Large mammalian herbivores not only depend on plant communities for their existence but cause major changes in plant community composition and structure. These changes have direct consequences for ecosystem processes, but recent studies of ungulate-ecosystem relations show widely divergent ungulate effects in different ecosystems. We reviewed studies of ungulate effects on plant community composition to gain insight into potential mechanisms of ungulate-induced changes in both community composition and ecosystem processes. Our analysis of these studies is based on the premise that the effect ungulates exert on plant communities depends on the balance between (1) feeding selectivity of herbivores (i.e., degree to which dif- ferent plant species or ecotypes experience different levels of tissue loss), and (2) differences among plant species in their ability to recover from tissue loss. A large number of studies clearly show that selective ungulate herbivory leads to the dominance of unpalatable, chemically defended plant species in communities. However, many studies have also demonstrated that intensive long-term herbivory does not lead to the invasion of unpalatable species into the community, and can even increase the dominance of highly palatable species. Our review indicates that high levels of nutrient inputs or recycling and an intermittent temporal pattern of her- bivory (often due to migration) are key factors increasing the regrowth capacity of palatable species and hence maintaining their dominance in plant communities supporting abundant herbivores. Key factors limiting un- gulate foraging selectivity, again limiting herbivore-induced dominance of slow-growing, unpalatable species, include herding behavior, early growing season and postfire herbivory, asynchronous phenology of palatable versus unpalatable species, and low relative abundance of unpalatable species. Our review indicates differences among ecosystems in the role played by ungulate herbivory result from the relative strength of these factors enhancing plant tolerance to herbivory and limiting foraging selectivity. Anthropogenic changes in these factors (e.g., alteration of migration patterns) therefore have the potential to significantly alter the effects of ungulates on plant communities and ecosystem processes.
Fire and grazing are interactive disturbance processes that are important to the structure and function of grassland ecosystems. Studies of nitrogen (N) availability report different effects following grazing and fire. However, these studies have largely neglected the spatially controlled interaction between fire and grazing. The objective of our work was to evaluate an application of the fire-grazing interaction model on N availability in a tallgrass prairie. We compared patches within a shifting mosaic landscape where each patch varied in time since focal disturbance (fire and intense grazing disturbance). We also evaluated N availability on a burned and grazed landscape where fires and moderate grazing occurred annually and uniformly across the entire landscape. These treatments were both burned and grazed where the only difference was spatial and temporal variability in fire application and grazing disturbance. Samples were collected from upland sites in May of 2003 and 2004. Total soil inorganic N (NH4+-N + NO3--N) and a growth chamber experiment with hard red winter wheat (Triticum aestivum L. cv. Jagger) were used to evaluate potential N availability. Our study produced patterns of N availability that are more similar to studies of grazing lawns where N availability is enhanced by focal grazing than from studies of fire without grazing. Overall, our study demonstrates that fire and grazing are interactive. Unburned patches have minimal grazing pressure and low N availability. Fire-grazing interaction may provide a management alternative that enables sustainable livestock production, through increased carrying capacity in focally disturbed patches, concomitant with biological diversity in tallgrass prairie.
In many rangeland ecosystems, the role of fire on vegetation dynamics has been the object of detailed studies. In Argentina, and especially in Patagonia, the knowledge of how fire changes vegetation is scarce. In 3 areas affected by wildfires on different dates (in 1988, 1994, and 1998), we determined the structure of the vegetation (plant cover, density, and biomass) and compared it with that of nearby unburned areas. Based upon these data, we present a qualitative state-and-transition model of this rangeland. For the sites burned in 1988 and 1994, aerial biomass, density, and cover of perennial grasses were significantly greater (P < 0.05) for burned than for unburned areas. For the site burned in 1998, although there were no significant differences in perennial grass biomass and cover, density was significantly greater compared to the unburned area. Total shrub cover was significantly lower (P < 0.05) in burned than in the unburned areas, attaining 49.7%, 15.0%, and 5.5% of that of the unburned areas for the sites burned in 1988, 1994, and 1998, respectively. Similar to cover, density for most shrubby species was significantly greater in the unburned than in the burned areas for the 3 sites. Grazing and fires of different intensities, combined with variable rainfall, makes the prediction of postfire vegetation changes difficult. However, it can be generalized that fire changed the vegetation from shrub-dominated steppes, a persistent state in northeastern Patagonia, into a grass-dominated transient state. This change has persisted for more than 10 years after a fire event. From the perspective of sheep raising, the significant postfire increase in perennial grass biomass represents a substantial improvement in the condition of these rangelands. However, repeated fires would be necessary to control the sprouting shrubs and maintain the grass-dominated state.
Although remote sensing has many potential applications for range management, its use by range managers thus far has been limited. To investigate the factors that encourage use of remote sensing and to examine its influence on decision making by individuals who manage privately owned rangeland, we evaluated the decision-making processes of 3 ranch owners and 1 professional ranch manager who were introduced to remote sensing while collaborating with us in a rangeland stewardship program in California. Two of the participants had extensive ranching experience (11 to > 20 years) and managed large cattle ranches (1 000 to > 2 000 ha), and 2 had less experience and managed smaller sheep ranches (< 200 ha). During the 5-year program, the participants implemented a series of new management practices, including prescribed burning, rotational grazing, and seeding of native grasses, with the aim of reducing noxious weeds and increasing productivity. We used remote sensing to quantify the effect of these practices and provided ranch-wide remote sensing analyses to each manager on a password-protected Web site. Using case study methodologies, we found that managers of larger, commercially active ranches found the experimental use of remote sensing to be a highly positive experience that convinced them that this technology could help address difficult management situations and increase ranch profitability. This suggests that the broad use of remote sensing by managers of privately held, commercial rangelands may be limited in part by the simple lack of opportunity to test these technologies. Programs that assist ranchers in obtaining appropriate remote sensing products thus may be a cost-effective way to enhance conservation on private rangelands. Our findings suggest that voluntary self-analysis by ranchers of the landscape dynamics of their own properties is likely to lead to more engaged conservation efforts than will top- down prescriptions.
Flea beetles (Aphthona spp.) are biological control agents introduced from Eurasia to reduce the cover and density of leafy spurge (Euphorbia esula L.). The potential for the introduced beetles to use alternate hosts for feeding and development in North America is slight; however, it is possible. Species at highest risk are native species closely related, ecologically and taxonomically, to leafy spurge. A native spurge, Euphorbia brachycera Engelm. is consubgeneric and sympatric with leafy spurge throughout the northern Great Plains of the United States, and was not included in prerelease host-specificity testing for Aphthona nigriscutis or Aphthona lacertosa. The objective of this study was to evaluate the actual and potential ecological overlap among leafy spurge, flea beetles, and E. brachycera. Wide-ranging and intensive field surveys indicate that E. brachycera is found well within the range of leafy spurge and flea beetles. E. brachycera occurs infrequently, in low densities, in areas with a high percentage of bare ground, and with a root system dissimilar to leafy spurge. Flea beetles released directly into populations of E. brachycera failed to persist beyond a single field season and plants showed no evidence of feeding by beetles. Our results suggest that the potential for flea beetles to host-shift is low due to differences in growth habit and root morphology between E. brachycera and leafy spurge.
Rangeland managers often must decide whether to suppress dicotyledonous weed populations with expensive and time-consuming management strategies. Often, the underlying goal of weed suppression efforts is to increase production of native forage plants. Many managers suppress weeds only when they feel the unwanted plants are substantially impacting their forage base. Currently, intuition and guesswork are used to determine whether weed impacts are severe enough to warrant action. We believe scientific impact assessments could be more effective than these casual approaches to decision making. Scientific approaches will necessitate data on weed abundances because the severity of a weed's impact is highly correlated with its abundance. The need for weed abundance data poses major obstacles because gathering these data with readily available techniques is time consuming. Most managers cannot or will not spend a lot of time gathering vegetation data. In this paper, we explore a rapidly measured index (<2 minutes per sample location) that is highly correlated with weed (i.e., leafy spurge Euphorbia esula L.) abundance per unit area. This index is based on the light attenuation leafy spurge causes. After measuring light attenuation in plots planted to leafy spurge and grasses, we developed a probabilistic model that predicts leafy spurge impacts on forage production. Data from experiments where herbicides suppressed leafy spurge provided an opportunity to evaluate prediction accuracy of the model. In each case herbicide experiment data fell within the range of values (i.e., credibility intervals) the model predicted, even though the model development experiments were separated from the herbicide experiments by several hundred kilometers in space and 4 years in time. Therefore, we conclude that the model successfully accounts for spatial and temporal variation. We believe light attenuation could help natural resource managers quickly quantify some kinds of weed impacts.
Vegetative differences and changes were evaluated over a 6-year period (1999-2004) on adjoining conservatively grazed and grazing-excluded (22 years) shortgrass rangelands in northwestern New Mexico. Autumn total perennial grass and blue grama (Bouteloua gracilis [Willd. ex Kunth] Lag. Griffiths) standing crop did not differ on grazed and grazing-excluded areas when data were averaged across years. There were no long-term differences in vegetation basal cover or composition between the grazed and grazing-excluded areas. Plant community similarity values between the grazed and grazing-excluded areas were 80% and 93% during the first 2 years (1999-2000) and last 2 years (2003-2004) of study, respectively. Climatic conditions had more impact on vegetation composition of the 2 areas than livestock grazing. Similarity values between 1999-2000 and 2003-2004 periods were 52% and 64% for the grazed and grazing-excluded plant communities, respectively. At the beginning of our study, blue grama productivity was depressed on the grazed area compared to the exclosure, but after 3 years of conservative winter grazing, it was similar on the 2 areas. Our study indicates there is no benefit to blue grama rangelands from long-term rest from the standpoint of vegetation composition.
QuickBird satellite imagery was evaluated for differentiating among rangeland cover types on the Welder Wildlife Refuge in south Texas. The satellite imagery had a spatial resolution of 2.8 m and contained 11-bit data. Four subsets of the satellite image were extracted and used as study sites. Field spectral measurements made among the dominant vegetation types showed significant differences in visible and near-infrared reflectance. Unsupervised classification techniques were used to classify false color composite (green, red, and near-infrared bands) images of each study site. Accuracy assessments performed on the classification maps of the 4 sites had overall accuracies ranging from 79% to 89%. These results indicate that QuickBird imagery can be a useful tool for identifying rangeland cover types at a regional level.
Housing developments are replacing ranches in the southwestern United States, with potentially significant but little-studied ecological effects. We counted grasshoppers (Orthoptera: Acrididae) and measured vegetative cover for 2 years in a grassland and mesquite/oak savanna in southeastern Arizona, on 48 transects that were grazed by livestock, embedded in low-density housing developments, or both, or neither. Grasshopper species richness was unrelated to grazing or development, but grasshopper abundance was much higher on exurban transects where homeowners kept livestock than in the other areas. Forb canopy and basal area also were highest in grazed exurban areas, perhaps because exurban grazing was relatively patchy, frequently involved horses, and created disturbances more conducive to forb establishment than did relatively uniform grazing on nearby ranches. Abundance patterns of 3 grasshopper subfamilies were generally consistent with their known habitat preferences. Counts of grass-feeding Gomphocerinae were relatively high in ungrazed and unburned areas, and positively correlated with grass cover. Numbers of forb- and mixed-feeding Melanoplinae were positively correlated with forb cover across all transects, and melanoplines dominated counts on grazed exurban properties. Band-winged grasshoppers (Oedipodinae) prefer areas of sparse vegetation, and their numbers were negatively correlated with height of ground vegetation and positively associated with the presence of livestock, in both exurban and undeveloped landscapes. Overall, our results suggest that heterogeneous landscapes in exurban areas that included small livestock pastures had higher grasshopper densities than either ungrazed areas or large cattle ranches.
The distribution and abundance of flowering leafy spurge (Euphorbia esula L.) can be determined with hyperspectral remote sensing, but the availability of hyperspectral sensors is limited. Hence, the Landsat 7 Enhanced Thematic Mapper Plus (ETM+) and System Pour d’Observation de la Terre (SPOT) 4 imagery were acquired to test the ability of these sensors to detect leafy spurge. The green:red band ratio was the vegetation index with the highest correlations to flowering leafy spurge cover, but the correlations were weak and not useful for predictions. With Airborne Visible Infrared Imaging Spectrometer (AVIRIS) data, the green:red band ratio was also weakly correlated to flowering leafy spurge cover, although the output from a hyperspectral unmixing algorithm was highly correlated with cover using the same data, indicating simple indices have limited power for detecting leafy spurge. Canopy reflectance modeling using the Scattering by Arbitrarily Inclined Leaves (SAIL) model suggests the weak correlations were caused by variations in leaf area index. It is important to develop spectral libraries in order to use canopy reflectance simulation models that can reduce the time and effort of remote sensing analysis for detecting leafy spurge and other invasive weeds.
Heterogeneous distribution of ungulates about the landscape can be a particularly vexing problem for resource managers. Although livestock preferences for leaves over stems among plants and patches of herbage are well documented, effects of senescent for- age (herbage supporting both green and cured materials) on cattle distribution and nutrition at pasture scales have not been investigated. Our primary objective was to determine the proportions of time cattle spent foraging within senescent and con- ditioned sectors (areas supporting only current season’s herbage) of pastures. Other endeavors included the following: comparing velocities of foraging cattle in conditioned and senescent sectors, determining diet quality of cattle confined to conditioned and senescent treatments, and quantifying levels of forage utilization by cattle in conditioned and senescent treatments. Global positioning system collars were used to track cattle movement and activity in treated crested wheatgrass (Agropyron desertorum [Fisher ex Link] Schultes) pastures. Over a 7-day trial, cattle were found in senescent and conditioned areas 41% and 59% of the time, respectively. When cattle were grazing, 32% of observations were in senescent sectors and 68% were in conditioned areas. With a decline in standing crop in the conditioned treatment, cattle switched to senescent herbage (73% of observations) on day 7. Standing crop was reduced by 13% to 40% in conditioned sectors and increased by 10% in senescent areas. Despite dispa- rities in the crude protein of standing crops in senescent (x= 6.5%) and conditioned (x= 11.3%) treatments, cattle confined to treatment harvested diets of similar quality (x= 13.6% CP) at turn-out. Cattle walked farther when making transitions between treatments, and walked farther each day as the trial progressed. A preference of foraging cattle for portions of pastures grazed (con- ditioned areas) during the previous growing season suggests that utilization patterns established by livestock are self-sustaining.
The red imported fire ant (Solenopsis invicta Buren) is an invasive pest that causes ecological disturbance and economic damage to habitats worldwide. Since its introduction to the United States 75 years ago, the ant has spread across the southeast through Texas into California and causes over 6.5 billion in damages. Conventional control techniques have not proven effective or long-lasting, leading to a search for alternative methods. We examined the ability of WW-B.Dahl Old World bluestem (Bothriochloa bladhii [Retz] S.T. Blake), increasingly used in pastures, to limit or reduce ant infestations. Pastures planted with WW-B.Dahl had about one-third the fire ant mounds found in adjacent pastures of native grass or coastal bermuda (Cynodon dactylon [L.] Pers) grass, and the difference is statistically significant (P = 0.0006). No difference was found in the number of ants collected in bait cups or in mound vitality ratings, suggesting that more than one measure of ant infestation is needed to accurately assess ant populations. A reduction in fire ant mounds can improve the efficiency of haying operations and reduce wildlife impacts, suggesting broad uses for WW-B.Dahl in ant-infested areas.
We compared calibration equations for estimating herbage standing crop (HSC) from comparative yield (CY) rank or stubble height (SH) to determine 1) if CY rank is a better estimator than SH of standing crop, 2) if addition of SH to CY rank will improve the estimation of standing crop, 3) if there is a seasonal effect on CY rank or SH, and 4) if botanical composition influences the prediction of HSC from CY. The results of this study indicate that CY is a slightly better predictor of HSC than is SH. Addition of SH to CY did not improve the prediction of HSC. Models that predict HSC from CY in summer were weaker than models for winter, early spring, and late spring. Thus the CY method can be used with confidence throughout the year. The presence of filaree (Erodium cicutarium L.) in winter and early spring resulted in steeper calibration equations than were present in nonfilaree quadrats.
Juniper (Juniperus spp.) encroachment in grasslands usually progresses toward a stable woody state of mature trees that requires a significant disturbance to shift succession in another direction. Fire alone is often inadequate to shift succession in dense stands of mature juniper and must be preceded by a mechanical treatment such as chaining to reduce juniper competition and increase herbaceous growth that fuels a subsequent fire. However, little long-term data are available that measure combined effects of mechanical and fire treatments on restoration of juniper-dominated grasslands. Here, on a degraded (40% bare ground) north Texas site dominated by redberry juniper (Juniperus pinchotii Sudw), we quantify long-term herbaceous responses to mechanical chaining followed by fire. Two types of chaining, ground-level and elevated, were evaluated and all chained plots were burned 4 years after chaining. Herbaceous and woody responses were measured for several years after both chaining and fire treatments and compared to untreated controls. At study termination, both of the chaining + fire treatments reduced juniper cover from 32% to < 6%, but mortality was < 10%, because most plants basal-sprouted. Total grass production did not increase in chained treatments over the untreated until 3 growing seasons after chaining. Grass production declined the first growing season following the fire treatment, but increased in treated plots to 3 times the untreated the second and third year postfire. Total grass cover in treated plots did not increase over the untreated until the second year after the fire treatment. There was no difference in juniper or herbaceous responses between the 2 chain types. Results suggest increases in herbaceous production from chaining alone were due to increased growth of existing vegetation patches whereas the fire treatment appeared to stimulate herbaceous recruitment into bare soil areas.
Leafy spurge is an exotic, noxious, perennial weed which is widely established in the north central United States and is an especially serious problem in the northern Great Plains. In 1997, the Agricultural Research Service and Animal and Plant Health Inspection Service, US Department of Agriculture, initiated a major Integrated Pest Management (IPM) research and demonstration project, The Ecological Area-wide Management (TEAM) Leafy Spurge (TLS), to develop and demonstrate ecologically based IPM strategies that can produce effective, affordable leafy spurge control. A key component of the TLS project was expanding the use of biological control agents. To assess the level of insect utilization and implementation and the level of current and perceived future control of leafy spurge as a result of biological control agents, a mail survey of 468 individuals that obtained biological control agents (insects) at TLS-sponsored events and of all the county weed boards in North Dakota, South Dakota, Montana, and Wyoming was conducted. Forty-six percent of the landowner/land managers and 70% of the county weed boards responded to the questionnaire. Respondents reported basic information about the number and characteristics of release sites, and characteristics of the leafy spurge stands, as well as the level of control to date and perceived level of eventual control.
Data on the effects of livestock grazing on soil nutrient availability are virtually nonexistent for meadow systems. We measured the effect of livestock grazing on soil, emphasizing soil-solution chemistry, in a Sierra Nevada riparian meadow. Treatments were livestock exclusion (begun in 1989) and grazing to leave 1 000 kg ha-1 of vegetation. Ceramic tension lysimeters were placed in the treatments (2 replicates) by landscape position (stream edge, midfloodplain, and forest edge), and by depth (approximately 0.1, 0.6, and 1.2 m below the soil surface). Lysimeter water was extracted twice monthly in April, May, and June of 1990 through 1993, and cations and anions were quantified. In addition, KCl-extractable NO3 and NH4+– bicarbonate-extractable ortho-P; available Mn, Cu, Fe, and Zn; and root-length density (RLD) were quantified in soils by treatment, landscape position, and soil depth in July 1991 and September 1993. RLD was not affected by grazing. Significant (P 0.05) treatment effects were largely limited to the forest edge. The grazed treatment had greater lysimeter-extractable Na+, Ca+2, Mg+2, and NO3; higher pH; and less K+ and NH4+ than the excluded treatment. Compared with corresponding excluded treatments, bicarbonate-extractable P was significantly greater on the grazed forest edge, and available Mn was significantly greater at the grazed stream-edge position in 1991. Extractable NO3 was significantly higher in the 0-25 cm depth increment of the grazed treatment, and available Zn was significantly greater on the grazed midfloodplain position in 1993. Grazing did not result in more anoxic soil conditions than the excluded treatment. Grazing effects were most pronounced at the forest edge, possibly as a result of spatial transfer of nutrients via cow urine and feces. Management goals to sustain high-elevation meadows should emphasize maintenance of high RLD to sequester soil nutrients.
The Wyoming big sagebrush (Artemisia tridentata ssp. Wyomingensis [Beetle A. Young] S.L. Welsh) alliance is the most extensive of the big sagebrush complex in the Intermountain West. This alliance provides critical habitat for many sagebrush obligate and facultative wildlife species and serves as a forage base for livestock production. There is a lack of information that describes vegetation cover values, characteristics, diversity, and heterogeneity of the Wyoming big sagebrush alliance. This study describes vegetation cover values and defines distinct associations for intact, late-seral Wyoming big sagebrush plant communities across part of its northwestern range. We sampled 107 Wyoming big sagebrush plant communities. Total herbaceous cover values were variable among sites with differences between sites exceeding 700%. Mean sagebrush cover was 12.3% with 90% of the sites producing 6% to 20% cover. Tall forb (.18 cm) cover averaged 1.9% and 90% of the sites varied between 0.2% and 5.6% cover. Five associations delineated by dominant perennial bunchgrass species were identified: ARTRW8 (Wyoming big sagebrush)/PSSP6 (Pseudoroegneria spicata [Pursh] A. Löve, bluebunch wheatgrass), ARTRW8/ACTH7 (Achnatherum thurberianum [Piper] Barkworth, Thurber’s needlegrass), ARTRW8/FEID (Festuca idahoensis Elmer, Idahofescue), ARTRW8/HECO26 (Hesperostipa comata [Trin. Rupr.] Barkworth, needle-and-thread), and ARTRW8/PSSP6-ACTH7 (a codominance of bluebunch wheatgrass and Thurber’s needlegrass). Our results suggest when the vegetation cover values proposed for sage-grouse are applied as requirements at or above the stand level, they exceed the ecological potentialof many of the sites sampled.
Soil aggregate stability (AS) has been promoted as a primary indicator of soil-surface function and a key metric in state-and-transition models. There are few studies, however, that relate indices of AS to the process of grassland degradation. In a Chihuahuan Desert rangeland, we measured variation in AS across vegetated-bare patch boundaries within six plot types reflecting a hypothesized fragmentation/transition sequence. We also examined wetting front depth and pH along this sequence. We found that AS exhibited consistent and interpretable variation across the patch boundaries of the different plot types. Average AS was highest in grass patches adjacent to small to medium-sized (0.5-1.5 m) bare patches and was low in grass patches adjacent to large (> 3 m) bare patches. AS of bare ground was also lowest when bare patches in continuous grassland were large and when bare ground formed an interconnected matrix. Wetting depth after a large storm decreased and pH increased along the fragmentation sequence. The results suggest that AS has interpretable relationships with grassland fragmentation and transitions among states. Careful attention to patchiness within states and stratification, however, is important and simple classifications of strata, such as ‘‘bare interspace’’ and ‘‘plant,’’ may not be sufficient to document variation in soil function.
Around the world rangelands that have been degraded, such as historical desert grasslands now dominated by woody shrubs, are resistant to restoration efforts. The goal of this descriptive research was to examine the potential for black grama(Bouteloua eriopoda [Torr.] Torr.) recovery by remnant plants in a degraded area as a function of plant location across a landscape. Our objectives were 1) to document the historical dominant vegetation as a perennial grassland and determine broad-scale changes in dominance through time and 2) to examine fine-scale patterns of black grama presence and basal area with respect to microenvironmental conditions that indicate the landscape positions favorable for restoration. Historical vegetation maps starting in 1858, a field survey in 2002-2003 of the location of all individual black grama plants in a 29-ha area, and spatial data layers in a geographic information system were used to address these objectives. Upland grasses, including black grama, dominated the study site in 1858, although tarbush (Flourensia cernua DC.) was the dominant species by 1915, and creosotebush (Larrea tridentata [DC.] Cov.) is the current dominant. A total of 3 334 black grama plants were found for an average density of 0.01 plants . m(-2). High spatial variation was found in the occurrence and basal area of black grama plants that was related to water availability rather than livestock grazing: most plants were found in or adjacent to an arroyo (67%), at a northern aspect (47%), and outside experimental exclosures established in 1930 (43%). Largest average basal areas were found in the livestock exclosure, and in general, average basal area was not related with aspect or canopy microsite. These remnant plants can be used as propagule sources in restoration efforts, and information on microsite conditions for black grama survival can be used to improve restoration potential for similar sites.
Drought is an inherent trait of most rangelands and sound management necessitates managers address two fundamental questions when facing a drought situation. The first question is, ‘‘what is the probability that a useful amount of precipitation will be received over the period of concern?’’ and the second question is, ‘‘if it does rain, what will the impact be in terms of quantity and quality of herbage produced?’’ The objective of this study was to address the second question. Our hypothesis was that herbage growth response to above normal summer precipitation (i.e., 23 in July and August) would be limited in the northern Great Plains because of a general absence of productive warm-season species. Study plots were twelve 5 X 10-m non-weighing lysimeters. Treatments were: 1) simulated (i.e., rainout shelter imposed), severe spring drought (i.e., 1 May - 1 July) followed by ambient precipitation thereafter; 2) simulated, severe spring drought followed by ambient precipitation thereafter plus summer irrigation (i.e., July and August); 3) ambient precipitation only; and 4) ambient precipitation plus summer irrigation. Results indicated substantial herbage production can be expected in this region during summer when precipitation is well above average because of the positive growth response of blue grama (Bouteloua gracilis [H.B.K.] Lag. ex Griffiths), the dominant warm-season grass growing in this region. However, results also showed that level of production in the study situation (i.e., spring drought, wet summer) was only about 50% of that attained in a normal (i.e., wet spring/dry summer) year. Moreover, long-term weather data shows the probability of receiving 23 normal precipitation in both July and August (i.e., our irrigation treatments) is < 1%. Thus, although these rangelands possess the capacity to respond favorably to summer precipitation, the low probability of receiving substantial levels of summer precipitation ensures levels of ecological and economic risk remain high.
