Effects of Grazing on Abundance and Distribution of Shortgrass Steppe Consumers

Although livestock are the most obvious consumers on the shortgrass steppe, they are certainly not the only consumers. However, livestock may influence the other consumers in a number of different ways. They may directly compete for food resources with other aboveground herbivores. There is behavioral interference between livestock and some species of wildlife (Roberts and Becker, 1982), but not others (Austin and Urness, 1986). The removal of biomass by livestock alters canopy structure (physiognomy) and influences microclimate. Bird, small-mammal, and insect species can be variously sensitive to these structural alterations (Brown, 1973; Cody, 1985; MacArthur, 1965; Morris, 1973; Rosenzweig et al., 1975; Wiens, 1969). There are both short- and long-term effects of grazing on plant community species composition, primary production, and plant tissue quality. Belowground consumers can also be affected by the effects of grazing on soil water infiltration, nutrient cycling, carbon allocation patterns of plants, litter accumulation, and soil temperature. The overall effects of livestock on a particular component of the native fauna can be negative or can be positive through facilitative relationships (Gordon, 1988). In this chapter we assess the effects of cattle grazing on other above- and belowground consumers, on the diversity and relative sensitivity of these groups of organisms, and on their trophic structure. We first present some brief background information on plant communities of the shortgrass steppe and on the long-term grazing treatments in which many of the studies reported herein were conducted. Details on the plant communities are presented by Lauenroth in chapter 5 (this volume), grazing effects on plant communities by Milchunas et al. in chapter 16 (this volume); and grazing effects on nutrient distributions and cycling by Burke et al. in chapter 13 (this volume). The physiognomy of the shortgrass steppe is indicated in its name. The dominant grasses (Bouteloua gracilis and Buchloë dactyloides), forb (Sphaeralcea coccinea), and carex (Carex eleocharis) have the majority of their leaf biomass within 10 cm of the ground surface. A number of less abundant midheight grasses and dwarf shrubs are sparsely interspersed among the short vegetation, but usually much of their biomass is within 25 cm of the g round. Basal cover of vegetation typically totals 25% to 35%, and is greater in long-term grazed than in ungrazed grassland. Bare ground (more frequent on grazed sites) and litter-covered ground (more frequent on ungrazed sites) comprise the remainder of the soil surface (Milchunas et al., 1989).

Soil Organic Matter and Nutrient Dynamics of Shortgrass Steppe Ecosystems

Since the days of the IBP, there has been a strong emphasis on research about the biogeochemistry of shortgrass steppe ecosystems (e.g., Clark, 1977; Woodmansee, 1978). A major theme has been seeking to understand spatial and temporal patterns and controls of biogeochemical pools and fluxes at scales that span from several centimeters to hundreds of kilometers, and from hours to millennia. The synthesis of this work has resulted in a conceptual framework regarding the biogeochemical dynamics of the shortgrass steppe, with two key components:… 1. Spatial and temporal patterns are controlled by five 1. major factors: climate, physiography, natural disturbance, human use, and biotic interactions. Plants are the most important biotic component. The interaction of these factors as they change in time and space determines the distribution and size of biogeochemical pools and the rates of biogeochemical processes. 2. Carbon (C), nitrogen (N), and other associated biologically active elements are overwhelmingly located belowground, with more than 90% found in soils (Burke et al., 1997a). This distribution determines the biogeochemical sensitivity of the shortgrass steppe to perturbations…. These ideas have been synthesized in the development of the CENTURY ecosystem simulation model, originally developed for grasslands and agroecosystems in the shortgrass steppe region of the western Great Plains (Parton et al., 1987, and chapter 15, this volume). The model represents complex interactions among the five controlling factors to simulate C and N cycling, and has served as an organizing framework for developing hypotheses and for evaluating questions that are dif. cult to address in the field (Parton et al., chapter 15, this volume). The objectives of this chapter are to describe how nutrient pools and fluxes are distributed in the shortgrass steppe, to characterize how the five controlling factors interact to create spatial and temporal patterns, and to evaluate the potential future changes to which the biogeochemistry of the shortgrass steppe may be particularly vulnerable.

Net Primary Production in the Shortgrass Steppe

Net primary production (NPP), the amount of carbon or energy fixed by green plants in excess of their respiratory needs, is the fundamental quantity upon which all heterotrophs and the ecosystem processes they are associated with depend. Understanding NPP is therefore a prerequisite to understanding ecosystem dynamics. Our objectives for this chapter are to describe the current state of our knowledge about the temporal and spatial patterns of NPP in the shortgrass steppe, to evaluate the important variables that control NPP, and to discuss the future of NPP in the shortgrass steppe given current hypotheses about global change. Most of the data available for NPP in the shortgrass steppe are for aboveground net primary production (ANPP), so most of our presentation will focus on ANPP and we will deal with belowground net primary production (BNPP) as a separate topic. Furthermore, our treatment of NPP in this chapter will ignore the effects of herbivory, which will be covered in detail in chapter 16. Our approach will be to start with a regional-scale view of ANPP in shortgrass ecosystems and work toward a site-scale view. We will begin by briefly placing ANPP in the shortgrass steppe in its larger context of the central North American grassland region. We will then describe the regional-scale patterns and controls on ANPP, and then move to the site-scale patterns and controls on ANPP. At the site scale, we will describe both temporal and spatial dynamics, and controls on ANPP as well as BNPP. We will then discuss relationships between spatial and temporal patterns in ANPP and end the chapter with a short, speculative section on how future global change may influence NPP in the shortgrass steppe. Temperate grasslands in central North America are found over a range of mean annual precipitation from 200 to 1200 mm.y–1 and mean annual temperatures from 0 to 20 oC (Lauenroth et al., 1999). The widely cited relationship between mean annual precipitation and average annual ANPP allows us to convert the precipitation gradient into a production gradient (Lauenroth, 1979; Lauenroth et al., 1999; Noy-Meir, 1973; Rutherford, 1980; Sala et al., 1988b).

Ecology of Mammals of the Shortgrass Steppe

At first glance, the shortgrass steppe seems to offer little in the way of habitat for mammals. The expansive rolling plains, with little topographic relief or vegetative cover, provide minimal protection from predators or the harsh weather typical of the region. The short stature of the dominant native grasses prevents the development of any significant litter layer, and although snowfall can often be significant, too little accumulates to form the subnivean habitats that support small mammal populations in forests and more productive grasslands in winter. As a consequence, ecologists have typically considered the vertebrate fauna of the shortgrass steppe to be depauperate compared with other Great Plains grasslands, a hardy collection of generalists living in sparse populations. Although this characterization may generally be accurate, it has led mammalian ecologists to overlook the fauna of the shortgrass steppe in favor of that of other grasslands. It is precisely these circumstances, however, that suggest that a long-term approach may be necessary to understand the dynamics of mammal populations here. Relatively few such studies have been completed to date, but we can use the comparative and experimental results that are available to begin to determine what factors might be important. Here we review research on mammals in the shortgrass steppe, with the goal of identifying the general patterns and processes that contribute to them. Our review is roughly divided into four parts. We begin by describing the mammal communities and their broad habitat associations in shortgrass steppe environments. We then review the history of mammal research in the region to synthesize what these studies (many unpublished) have taught us about the most important determinants of the distribution and abundance of native species. Studies of mammal\ populations in the northern shortgrass steppe have spanned nearly 40 years, and we next describe some major patterns that have emerged from studies during this period. Much of this past research focused on the role of mammals in the structure and function of shortgrass steppe ecosystems, and we revisit this issue in some detail, with special emphasis on the important and sometimes controversial role of prairie dogs and other burrowing rodents. Finally, we end by considering how humans, and especially agriculture and its related activities, affect the diversity, abundance, and persistence of resident mammal populations.

Effects of Grazing on Vegetation

Grazing by large native ungulates and semiaridity are the two main forces that have had a large infuence in shaping the current-day structure of the shortgrass steppe ecosystem (Milchunas et al., 1988). With the uplift of the Rocky Mountain chain during the Miocene (approximately one million years ago), forests of the Great Plains were gradually replaced by grasslands (Axelrod, 1985). Large grazing and browsing animals inhabited the Great Plains during the middle to late Pleistocene, as did grasses of the genera Stipa, Agropyron, Oryzopsis, and Elymus (Axelrod, 1985; Stebbins, 1981). Bison occurred both east and west of the Rockies during the Wisconsin glacial period in the latter part of the Pleistocene (Wilson, 1978). During the early Holocene, approximately 10,000 years ago, bison and grasses of the genera Bouteloua, Buchloë, Andropogon or Schizachyrium, and Sorghastrum concomitantly increased throughout the Great Plains (Stebbins, 1981), but bison did not proliferate west of the continental divide (Mack and Thompson, 1982; Van Vuren, 1987). The natural shift in fauna from horses, pronghorn, and camels to bison and wild sheep from Eurasia is thought to have favored the spread of shortgrasses such as Bouteloua and Buchloë (Stebbins, 1981). Furthermore, grassland flora east and west of the Rocky Mountains probably had separate origins (Leopold and Denton, 1987). The shortgrass steppe is unique from other North American semiarid ecosystems in having bison play an important role. Bison did not proliferate west of the Rocky Mountains as they did on the Great Plains to the east. This is due in part to a lack of coincidence in timing of bison lactation and the phenological development of C3 grasses in the more Mediterranean–like climate west of the Rockies, in contrast to the mix of C3 and C4 grasses and pattern of spring–summer precipitation on the Great Plains (Mack and Thompson, 1982). Other explanations for the low numbers of bison west of the Rocky Mountains include physiographic barriers restricting immigration (Kingston, 1932), low p rotein content of forage (Daubenmire, 1985; Johnson, 1951), heavy snowfall as a cause of mortality (Daubenmire, 1985), and low aboveground primary production coupled with disjunct suitable habitat (Van Vuren, 1987). Bison a lso did not prosper in the southwestern United States, nor did a large herbivore fauna develop in South America (Stebbins, 1981).

The Future of the Shortgrass Steppe

Where lies the future of the shortgrass steppe? In prior chapters we have described the remarkable resilience of the shortgrass steppe ecosystem and its organisms to past drought and grazing, and their sensitivity to other types of change. Emerging from this analysis is the idea of vulnerability to two main forces: future changes in precipitation or water availability, and direct human impacts. What are the likely changes in the shortgrass steppe during the next several decades? Which of the changes are most likely to affect major responses in the plants, animals, and ecosystem services of the shortgrass steppe? In this chapter we evaluate the current status of the shortgrass steppe and its potential responses to three sets of factors that will be driving forces for the future of the steppe: land-use change, atmospheric change, and changes in diseases. Referring to the early 1900s, James Michener in his novel Centennial (1974) wrote the following:… The old two-part system that had prevailed at the end of the nineteenth century— rancher and irrigator—was now a tripartite cooperation: the rancher used the rougher upland prairie; the irrigation farmer kept to the bottom lands; and the drylands gambler plowed the sweeping 0 eld in between, losing his seed money one year, reaping a fortune the next, depending on the rain. It was an imaginative system, requiring three different types of man, three different attitudes toward life. . . . (p. 1081)… Even today, because of the strong water limitation for cropping, the shortgrass steppe remains relatively intact, or at least unplowed, in contrast to other grassland ecosystems (Samson and Knopf, 1994). More than half of the shortgrass steppe remains in untilled, landscape-scale tracts, compared with only 9% of tallgrass prairie and 39% of mixed-grass prairie (The Nature Conservancy, 2003). These large tracts, including those in the national grasslands (Pawnee, Cimarron, Comanche, and Kiowa/Rita Blanca), provide the greatest opportunity for preserving key ecological processes and biological diversity.

Cattle Grazing on the Shortgrass Steppe

Cattle are the primary grazers on the shortgrass steppe. For example, during the late 1990s, 21 shortgrass counties in Colorado reported about 2.36 million cattle compared with 283,000 sheep (National Agricultural Statistics Service, USDA, 1997a), 60,000 pronghorn antelope, and a few thousand bison (Hart, 1994). Assuming one bison or five to six sheep or pronghorn consume as much forage as one bovine (Heady and Child, 1994), cattle provide about 97% of the large-herbivore grazing pressure in this region. The ratio of cattle to other grazers is even greater in the remainder of the shortgrass steppe. In 1997, the three panhandle counties of Oklahoma reported 387,000 cattle and only 1300 sheep, whereas the 38 panhandle counties of Texas reported 4.24 million cattle and 14,000 sheep (National Agricultural Statistics Service, USDA, 1997b,c). How ever, only a bout half the cattle in the panhandle counties of Texas and Oklahoma graze on rangeland the remainer are in feedlots. Rangeland research on the shortgrass steppe (Table 17.1 describes the parameters of the major research stations in the shortgrass steppe) has included a long history of both basic ecology and grazing management. The responses of rangeland plant communities to herbivory are addressed by Milchunas et al. (chapter 16, this volume) and to disturbance are discussed by Peters et al. (chapter 6, this volume). Here we focus on research pertaining to three management practices important to cattle ranching on shortgrass steppe: stocking rates, grazing systems, and extending the grazing season via complementary pastures and use of pastures dominated by Atriplex canescens [Pursh] Nutt (fourwing saltbush). Stocking rate, de. ned as the number of animals per unit area for a speci. ed time period, is the primary and most easily controlled variable in the management of cattle grazing. Cattle weight gain responses to stocking rate or grazing pressure (animal days per unit of forage produced) have been quanti. ed in several grazing studies on the shortgrass steppe (Bement, 1969, 1974; Hart and Ashby, 1998; Klipple and Costello, 1960). Average daily gains per animal are better estimated as a function of grazing pressure, rather than stocking rate, as forage production is highly variable in this semiarid environment (Lauenroth and Sala, 1992; Milchunas et al., 1994).

The Shortgrass Steppe and Ecosystem Modeling

Ecological modeling has played a key role in scientific investigations of the SGS LTER during the past several decades. The SGS LTER site, focused initially on the Central Plains Experimental Range (CPER), was the main grassland research site for the Grassland Biome component of the U.S. IBP effort (Lauenroth et al., this volume, chapter 1). Initial development of ecosystem models occurred from 1 970 to 1 975 as p art of t he I BP . All the U.S. I BP projects (grassland, tundra, desert, deciduous forest, and coniferous forest biomes) included research on the development of ecosystem models, with the goals of using models to help formulate and interpret field experiments, and of projecting the impact of changes in management practices on ecosystem dynamics. Models were developed as part of the Grassland Biome project (Bledsoe et al., 1971; Innis, 1978), and included modeling specialists who worked with research biologists on the development and formulation of the ecosystem models. The modeling activities of t he U.S. IBP Grassland Biome project included developing the ELM Grassland model (Innis, 1978). The ELM model was a complex process-oriented model that was intended to be used at all the Grassland Biome sites in the United States. This model was developed by postdoctoral fellows who were to formulate the different submodels, and then link the submodels using software that was developed as part of the program. The submodels included a plant production submodel, a cattle production submodel, a linked nutrient cycling and soil organic matter submodel, a grasshopper dynamics submodel, and a soil temperature and water submodel. Biophysical and biological data from the different sites were collected to develop and test the model. Model development was constrained by lack of knowledge about the biological processes that control ecosystem behavior, and by lack of appropriate data to test the ability of the model to simulate ecosystem responses to changes in grazing and fertility management practices. However, the ELM Grassland model was quite successful at investigating the interactions of different components of the ecosystem, and at helping to formulate new research efforts.

Land-Use History on the Shortgrass Steppe

As described in chapter 1 of this volume, the grasslands of central North America began to expand at the end of the Wisconsin period (about 10,000 years BP), and continued their expansion through the warming trend that persisted until about 3000 years BP, occupying their maximum territory at that time (Dix, 1964). Currently, the region still supports trees on escarpments, along streams, and at other sites protected from fire, but centuries ago, fires caused by lightning or kindled by Native Americans may have eliminated relict stands of forest and savanna on the open plains. Large browsers and grazers also may have played a part in eliminating trees as well as grasses sensitive to grazing pressure (Axelrod, 1985). Throughout millennia, bison in particular were likely to have shaped the plant communities of the shortgrass steppe, and thus were an essential component of the system (Larson, 1940). Bison appeared as early as 300,000 years BP; bison, mammoths, mastodons, camels, horses, and other grazers were numerous by 20,000 years BP. Humans arrived in North America perhaps as early as 60,000 years BP, but certainly by 15,000 years ago. Fires and bison may have achieved maximum impact as recently as the past 500 years (Axelrod, 1985; Looman, 1977). The roles of climate, fire, and grazing in the development of North American grasslands have been examined by Ellison (1960), Coupland (1979), Dyer et al. (1982), Anderson (1982), and Tetlyanova et al. (1990). The earliest known human sites on the shortgrass steppe date to about 13,000 years BP (Wedel, 1 979) and a re f ound i n the vicinity o f fossil g lacial l akes. The population of these mammoth hunters was apparently sparse and scattered. Soon after 11,000 years BP, many of the large mammalian species such as the mammoth, native horse, camel, and ground sloth vanished, and the hunters turned to bison. Bone beds representing mass kills of bison have been found below buffalo jumps (Fig. 4.1) and even in the remains of wood or stone corrals, but single kills must have been much more common.

Climate of the Shortgrass Steppe

The climate of a region involves the short- and long-term interaction among the atmospheric, hydrologic, ecologic, oceanographic, and cryospheric components of the earth’s environmental system (Hayden, 1998; Pielke, 1998, 20 01a,b). These interactions occur across a ll spatial and temporal scales, from turbulence generated by diurnal cycles at a landscape scale, to globalscale circulation. The establishment of particular ecosystem types is associated with a nonlinear feedback between the atmosphere and the underlying vegetation (Pielke a nd Vidale, 1995). Wang a nd E ltahir (20 0 0) and Claussen (1998) have demonstrated that vegetation patterning cannot be accurately simulated in a model unless vegetation–atmosphere feedbacks are included. In this chapter we summarize the climate system of the shortgrass steppe. This is a region of large seasonal contrasts, and of interannual and longer term variability. It is also a region that has undergone major human impacts during the past 150 years. We present both average conditions and examples of extreme events in the shortgrass steppe to illustrate the variable climate of this interesting ecosystem. Geographic factors play a large role in determining the climatic characteristics of the shortgrass steppe (Lauenroth and Burke, 1995; Lauenroth and Milchunas, 1992; Lauenroth et al., 1999). Key factors for this region include its mid-latitude position, its relatively high elevations, its interior continental location, and its proximity to the Rocky Mountains, a substantial north–south-oriented mountain barrier immediately to the west. Air masses affecting the region consist of continental polar air from the north, humid continental air masses from the east, humid subtropical air masses from the southeast and south, and Paci8 c maritime air masses from the west. The latter can be signi8 cantly modi8 ed as they cross a series of mountain ranges and interior dry regions before reaching the shortgrass steppe region. Each of these geographic and atmospheric features contributes to the climate of the region. Latitude determines day length and sun angle, and, hence, solar insolation. This, in turn, greatly affects air temperature. Upper level westerly winds increase over the mid-latitudes in the fall and winter in response to strengthening north–south temperature gradients in the atmosphere. Paci8 c air masses are carried eastward over the Rocky Mountains, depositing considerable cool-season precipitation in the mountains, but rarely on the shortgrass steppe.