3
Current Methods of Rangeland Assessment

The current theory and practice of rangeland assessment have a long history that is closely related to the ways that rangelands were used and studied. The nineteenth century was a period of exploration and development of the rangelands of the western United States. The need for systematic methods of rangeland assessment first became apparent when Jared Smith was sent by the U.S. Botanical Survey in 1895 to study the causes of the deterioration of western rangelands that had been widely reported in the late 1880s. He reported that:

The shortage of cattle all through the west is due to the fact that ranges were stocked up to the limit that they would carry during the series of exceptionally favorable years preceding the years of drought. Then followed the bad years when the native perennial grasses did not get rain enough to more than keep them alive. The cattle on the breeding grounds of the West and Southwest died by the thousands of thirst and starvation (Smith, 1896:322-323).

Such early investigations, however, were not based on a unifying science that could systematize the data collected to assess rangelands or relate the effects of livestock grazing to the rangeland deterioration that was evident in the late nineteenth century. The need for more thorough assessments was evident.

DEVELOPMENT OF CURRENT THEORY AND PRACTICE OF RANGELAND ASSESSMENTS

Between 1890 and 1905, 11 state agricultural experiment stations published 879 range management-related bulletins dealing with the control of weeds, pests, poisonous plants, soil moisture, fertility, conservation, rangeland inventory and analysis, water use, fencing, and other topics (Beetle, 1954). The U.S. Forest Service (USFS), which was formed in 1905,

   

Scarlet globemallow (Sphaeralcea coccinea)



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands 3 Current Methods of Rangeland Assessment The current theory and practice of rangeland assessment have a long history that is closely related to the ways that rangelands were used and studied. The nineteenth century was a period of exploration and development of the rangelands of the western United States. The need for systematic methods of rangeland assessment first became apparent when Jared Smith was sent by the U.S. Botanical Survey in 1895 to study the causes of the deterioration of western rangelands that had been widely reported in the late 1880s. He reported that: The shortage of cattle all through the west is due to the fact that ranges were stocked up to the limit that they would carry during the series of exceptionally favorable years preceding the years of drought. Then followed the bad years when the native perennial grasses did not get rain enough to more than keep them alive. The cattle on the breeding grounds of the West and Southwest died by the thousands of thirst and starvation (Smith, 1896:322-323). Such early investigations, however, were not based on a unifying science that could systematize the data collected to assess rangelands or relate the effects of livestock grazing to the rangeland deterioration that was evident in the late nineteenth century. The need for more thorough assessments was evident. DEVELOPMENT OF CURRENT THEORY AND PRACTICE OF RANGELAND ASSESSMENTS Between 1890 and 1905, 11 state agricultural experiment stations published 879 range management-related bulletins dealing with the control of weeds, pests, poisonous plants, soil moisture, fertility, conservation, rangeland inventory and analysis, water use, fencing, and other topics (Beetle, 1954). The U.S. Forest Service (USFS), which was formed in 1905,     Scarlet globemallow (Sphaeralcea coccinea)

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands Timeline of Rangeland Classification and Inventorying in the United States Rangeland Use and Management Year Inventory Spanish settlers introduce sheep and cattle into California rangelands. 1500   British settlers expand livestock grazing to more rangelands in the western United States. 1800   Major increase in livestock on rangelands in Texas, Kansas, Nebraska, Oklahoma, and the Great Plains. 1860   Overgrazing of the Great Plains and the Great Basin is accompanied by drought and severe erosion. 1880     1895 Jared Smith conducts a survey of depleted western rangelands for the U.S. Botanical Survey. He reports that overgrazing and drought are responsible for widespread rangeland degradation.     U.S. Department of Agriculture begins to publish reports on forage condition and grazing problems in the western United States. U.S. forest reserves are established, bringing part of western rangelands under the jurisdiction of the U.S. Department of Agriculture. 1897 Surveys of forest reserves begin.   1899 H.C. Cowles describes plant succession on the sand dunes of Lake Michigan.   1900 State agricultural experiment stations begin to issue reports on the condition of rangelands and range management practices.

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands Rangeland Use and Management Year Inventory U.S. Department of Agriculture establishes regulation of grazing on forest reserves. 1901   The U.S. Forest Service (USFS) is formed and the National Forest system is created. 1905 National forests are surveyed.   1905 F. E. Clements begins adapting succession to Great Plains grasslands.   1910 A. W. Sampson begins the first ecologically based range work of Utah.   1917 J. T. Jardine develops the first scientific survey method.   1923 A. W. Sampson introduces the use of succession as a way to assess grazing capacity. Dust Bowl ravages western rangeland and cropland. 1930     1933 A. R. Standing proposes quantifying rangeland assessments by measuring volume of vegetation rather than estimating plant cover. The Taylor Grazing Act withdraws all remaining public land that is not under the jurisdiction of the USFS or other federal agencies into grazing districts under the jurisdiction of the Bureau of Land Management (BLM). BLM is charged with halting overgrazing and soil deterioration. 1934   The Soil Conservation Service (SCS) is created. Its mandate is to inventory soil and water resources and to assist farmers and ranchers with ending erosion. 1935  

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands Rangeland Use and Management Year Inventory   1936 The U.S. Senate issues a report on the condition of rangelands and the causes of rangeland deterioration.     The interagency range survey technique is standardized in the late 1930s and used by BLM, SCS, and USFS to inventory rangelands.   1938 J. E. Weaver and F. E. Clements publish Plant Ecology, extending successional concepts explicitly to rangeland ecosystems. Plant Ecology becomes the standard text for the emerging field of range management.   1949 E. J. Dyksterhuis solidifies the use of successional stages as measures of rangeland condition by proposing the measurement of range condition as a departure from climax vegetation for a specific range site. Dyksterhuis' range site and condition method becomes the basis for SCS rangeland assessments.   1950 USFS develops the Parker three-step method of rangeland assessment.     BLM continues to use the interagency survey method. The Wilderness Act is passed. 1964   The National Forest Management Act mandates multiple use and sustained-yield policies for national forest management.       1968 R. Daubenmire describes habitat types, which become the basis for USFS forest and rangeland classifications.

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands Rangeland Use and Management Year Inventory National Environmental Protection Act requires all federal agencies to write environmental impact statements on major federal actions. 1969   The Endangered Species Act requires federal agencies to protect listed wildlife species. 1973   The National Resources Defense Council v. Morton requires environmental impact statements on all local grazing programs administered by BLM. 1974   The Resources Planning Act requires USFS to inventory national forests every 10 years.     Congress passes the Forest and Rangeland Renewable Resources Research Act, which provides the information needed for Resources Planning Act implementation.       1975 The first Resources Planning Act Assessment of resources, including rangelands on national forests, is published. Congress passes the Federal Land Policy and Management Act, which requires BLM to prepare an inventory of the resources on federal lands under BLM's jurisdiction. 1976   The Soil and Water Resources Conservation Act is passed, requiring SCS to inventory soil, water, wildlife habitat, and related resources on nonfederal lands. SCS establishes the National Resources Inventory to carry out its mandate. 1977  

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands Rangeland Use and Management Year Inventory The Public Rangeland Improvement Act calls for improvements in soil quality, wildlife habitat, watersheds, and vegetation on federal rangelands and requires inventories of federal rangelands. 1978     1979 BLM develops the soil-vegetation inventory method; the method is tested but is not formally adopted by BLM.   1980 The second Resources Planning Act Assessment is published.   1981 The first Resource Conservation Act Appraisal of soil and water resources on nonfederal lands based on the 1977 National Resources Inventory is published by SCS.   1983 BLM adopts SCS range site and range condition procedures for assessing rangelands.     The Society for Range Management recommends that SCS, BLM, and USFS adopt common terminology to classify rangelands and make ecological status ratings.   1985 The National Resources Defense Council and the National Audubon Society assemble data on the rangelands under BLM's jurisdiction and report that many rangelands are in unsatisfactory condition.   1987 The second Resources Conservation Act Appraisal of soil and water resources on the 1982 National Resources Inventory, is published by SCS.

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands Rangeland Use and Management Year Inventory   1989 The Society for Range Management assembles data from SCS, BLM, and USFS in an attempt to make a national assessment of rangelands. The society reports that the data available for federal rangelands are not adequate for a national assessment. The USFS publishes an assessment of the range resources of the United States.   1990 The third Resources Planning Act Assessment is published. recognized the need to develop a scientifically credible and economically feasible method of surveying rangelands to carry out its mandate. Since the goal of most of the early rangeland professionals was to provide high-quality livestock forage, the techniques and systems they developed for rangeland assessments concentrated on the effects of livestock grazing on forage production. The first formal attempt to develop a scientific rangeland survey method was made by James L. Jardine on the Conconino National Forest in 1910 (Chapline and Campbell, 1944). Early Development Of Survey Methods Jardine's range reconnaissance method involved a careful visual examination of the rangeland to provide a written record of the rangeland's resources. He recorded the following data: (1) a topographic map showing watering places, roads, fences, and cabins; (2) a classification of the rangeland into 1 of 10 grazing or vegetation types; (3) the percentage of the rangeland covered by each forage species; (4) a descriptive report of each grazing or vegetation type, including the suitability of each type for each kind of grazing animal; (5) a map of the timber; and (6) samples of the major species present on the rangeland (Jardine and Anderson, 1919). Jardine's survey method was highly credible in its time, but it had several shortcomings regarding forage availability estimates. For example, it was based on estimates of the ground cover of each species rather than on direct measurements of the volume or weight of the forage pro-

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands duced by each plant species, and it therefore did not give an accurate measurement of productivity or yield. Standardization of Rangeland Surveys In 1933, Standing introduced the concept of using measured volumes of vegetation rather than visual estimates of cover (Standing, 1933). During the 1930s, other modifications were made to the Jardine method, and these were finally standardized as the interagency range survey technique used by the Bureau of Land Management (BLM) and USFS. Although more quantitative than the original reconnaissance method, the interagency survey depended heavily on palatability factors and other subjective criteria for estimating forage production or carrying capacity. This method assessed, almost exclusively, forage production and livestock carrying capacity. Few if any data were collected on soil conditions, wind and water erosion, or other factors that would allow a more comprehensive evaluation of rangelands. More important, the method was not linked to any theoretical base that suggested how the forage composition data that were collected could be interpreted as indicators of ecological conditions on rangelands. Forage production, rather than the state of rangeland ecosystems, was evaluated. New Theoretical Foundation for Rangeland Surveys At the same time that Jardine was developing his method for evaluating rangelands, ecologists were developing theories of community dynamics (how plant communities develop and change) that would provide the foundation for new methods for evaluating rangelands. SUCCESSION AND CLIMAX COMMUNITIES F. E. Clements of the University of Nebraska, Lincoln, was extremely influential in the study of succession in the Great Plains grasslands. His numerous publications on plant succession and ecology formed a major source of information for resource managers. The textbook Plant Ecology, which Clements wrote with his colleague J. E. Weaver (Weaver and Clements, 1938), became a standard in the field. Students from the ''Nebraska school of ecology'' such as E W. Albertson, E. J. Dyksterhuis, A. W. Sampson, and L. A. Stoddart became leaders in the young science of rangeland management and brought the Clementsian model of community change into the new field. The Clementsian model dominated much of the early literature in the field. Clements developed a theory of vegetation dynamics and a quantita-

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands tive method to test his theory. To Clements, the climax theory rested on the assumption that vegetation could be classified into formations that represented a group of plant species that acted together as if they were a single organism. He wrote, ''As an organism, the formation arises, grows, matures, and dies.... each climax formation is able to reproduce itself, repeating with essential fidelity the stages of its development" (Clements, 1916:3). The climax formation was "the climax community of a natural area in which the essential climatic relations are similar or identical" (Clements, 1916:126). (A climax community is the assemblage of plant species that most nearly achieves a long-term steady state of productivity, structure, and composition on a given site [Tueller, 1973].) Clements believed that all successional units within a climatic region developed along one linear path toward a plant community climax that was determined by climate (a climatic climax community). Thus, within a climatic region, a group of plant species would be identified as the climax vegetation, and all sites within that region could be compared with the climax plant species to determine where in the successional path the site was. This theory of vegetation dynamics has been referred to as the monoclimax theory. Clements' method of vegetation analysis involved the use of permanently located quadrats (a plot, usually rectangular, used for ecological and population studies). The species of vegetation in the quadrat was carefully plotted on a map. Changes in vegetation were determined by periodically replotting on a map the species that were present. The concept of successional change in rangeland ecosystems was to become the fundamental basis of the methods used today to inventory and classify rangelands. Rangelands would be classified on the basis of differences in climax plant community composition and assessed on the basis of the divergence of the current plant composition from the climax plant community composition. SUCCESSIONAL STAGES AND RANGELAND ASSESSMENT Sampson (1917) provided what was perhaps the first published reference on the utility of successional stages in rangeland assessment. Then, in 1923, Sampson wrote about the need to move from the old method of determining grazing capacity, which used palatability factors and visual estimates of forage composition, to a new method based on observation of the succession of conspicuous vegetation, that is, the replacement of one set or type of plants by another (Sampson, 1923). Sampson studied community development in the Watasch Mountains in Utah and classified four developmental stages: the climax herbaceous stage, the mixed grass and weed stage, the late weed stage, and the early

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands weed stage. Although Sampson acknowledged that the Watasch Mountain climax species were not found everywhere, he noted that the character of growth and the habitat requirements of the plants of the different stages were generally the same on native pasturelands. In describing forage production during these four stages, he noted that the climax and the mixed grass and weed stages produced the most forage in terms of quantity and quality (Sampson, 1923). Sampson noted that the use of successional units to develop a rational grazing plan presumed a detailed knowledge of the successional stages in the development of the vegetation (Sampson, 1923). To obtain this information, he recommended the use of quadrats. However, the great amount of tedious work involved in the mapping and the subsequent synthesis of the data led Sampson to recommend that the person working in the field record the percent cover of all plants of each species within each of the 100 cells that divided the chart quadrat rather than mark the specific location of each plant within each cell. This cover estimate was then multiplied by the palatability of the cover to determine forage yield. Sampson's work was instrumental in bringing successional theory and practical grazing management together. SUCCESSIONAL STAGES AS CONDITION CLASSES Sampson's ideas spawned much research into using successional stages as indicators of the status of rangelands. A number of rangeland scientists experimented with methods that could be used to determine the relationship of successional stages to rangeland condition in, for example, Colorado (Hanson et al., 1931), Kansas (Albertson, 1937), Nebraska (Weaver and Fitzpatrick, 1932), North Dakota (Hanson and Whitman, 1938; Sarvis, 1920, 1941), and the intermountain region (Sampson, 1919, 1923). In 1949, E. J. Dyksterhuis published a landmark paper that was to solidify the contribution of successional theory to the assessment of rangelands. Dyksterhuis refined the climatic climax community described by Clements (1916), proposing that different climaxes coexist as a function of soil or topographic or geographic differences within a similar climate. Dyksterhuis defined those areas that support a unique climax community as a range site. Each site—defined by its climax plant community, soil, and climatic environment—would support a characteristic assemblage of plants, and this vegetation would persist unless it was disturbed by grazing, fire, drought, or other factors. Vegetation would develop toward this climax plant community through successional processes once disturbances (wind, drought, fire) ceased. Grazing drove the plant composition toward the early stages of succession, whereas natural successional pro-

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands cesses drove plant composition toward a climax community. By adjusting the grazing pressure or the duration or season of use, rangeland managers could maintain rangelands at any stage of succession. Dyksterhuis proposed a quantitative system for assessing whether a rangeland was at an early or late stage of succession by analyzing the behaviors of three classes of plant species: decreasers, increasers, and invaders. As livestock grazing drove the plant composition toward earlier stages of succession, certain plants were thought to decrease in abundance. These decreasers were replaced by other plants that initially increased in abundance. Those increaser plants were thought to decrease in number and abundance if grazing pushed the plant composition to even earlier stages of succession. The plants that replaced the increasers were called invaders. The successional stage that the rangeland was in could then be determined by what proportion of the vegetation, measured by percent composition by weight, was decreasers, increasers, or invaders. If most of the plants were decreasers, the rangeland was thought to be in a late successional stage; if most plants were invaders, the rangeland was considered to be in a very early stage of succession. Dyksterhuis also proposed that the condition of rangelands improved as succession progressed. Later successional stages were thought to provide better forage and to be more stable and productive plant communities. The condition of a rangeland could therefore be determined by the climax plant community of the site. The greater the proportion of increasers or invaders, the poorer the condition. The greater the proportion of decreasers, the better the condition. ADOPTION OF THE SUCCESSION-RETROGRESSION MODEL BY FEDERAL AGENCIES Dyksterhuis's use of successional stages as the measure of the condition of rangelands had great appeal. His concept not only proposed a systematic way of evaluating the condition of rangelands but also explained the effects of grazing on rangeland vegetation and provided the basis for changes in grazing management. Estimations of livestock carrying capacity were linked to range sites, condition classes, and successional stages. By 1950, the measurement of range condition (Soil Conservation Service [SCS]) as the degree of departure from climax plant community (SCS) vegetation of a defined range site and the succession-retrogression model of rangeland development became the standard concept in U.S. rangeland management. All major inventory and classification methods in use today are modifications of that basic concept. The concept was adopted to varying degrees by the USFS, BLM, and

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands Land Management [1984], and U.S. Department of Agriculture, Soil Conservation Service [1989a]). The development of the current methods for evaluating the ecological state of rangelands on the basis of the departure from climax vegetation and the succession-retrogression model of rangeland change can be viewed as the first approximation of rangeland health. There were and are reasons to consider rangelands that contain climax vegetation healthy, as defined by the committee. The process of ecosystem development—that is, succession—was thought to culminate in maximum stability, productivity, diversity, and other presumed desirable qualifies (Stoddart et al., 1975). Communities in the early stages of succession were thought to be characterized by less complex energy flows and more open nutrient cycles and to be more vulnerable to invasion by exotic species (Odum, 1969). Assessments of range condition (SCS) and ecological status (USFS and BLM) have produced a wealth of useful data and research that has provided the underpinning for efforts to manage the impact of grazing on both federal and nonfederal rangelands. Range condition (SCS) and ecological status (USFS and BLM) ratings, however, are not sufficient measures of rangeland health, as defined in this report. The committee identified three problems with range condition (SCS) and ecological status (USFS and BLM) that limit the utility of these methods as measures of rangeland health: (1) use of climax plant community (SCS) or potential natural community (USFS and BLM) composition as standards, (2) the difficulty in identifying thresholds of change, and (3) the reliance on changes in plant composition and production as the sole indicator of change in the ecological state of rangelands. STANDARDS FOR ACCEPTABLE CONDITIONS The current methods of assessing range condition (SCS) or ecological status (USFS and BLM) establish a benchmark plant community against which current plant composition and production are compared. Condition and status ratings are a measure of how closely the current vegetation resembles the defined benchmark plant community. LINKING SUCCESSIONAL STAGE TO CONDITION AND STATUS RATINGS The linking of successional stages to range condition (SCS) classes has confused the interpretation of the results of range condition (SCS) and ecological status (USFS and BLM) surveys. Smith (1989) has observed that acceptance of the view that succession is ecosystem development that culminates in maximum stability, productivity, diversity, and other presumed desirable qualities "really leaves one with little choice but to manage for near climax, or admit that the goal is a second rate, degenerated ecosystem" (Smith, 1986:120). The public, the U.S. Congress, and environmental in-

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands terests are understandably concerned with reports that 6 and 16 percent of lands managed by BLM are in fair or poor condition, respectively. Concern that poor or fair condition does indicate rangeland degradation is reinforced when decreases in the amount of rangelands in poor or fair condition are reported as evidence of agency success in meeting mandates to improve rangelands (see U.S. Department of the Interior, Bureau of Land Management [1990], for example). Rangeland managers and livestock producers respond that fair and poor conditions do not necessarily indicate a problem or the need for changes in management practices on a particular site; they only indicate that rangeland's stage of succession. BLM and USFS have eliminated the terms excellent, good, fair, and poor and have adopted terminology that reflects successional stages. The problem that remains, however, is in determining whether there is a cause for concern about any one of the successional stages that a rangeland may be in. The relationship between successional stages and the stability of a site's soil and the integrity of its ecological processes—that is, its health—is uncertain. Spence (1938) noted that the soil, water, and productiveness of a rangeland are conserved when it contains its climax vegetation but that rangelands in earlier stages of succession can also conserve these values. Spence also noted that species that are not part of the climax vegetation can also conserve the soil, water, and productiveness of rangelands. Lauenroth (1985) concluded that the Similarity of a rangelands' vegetation to climax vegetation does not necessarily indicate where site degradation is occurring or where it might occur. Different combinations of plants with the same degree of similarity to an established benchmark community may differ in the effectiveness with which they protect the site from accelerated soil erosion. Risser (1989) noted that a rangeland's vegetative composition may not reflect the erosional status of the soil mantle, and Wilson (1989) reported that there is no clear relationship between changes in plant composition and soil erosion on rangelands studied in Australia. Smith (1989) also noted climax vegetation is not the only type of vegetation that furnishes adequate soil protection. Other investigators have noted that seedlings of exotic species, such as crested wheatgrass (Agropyron cristatum), that are completely dissimilar to the climax vegetation can offer adequate protection from erosion (Dormaar et al., 1978). The committee inspected a site near Reno, Nevada, that illustrates the difficulty in clearly relating the degree of similarity to an established benchmark plant composition to the degree to which the soil and ecological processes are being conserved. The existing plant composition was dominated by desert needlegrass (Stipa speciosa) and Wyoming big sagebrush (Arteraisia tridentata ssp. wyomingensis ) but included a number of

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands other species such as Indian rice grass (Oryzopsis hymenoides) and Utah juniper (Juniperus osteosperma). The percent species composition was similar enough to the climax plant community (SCS) so that the rangeland could be rated in the excellent condition class on the basis of the rangeland's plant composition alone. The range condition (SCS) of the site was reduced to good, however, because of serious wind erosion. Wind erosion on the site was severe enough that many plants were being damaged by abrasion, and some were being buried by drifting soil. It appeared that the amount of plant cover on the site had declined, exposing the soil to wind erosion. The decline in plant cover, however, appeared to have been evenly distributed among all species present on the site, so that the percent composition remained the same as that which would be expected in the climax plant community (SCS). Similarity to a benchmark plant composition was not, in this case, a sensitive indicator of other changes that were occurring on the site. The similarity of current plant composition and biomass production to that of a climax plant community (SCS) or potential natural community (USFS and BLM) should not be used as the primary standard of rangeland health. The degree of similarity to the plant composition and annual biomass production of a climax plant community (SCS) or potential natural community (USFS and BLM) alone is not a sufficient measure of rangeland health. The evaluation of rangeland health will require additional and different criteria and indicators. DIFFICULTIES IN COMPARING DIFFERENT SITES The climax plant community (SCS) and the potential natural community (USFS and BLM) is different for each site. Similarly, the plant composition and biomass production of successional stages also may differ between sites. Since the benchmark against which current plant composition is compared is different for different sites, comparing the results of condition or status ratings between sites is difficult. It is possible, for example, that an early successional stage on one site may closely resemble a later successional stage on a different site. Plant composition and production may be very similar between the two plant communities on the two sites, but they may receive different condition or status ratings since the benchmarks against which they are being compared differ. Laycock (1989) cited an example of such a problem in Idaho: The Artemisia tridentata ssp. wyorningensis/Poa secunda [Wyoming big sagebrush/sandberg bluegrass] habitat type occurs in areas in western Idaho where precipitation is less than 18 cm annually (Hironaka et al., 1983). The dominant plant is the Wyoming big sagebrush with an understory of sandberg bluegrass and scattered other species. This [Wyoming

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands big sagebrush/sandberg bluegrass] vegetation mix is identical to a rather severely degraded ("poor condition") stage of the Artemisia tridentata wyomingensis/Stipa thurberiana [Wyoming big sagebrush/Thurbers needlegrass habitat] type that occurs over a wide area of southern Idaho on better sites and higher precipitation where grazing has removed the [needlegrass] and other desirable species. It may also resemble other sagebrush habitat types in lower stages of condition. (Laycock, 1989:5) The sites described by Laycock had essentially the same plant compositions and production, but they would have different ratings since the benchmarks against which they were compared differed. This difference may be important because the relationship between a similarity rating and the conservation of soft and ecological function is not well understood. One site that supports 40 percent of its expected climax plant community (SCS) or potential natural community (USFS and BLM) composition may still have enough soil cover to be protected from water and wind erosion, whereas another site, also supporting 40 percent of its expected climax plant community (SCS) or potential natural community (USFS and BLM) composition, might be eroding at rates that will lead to serious site degradation, it might have a compacted soil that restricts infiltration, or it might have nutrient cycles that are interrupted by a lack of litter cover and lack of incorporation of organic matter into the soil surface. The use of similarity to a defined benchmark plant community, whether defined as climax plant community (SCS) vegetation or potential natural community (USFS and BLM), also imposes the difficulty of direct comparisons between sites with different benchmark plant communities. The lack of a single, dearly defined standard that does not differ from site to site is a fundamental limitation to the use of current methods for assessing rangeland health. This problem has and continues to confuse the public, the U.S. Congress, livestock producers, and rangeland scientists themselves. LIMITATIONS OF SUCCESSION-RETROGRESSION MODEL The successional model that supports the current range condition (SCS) and ecological status (USFS and BLM) methods postulates that succession toward the climax plant community (SCS) or potential natural community (USFS and BLM) is a process that moves through recognizable and predictable stages. The effects of grazing, drought, and other disturbances are also thought to produce recognizable and predictable changes in composition toward early stages in succession. The vegetation that is found on any particular rangeland is the product of the equilibrium between the two forces of succession toward the climax plant community and retrogression toward earlier successional stages.

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands This rangeland succession model assumes that range condition (SCS)—that is, the successional stage of a particular rangeland—will change back and forth along the successional gradient characteristic of that site in response to changes in management. The main tool of rangeland managers is to adjust the stocking rate, species of grazing animal, the duration of grazing, and the season that the rangeland is used to achieve an equilibrium between the opposing forces of succession and retrogression. Once this equilibrium is achieved, it will tend to remain stable. The manipulation of livestock grazing allows rangeland managers to maintain a particular plant composition at some point along the successional gradient characteristic of that site. The choice of that point along that gradient that should be maintained depends on whether the goal of management is to maximize or optimize the production of livestock, wildlife, or some other product or value. DOES SUCCESSION OCCUR ON ALL RANGELANDS? Current ecological research has questioned whether the concept of well-defined, predictable, and reversible changes along a successional gradient holds for all or the majority of rangelands. Ecologists working in rangeland ecosystems have developed theories that allow for multiple equilibria and for transitions between alternative vegetational states that are not easily reversible (Friedel, 1991; Westoby et al., 1989). These theories, which were discussed in more detail in Chapter 2, also allow for the existence of transitional states that represent the process of change from one state of equilibrium to another. The specific outcome of the change will depend on events that occur while the rangeland is in that transitional state rather than on a predictable succession from one state to another. Investigators have attempted to describe the mechanisms that produce such complex dynamics on rangelands. In some cases, the random occurrence of fire, drought, or changes in grazing systems have produced changes in rangelands that do not appear to follow a readily discernible successional sequence (Friedel, 1991; Laycock, 1989; Westoby et al., 1989). Sharp et al. (1990), for example, reported 40 years of data from a salt desert shrub rangeland in south-central Idaho that illustrate this phenomenon. Three different plant communities have occurred on this site, which has not been grazed since 1945. During periods of normal precipitation and no buildup of a naturally occurring scale insect (Orthisae sp.), the site is dominated by shadscale (Atriplex confertifolia) in association with bottlebrush squirreltail (Sitanion hystrix), sandberg bluegrass (Poa secunda), globe mallow (Sphaeralcea coccinea), and other plants. Following a cyclic outbreak of the scale insect, most of the shadscale dies or is reduced in size and vigor. If precipitation is normal in the years following the out-

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands break, the site becomes dominated by bottlebrush squirreltail, production is high, and shadscale may eventually recover. If precipitation is below average following the insect outbreak, several species codominate the site. The species that becomes more important depends on the timing of precipitation events. The transition from one complex to another appears to be due more to the vegetation's adaptation to episodic events than to a linear successional development. It could be argued, however, that the desert sites described simply have not had time to recover from historic overgrazing and to reestablish the potential natural community (USFS and BLM). In any case, the kinds of vegetation dynamics described above are difficult to incorporate into existing succession-retrogression models of rangeland development. Risser (1989) summarized the questions about the succession-retrogression model raised by ecologists as including some of the following ideas. Biological communities may go through different pathways yet reach a similar climax or terminal state. Depending on disturbances during succession, the system may proceed through a new set of seral stages. The initial conditions at the beginning of a successional sequence can cause quite different outcomes even on apparently equivalent sites. The outcome of successional sequences is determined by the characteristics of the interacting plant and animal populations and the present and preceding environment, not by predetermined organismic controls. Certain ecosystem-level characteristics, such as the ability to absorb or release nutrients, are characteristic functions of systems in early and late stages of succession. The terminal stage may not be the most productive, stable, or diverse community. Risser concluded that it is important to recognize that a simple linear successional sequence is not always an adequate representation of the conditions observed in the field (Connell and Slatyer, 1977; Drury and Nisbet, 1973; Gutierrez and Fey, 1975; McIntosh, 1980; Odum, 1969 [as cited by Pisser, 1989]). Traditional successional theory implies that a site that has retrogressed can recover if the process is reversed. This is not possible or is very slow, however, if severe soil erosion, invasion of a new and very dominant species, or change from a fire-dependent to a fire-safe plant community has resulted in near-permanent changes in the abiotic or biotic community. The succession-retrogression model of how rangelands develop and change, which is the foundation of current rangeland classification and assessment methods, should be modified or new models developed to assist in assessing whether rangelands are approaching thresholds of change.

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands Current range condition (SCS) and ecological status (USFS and BLM) ratings are founded on the concept of well-defined, predictable, and reversible changes along a successional gradient that holds for all or the majority of rangelands. An evaluation of rangeland health requires consideration of additional processes of ecosystem change and the reversibility of those changes. Several alternative models of rangeland change have been proposed, but no single model has received widespread acceptance. An accelerated effort by ecologists is needed to develop and test models of rangeland change that will assist in identifying rangelands that are approaching thresholds of change. MULTIPLE INDICATORS ARE NEEDED Current range condition (SCS) or ecological status (USFS and BLM) ratings rely nearly exclusively on measurements of plant composition and annual biomass production. Problems such as soil erosion, disruption of nutrient cycling, or other ecological attributes of rangelands are not primary considerations in assessing range condition (SCS) or ecological status (USFS and BLM). Changes in other important attributes of an ecosystem may not be detected by measuring the plant composition and production alone. The difficulty of assessing ecosystems using only one index has long been recognized. Ellison (1949) wrote that the soft, plants, animals, topography, and climate develop together and are knit together into an integrated whole. He also noted that vegetative and soft trends do not always parallel each other and that they may be widely divergent on eroding rangelands. He suggested that "if the rangeland manager is impressed by evidence of change in vegetation and not by evidence of soil erosion, he may be led astray" (Ellison, 1949:794). The effectiveness of vegetation in protecting soil is more a function of effective soil cover than plant composition, since effective soft cover is more closely tied to the type and pattern of cover than it is to plant composition. Slight changes in plant populations and litter may produce accelerated erosion on steep, erodible soils, whereas complete changes in species and life-forms, such as in artificially established annual grasslands, may result in a vegetation type that still provides good soil protection. Erosion hazard on some soils may be quite insensitive to changes in vegetation type if the site is very flat or very rocky, whereas on steep day slopes the kind or amount of vegetation may be critical (Smith, 1989). Loss of minor species may not be indicated by a change in range condition (SCS) or ecological status (USFS and BLM) rating if these species make up a small percentage of the plant composition and annual biomass production of the climax plant community (SCS) or potential

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands natural community (USFS and BLM). The loss of minor species, however, may indicate change in nutrient cycles caused by reduced diversity in rooting depth or changes in energy flow because of a reduced period during which the remaining plants photosynthesize. Tueller (1973) described a process of site degradation that began with the loss of plant vigor and seed production and that led to the death of individual plants and a reduction in litter cover and plant density. These changes caused changes in plant cover, distribution, and potential for reproduction. Total biomass production or the annual production of individual species was reduced. Further deterioration led to reduced litter accumulation, the formation of soil crusts that retarded germination, and altered plant growth forms. Reduction in soil cover and litter led to soil erosion and the disruption of nutrient cycles. Eventually, the site potential was seriously impaired. The process of site degradation described by Tueller (1973) is driven by a complex of interacting factors, with no single factor predominating. Changes in species composition, plant density or frequency, distribution and cover of litter, soil erosion, total biomass production, plant vigor, and seedling recruitment, among other factors, are all apparent at different points in the degradation process. No single factor alone can completely describe such a process. Tueller for instance, listed 16 separate factors that can serve as useful indicators of site degradation (Tueller, 1973). The problem is not that plant composition and biomass production are unimportant attributes of rangeland ecosystems; rather, the problem is that they are typically the only attributes measured. A system that used only plant density, erosion, litter cover, seedling density, or compaction as a single measure of range condition (SCS) or ecological status (USFS and BLM) would have the same problem. The evaluation of rangeland health will require analysis of attributes in addition to plant composition. A comprehensive evaluation of rangelands should be based on the preponderance of evidence derived from sampling multiple attributes related to ecological function and soil stability. Range condition (SCS) or ecological status (USFS and BLM) ratings based primarily on changes in plant composition and annual biomass production alone are not sufficient measures of rangeland health. Plant composition and biomass production will usually change if soil erosion accelerates, the soil continues to compact, or seedlings fail to become established. However, there may be significant lags between the onset of rangeland degradation and a change in any single indicator of rangeland health. Any single measure will have different sensitivities to different stresses. The measurement of multiple attributes increases the probability that rangeland degradation will be detected early enough for corrective measures to be taken.

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands Site Potential and Resource Values In current theory and practice, the rangeland site classification and the definition of site potential are nearly synonymous. The kinds and amounts of vegetation produced in the climax plant community (SCS) or potential natural community (USFS and BLM) are considered the best measure of the potential productivity of a site. Overgrazing, accelerated erosion, or other influences that result in loss of the capacity to produce the plant composition and annual biomass production characteristic of the climax plant community (SCS) or potential natural community (USFS and BLM) are thought to have caused a loss in site potential. DESIRED PLANT COMPOSITION Many observers have argued that the plant composition and production desired on a site should depend on how the site is to be used. In its National Range Handbook (U.S. Department of Agriculture, Soil Conservation Service, 1976), SCS states that although the climax plant community (SCS) describes the site potential of a particular site, the goal of management is not necessarily always to achieve climax plant community (SCS) composition and production. Other seral (successional) stages may be better for particular uses. Pisser (1989) stated that the climax vegetation for a given site may not by the most productive or desirable type of vegetation for livestock forage production and that climax vegetation may not be the most appropriate goal of rangeland management if the management objectives include multiple uses or values. Such multiple uses or values may include wildlife, water quality or quantity, recreation, and livestock grazing. RESOURCE VALUE RATING Wilson (1989) suggested the need to first define the land use objectives for rangelands. This should be followed by a description of the vegetative structure that will maximize those objectives. The Society for Range Management's Range Inventory Standardization Committee similarly recommended that a system of resource value ratings be used as a measure of the value of vegetation or other features for a particular use (Society for Range Management, Range Inventory Standardization Committee, 1983). The resource value rating would be based on the particular species present, growth forms and foliage types, or other criteria. Each use may have a separate resource value rating. The resource value rating was to be a measure of the suitability or usefulness of the vegetation of an ecological site (BLM) for a specific use.

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands Most recently, the Society for Range Management's Task Group on Unity in Concepts and Terminology recommended that management objectives should be defined in terms of a desired plant community for each ecological site (BLM). The desired plant community should be defined as follows: ''of the several plant communities that may occupy a site, the one that has been identified through a management plan to best meet the plan's objectives for the site'' (Society for Range Management, Task Group on Unity in Concepts and Terminology, 1991:10). USFS has proposed adoption of a system of resource value ratings. Resource value ratings require that goals be set for the vegetation needed to meet a particular use. Those goals will be a desired plant community. Because goals will vary from location to location, different desired plant communities may be described for different locations on the same site. Management would be designed so that the plant composition could be changed to reflect that described for the desired plant community. This may or may not represent a change toward the climax plant community (SCS) or potential natural community (USFS and BLM). DISTINCTION BETWEEN RANGELAND HEALTH, SITE POTENTIAL, AND RESOURCE VALUES Rangeland health assessments should be separate and independent of assessments for determining the proper use of particular rangelands . It is essential that there be a clear understanding of the distinction between rangeland health, as defined in this report, and site potential or resource value ratings. Rangeland health, as used in this report, is intended as a minimum ecological standard, independent of the rangeland's use and how it is managed. The particular mix of commodities and values produced by a rangeland depends on how it is used and managed. If rangeland health is conserved, then the capacity of the site to produce different mixes of commodities and values is conserved. The determination of which uses and management practices are appropriate may require the evaluation of data different from those used to evaluate rangeland health. No single index will meet all the needs of rangeland inventory, classification, and management. Rangeland health is a measure of the integrity of the soil and the ecological processes of a rangeland. Loss of rangeland health causes a loss of capacity to produce resources and satisfy values. Rangeland health is not, however, an estimate of the kinds or amounts of resources that a rangeland produces, nor is it an evaluation of the different uses of a site. Two rangelands may have the potential to produce different commodities and values, but both can be equally healthy if the integrity of the soil and ecological function are conserved.

OCR for page 51
Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands The protection of rangeland health should serve as the minimum standard for management. If rangeland health is sustained then decisions about the appropriate plant community composition and production can be made depending on the desired rangeland use. Most important, the conservation of rangeland health preserves the option to change the use and management of a site as the desired resources and values change. Apparent Trend A one-time measure of most rangeland characteristics is only that—a picture of the situation at the time of measurement. Without a previous measurement with which the current measurement can be compared, the range manager's ability to interpret whether the management program is succeeding or failing is limited. Personnel and budget constraints and inconsistencies in the indicators measured at different times, however, have limited the number of sites where trend can be determined from comparable data collected at different points in time. To compensate for this problem, range managers have attempted to determine apparent trend; that is, they evaluate site characteristics that indicate whether an area is improving or deteriorating. Factors such as accelerated erosion, for example, have been used to indicate a downward trend. If the soil is eroding at an accelerated rate, then the productive capacity of the site is probably being lost. The accumulation of litter is viewed as a sign of an improving rangeland because it is a sign that the amount of plant material needed to protect a site from erosion is increasing. The presence of seedlings of desirable plants was also interpreted as an upward apparent trend because those seedlings indicated that the plant composition was evolving toward the climax plant community (SCS) or potential natural community (USFS and BLM). The presence of seedlings of undesirable plants represents a downward trend. Plant vigor has also been used to judge apparent trend. Vigorous dominant plants in a climax plant community (SCS) or potential natural community (USFS and BLM) indicate an upward apparent trend, whereas the presence of weak, deformed plants indicates a downward trend.