Inventorying and Monitoring Rangeland Health
Rangeland inventories have been initiated for a variety of reasons: scientific curiosity, economic necessity, and legal mandates. Ownership patterns in the western United States have influenced the demand for and the type of information needed from rangelands. Increased awareness of environmental issues has also influenced the type of information demanded for range management. These different incentives have resulted in a variety of inventory methods at a variety of spatial scales. Site-specific inventories, subject-specific inventories, and national inventories currently exist.
PAST INVENTORIES OF RANGELANDS
The first recorded surveys of U.S. rangelands are fragmentary comments from the letters and journals of explorers, trappers, members of military expeditions, and missionaries. They are based on visual observations of what were, to them, vast and foreign landscapes. Early military and exploratory expeditions and the medical and scientific plant collectors who accompanied them provided much information about western U.S. rangelands (Chapline and Campbell, 1944). Although few of these people described the vegetation or the grazing resource sufficiently for a credible analysis of the soft and plant conditions prior to settlement by Europeans, their records served to entice additional surveys.
The First Surveys
Botanical surveys were formalized in the 1800s. Created in 1869, the Division of Botany of the U.S. Department of Agriculture (USDA) initiated a number of surveys of plants and catalogued the flora and poisonous plants of the western United States (Chapline and Campbell, 1944; Wass-
er, 1977). During the same period, land surveyors were working on western lands, and some of their notes are fairly detailed in their descriptions of vegetation. The U.S. Bureau of the Census kept track of records of livestock populations across the United States.
During the 1800s, landholdings in the western United States were a changing mix of public domain and private land. Livestock owners developed operations based on the intermixes of these lands and the seasonal availability of forage. Western rangelands were being changed from vast and foreign landscapes to important commercial resources. Western livestock operations flourished during the period from 1867 to 1887 as railroads connected these grazing lands with the expanding cities of the Midwest and the East (Rowley, 1985).
In the late 1800s, the dramatic expansion of the livestock industry combined with equally dramatic droughts, and severe winters led to widespread degradation of rangelands. In 1895, letters from individuals about the declining range conditions in Texas were sent to USDA, and plans were made for an experiment station in Abilene, Texas. The Division of Agrostology within USDA was created in 1895 to investigate grass, forage, and range management (Wasser, 1977). Although it lasted for only 6 years, this division published a number of reports on the forage conditions and grazing problems of the western United States (Chapline and Campbell, 1944). These surveys were localized descriptions of rangeland vegetation and management.
FOREST RESERVE SURVEYS
In 1891, the U.S. Congress authorized the President to set aside national forests from the unreserved public domain.1 This act was the first step in closing America's vast open land frontier. This act also marked a change in the kind of information the government needed about these lands—it shifted from landholding to land management information.
The management direction of these forest reserves was not officially set until 1897, when Congress decreed that national forests were established to ''improve and protect the forest ... or for the purpose of securing favorable conditions of water flows, and to furnish a continuous supply of timber.''2 In 1901, USDA developed principles to regulate grazing within U.S. forest reserves. Grazing was allowed ''where it is shown after careful examination, that grazing is in no way injurious to or preventive of the conservation of the water supply" (Rowley, 1985:40).
In 1905, a letter from the secretary of USDA to the director of the newly created U.S. Forest Service (USFS) indicated that the rangelands were to be appraised as to their carrying capacity—that is, the maximum limit of livestock numbers that could be grazed on that land—to divide
the land and set limits for grazing permits (Rowley, 1985). This need for information to determine grazing capacity led Jardine and others to develop methods of surveying forest reserves (Chapline and Campbell, 1944). These methods were, by necessity, limited because surveys of extensive areas were required. Although these determinations of carrying capacity were a regular responsibility of forest managers, only the annual numbers of permits and numbers of livestock that grazed the forest reserves were published in the annual reports of the USFS (Rowley, 1985).
STATE EXPERIMENT STATION SURVEYS
Around the early 1900s, states were establishing state experiment stations in response to the research and management needs of the agricultural industries within their borders. This activity also resulted in localized inventories of rangelands, such as the Washington Agricultural Experiment Station bulletin by John Cotton on the range conditions of central Washington State (Chapline and Campbell, 1944).
BLM GRAZING DISTRICT SURVEYS
In 1934, the U.S. Congress passed the Taylor Grazing Act,3 thereby putting an end to the status of unreserved federal lands as unregulated common lands. The Taylor Grazing Act was intended to "stop injury to the public grazing lands by preventing overgrazing and soil deterioration, to provide for their orderly use, improvement, and development, to stabilize the livestock industry dependent upon the public range, and for other purposes."4 Virtually all remaining land in the public domain—about 69 million hectares (170 million acres)—outside Alaska was withdrawn into grazing districts under the jurisdiction of the Bureau of Land Management (BLM) (Peffer, 1951; Voss, 1960). Although the act's preamble made it clear that rangeland improvement was a primary congressional goal, the U.S. Congress provided no precise guidelines in the act for achieving that aim. The Taylor Grazing Act addressed range inventories only indirectly by referring to grazing capacities.
SURVEYS OF NONFEDERAL RANGELANDS
In 1935, the U.S. Congress, responding to the devastation of the Dust Bowl of the early 1930s, established the Soil Conservation Service (SCS) as part of USDA to carry out programs for the control and prevention of soil erosion on public and private lands.5 In enacting this legislation, which was known as SCS's "Organic Act," Congress officially recognized the need to gather soil, vegetation, and other resource data on nonfederal
rangelands and specified the purposes for which those data were to be used: to end soil erosion and to preserve natural resources.6 Toward this end, SCS field technicians developed a system of range condition classification that field staff and ranchers could understand and use in the development of ranch conservation plans to minimize soil erosion (Helms, 1990). Although these determinations of range condition were a regular responsibility of SCS field technicians, no data from these individual conservation plans were summarized in a report on the conditions of nonfederal rangelands.
CONGRESS MANDATES SURVEY OF WESTERN RANGELANDS
In 1936, Resolution 289 (U.S. Congress, Senate, 1936) requested from USDA a report summarizing information on western rangelands. The Congress justified its request in the resolution by stating that large parts of the western range had been subject to unrestricted use since settlement and were commonly believed to be seriously depleted, that the range resource constituted one of the major sources of wealth to the nation and, that USDA had accumulated a large amount of information on the condition of the range resource, the factors causing degradation, and the social and economic importance of rangelands.
The request for this report points to the lack of a national-level inventory or a mechanism to synthesize data collected by range managers to regulate grazing or evaluate management practices. Although the base data were unpublished, summary tables were presented within the report. Forage depletion by vegetation type was estimated in terms of decline from the original forage values (pristine condition). Results were summarized by the following depletion classes: 0-25, 26-50, 51-75, and 76-100 percent. Remnants of pristine range and protected areas—such as cemeteries and railroad rights of way—were used for comparison. Thus, the report represented the professional opinion of range scientists about the degree of depletion of both federal and nonfederal rangelands.
More Recent Surveys
From 1936 to 1966, range conditions were not surveyed at the national level (Box, 1990). In the early 1960s, the Public Land Law Review Commission was established to study the management of and whether the public lands should be kept in public ownership. As part of their deliberations, they commissioned Pacific Consultants, a private consulting firm, to do a nationwide range survey. Pacific Consultants, however, did not conduct original on-site field surveys. Instead, they collected information from the files of the federal agencies and other sources and made an assessment of U.S. rangelands. In 1972, the USFS made an assessment of the condition of the federal rangelands. It appears the assessment was done by using data already collected by federal agencies and perhaps used previously in the survey by Pacific Consultants (Box, 1990).
Environmental Legislation Increases Information Needs
The wave of federal environmental legislation that commenced with the Wilderness Act of 19647 included several laws that contained significant inventory mandates for SCS, USFS, and BLM. Although some of these laws required more detail from agencies with respect to their on-the-ground management, other laws expanded the role of federal agencies in inventories at the national level. During this period, the U.S. Congress also enacted several laws with peripheral, but nonetheless important, relevance to rangeland inventories.
NATIONAL ENVIRONMENTAL POLICY ACT
The first of these laws, the National Environmental Policy Act of 1969,8 requires all federal agencies to write environmental impact statements on all proposals for major federal actions that significantly affect
the human environment. In 1974, a court ordered BLM to prepare such statements for all of its local grazing programs according to an agreed-upon schedule.9 That ruling was premised on and impelled by the evidence before the court of extensive resource damage on BLM-managed rangelands caused by livestock grazing. The National Environmental Policy Act brought the need for comprehensive inventorying and monitoring of rangelands into stark relief because it is difficult, if not impossible, to assess the environmental consequences of grazing without conducting baseline inventories and incorporating the resultant information into environmental impact statements. Although by 1988 BLM did complete the 144 site-specific grazing environmental impact statements required by law, it did not do all of the necessary site-specific inventories, in part because of budgetary problems. Data on 16 percent of BLM-man-aged lands were more than a decade old (U.S. General Accounting Office, 1988a). In 1985, an examination of 116 environmental impact statements indicated that BLM had used different terminologies and methodologies to assess rangelands (Wald and Alberswerth, 1985). Such methodological differences make it difficult to compare the results of rangeland assessments across the United States.
CLEAN WATER ACT
The Clean Water Act of 1972,10 as amended in 1987, requires states to develop and employ the best management practices to control nonpoint sources of water pollution, which include range livestock grazing, silvicultural activities (silviculture is a branch of forestry that deals with the development and care of forests), and other agricultural practices (Thompson, 1989; Whitman, 1989). The Clean Water Act requires each state to adopt water quality standards for its waters. Because federal agencies must comply with those standards,11 they need to obtain water quality information. The act's requirements have had some impact on timber harvesting on public lands12 and may have important implications for the kind of information needed from surveys of federal and nonfederal rangelands.
ENDANGERED SPECIES ACT
The Endangered Species Act of 197313 requires federal agencies to protect the wildlife species listed in the Act and their habitats in several distinct ways (Coggins and Russell, 1982). Compliance with the Act's requirements and implementing regulations14 may well necessitate special resource inventories. The Act has also had an effect on timber harvests,15 but as yet it has not been a major influence on rangeland inventories or monitoring.
Legislation Spurs Development of National Inventories
The U.S. Congress passed a series of laws in the 1970s that redefined the responsibilities of federal agencies to conduct and analyze data on natural resources on both federal and nonfederal lands. The Resources Planning Act, the Forest and Rangeland Renewable Resources Planning Act, and the Soil and Water Resources Conservation Act mandated efforts by USFS, BLM, and SCS to inventory and monitor rangelands.
RESOURCES PLANNING ACT
In 1974 Congress passed the Resources Planning Act of 1974,16 in which it directed the USFS to conduct and update every 10th year a detailed renewable resource assessment. Each assessment must include the following, among other things: "an inventory, based on information developed by the Forest Service and other Federal agencies, of present and potential renewable resources, and an evaluation of opportunities for improving their yield of tangible goods and services, together with estimates of investment costs and direct and indirect returns to the Federal Government."17 The assessments and inventories on which they are based are intended to provide information and analysis for use in the development of national-level resource policies and programs as well as site-specific management plans.18
The Resources Planning Act specifically directs the USFS to analyze the present and anticipated uses of, demand for, and supply of the nation's renewable resources. Timber, wildlife, range, water, minerals, and lands have been analyzed in the three assessments completed since the Act was passed. The need to collect more data to support the Resources Planning Act has been recognized (McClure et al., 1979) and has led the USFS to expand the timber survey, now called the forest inventory (U.S. Department of Agriculture, U.S. Forest Service, 1985). However, this inventory surveys only pinyon-juniper rangelands (a habitat type dominated by one of several species of pinyon pine [Pinus spp.] or juniper [Junipe-rus spp.]) and some shrublands. The 1989 assessment analyzed the federal and nonfederal rangelands from a national perspective, reporting data on the condition of rangelands gathered by USFS, BLM, and SCS (Joyce, 1989).
FOREST AND RANGELAND RENEWABLE RESOURCES PLANNING ACT
Congress passed the Forest and Rangeland Renewable Resources Research Act19 to complement the policies and direction set forth in the Resources Planning Act. The Act includes a provision intended to "en-
sure the availability of adequate data and scientific information for development" of the assessment.20 It requires the USFS to keep current a comprehensive survey and analysis of the present and prospective conditions of and requirements for renewable resources of the forests and rangelands of the United States; the supplies of such renewable resources, including a determination of the present and potential productivity of the land; and any other facts that may be necessary and useful in the determination of ways and means needed to balance the demand for and supply of these renewable resources, benefits, and uses in meeting the needs of the people of the United States. 21
SOIL AND WATER RESOURCES CONSERVATION ACT
The legislative mandate to SCS for national-level inventories is similar to that of the USFS. In 1977, Congress, recognizing a lack of data necessary to facilitate "informed long-range policy decisions with respect to the conservation and improvement of our country's soil and water resources" enacted the Soil and Water Resources Conservation Act.22 The Act made the SCS responsible for carrying out a "continuing appraisal of the soil, water, and related resources" of the United States.23 The phrase ''soil, water, and related resources" is defined broadly as ''those resources which come within the scope of the programs administered and participated in by the Secretary of Agriculture through the Soil Conservation Service."24
In the Soil and Water Resources Conservation Act, Congress declares that "[r]esource appraisal is basic to effective soil and water conservation" and that a coordinated appraisal program is "essential" since decisions affecting soil and water resources are made by a variety of individuals and administrative agencies and may affect other decisions and programs. 25 The appraisal "shall" include
data on the quality and quantity of soil, water, and related resources, including fish and wildlife habitat;
the capability and limitations of those resources for meeting current and projected demands on the resource base; [and]
the changes that have occurred in the status and condition of those resources resulting from various past uses, including the impact of farming technologies, techniques, and practices.26
Regarding the sources of information to be used in the appraisal, the Act states that the SCS shall use not only data collected under the Soil and Water Resources Conservation Act but also "pertinent data and information collected by the Department of Agriculture and other Federal, State, and local agencies and organizations."27
NATIONAL RESOURCES INVENTORY
The appraisal mandated by the Soil and Water Resources Conservation Act, now termed the National Resources Inventory, must be carried out at periodic intervals. The next of these is due by December 31, 1995.28 The secretary of USDA may require that interim appraisals be made. A description of the current National Resources Inventory is given later in this chapter.
There have been two appraisals since passage of the Soil and Water Resources Conservation Act in 1977. Range condition measures have been reported for nonfederal U.S. rangelands as part of both appraisals. The most recent assessment estimated the areas of nonfederal rangelands that are in excellent, good, fair, and poor range condition on the basis of inventory data from 1982 (U.S. Department of Agriculture, Soil Conservation Service, 1989a).
ENVIRONMENTAL MONITORING AND ASSESSMENT PROGRAM
In 1988, the U.S. Environmental Protection Agency (EPA) developed the Environmental Monitoring and Assessment Program (EMAP) in response to a recommendation from the EPA Science Advisory Board that such a program be implemented to monitor ecological status and trends that would identify emerging environmental problems before they reach crisis proportions (U.S. Environmental Protection Agency, 1992). EMAP will coordinate the research and the monitoring and assessment efforts needed to both document the current condition of ecological resources and predict the effect of different management alternatives on those resources.
The EMAP program has three broad objectives: (1) estimate the current status, extent changes, and trends in indicators of the condition of U.S. ecological resources; (2) monitor indicators of pollutant exposure and habitat condition and seek associations between human-induced stresses and ecological condition; and (3) provide periodic statistical summaries and interpretive reports on ecological status and trends to resource managers and the public (U.S. Environmental Protection Agency, 1992). EMAP will coordinate several long-term monitoring efforts to collect data at regional scales from eight resource categories: arid lands, agricultural systems, forests, lakes and streams, the Great Lakes, inland and coastal wetlands, estuaries, and coastal waters. The monitoring program includes collection and interpretation of field data as well as statistical analysis and sampling design and development of ecological indicators. The arid lands program, which includes rangelands, is currently undertaking research to develop ecological indicators and rangeland classification systems and to explore the use of remote sensing to monitor changes in arid lands.
Inventory Mandates for Land Management and Planning
Both the USFS and BLM have been given inventory responsibilities in connection with their land use planning and management duties. Three major pieces of legislation, two enacted in 1976 and one enacted in 1978, provide the framework for management, inventory, and planning for both agencies.
FEDERAL LAND POLICY AND MANAGEMENT ACT
The Federal Land Policy and Management Act of 197629 is called BLM's "Organic Act," but sections 1751 through 1753 deal with range management by both BLM and USFS. In essence, these sections declare that "a substantial amount of the Federal rangelands is deteriorating in quality,"30 call for improvement in current conditions,31 provide funds for "range improvements,"32 authorize adjustments to grazing privileges depending on the condition of the rangeland,33 and sanction other management actions.
Concerning inventories of rangelands, Congress stated: "The [BLM] shall prepare and maintain on a continuing basis an inventory of all public lands and their resource and other values..., giving priority to areas of critical environmental concern."34 That inventory is to form the basis of land use plans that are required to produce multiple-use, sustained-yield management of the surface resources of BLM-managed public lands.35 Livestock management of these lands is but a subset of multiple-use management,36 which the Federal Land Policy and Management Act defines to mean management to produce a harmonious and optimum combination of outdoor recreation, range, wildlife, watershed, and timber production. The Act specifies that economic optimization is not the goal.37
SOIL-VEGETATION INVENTORY METHOD
In 1977, BLM adopted the site inventory method, which was later expanded into the soil-vegetation inventory method, to standardize rangeland sampling for environmental impact statement and allotment management plan reporting. Within this inventory, soil maps, site descriptions, and aerial photographs were used to classify rangelands for survey purposes (Wagner, 1989). Ground checks determined range site, woodland type, or forest type, as well as soil type. Sampling units called site write-up areas were defined on the basis of the vegetation communities that were present on the site. Data on basal and canopy ground covers and vegetative production were collected on site write-up areas and combined with data from additional studies on phenology (a branch
of science that deals with the relations between climate and periodic biological phenomena), climate, actual use, and utilization (Wagner, 1989).
The soil-vegetation inventory method was intended to be the official inventory method for basic inventories of soil and vegetation by BLM, but it was not to preclude the use of site-specific studies for special purposes. The soil-vegetation inventory method, however, was discontinued in the early 1980s and never became the official standardized inventory system for federal rangelands managed by BLM.
NATIONAL FOREST MANAGEMENT ACT
Like the BLM, the USFS is also required to manage livestock within the context of multiple uses and sustained yields on the lands within its jurisdiction.38 The National Forest Management Act of 197639 directs USFS to "develop and maintain on a continuing basis a comprehensive and appropriately detailed inventory of all National Forest System lands and renewable resources."40 This inventory, like the BLM inventory, is to be kept current "so as to reflect changes in conditions and identify new and emerging resources and values."41 As is the case with BLM, the inventory information is to be used in the development of land use plans.
Within the national forest system, inventories exist for range; timber;
soils and geology; natural water occurrences, including quality and quantity and wetlands and floodplains; existing plant life, including threatened and endangered species; existing fish and wildlife, including threatened and endangered species; habitat conditions for selected vertebrate or invertebrate species; and quantitative data for determining species and community diversity (U.S. Department of Agriculture, U.S. Forest Service, 1985). The earliest rangeland inventory was done on the Coconino National Forest in 1911 in response to a question regarding the livestock carrying capacity of the rangeland (U.S. Department of Agriculture, U.S. Forest Service, 1985). Methodologies for inventorying rangelands have varied since then, but they have primarily focused on the vegetation within allotments. Information needs have included determination of the suitability of lands for livestock grazing, the kinds of plant communities present, the ecological status of those communities, livestock forage conditions, the status of soil cover and soil stability, the trends in ecological status and soil stability, the capability of the lands to produce suitable food and cover for wildlife, and the effects of grazing. Current inventory procedures are described later in this chapter. Development of specific stands and guidelines for range inventory are the responsibility of the regional forester; thus, variations between regions are permitted.
PUBLIC RANGELANDS IMPROVEMENT ACT
The Public Rangelands Improvement Act of 1978,42 which is also applicable to both BLM and USFS, directly addresses the issue of range condition measurement.43 The act's policies include a federal commitment to ''inventory and identify current public rangeland conditions and trends as part of the inventory process required by [the Federal Land Policy and Management Act]," and to "manage, maintain, and improve the condition of the public rangelands so that they become as productive as feasible for all rangeland values in accordance with management objectives and the land use planning process."44 Congress thereafter defined its terms as follows45
The term "range condition" means the quality of the land reflected in its ability in specific vegetative areas to support various levels of productivity in accordance with range management objectives and the land use planning process, and relates to soil quality, forage values (whether seasonal or year round), wildlife habitat, watershed and plant communities, the present state of vegetation for that site, and the relative degree to which the kinds, proportions, and amounts of vegetation in a plant community resemble that of the desired community for that site.
The term "native vegetation" means those plant species, communities, or vegetative associations which are endemic to a given area and which
would normally be identified with a healthy and productive range condition occurring as a result of the natural vegetative process of the area.46
Carrying out one of its policies, Congress said that both agencies shall update, develop, and maintain on a continuing basis... an inventory of range conditions on the public rangelands, and shall categorize or identify such lands on the basis of the range conditions and trends thereof as they deem appropriate. Such inventories... shall be kept current on a regular basis so as to reflect changes in range conditions; and shall be available to the public.47
The Act directs that when grazing is allowed on BLM-administered public lands, the management goal "shall be to improve the range conditions of the public lands so that they become as productive as feasible for all rangeland values."48
Statutory Requirements for Inventories and Monitoring
The SCS, USFS, and BLM are required to inventory rangelands for all resources, albeit for a variety of purposes. The laws cited in the previous sections mandate, request, or imply the need for significant amounts of information for managing, planning, and inventorying rangelands. An assessment of rangelands is a statutorily essential component of the inventories of all three agencies, and all three agencies are specifically required to consider watershed, recreation, and wildlife as part of either their inventory responsibilities (SCS) or other duties (SCS, USFS, and BLM). The purpose of inventories is to provide the necessary planning and management information. Recent legislation has increased the breadth and depth of the information required from inventories. Congress has left federal agencies considerable discretion to devise and adopt standards and methods to inventory and monitor rangelands. Congress has made clear, however, that rangeland inventory and monitoring systems should
cover all rangeland resources, specifically including watersheds, and should emphasize determination of the quality of the land;
relate to the congressional mandate to achieve sustained yields or health of all renewable resources;
lead to improvement in the status of rangelands; and
be useful for planning and management purposes as well as for purely informational purposes.
CURRENT INVENTORYING AND MONITORING SYSTEMS
There are great differences between the inventory and monitoring systems currently used on nonfederal versus federal rangelands.
SCS surveys all nonfederal lands every 5 years as part of the National Resources Inventory, which is mandated by the Soil and Water Resources Conservation Act of 1977. In 1987, nearly 163 million hectares (402 million acres) of nonfederal lands, excluding Alaska, were classified as rangelands. The information gathered through the National Resources Inventory process is used in the Soil and Water Resources Conservation Act appraisal published by SCS every 10 years. The data used in the National Resources Inventory are collected by census, area sampling, and point sampling methods (U.S. Department of Agriculture, Soil Conservation Service, 1989b). (Area sampling involves stratification of the land base and selection of a primary sampling units from each stratum. Point sampling is the collection of data such as plant composition, slope, and soil type at points designed within each primary sampling unit.) Census data are used to identify nonfederal lands, which are the lands the SCS then inventories for the National Resources Inventory using area and point sampling methods. Major land resource areas were the basis for stratification in the 1977, 1982, and 1987 National Resources Inventories; and the 1982 inventory was statistically reliable to that level.
For the United States as a whole, nearly 350,000 permanent primary sampling units have been established, and a random selection is sampled for each National Resources Inventory. Each of the three National Resources Inventories conducted by SCS has used a different number of primary units: 77,000 sites in the 1977 inventory, more than 300,000 in the 1982 inventory, and 100,000 in the 1987 inventory (U.S. Department of Agriculture, Soil Conservation Service, 1987).
Primary sampling units come in four sizes: 16, 40, 65, and 299 hectares (40, 100, 160, and 640 acres, respectively). For each primary sampling unit there are computer-selected points from which data are gathered. For example, there are two such points on 16-hectare (40-acre) primary sampling units and three on 65-hectare (160-acre) primary sampling units. Data are collected over the entire primary sampling unit (area sampling) and at each designated point in the primary sampling unit (point sampling). Both kinds of data are then expanded to produce totals for a state.
Instructions for the 1987 National Resources Inventory required each randomly selected primary sampling unit to be located on an aerial photograph or map. They directed that specified data—those for urban and built up lands, farmsteads, critically eroding areas, and water bodies—within the sampling area be delineated from photographs on a preprinted work sheet developed specifically for the primary sampling unit and that inventory (U.S. Department of Agriculture, Soil Conservation Service, 1987). Because the 1987 survey was designed to update the 1982 National
Resources Inventory (U.S. Department of Agriculture, Soil Conservation Service, 1987), the work sheet included data obtained during the previous effort(s). The instructions required data from previous inventories to be verified and revised and identified the new data that were to be obtained. The instructions prescribed how each item was to be addressed, including the way that these measurements were to be validated in the field.
The instructions for point sampling required the collection of more detailed information for each primary sampling unit—range site (SCS) and range condition (SCS), apparent trend (SCS), kinds of crops, types of conservation practices in use and needed, soil characteristics, erosion, and wetlands either during actual visits to established points or through the use of maps, photographs, or remote sensing. They called for range condition (SCS) to be determined by the SCS methods described in Chapter 3 of this report.
The systematic collection of range condition (SCS) data as part of the National Resources Inventory allows SCS to make estimates of the area of nonfederal rangelands in excellent, good, fair, or poor range condition (SCS) as determined by their resemblance to the defined climax plant community (SCS) thought to be characteristic of that site. The National Resources Inventory system permits national-level assessments of range condition (SCS) on nonfederal rangelands. Data are collected by a statistically valid sampling scheme and by using standardized definitions and methodologies. The assessment of range condition (SCS) that is part of this system is not an adequate assessment of rangeland health, but the system to collect, analyze, and aggregate data that could be used to assess the health of nonfederal rangelands is in place.
There is no statistically designed survey of federal rangelands comparable to the National Resources Inventory of nonfederal rangelands. Each agency responsible for managing federal rangelands collects data on the rangelands under its jurisdiction using the methods selected by each agency.
BUREAU OF LAND MANAGEMENT
BLM has responsibility for managing 109 million hectares (270 million acres) of land, approximately 69 million hectares (170 million acres) of which have been classified as rangeland (U.S. Department of the Interior, Bureau of Land Management, 1989). BLM recognizes that monitoring and evaluation are essential management functions in (1) establishing and evaluating progress in meeting resource management objectives, (2) developing management plans, (3) preparing environmental analyses, and
(4) supporting decision making (U.S. Department of the Interior, Bureau of Land Management, 1985a:Section 4400.06). BLM's national policy calls for the rangelands under its jurisdiction to be inventoried and classified according to their ecological status (USFS and BLM). However, BLM does not gather data primarily for a national inventory. Most of the available BLM rangeland data are used to manage individual grazing allotments. These data are not collected by a statistically designed sampling method that would allow confident aggregation of results on a national basis. BLM's ecological status (USFS and BLM) and trend data have, however, been combined in the past and have been used in national reports, including the assessment required by the Resource Planning Act of 1974.
U.S. FOREST SERVICE
USFS has responsibility for managing 74 million hectares (182 million acres) of land, of which about 16 million hectares (40 million acres) are classified as rangelands (U.S. Department of Agriculture, U.S. Forest Service, 1989b). Management plans are developed by USFS personnel for units of land such as livestock grazing allotments, timber management units, or watersheds. All of the units and management activities are combined into forest plans for each national forest.
Standards for ecosystem classification and interpretation for use by all units of the national forest system have been recently adopted (U.S. Department of Agriculture, U.S. Forest Service, 1991b). Regional offices will continue to have the flexibility they need to select the methods that they consider appropriate to meet the standards. The new policy states that USFS shall use ecological type classification to "coordinate and integrate resource inventories to stratify land and resource production capability and make predictions and interpretations for management" (U.S. Department of Agriculture, U.S. Forest Service, 1991c:Section 2060.3). The purposes of this new policy are (1) to provide an integrated ecosystem classification based on potential natural community, soils, and physical characteristics; (2) to provide a unifying ecosystem framework for use in land and resource management;" and (3) to develop an ecologically based information system to aid in evaluating land capability, interpreting ecological relationships, and improving multiple use management (U.S. Department of Agriculture, U.S. Forest Service, 1991c:Section 2060.2).
In addition to utilizing information regarding ecological type (USFS) and ecological status (USFS and BLM), the USFS system also includes resource value ratings for livestock, wildlife, and other uses. These ratings are intended to permit the agency to assess the degree to which the vegetation on a site satisfies or meets previously established vegetation management objectives (Society for Range Management, 1989:5).
Current USFS inventory procedures, however, do not produce statistically reliable estimates of the proportion of rangelands in each ecological status (USFS and BLM) class. Different methods are used to measure ecological status in different USFS regions. In addition, the ecological status (USFS and BLM) data are collected as part of ongoing USFS management activities rather than as part of a representative sampling program. The differences in methods used and the absence of a statistically reliable sampling design do not allow confident compilation of USFS ecological status (USFS and BLM) data at the national level.
Current Inventories are Inadequate
Recent U.S. General Accounting Office reports suggest that both BLM and USFS are limited in their ability to obtain adequate inventory information. About two-thirds of BLM allotments and one-fourth of USFS allotments did not have management plans, and data for another 16 percent of BLM-managed rangelands and 31 percent of USFS-managed rangelands were more than a decade old (U.S. General Accounting Office, 1988a). The USFS's forest planning process rests on the availability of information on all resources, and insufficient or inadequate data hamper this process (U.S. Department of Agriculture, U.S. Forest Service, 1990a). Fosburgh (1986) noted the lack of an adequate information flow between forest planning processes and the national-level planning process as envisioned in the Resources Planning Act and suggested that this was a failure of the Act.
Professional organizations such as the Society for Range Management (1989) have tried to bring the various reports of the SCS, BLM, and USFS together into a single document. Unfortunately, the data for those reports were originally gathered for different purposes, at different times, by using different techniques and formats. In this example, USFS reports ecological status on rangelands, whereas SCS and BLM report range condition.
Conservation groups such as the National Wildlife Federation and the Natural Resources Defense Council have examined data reported by BLM for the portion of U.S. rangelands under its jurisdiction (Wald and Alberswerth, 1989). Oversight agencies such as the U.S. General Accounting Office (1988a,b) have made their own assessments. The U.S. General Accounting Office assessments are in response to House and Senate requests for information. For the U.S. General Accounting Office study on the management of grazing allotments (U.S. General Accounting Office, 1988a), a questionnaire was used to obtain data from BLM and USFS because it was considered impractical to conduct on-site visits at more than a few field offices.
Some individuals (Box, 1990; Box et al., 1976) have applied their professional judgments to the various historical surveys to try to assess historical changes in rangelands. These data sources range from documents such as the 1936 Senate report (U.S. Congress, Senate, 1936) to compilations of data from different agencies, and they cannot be considered adequate surveys of the nation's rangelands. Current rangeland inventories simply do not provide the data needed to support national assessments of rangeland health.
NATIONAL SYSTEM OF INVENTORYING AND MONITORING RANGELAND HEALTH IS NEEDED
The structure of an inventory involves consideration of its appropriate geographical scope, the intensity of sampling within that geographical scope, the kind of land to be sampled, and the attributes to be sampled. The questions that are asked at the national level require a different level of sampling intensity than the questions that are asked about a specific allotment on USFS- or BLM-administered lands. The questions asked about rangelands involve both rangeland monitoring and rangeland inventorying.
Inventories establish the status of a rangeland's resources at a given point in time. Resources are evaluated on the basis of either direct measurement or statistical inference taken from sampling resources indicators. Monitoring measures changes in the status of selected resources or indicators over time. It may involve a complete reinventory, but it more commonly involves repeated measurements using the same methods on selected areas that represent larger areas (Society for Range Management, Range Inventory Standardization Committee, 1983). Monitoring measures selected attributes on rangelands that can be accurately remeasured to determine changes in those attributes. The attributes selected (that is, vegetation, soil, or animals) determine the environmental changes that can be detected. Although rangeland inventories have been implemented, few would qualify as rangeland monitoring.
A national-level assessment of rangeland health requires the following:
adoption of a standardized and consistent definition of rangeland health and of measurable indicators of change in rangeland health;
consistent and well-correlated classification of federal and nonfederal rangelands;
collection of data by the same or similar methods that will enable the data to be combined on a national level;
collection of data on the basis of a statistically valid sampling
scheme that enables data to be evaluated at the state, regional, and national levels; and
periodic and consistent repetition of sampling to detect trends in the measures used to evaluate rangeland health.
A system that can be used to produce statistically reliable estimates of the health of rangelands is not in place for either federal or nonfederal rangelands. There is an urgent need to develop a system to inventory rangeland health to judge whether current management and use of the nation's federal and nonfederal rangelands are adequately conserving the capacity of rangelands to produce commodities and satisfy values. The fundamental output of a national monitoring system should be the collection and reporting of a set of data of measurable indicators of rangeland health. These data can be used to estimate the proportion and distribution of federal and nonfederal rangelands that are healthy, at risk, or unhealthy and to determine whether current use and management are conserving the productive capacity of the nation's rangelands.
Convene an Interagency Task Force
The secretaries of USDA and DOI should convene an interagency task force to develop, test, and standardize indicators and methods for inventorying and monitoring rangeland. health on federal and nonfederal rangelands.
Standardized indicators are needed for each of the criteria recommended in Chapter 4: degree of soil movement by wind and water, distribution of nutrients and energy in space and time, and plant demographics. The indicators suggested in Chapter 4 should serve as a useful starting point in the development of standardized indicators of rangeland health, but they must be refined and tested.
Standardized methods to measure indicators and to classify rangelands as healthy, at risk, and unhealthy are also needed. The preliminary decision rules described in Chapter 4 (see Table 4-7) should serve as a useful starting point for determining when a rangeland is healthy, at risk, or unhealthy, but these classification decisions must be refined and tested.
The multiagency task force should coordinate federal efforts, including EPA's EMAP, leading to
a set of indicators that should be included in a minimum data set for inventorying and monitoring rangeland health,
standard methods of measuring indicators of rangeland health,
a series of field tests to validate the indicators and methods selected, and
quantification of the correlation between measures of rangeland health and range condition (SCS) or ecological status (USFS and BLM).
National Sampling System Needed
The secretaries of USDA and DOI should develop coordinated plans for implementing a sampling system on federal and nonfederal rangelands that will produce estimates of the proportion of healthy, at-risk, and unhealthy rangelands that are significant at an appropriate substate level.
A national sampling system that coordinates the activities of the USDA, DOI, and EPA is needed to collect, analyze, and aggregate data to determine the proportion of federal and nonfederal rangelands that are healthy, at risk, or unhealthy. The National Resources Inventory, conducted by SCS, provides a statistically valid sampling design for nonfederal rangelands. The addition of standardized indicators of rangeland health to the National Resources Inventory can produce statistically valid estimations of the proportions of nonfederal rangelands that are healthy, at risk, or unhealthy.
No comparable sampling program is in place on federal rangelands. Most of the data collected are for management purposes rather than for inventorying and monitoring purposes. The development of a coordinated sampling system for both federal and nonfederal rangelands is urgently needed.
Periodic Sampling Needed
The secretaries of USDA and DOI should develop coordinated plans for implementing periodic sampling of federal and nonfederal rangelands to determine changes in the proportions of healthy, at-risk, and unhealthy rangelands.
Periodic monitoring must be a fundamental part of a valid national system for evaluating rangeland health. The periodicity of repeated sampling should reflect the rapidity of change within the indicators selected to monitor rangelands and the degree of degradation that a change implies to give adequate early warning of increases in the area of unhealthy rangelands. Monitoring should be periodic enough such that a rangeland would not slip from a healthy to an unhealthy state between sampling periods.
Transition to Rangeland Health
Indicators of soil surface condition should be added to all current and ongoing range condition (SCS) and ecological status (USFS and BLM) assessments, and any other ongoing efforts to assess rangelands, as a first step toward a more comprehensive evaluation of rangeland health.
There is much experience with the use of soil surface characteristics as indicators of soil stability and watershed function. The addition of indica-
tors of soil surface condition to all current and ongoing efforts to assess rangelands would be a useful first step toward a more comprehensive system of evaluating rangeland health—a step that should be taken immediately. Data on the soil surface condition of rangelands should also be collected as part of the National Resources Inventory.
All current and ongoing rangeland assessments done as part of Resources Conservation Act (RCA) appraisals, Resources Planning Act (RPA) assessments, national forest planning, USFS and BLM land use and allotment planning, and environmental assessments should be based on the analysis of multiple ecological attributes.
SCS, USFS, and BLM should analyze multiple ecological attributes of rangelands as part of current rangeland assessments and appraisals. Currently, plant composition and, in some cases, biomass production are the only attributes systematically used in rangeland assessments. These data alone are not sufficient for assessing rangeland health. Plant composition and production data collected as part of range condition (SCS) and ecological status (USFS and BLM) ratings should be analyzed in conjunction with information collected on indicators of soil surface condition as recommended above and all other available information on erosion rates. Using these multiple indicators, the agencies could begin to assess soil stability and watershed function, distribution of nutrients and energy, and presence of functioning recovery mechanisms as a means of identifying rangelands at greater risk of loss of health. This analysis should be part of conservation planning or management of grazing allotments as well as national appraisals and assessments.
These assessments would not provide a complete assessment of rangeland health, but they would represent progress toward measuring and analyzing multiple ecological attributes within each agency. They would also help guide national policy for managing federal and nonfederal rangelands in the interim while more comprehensive and systematic assessments of rangeland health are developed.
Basic data on soil surface conditions, erosion rates, plant composition, and biomass production assembled and used to assess rangelands as part of RCA appraisals, RPA assessments national forest planning, environmental assessments, and other assessments of federal and nonfederal rangelands should be made available to the public and the scientific community for independent review.
Independent review will increase the understanding of and confidence in the results of the assessments of federal and nonfederal rangelands. Publication of basic data will provide a data set for scientific evaluation of the utility of alternative indicators of soil stability and watershed function, distribution of nutrients and energy, and presence of recovery mechanisms as measures of rangeland health. The availability
of basic data, for example, used to estimate erosion rates as part of the National Resources Inventory has allowed scientists to test the effect of alternative agricultural policies and crop management practices on erosion rates. Independent review of these basic data has increased the confidence of estimates of erosion reductions expected from changes in farming practices. It is important that basic data on multiple ecological attributes of federal and nonfederal rangelands be made available to both the public and the scientific community to accelerate the transition to comprehensive methods for assessing rangelands.
SCS, USFS, and BLM should continue current and ongoing range condition (SCS) and ecological status (USFS and BLM) ratings while the transition to rangeland health assessment is made.
The data that have been and continue to be collected for range condition (SCS) and ecological status (USFS and BLM) assessments should provide a critical historical data set for use in judging changes in rangeland conditions. As a transition is made to national-level inventorying and monitoring of rangeland health as recommended here, it is imperative that this information not be lost. The committee strongly recommends that current and planned monitoring efforts that use range condition (SCS) and ecological status (USFS and BLM) move ahead and be augmented by the collection of additional data for evaluating rangeland health.