Before we go further, we need to define some critical terms. First, one needs to realize that biodiversity entails many different things to different interest groups (West, 1995). To some, it is mainly genetic material. To others, it is taxonomic richness, usually species, of biota within plots or more abstract communities and landscapes. To still others, it is properly functioning ecosystems, including indigenous human cultures living in sustainable ways. All these views are legitimate and have to be respected in democratic societies.
Even though scientists of various kinds are pushing broadened views of biodiversity, the public activists are, as reflected in legislation, budgets, and activity, favoring the charismatic megafauna, the warm, fuzzy, and appealing organisms, particularly the vertebrates, not the little things that run the world (Wilson, 1987). Administration of the Endangered Species Act (ESA), the strongest environmental law in the U.S., currently only impacts what can be done to listed species and their habitats, including activities on privately held lands and waters.
It is becoming obvious that far more than scientific information is involved in what is being done about biodiversity. Stances about biodiversity inevitably involve one's personal and professional ethics (Coufal, 1997). Thus, this is a topic that will inevitably cause philosophical reflection, as well as scientific and managerial action.
The second term deserving further definition is rangelands. Some prefer a strictly use-oriented definition. In that sense, rangelands are agroecosystems since they are all lands with self-sown vegetation used for livestock grazing. That is the oldest definition that still prevails in developing countries. This traditional definition also applies to a wide array of ecosystem types where livestock grazing has and could occur, including recently cut forests, tundras, and marshes. The majority of rangelands, however, occur where grasslands, shrub steppes, deserts, woodlands, or savannas prevail, in other words, most of the untilled or undeveloped western U.S. (about 70% of the area). Rangeland managers and scientists are thus more familiar with drier and less fertile systems than most foresters, wildlife biologists, and agronomists. Whereas most of such lands were recently seen primarily as sources of food and fiber, in developed countries many of them are being increasingly dedicated to sustaining other values that are now prized more highly in industrialized societies. We thus have to contrast how rangeland biodiversity is being considered in the developed compared with the developing world.
My focus here will be on the drier parts of the world where self-sown vegetation is managed extensively based on ecological principles. Agronomic principles rarely apply to these lands: the costs of attempting to till, seed, fertilize, treat with pesticides, and use other means of strong manipulative control to enhance production of food and fiber rarely justify their expenditure because plant responses are fundamentally low due to meager precipitation, salty, steep, and rocky soils, etc. The previous lack of such treatments is the major reason that rangelands are now seen as valuable repositories of biodiversity. That is, most of these rangelands have not yet been simplified and homogenized by intensive agricultural activities (Matson et al., 1997). There are some important exceptions, however, such as the Conservation Reserve Lands (Allen, 1995), which are former croplands that could become rangelands and/or wildlife reserves, depending on Congress' budget setting.
A thorough review of all aspects of biodiversity in all kinds of rangelands around the world would be impossible for several reasons. First of all, not all aspects of biodiversity have been thoroughly studied in all kinds of rangelands. The genetics of even dominant plants and vertebrates, and anything about invertebrates, microbes, ecosystem functions, and feedbacks, have rarely been studied. Second, even the information that does exist cannot all be summarized in the space available here. Therefore, what I have chosen to do is exemplify how biodiversity issues interact with science and policy in one ecosystem type (sagebrush steppe) well known to the author. I will bring in ideas and experimental results from other contexts as well and discuss how they might apply to sagebrush steppe. In that way I can give a more-focused introduction to the topic at hand.
Shrub steppes are ecosystems with organisms and life-forms of both deserts and grasslands. Although, on average, they are drier than most grasslands and wetter than deserts, the variation in climate is high (coefficients of variation in total annual precipitation usually exceed 30%). Thus, some years have grasslandlike climate whereas other years are desertlike. This climatic variation is probably the main reason for the mix of grassland and desert life-forms in making up shrub steppes. Another result of the high climatic variation is the inherently low stability of these systems under disturbance (Archer and Smeins, 1991).
Because the environmental conditions of the sagebrush steppe are harsh and highly variable over time and space, the dominant organisms are few and widely distributed. This belies the probable high degree of intraspecific ecotypic and genetic variation, which has barely been studied. Once these patterns are understood, variations in autecological and ecophysical responses and synecological interactions will be more comprehendible.
Location, Ownership, and Land Uses
Sagebrush steppe occurs wherever there is or once was vegetation with shared dominance by sagebrushes (woody Artemisia spp.) and bunchgrasses (West and Young, 1998). This system occurs mostly in the lowlands of the northern part of the Intermountain West. Sagebrush steppe once occupied about 45 x 106 ha there (West and Young, 1998). About 20% of this ecosystem type passed into private ownership with the Euroamerican settling of the West (Yorks and McMullen, 1980). The remaining 80% is managed by various agencies of the U.S. and state governments. This circumstance makes the management of these lands much more difficult than those under private ownership. Many interest groups, including those championing biodiversity, can and do politically influence management policies on these public lands.
About half of the original sagebrush steppe area now in private ownership has been converted to either dryland or irrigated agriculture over the past 150 years. The approximately 90% remaining untilled lacks irrigation water or is too steep, rocky, or shallow soiled for annual cultivation. The dominant historical uses of these wildlands by human societies have been first hunting and gathering and then livestock grazing.
The prevailing climate in sagebrush steppe is temperate, semiarid (mean annual precipitation of 20 to 40 cm) and continental (cool, wet winters and springs and warm, drier summers and autumns). Mean annual temperatures range from 4 to 10°C. Winters are cold enough so that snow packs of 50 to 100 cm are common. Snowmelt is usually gradual and thus most of the moisture therein becomes stored at depth in the soil. Native plant growth occurs largely from April to July, the only part of the year when both temperatures and soil moisture are favorable. Summer precipitation is rarely enough to carry herbaceous plant growth throughout the summer. Early fall precipitation is not dependable and by October temperatures are usually too cool to allow much regreening of grasses (West and Young, 1998).
The major woody dominants here are woody Artemisia, collectively known as the sagebrushes. These are shrubs derived from progenitors which came from Eurasia over the Bering Land Bridge and have subsequently radiated into about 13 species (McArthur, 1983). Furthermore, the major species, Artemisia tridentata (big sagebrush), has at least five relatively easily recognizable subspecies that should be used in separating out different ecological sites (McArthur, 1983).
The sizes and degrees of dominance of the sagebrush species vary greatly with both site and disturbance history. Sagebrush density is generally greater, but height lower, on more xeric sites. Sagebrush also increases in abundance following excessive livestock grazing in the spring (West and Young, 1998). Livestock grazing also reduces the chance of fires by removal of fine fuels in the interspaces connecting the clumps of shrubs. Fire formerly kept the sagebrush steppe more frequently burned (60 to 110 year return interval) (Whisenant, 1990) and less dominated by sagebrush because most species of sagebrush do not resprout after fire, but have to regenerate from seed (Blaisdell et al., 1982).
Even when sagebrush is dominant, a moderate number of other plant species are found associated with it. On relict (naturally ungrazed by livestock) sites in central Washington, Daubenmire (1970) found an average of 20 vascular plant species in 1000-m plots. Tisdale et al. (1965) found a range of 13 to 24 vascular plant species on three relict stands in southern Idaho. Zamora and Tueller (1974) found a total of 54 vascular plant species in a set of 39 late seral stands in the mountains of northern Nevada. Mueggler (1982) found between 24 and 41 vascular plant species in a set of 68 0.05-ha lightly grazed macroplots in sagebrush steppe of western Montana.
The vertical and horizontal plant community structures are remarkably similar in all relatively undisturbed examples of this ecosystem type. The shrub layer reaches approximately 0.5 to 1.0 m in height. The shrubs have a cover of about 10 to 80%, depending on site and successional status. The grass and forb stratum reaches to about 30 to 40 cm during the growing season. Herbaceous cover also varies widely depending on site and successional status. On relict sites, the sum of cover values usually exceeds 80%, and can approach 200% on the most mesic sites (Daubenmire, 1970).
The herbaceous life-forms most prevalent on relict sites are hemicryptophytes (Daubenmire, 1975). The proportion of therophytes increases markedly with disturbance. The proportion of geophytes is around 20%. A microphytic crust dominated by mosses, lichens, and algae is commonly found where litter from perennials is not excessive (West, 1990). Sagebrushes have both fibrous roots that can draw water and nutrients near the surface and a taproot that can function from deep in the soil profile. Near the end of the growing season for grasses, sagebrushes nocturnally water from more than 90 cm and excrete it in the upper part of the soil profile at night (Caldwell and Richards, 1990). This hydraulic can help the grasses stay active longer than possible on their own.
Perennial grasses associated with Artemisia vary greatly throughout the region. The C bunchgrasses (Agropyron spicatum, Festuca idahoensis, Stipa spp., Sitanion hystrix, Poa spp.) dominate the herbaceous layer in the north and western parts of the type. C sod grasses (e.g., Agropyron smithii, Hilaria jamesii) become more common in the south and east where more growing season precipitation occurs (West, 1979).
Total aboveground standing crop phytomass within the sagebrush steppe type varies between about 2000 to 12,000 kg/ha, depending on site differences, successional status, and age of the brush (West, 1983). Litter standing crops are about one half the live nonwoody material (West, 1985). Belowground phytomass is similar in magnitude to that aboveground. Annual net aboveground primary production varies between about 100 and 1500 kg/ha, depending on site, successional status, stand age, and preceding climatic conditions (Passey et al., 1982).
Plant ecologists have long assumed that communities that are floristically richer stabilize primary production in the face of variable climate (Chapin et al., 1997). Indeed, Passey et al. (1982) in their discussion of long-term data gathered from ungrazed sagebrush steppe relicts conclude that each year brings both unique dominance-diversity and production relationships. They attribute this to differing phenologies, rooting patterns, and green leaf persistence. Harper and Climer (1985) reanalyzed the Passey et al. (1982) data set and concluded that variation in plant community production was more positively related with floristic richness than either average precipitation or precipitation of a given year. Tilman et al. (1996) have shown that greater species richness in tall grass prairie leads to greater production during drought than in more depauperate stands created by adding nutrients.
Any landscape within which sagebrush steppe is the matrix is a patchwork of stands of differing species composition and shrub or other growth form dominance. The mix of plant species and growth forms is dependent on ecological site potential and time since particular disturbances. Fires, grazing by both native and introduced vertebrates and invertebrates, as well as unusual climatic events such as deep soil freezing before snowpack accumulation and unusually heavy precipitation and consequent soil anoxia, all contribute to resetting the successional clock (West and Young, 1998). Livestock grazing on these rangelands usually takes place in large paddocks with only one or a few watering points. The parts most distant from water thus are less grazed and of higher seral status (Hosten and West, 1996). This creates a patchwork of differing seral statuses across the landscape (Laycock et al., 1996).
The native vertebrates using this ecosystem type are a mixture of grassland and desert species. Maser et al. (1984) grouped the vertebrates of sagebrush steppe in southeastern Oregon into 16 life-forms and related them to vegetation structure and other features of habitat. The vertebrate community is more diverse when the vegetation has the greatest structural diversity (Parmenter and MacMahon, 1983). Neither shrub-dominated nor grass-dominated situations favor as many different kinds of vertebrates as do the mixtures. A few such as voles (Microtus montanus) can influence the structure by girdling the shrubs (Mueggler, 1967; Parmenter et al., 1987).
Over 1000 species of insects have been observed on a sagebrush-grass site in southern Idaho (Bohart and Knowlton, 1976). Wiens et al. (1991) recently identified 76 taxa of invertebrates on sagebrush alone in central Oregon. Relatively little is known about the habitat preferences, trophic relationships, and other aspects of the roles of invertebrates in this ecosystem type. Only a few — thrips, webworms, grasshoppers, cicadas, aphids, and coccids (Kamm et al., 1978; West, 1983) — are known to be irruptive and visibly alter vegetation structure.
Very little is known about microbes and the decomposition process in this ecosystem type. Initial studies of the nitrogen (West and Skujins, 1978) and phosphorus (West et al., 1984a) cycles showed that available forms of these elements may limit plant production in wetter than average years. Allelochemics from sagebrush and the high C:N ratios of its litter may inhibit some decomposition and nitrogen-cycling processes, perhaps indirectly strengthening sagebrush dominance in this ecosystem type (West and Young, 1998). Changes in litter quality can lead to degradation of soil organic matter in such systems (Lesica and DeLuca, 1996). Global environmental changes may produce some unexpected interactions among plants, soil microbes, and soil degradation (West et al., 1994).
Interactions among Plants, Animals, and Humans
The pristine sagebrush steppe evolved with large browsers (megafauna), most of which had disappeared by about 12,000 years ago (Mehringer and Wigand, 1990; Burkhardt, 1996). The loss of the megafauna is inextricably linked to simultaneous increases in human hunting and climatic warming (Grayson, 1991). Remaining graminivores were few in the pre-European system (Mack and Thompson, 1982; Harper, 1986). The small populations of aboriginal hunters and gatherers of the mid-Holocene probably influenced the vegetation largely by burning. It took European colonization to change drastically the native vegetation and the wildlife habitat it provides (Young, 1989).
The pre-European era livestock grazing capacity, when shrubs were fewer and grasses more prevalent, was estimated to be 0.83 animal unit months (AUM)/ha (McArdle and Costello, 1936). Because sagebrushes are usually unpalatable to livestock, whereas herbs are palatable, uncontrolled livestock use led to a decline of herbs and increase in brush. Carrying capacities declined to an average of 0.27 AUM/ha in the 1930s (McArdle and Costello, 1936), but had improved slightly to 0.31 AUM/ha by 1970 (Forest-Range Task Force, 1972).
Livestock populations built up rapidly near the end of the 19th century. Griffiths (1902) judged that the grazing capacity of these rangelands had been exceeded by 1900. Hull (1976) examined historical documents and concluded that major losses of native perennial grasses and expansion of shrubs took only 10 to 15 years after a site was first grazed by livestock.
The native grasses are extremely palatable, especially when green. They die easily when grazed heavily in the spring (Miller et al., 1994). In addition, they rarely produce good seed crops (Young, 1989).
The only time the grasses and forbs have an advantage over brush is when sites are burned. However, on the sites with heavy historical livestock use, both remaining native herbaceous perennials and their seed reserves have been greatly diminished (Hassan and West, 1986). In addition to tall, thicker sagebrush, grazing-induced freeing of space and resources gave opportunities for the invasion of aggressive Eurasian plants. The advent of introduced winter annual grasses, notably Bromus tectorum in the 1890s (Mack, 1981), and the continuous, fine, and early-drying fuels they provide has led to seasonally earlier, more frequent (less than 5 years), and larger fires (Whisenant, 1990). After repeated fires, combined with unrestricted grazing, any remaining native vegetation becomes easily replaced by other, even more noxious introduced annuals, such as medusahead (Taeniatherum caput-medu-sae), knapweeds (Centaurea spp., Acroptilion spp.), and yellow star thistle (Centaurea solstitialis). The result has been a considerable decrease in plant species structural and floristic diversity, average forage production, and nutritional value to vertebrates (Billings, 1990; Whisenant, 1990). This simplification of self-sown vegetation results in much more frequent bare ground and accelerated wind and water erosion (Hinds and Sauer, 1974). Variability in plant production goes up several orders of magnitude after replacement with annuals (Rickard and Vaughn, 1988).
Wildlife responds dramatically to these changes in vegetation structure (Maser et al., 1984). For instance, the pigmy rabbit (Brachylagus idahoensis) is a threatened species that prefers the tallest, densest stands of Basin big sagebrush. Sites occupied by this plant have been widely converted to intensive agriculture. Thus, the range of this sensitive animal has been reduced and its abundance greatly diminished.
Another native herbivore of special interest in the sagebrush steppe is the sage grouse (Centrocercus urophosianus). This is a large galliform with a unique digestive system that has coevolved with Artemisia. The mature birds survive the less hospitable times of the year by eating the twigs of sagebrushes, especially the low sagebrushes found on windswept ridges. There are, however, other requirements during other parts of their life cycle. During March and April, the males gather on open areas without brush (called leks) and display themselves to the females. Only about half of the males survive raptor predation and intraspecific fighting during this about 2-week mating period. The females fly to the most productive interfluvial areas to nest and raise the chicks. For the first 6 weeks of life, the young birds require a high protein diet made up of insects and forb buds. These are most abundant in fresh burns and in riparian corridors.
Sage grouse were very abundant in the region when Europeans first arrived and have remained abundant enough to be an important game bird until recent decades. Unfortunately, they are now being considered for placement on the endangered lists in several Intermountain states. Wildlife and conservation biologists find it tempting to single out the range livestock industry for causing this problem. However, sheep, which prefer forbs over other types of forage, were much more abundant on these rangelands up to about 1960, but have since declined to a tiny fraction of their former abundance. Sheep do, however, eat some sagebrush, particularly in the fall and winter. The amount of time cattle are permitted on public lands of the sagebrush steppe has also been declining since about 1964, well before sage grouse populations crashed. The amount of perennial cover on much remaining sagebrush steppe has been increasing of late because of reduced livestock grazing and more effective fire control. There is now probably more sagebrush than necessary for optimum sage grouse use in most portions of the sagebrush steppe. Several other possible influences have also been increasing of late, such as vehicular access and nonhuman predators. Coyotes, foxes, skunk, racoons, corvids (jays, magpies, crows, and ravens), and raptors (eagles, hawks, and owls) have all been increasing because of less shooting and pesticide use and could be taking more eggs and chicks, as well as adults. The thickened brush could be making predator stalking and capture easier.
Because of passage of laws such as the ESA and National Forest Management Act, the interests of wildlife, particularly the rare, endangered, and threatened vertebrates, can take precedence over optimal livestock grazing on publicly owned rangelands in the U.S. This is the reason that the U.S. Forest Service and Bureau of Land Management currently strives to leave about 15 to 20% of the mature sagebrush cover intact across the landscape rather than burning or using herbicides to reach the 100% kill they once strived for in the 1940s and 1950s when the nation demanded more red meat.
There has already been a vast replacement of native plant species by Eurasian plant invaders in sagebrush steppes. More is expected, especially if global warming materializes. Controlling fires entirely is an impossibility. Reductions or even complete removal of livestock will not result in a rapid return to the vegetation that occurred before European colonization (Miller et al., 1994). Sheep, grazed during the fall, because they utilize some sagebrushes and can do little damage to the herbaceous understory during that time of year, can actually enhance floristic richness (Bork et al., 1998).
Our major means of obtaining greater dependability of forage production and soil protection on severely degraded sagebrush steppe sites, while at the same time reducing the chance of fire, has been to plant Eurasian wheatgrasses and ryegrasses (Asay, 1987). However, this can only be done easily on relatively level sites with deep, largely rock-free soils. Environmental and archaeological interest groups have recently stopped these procedures, however. Environmentalists object to using any introduced species, regardless of their ability to grow rapidly and protect the soil. Archaeologists object to the physical disturbances to archaeological objects and strata. Native species have been repeatedly tried in plantings, but rarely grow early and rapidly enough to outcompete the introduced annuals. Because environmentalists have prevailed, public land managers are no longer daily involved in proactive management or ecosystem repair here.
Let us now turn to other possible ways to conserve remaining community diversity, alter existing stands, or rehabilitate degraded sagebrush steppe stands. Figure 1 will be used to guide the following discussion. This figure is a state-and-transition model (Laycock, 1995) thought to accommodate better our current understanding of degradation and successional processes in sagebrush steppe than the simpler, linear models of the past with one end point (the climax).
Preservation of Relatively Unaltered Ecosystems
Pristine, relictual areas (State I in Figure 1) no longer exist nor are probably recoverable. The reasons for this view are
1. Humans (indigenous peoples) are no longer hunting, gathering, and burning these areas. The previous fire regimes are no longer in place and as the vegetation changes in response to less frequent fires, the hydrologic and nutrient cycles are being altered, as is the habitat for numerous animals and microbes.
2. The present climate is warmer and drier than the cooler, wetter Little Ice Age climate which prevailed up to about 1890. Thus, only heat- and drought-tolerant species may thrive now under global warming.
3. Atmospheric CO has increased about 20% during the past century, altering the competitive balances in this vegetation as well as changing the nutritional qualities of the phytomass and litter (Polley, 1997).
4. About 15% of the flora is now new to the region.
Since we can reverse none of these influences, at least in the short term, we should learn to live with what remains and manage it toward the desired plant communities we choose for each circumstance.
There are, however, some remnants of these landscapes that have escaped direct human influences. These relics exist because they have no surface water, are surrounded by difficult topography, or protected in special-use areas (e.g., military reservations). I place these in State II of Figure 1. Tisdale et al. (1965) describe an example. I estimate that less than 1% of the sagebrush steppe that remains has avoided the direct impact of any livestock. Even these relicts are, however, incomplete because of lack of indigenous humans and lengthened fire frequencies. Relicts are influenced by air pollutants, climatic change, and invasion by exotics (Passey et al., 1982). Most of the existing late seral sagebrush steppe (State II in Figure 1) has had light livestock use. Even light livestock use puts inordinate pressure on a few highly palatable species (ice cream plants), partially explaining the lack of a return arrow from State I to State II.
In some places, feral horses and burros now put considerable pressure on such rangelands, but are protected by federal law on most public lands. I estimate that about 20% of the remaining sagebrush steppe is in State II.
The perceived will of a majority of Americans now is to identify these remaining State II areas, especially those on public lands, and protect them from being developed. Some advocate all such areas be reserved (Kerr, 1994), whereas others (Bock et al., 1993) propose that 25% have livestock excluded. Rose et al. (personal communication) have, however, recently demonstrated that lightly grazed sagebrush steppe has higher species richness than adjacent exclosures dating to 1937. Others propose restoration efforts to bring further-degraded systems back to States I or II (Dobson et al., 1997). State II areas serve as the parts catalogue for restoration efforts. The Gap Analysis Program (GAP) of the U.S. Fish and Wildlife Service (Scott et al., 1993) and the various natural heritage programs initiated by the Nature Conservancy are well under way to put these views in action. These efforts are, however, not without attack from both political and scientific groups (Machlis et al., 1994; Short and Hestbeck, 1995).
I expect to see physical modifications to enhance production of food and fiber (formerly called range improvements) to be more spatially limited than in the past because such actions on public lands or with public monies require environmental assessments or impact statements and thus public scrutiny and debate. The remaining relatively unaltered areas on public lands will probably be consciously protected to provide the later seral condition patches necessary to hold a broader spectrum of all species, and meet the special requirements for some featured species such as sage grouse and pigmy rabbit (Call and Maser, 1985). Of special concern are other sagebrush bird obligates that are also apparently declining: sagebrush sparrow
(Amphispiza belli), sage thrasher (Oreoscoptes montanus), and Brewer's sparrow (Spizella breveri).
Rangeland managers in the past strove to reduce the limitations of the land for producing livestock. These limitations were mainly topography, forage availability, and water. For example, trails were constructed into areas where topographic breaks limited livestock access. Natural water was supplemented by development of springs, building stock tanks and small dams, drilling wells, piping and hauling water. Fences were constructed and salt distributed to control livestock movement and institute grazing management systems (e.g., rest-rotation grazing). All these improvements were designed to distribute livestock utilization more uniformly across the land, gain greater efficiency of food and fiber production, and divert livestock from the especially sensitive riparian areas (Elmore and Kauffman, 1994; Laycock et al., 1996). The net result has been progressively more widespread intensive use of a landscape that has become partially tamed from the wild. These assumptions need to be reexamined in the light of biodiversity concerns. Let us continue our consideration of these relationships in the sagebrush steppe.
Because livestock grazing of native sagebrush steppe usually avoids the unpalatable forages, particularly woody species, they are freed from competition and dominance becomes concentrated in the few woody plants on areas with a history of heavy livestock grazing (T2), but not recent fire (State III, Figure 1). About 30% of this ecosystem type is estimated to exist currently in this state. Most of these stands can stay stagnated for decades (Rice and Westoby, 1978; West et al., 1984b; Sneva et al., 1984; Winward, 1991). The dense, competitive stands of excess sagebrush prevent the herbaceous species from recovering. Such brush-choked stands are usually chosen by both livestock and wildlife managers for manipulation to diversify vegetation structure. This enhances it for livestock or native animals in spots, concentrating livestock use, reducing their pressure elsewhere, while simultaneously advantaging some wildlife species through vegetation modifications via grazing systems, prescribed burning, brush beating, or chaining (T3). For example, sheep grazing in the fall, because they consume more sagebrush then (Bork et al., 1998), can be used to obtain a reversal from State III to State II. Prescribed burning (Harniss and Murray, 1973) can also be applied to stands with sufficient remnant populations of native herbs to quickly recover following brush kill. Rest from livestock use, such as with a rest-rotation grazing system or winter only use (Mosely, 1996), will often allow a slower return to State II from State III. Reduction of brush also enhances water yields (Sturges, 1977), and some seeps, springs, and streams reappear. When phenoxy herbicides are used alone (Evans et al., 1979) (T4) or in conjunction with fire, the community becomes dominated with native grass (State IV, Figure 1) because the chemicals impact all broad-leafed species. The conversion only slowly returns (T6) to State II with judicious grazing and a secondary treatment with prescribed burning. About 5% of the remaining sagebrush steppe is now estimated to be in State IV. This is a short-lived state, especially under heavy grazing (T5). Mueggler (1982) found enhanced alpha diversity in moderately grazed sage brush steppe communities in western Montana following prescribed fire, 2,4-D, and brush-beating treatments. Summer fires can damage some of the grasses (Young, 1983), but encourage the resprouting rabbitbrushes (Chrysothamnus spp.) and horse-brushes (Tetradymia spp.) (Anderson et al., 1996).
If accelerated soil erosion does not ensue and the fundamental potential of the site does not change, then State III can be maintained or managed toward States II or IV. However, as herbaceous plants and litter in the interspaces between perennials are reduced, soil aggregate stability declines, infiltration of precipitation diminishes, overland flow increases, and soil erosion frequently increases (Blackburn et al., 1992). When a probable threshold of use is exceeded, the site can irreversibly change to one of lesser potential. This explains the dashed line and downward arrows below States III and V as the only believable transitions. This is where the syndrome of desertification is most evident. All the former states can be dealt with via soft energy management approaches. Once this threshold is exceeded, however, subsequent management requires expensive, risky, hard energy solutions. Unfortunately, it is often easier to get political attention after major damage has been done rather than getting budgets and personnel to plan, monitor, and tweak the higher-condition, more-natural systems at opportune times.
The desertified sites with thickened brush have largely introduced annuals in their understory. I estimate that State V comprises about 30% of the current sagebrush steppe. Reduction or removal of livestock only hastens further degradation from State V because livestock remove part of the fuel load and thus reduce the chance of fire destroying the sagebrush and the spots of soil it protects.
If insufficient amounts of native herbs remain on sagebrush steppe, the usual land management agency response has previously been to replace them mechanically (T7) with introduced wheatgrasses and ryegrasses, especially crested wheatgrass (Asay, 1987). This has been done because the introduced perennial grasses are much more easily established than the native grasses and they grow quickly to provide more forage with a higher nutritional plane. The introduced perennial grass stands are also much more tolerant of subsequent heavy livestock use and have lasted for many decades (Johnson, 1986). There are some long-range concerns, however (Les-ica and DeLuca, 1996), because the introduced perennial grasses suppress the return of natives and richer plant species assemblages. Some large treatment areas have monocultures of Eurasian perennial grasses prevailing (State VI, Figure 1). I estimate about 5% of the original sagebrush steppe has already been transformed to State VI.
Wildlife biologists have noted declines in the numbers of birds (Olson, 1974; Reynolds and Trost, 1979; 1981), small mammals (Reynolds and Trost, 1980), and large reptiles (Reynolds, 1979) on such seedings of introduced grasses. It should be noted, however, that such studies present a worst-case scenario because samples came from the center of large treatments. Provision for increased diversity near edges (Thomas et al., 1979) is not usually mentioned in such studies. Present-day more-sensitized planners would provide for optimum edge effect and patchiness (McEwen and DeWeese, 1987). When society makes the investment in repairing severely damaged sagebrush steppe, creating perennial grass-dominated pastures of much greater productivity of species palatable to livestock (T7), this should compensate for livestock reductions and other management restrictions on lands where
States II, III, and IV (Figure 1) predominate. Because introduced grass pastures take heavy degrees of utilization in the spring, the native shrub steppe can support fall and winter grazing with less impact, especially on the native herbaceous perennials.
Introduced perennial grass plantings in the sagebrush steppe region, especially if grazed by livestock, will eventually experience shrub reinvasion (T8 to State VII) largely in response to intensity of livestock grazing. I estimate that about 5% of the remaining sagebrush steppe region is currently represented by shrub-reinvaded introduced wheatgrass/ryegrass pastures (State VII). Not all brush is now eliminated by re-treatment (T9). Herbicide use on public lands in the Pacific Northwest has been suspended by judicial decree. Prescribed burning of the coarser, introduced grasses is difficult and leaves patches where the shrubs prevail. There are, therefore, chances to enhance edge effects in large areas that were formerly homogenized. As in the untilled native areas, we could enhance wildlife habitat by providing a mix of successional stages or stand conditions, providing both cover and forage for either featured species or total species richness (Maser et al., 1984). For example, some success has been attained in creating alternate leks for sage grouse breeding following disturbances of development (Eng et al., 1979).
Despite greatly increased attention to fire prevention and control, much of the most-depauperized sagebrush steppe (State V) has been burned (T10) at least once during the past three decades and is now dominated by introduced annuals, mainly grasses such as cheatgrass and medusahead (State VIII, Figure 1). The Bureau of Land Management (M. Pellant, personal communication) estimates that about 3 million acres of public lands in Idaho, Utah, Oregon, and Nevada are now dominated by cheatgrass and medusahead. I estimate that about 25% of the total original sagebrush steppe has made this transition (T10, T11). Because of their short stature, restricted nutritional characteristics (short period of aboveground gree-ness), and greater susceptibility to recurring fires than sagebrush steppe, such areas are undesirable from all viewpoints. Without nutritional supplementation, livestock can graze State VIII only during the short, early spring plant-growing season (winter use is possible in the lower elevation areas near the Columbia River; Mosely, 1996). Only the most generalist animals, such as the introduced chukars (Alectoris chukar), horned larks (Eremophila alpestris), grasshoppers, and deer mice (Peromyscus maniculatus) seem to thrive on the annual grasslands (Maser et al., 1984). When such areas burn in early summer, soils are bared to wind and water erosion during the convectional storms of summer. The consequent needs for revegetation after fire are increasing while the budgets of federal land management agencies decline and pressure from environmentalists against active management increases.
Land dominated by annuals may provide fair watershed protection during years without fire and actually appears to be more productive of total plant tissues than the original sagebrush-native perennial grass and forb combination (Rickard and Vaughan, 1988). This is likely, however, to be only a temporary situation based on the priming effect of decomposing litter (Lesica and DeLuca, 1996) and the miner alization of nutrients from the enormous belowground necromass of the original system. When these reserves of nutrients and soil organic matter are finally respired away, the annual grasslands are likely to become much less productive. Similar transitions happened in the Middle East several millennia ago (Zohary, 1973). Many other, more noxious weeds from that region could find their way here, and we could witness a downward spiral of further degradation (T12).
Rather than allowing the annual grasslands derived from former sagebrush steppe (State VIII, Figure 1) to remain and the land to degrade further, some land managers are attempting to intervene. A joint U.S. Forest Service, Bureau of Land Management, Agricultural Research Service, and University of Idaho program is under way to reduce these threats (Pellant, 1990). The most notable component of this effort is the green-striping program most evident in southern Idaho. The basic approach is to begin breaking up the now vast stretches of cheatgrass and other annual dominance that have developed as fires have become earlier, larger, and more frequent (Whisenant, 1990). Land managers are attempting to break the area into smaller, burnable units, especially nearer to cities and towns. The approaches used thus far include planting strips of vegetation that stay green (and thus wetter and less burnable) longer than cheatgrass.
Although the introduced wheatgrasses and ryegrasses do stay green longer and burn less readily, because of coarser aboveground structure, they are not native and thus are rejected by some interest groups. Because the genetic biodiversity of the native plants is so primitively understood, the best that can be done is to gather such seed locally and plant it on comparable sites. Such seed sources are undependable, however; thus a root-sprouting big sagebrush is seen as a potentially better keystone species to put back in this area. A few sagebrushes may actually help sustain perennial grasses by harboring the predators on black grass bugs (Labops spp.) (Haws, 1987). Furthermore, total plant community production can be enhanced (Harniss and Murray, 1973) because sagebrushes help trap blowing snow (Sturges, 1977) and scattered sagebrushes moderate temperatures (Pierson and Wight, 1991), benefit the reestablishment of native herbs, and protect them from excessive utilization (Winward, 1991). Sagebrushes also harbor mycorrhizal fungi (Wicklow-Howard, 1989), which helps them extract nutrients from deep in the soil and recycle them to the surface through litter production (Mack, 1977; West, 1991).
Whether or not we can accomplish restoration of sagebrush steppe (T13, between States V and III in Figure 1) is highly questionable. Even where money is less limiting and topsoil is stockpiled on coal strip mines, early results are only partially encouraging (Hatton and West, 1987). We will have to know much more about how sagebrush steppe ecosystems are structured and function and obtain vast budgets and more trained personnel before such efforts are routinely successful. It is cheaper and more feasible to foster good stewardship of land having late seral vegetation (manage while in States I, II, III, or IV of Figure 1) rather than rely on restoration efforts after degradation has taken place (States V, VI, VII, and VIII of Figure 1).
Because biodiversity issues in sagebrush steppe are interconnected to multiple impacts and other ecosystem types over the entire region, the federal management agencies are attempting to address them in a holistic fashion. An important example of this is the proposal for ecosystem management in the interior Columbia Basin (Haynes et al., 1996). The documents generated (Quigley et al., 1997) appear to favor restoration practices. Environmentalists (e.g., Belsky, 1997), however, perceive little change in livestock grazing practices and intend to test the process judicially.
Was this article helpful?