction effects of livestock grazing and drought on rangeland ecosystems in general and quantity and quality of forage produced, root growth, soil water dynamics, water quality, and seed bank dynamics specifically. These studies are designed to produce data that will provide a greater understanding of these relationships and will allow development of management strategies that will promote efficient and ecological sound use of Northern Great Plains rangelands.


Key Scientists Associated with Rangeland Research Program at Fort Keogh1


Years at Fort Keogh3

Affiliation2 or Position

R.W. Collins

1930's - 1940's


L.C. Hurtt

1930's - 1940's


L. Ellison



E.J. Woolfolk

1930's - 1950's


C.E. Holscher

1940's - 1950's


W.R. Houston

1950's - 1960's


R.A. Peterson



M.J. Reed



L.R. Short

1940's - 1950's


R.S. White

1977 - 1987


P.O. Currie

1978 - 1989


M.R. Haferkamp

1988 - present


R.K. Heitschmidt

1990 - present


J.D. Volesky

1988 - 1989


M.M. Borman

1989 - 1990


M.G. Karl

1990 - present


Current Rangeland Research Support Staff

B. Bennett

1992 - present

Research Associate

C. Murphy

1978 - present

Bio. Lab. Tech.


1 This list of personnel is admittedly incomplete. The reason it is incomplete is because historically much of the range research that was conducted at Fort Keogh was conducted by non-resident scientists, and in many instances, by non-USDA Agricultural Research Service employees. Thus, detailed documentation of most cooperating scientists' careers at Fort Keogh are either totally absent or at best incomplete.

2 USDA Forest Service = USFS; Montana Agricultural Experiment Station = MAES; and USDA Agricultural Research Service = ARS

3 Many estimates based on publication records.


This Historical Perspective was published as part of the 1993 Field Day.


USDA, ARS Fort Keogh Livestock and Range Research Laboratory
243 Fort Keogh Rd., Miles City, MT  59301-4016
Phone: 406-874-8200, Fax:  406-874-8289