• Landscape Factors Influencing the Abundance and Dominance of the Invasive Plant Potentilla recta

      Endress, Bryan A.; Naylor, Bridgett J.; Parks, Catherine; Radosevich, Steven R. (Society for Range Management, 2007-05-01)
      Little is known about the relative importance of environmental, biotic, historical, and spatial factors that influence invasive plant abundance, dominance, and distribution across landscapes. We identified factors that influence the abundance and dominance of Potentilla recta L. (sulfur cinquefoil) in bunchgrass grasslands of northeastern Oregon to better understand the conditions under which it becomes a major component of plant communities. We estimated P. recta stem density and dominance from field measurements across the landscape and used classification and regression tree analyses to assess the importance of environmental, biotic, spatial, and historical factors in explaining P. recta presence, stem density, and dominance. Plots were sampled within a systematic grid with 250-m spacing within our 6.5-km2 study landscape. At each sample point we recorded P. recta presence, stem density, and dominance as well as 11 biological, environmental, spatial, and historical variables. P. recta was widely distributed, with stem densities in occupied plots averaging 5.8 stems N m-2 and dominance values ranging from 1% to 52%. Percent cover of bare ground was the most important variable to predict the presence of P. recta, though the model fit was poor, likely because the entire study area is suitable for P. recta establishment. A strong relationship between P. recta dominance and habitat type (r2 = 67.5%) was found, with dominance greatest in old fields on relatively flat slopes (mean dominance of 34.1%). Dominance estimates were < 1% in plots located in forest, shrub, and grassland habitats. Factors that make old fields susceptible to dominance remain unknown, though microsite conditions that increase P. recta seedling survival rates and limited native propagule availability due to previous cultivation may be involved. Since old fields are found throughout the region, are highly susceptible to P. recta invasion, and represent a source of seeds, containment and restoration activities should focus on these areas. 
    • Influence of Plant Functional Group Removal on Inorganic Soil Nitrogen Concentrations in Native Grasslands

      Davies, Kirk W.; Pokorny, Monica L.; Sheley, Roger L.; James, Jeremy J. (Society for Range Management, 2007-05-01)
      High plant functional group diversity has been hypothesized to reduce resource concentrations based on the assumption that species from one functional group acquire resources similarly to one another, while species from other functional groups acquire resources dissimilarly. To determine if functional groups use soil nutrients different from one another, we investigated the impact of removing individual functional groups on soil inorganic nitrogen (NO-3 and NH+4 ) concentrations in the Idaho fescue (Festuca idahoensis Elmer)/bluebunch wheatgrass (Pseudoroegneria spicata [Pursh] A. Löve) habitat type in Montana. Treatments were imposed by removing 1) all plant species (total plant removal), 2) shallow-rooted (< 15 cm) forbs, 3) deep-rooted (> 15 cm) forbs, 4) all forbs (total forb removal), 5) grasses, and 6) spikemoss, compared to intact control plots. Inorganic nitrogen was measured at 2 soil depths (0-15 cm and 16-40 cm) in the spring, summer, and fall 1 year after treatment imposition. The removal of individual functional groups generally increased soil NO-3 and NH+4 concentrations. Total plant removal increased NO-3 concentrations more than removing individual functional groups. Grass removal generally increased soil NO-3 concentrations in the 0-15-cm depth more than other functional groups removal. Whether the grass or total forb removal treatment resulted in greater soil NH+4 concentrations in the 0-15-cm depth depended on season. These results suggest that functional groups vary in their soil nutrient acquisition patterns and that increased functional diversity decreases soil nutrient concentrations. Therefore, maintaining or improving functional diversity may be a method to more fully utilize soil nutrients because functional groups can differ in their spatial and temporal acquisition of resources. 
    • Grazing-Induced Modifications to Peak Standing Crop in Northern Mixed-Grass Prairie

      Drener, Justin D.; Hart, Richard H. (Society for Range Management, 2007-05-01)
      Selective grazing can modify the productive capacity of rangelands by reducing competitiveness of productive, palatable species and increasing the composition of more grazing-resistant species. A grazing system (season-long and short-duration rotational grazing) 3 stocking rate (light: 16 steers 80 ha-1, moderate: 4 steers 12 ha-1, and heavy: 4 steers 9 ha-1) study was initiated in 1982 on northern mixed-grass prairie. Here, we report on the final 16 years of this study (1991-2006). Spring (April + May + June) precipitation explained at least 54% of the variation in peak standing crop. The percentage of variation explained by spring precipitation was similar between stocking rates with short-duration grazing but decreased with increasing stocking rate for season-long grazing. April precipitation explained the greatest percentage of the variation in peak standing crop for the light stocking rate (45%), May precipitation for the moderate stocking rate (49%), and June precipitation for the heavy stocking rate (34%). Peak standing crop was 23%-29% greater with light (1 495 +/- 66 kg ha-1, mean +/- 1 SE) compared to moderate (1 218 +/- 64 kg ha-1) and heavy (1 156 +/- 56 kg ha-1) stocking rates, which did not differ. Differences in peak standing crop among stocking rates occurred during average and wet but not dry springs. Neither the interaction of grazing system and stocking rate nor grazing system alone affected standing crop across all years or dry, average, or wet springs. Grazing-induced modification of productive capacity in this northern mixed-grass prairie is attributed to changes in species composition with increasing stocking rate as the less productive, warm-season shortgrass blue grama (Bouteloua gracilis [H.B.K.] Lag. ex Griffiths) increases at the expense of more productive, cool-season midheight grasses. Land managers may need to substantially modify management to offset these losses in productive capacity. 
    • Recovery of Big Sagebrush Following Fire in Southwest Montana

      Lesica, Peter; Cooper, Stephen V.; Kudray, Greg (Society for Range Management, 2007-05-01)
      Fire plays a large role in structuring sagebrush ecosystems; however, we have little knowledge of how vegetation changes with time as succession proceeds from immediate postfire to mature stands. We sampled at 38 sites in southwest Montana dominated by 3 subspecies of big sagebrush (Artemisia tridentata Nutt.). At each site we subjectively located 1 sample plot representing the burned area and an unburned macroplot in similar, adjacent, unburned vegetation. Canopy cover of sagebrush was estimated, and plants were counted in 10 microplots. Age and height of randomly chosen sagebrush plants in each size class were determined from 5 microplots. Average postfire time to full recovery of mountain big sagebrush (ssp. vasseyana [Rydb.] Beetle) canopy cover was 32 years, shorter for basin (ssp. tridentata) and much longer for Wyoming (ssp. wyomingensis Beetle Young) big sagebrush. Height recovered at similar rates. There was no difference in canopy cover or height recovery between prescribed fires and wildfires in stands of mountain big sagebrush. We found no relationship between mountain big sagebrush canopy cover recovery and annual precipitation, heat load, or soil texture. Nearly all unburned sagebrush macroplots were uneven-aged, indicating that recruitment was not limited to immediate postfire conditions in any of the subspecies. Average canopy cover of three-tip sagebrush (A. tripartita Rydb.) did not increase following fire, and many three-tip sagebrush plants established from seed instead of sprouting. Our results suggest that the majority of presettlement mountain big sagebrush stands would have been in early to midseral condition in southwest Montana assuming a mean fire interval of 25 years. Only long fire- return intervals will allow stands dominated by Wyoming big sagebrush to remain on the landscape in our study area. We speculate that effects of site-specific factors conducive to sagebrush recovery are small compared to stochastic effects such as fire. 
    • Habitat Effects on Condition of Doe Mule Deer in Arid Mixed Woodland-Grassland

      Bender, Louis C.; Lomas, Laurie A.; Kamienski, Tomas (Society for Range Management, 2007-05-01)
      Productivity of mule deer (Odocoileus hemionus Raf.) populations is closely linked to individual nutritional condition. We modeled body fat of individual does as a function of vegetation cover, composition, and water characteristics of their annual, summer, and winter home ranges in north-central New Mexico. We also modeled home range size as a function of the same characteristics. Levels of body fat were most closely and negatively related to the amount of pinyon-juniper in an individual deer’s annual home range (F1,21 = 7.6; P = 0.012; r2 = 0.26). Pinyon-juniper types provided little (combined ground cover of preferred forbs and shrubs = 5.7%) mule deer forage but were included in home ranges in excess of their availability on the landscape, likely because of security cover attributes. Proportion of grasslands in home ranges was most strongly related to both annual (F1,23 = 4.9; P = 0.037; r2 = 0.18) and summer (F2,25 = 5.7; P = 0.009; r2 = 0.31) home range sizes, and home ranges increased as the grassland component increased, indicating that this habitat type was providing little value to mule deer. Grassland (0.2% combined cover of preferred forb and shrub) and montane conifer (3.2% ground cover of preferred forb and shrub) habitat types similarly lacked preferred mule deer food, and grasslands also lacked cover. Most immediate gains in mule deer habitat in north-central New Mexico may be attained by management of pinyon-juniper communities to increase forage quantity and quality while maintaining cover attributes. Gains can also be realized in grasslands, but here management must establish both cover and forage. 
    • Standardized Ecological Classification for Mesoscale Mapping in the Southwestern United States

      Comer, Patrick J.; Schulz, Keith A. (Society for Range Management, 2007-05-01)
      Standardized ecological classification units form the foundation for effective data collection, assessment, and reporting on ecosystems. Attempts at regional land cover mapping often falter on this point or struggle along inefficiently. Over the past decade, NatureServe has worked with the Gap Analysis Program and others to map existing vegetation using the US National Vegetation Classification (US-NVC). US-NVC is a system of hierarchical structure and rules that are designed to provide a national classification of existing vegetation. Experience has demonstrated the need to develop map units at conceptual scales intermediate between the narrowly specific alliance (floristic) and the broadly generalized formation (physiognomic) levels of the US-NVC. NatureServe defined over 630 ‘‘mesoscale’’ vegetation-based units that are described across the lower 48 United States. These mesoscale classification units, which we term ‘‘terrestrial ecological systems,’’ are described using multiple plant communities that tend to co-occur based on recurrent similarities in environmental setting and ecological dynamics. By integrating environmental setting and ecological processes with vegetation into the concept of each unit, this classification system lends itself to biophysical modeling and robust characterization of wildlife habitat. These units apply well to land cover mapping and may be augmented with modifiers for specific variants in composition and structure resulting in robust, standardized maps. Regional-scale mapping of ‘‘near-natural’’ land cover was completed by the Southwest Regional Gap Analysis Project using 109 ecological system units, currently the most detailed regional land cover map of its kind. Terrestrial ecological system units provide a direct, systematic link to the US National Vegetation Classification and may also provide a useful framework for integration with ecological site concepts and descriptions. 
    • A Method for Landscape-Scale Vegetation Assessment: Application to Great Basin Rangeland Ecosystems

      Forbis, Tara A.; Provencher, Louis; Turner, Lee; Medlyn, Gary; Thompson, Julie; Jones, Gina (Society for Range Management, 2007-05-01)
      The growth of landscape-scale land management necessitates the development of methods for large-scale vegetation assessment. Field data collection and analysis methods used to assess ecological condition for the 47 165-h North Spring Valley watershed are presented. Vegetation cover data were collected in a stratified random design within 6 Great Basin vegetation types, and the probability of detecting change in native herbaceous cover was calculated using power analyses. Methods for using these quantitative assessment data are presented to calculate a departure index based on reference condition information from LANDFIRE (an interagency effort to map and model fire regimes and other biophysical characteristics at a mid-scale for the entire United States) Biophysical Setting models for the mountain big sagebrush (Artemisia tridentata Nutt. subsp. vaseyana [Rydb.] Beetle) vegetation type. For mountain big sagebrush in the North Spring Valley landscape, we found that the earliest successional classes were underrepresented and that mountain big sagebrush moderately invaded by conifers was more abundant than predicted by the LANDFIRE reference based on the historic range of variability. Classes that were most similar to the reference were mountain big sagebrush with the highest conifer cover and late development mountain big sagebrush with perennial grasses. Overall, results suggested that restoration or approximation of the historic fire regime is needed. This method provides a cost-effective procedure to assess important indicators, including native herbaceous cover, extent of woody encroachment, and ground cover. However, the method lacks the spatial information that would allow managers to comprehensively assess spatial patterns of vegetation condition across the mosaics that occur within each major vegetation type. The development of a method that integrates field measurements of key indicators with remotely sensed data is the next critical need for landscape-scale assessment. 
    • Prickly Pear Cactus Responses to Summer and Winter Fires

      Ansley, R. James; Castellano, Michael J. (Society for Range Management, 2007-05-01)
      Prescribed fire is used to reduce size and density of prickly pear cactus (Opuntia spp.) in many rangeland ecosystems. However, effects of dormant season fires (i.e., winter fires) are inconsistent. Thus, there is increasing interest in use of growing season (summer) fires. Our objective was to evaluate effects of fire season and fire intensity on mortality and individual plant (i.e., ‘‘motte’’) structure (area per motte, cladodes per motte, motte height) of brownspine prickly pear (O. phaeacantha Engelm.). The study had 4 treatments: no fire, low-intensity winter fire, high-intensity winter fire, and summer fire. Three sizes of prickly pear mottes were evaluated: small (0-20 cladodes per motte), medium (21-100), and large (101-500). At 3 years postfire, prickly pear mortality in the summer fire treatment was 100% in small mottes, 90% in medium mottes, and 80% in large mottes. Motte mortality increased in this treatment over time, especially in large mottes. Mortality from high-intensity winter fires was 29% and 19% in small and medium mottes, respectively, but no large mottes were killed. Motte mortality was < 10% in low-intensity winter fire and no-fire treatments. Summer fires reduced all motte structural variables to 0 in small mottes and nearly 0 in other motte size classes. High-intensity winter fires reduced some structural variables of medium and large mottes, but had no long-term negative effects on area per motte or cladodes per motte in surviving small mottes. Low-intensity winter fires had no long-term negative effects on motte structure in any size class. Rapid growth of mottes, and especially small mottes, in the no-fire treatment suggested that resistance to winter fires can occur rapidly. 
    • Cheatgrass Invasion in Salt Desert Shrublands: Benefits of Postfire Reclamation

      Jessoff, Brad D,; Anderson, Val Jo (Society for Range Management, 2007-05-01)
      In 1998, fires burned more than 11 330 ha of rangeland on Dugway Proving Ground in Utah’s west desert. Postfire revegetation was implemented in 2 affected salt desert shrub communities (greasewood; Sarcobatus vermiculatus Hook. and black sagebrush/ shadscale; Artemisia nova A. Nels; Atriplex confertifolia Torr. & Frem.) to deter cheatgrass (Bromus tectorum L.) encroachment. We monitored cheatgrass densities for 3 years after the fire in burned drill seeded, burned not-seeded, and unburned plots to assess the rate of invasion and determine the impact on cheatgrass of drill seeding perennial species. Cheatgrass invaded quickly in both shrub sites following the fires. In the greasewood site, drill seeded species germinated but did not establish. This was likely due to a combination of soil salinity and extremely dry weather conditions during the second year of the study. Drill seeded species in the black sagebrush site germinated and established well, resulting in the establishment of 16.5 perennial grasses m-2 and 1 356 shrubs ha-1. Cheatgrass densities were consistently lower in drill seeded versus not- seeded plots, although these were not always statistically different when Bonferroni comparisons were considered. The initial decrease in cheatgrass densities in drill seeded plots may have resulted from soil disturbance coupled with extremely low precipitation rather than competitive effects. Nevertheless, as seeded species mature and increase their competitive ability, we predict long-term suppression of cheatgrass in the absence of further disturbance. 
    • Key Attributes Influence the Performance of Local Weed Management Programs in the Southwest United States

      Hershdorfer, Mary E.; Fernandez-Giminez, Maria; Howery, Larry D. (Society for Range Management, 2007-05-01)
      In the southwestern United States, local weed management programs are increasingly important in weed prevention and control; however, little is known about the effectiveness of different local approaches to weed management. We surveyed coordinators of 53 local weed management programs in Arizona, Colorado, New Mexico, and Utah to determine how 4 key program attributes (interagency coordination, volunteer participation, regulatory authority and enforcement, and the state in which the program was located) were related with 4 performance measures: weed control, public education and outreach, weed monitoring, and integrated weed management. Based on the responses of 42 program coordinators (79%) we found that 1) weed programs that coordinated their activities with other organizations and those with citizen volunteers conducted more monitoring, but programs that did not coordinate or use volunteers treated more of their infested acreage; 2) programs that used a light-handed regulatory approach conducted more weed control than those with more punitive enforcement regimes or no enforcement authority; and 3) Colorado programs conducted more outreach and education than did programs in the other 3 states. Thus, although volunteer involvement and interagency coordination contributed to the performance of the local weed programs studied, particularly in monitoring, they have not compensated for the lack of locally enforceable weed regulations or adequate funding. Successful weed management in southwestern United States will require adequately funded, locally adapted approaches supported by locally enforceable weed regulations. 
    • Is Altering Grazing Selectivity of Invasive Forage Species With Patch Burning More Effective Than Herbicide Treatments?

      Cummings, D. Chad; Fuhlendorf, Samuel D.; Engle, David M. (Society for Range Management, 2007-05-01)
      Invasion of rangeland by exotic forage species threatens ecosystem structure and function and can cause catastrophic economic losses. Herbicide treatments often are the focus of management efforts to control invasions. Management with the fire-grazing interaction (or patch burning) might suppress an invasive forage species that has grazing persistence mechanisms developed apart from the fire-grazing interaction. We studied tallgrass prairies invaded by sericea lespedeza (Lespedeza cuneata [Dum.- Cours.] G. Don) to compare rate of invasion between traditional management and management with patch burning, to evaluate the effect of burn season on sericea lespedeza invasion within pastures managed with patch burning, and to correlate canopy cover of sericea lespedeza to canopy cover of other functional groups with and without herbicides. Sericea lespedeza canopy cover increased from 1999 to 2005 in both traditional- and patch-burn pastures, but sericea lespedeza increased from 5% to 16% canopy cover in traditionally managed pastures compared to 3% to 5% in the patch-burn pastures. Rate of increase in canopy cover of sericea lespedeza was less in patches burned in summer (0.41% year-1) than in patches burned in spring (0.58% year-1) within patch-burn pastures. Most plant functional groups, including forbs, were weak-negatively correlated with canopy cover of sericea lespedeza. Although herbicide application reduced mass of sericea lespedeza, other components of the vegetation changed little. Herbicide treatments temporarily reduced sericea lespedeza but did not predictably increase other plant functional groups. Patch burning reduced the rate of invasion by sericea lespedeza by maintaining young, palatable sericea plants in the burn patch, and could play a vital role in an integrated weed management strategy on rangelands. 
    • Runoff and Erosion After Cutting Western Juniper

      Pierson, Fredrick B.; Bates, Jon D.; Svejcar, Tony C.; Hardegree, Stuart P. (Society for Range Management, 2007-05-01)
      Western juniper (Juniperus occidentalis spp. occidentalis Hook.) has encroached on and now dominates millions of acres of sagebrush/bunchgrass rangeland in the Great Basin and interior Pacific Northwest. On many sites western juniper has significantly increased exposure of the soil surface by reducing density of understory species and surface litter. We used rainfall and rill simulation techniques to evaluate infiltration, runoff, and erosion on cut and uncut field treatments 10 years after juniper removal. Juniper-dominated hillslopes had significantly lower surface soil cover of herbaceous plants and litter and produced rapid runoff from low-intensity rainfall events of the type that would be expected to occur every 2 years. Direct exposure of the soil to rainfall impacts resulted in high levels of sheet erosion (295 kg ha-1) in juniper-dominated plots. Large interconnected patches of bare ground concentrated runoff into rills with much higher flow velocity and erosive force resulting in rill erosion rates that were over 15 times higher on juniper-dominated plots. Cutting juniper stimulated herbaceous plant recovery, improved infiltration capacity, and protected the soil surface from even large thunderstorms. Juniper-free plots could only be induced to produce runoff from high-intensity events that would be expected to occur once every 50 years. Runoff events from these higher-intensity simulations produced negligible levels of both sheet and rill erosion. While specific inferences drawn from the current study are limited to juniper-affected sites in the Intermountain sagebrush steppe, the scope of ecosystem impacts are consistent with woody-plant invasion in other ecosystems around the world. 
    • Wildlife Responses to Vegetation Height Management in Cool-Season Grasslands

      Washurn, Brian E.; Seamans, Thomas W. (Society for Range Management, 2007-05-01)
      Herbaceous vegetation comprises the main habitat type in cool-seasons grasslands and can be managed by various methods. We compared changes in plant communities and bird and mammal use of grasslands that were not managed, managed by mechanical methods (mowing), or managed by chemical methods (plant growth regulator). This 1-year study was conducted from May through October 2003 in Erie County, Ohio. Twelve circular 1.5 ha plots were established: 4 were not managed, 4 were mowed to maintain vegetation height between 9-15 cm, and 4 were sprayed with a plant growth regulator and mowed when vegetation exceeded 15 cm. We monitored vegetation growth, measured plant community composition, and observed all plots for wildlife activity each week. Vegetation in unmanaged plots was taller and denser (P < 0.001) than vegetation in mowed and growth regulator plots. Plant community characteristics differed among study plots (P < 0.001); managed plots had higher grass cover and lower woody cover than unmanaged plots. We observed more (P < 0.001) total birds per 5-minute survey in unmanaged than mowed or growth regulator plots. We observed more (P < 0.001) white-tailed deer (Odocoileus virginianus) in mowed plots than either control or growth regulator plots. We captured 13 small mammals in unmanaged plots and no small mammals in managed plots. Applying the plant growth regulator was not a cost-effective alternative to mowing for managing vegetation height in our study. Vegetation height management practices altered plant communities and animal use of grassland areas and thus might be useful for accomplishing species-specific habitat management objectives. 
    • Cattle Grazing Effects on Macroinvertebrates in an Oregon Mountain Stream

      Mclver, James D.; Mclnnis, MIchael L. (Society for Range Management, 2007-05-01)
      Cattle grazing effects on aquatic macroinvertebrates were assessed in a 4-year experiment of a mountain stream in northeastern Oregon. From 1996 through 1999, 10 cow-calf pairs were introduced into 6 experimental units along the stream for 42 days between July and September, and effects on aquatic macroinvertebrates were compared with 3 units in which no grazing occurred. Streambank and geomorphological variables were also measured to provide context for interpretation of effects on aquatic macroinvertebrates. Macroinvertebrate response to grazing was subtle, indicated by significantly lower abundance in grazed units. We measured more profound effects on streambanks: grazing caused an average decrease of 18% in bank length of the highest stability/cover class and caused an average increase of 8% in the lowest condition class over the course of each summer. By June of each following year, banks had recovered to their previous June condition, but grazing each summer caused a progressively larger decline in bank condition by September. Streambank effects were accompanied by an increase in cobble embeddedness over time in grazed units and were correlated with grazing-associated stream widening. Treatment effects were overwhelmed, however, by a profound decline in the abundance of most macroinvertebrates over the study period, with a drop in September 1999 to 14% of the initial September abundance of 1997. While the drop was more precipitous in grazed units, declines were common to all study units, suggesting that something more widespread affected the system during this time. Logging on lands just upstream of the study area in 1998 and 1999, in which trucks drove through the study stream without the benefit of a culvert, sent sediment plumes into the study area each of those 2 years and could have caused the precipitous decline in aquatic macroinvertebrates. 
    • Effects of Forage Management on Pasture Productivity and Phosphorus Content

      Haan, M. M.; Russell, J. R.; Kovar, J. L.; Powers, W. J.; Benning, J. L. (Society for Range Management, 2007-05-01)
      The objectives of the current study were to determine the amounts of above- and below-ground plant biomass production, P uptake by forage, and P concentration of cool-season grass forage as influenced by management and season. Five forage management treatments were evaluated over 3 years in smooth bromegrass (Bromus inermis Leyss) pastures. Management practices were: ungrazed (U), hay harvest/fall stockpile grazing (HS), rotational stocking to residual sward heights of 10 (10R) or 5 (5R) cm, and continuous stocking to maintain sward height at 5 cm (5C). Forage samples were hand-clipped within and outside grazing exclosures monthly from April through November of each year and analyzed for mass and P concentration. Root samples were collected at the initiation and completion of the study for determination of root length density (RLD) and root surface area density (RSAD). Phosphorus concentrations of forage outside the grazing exclosures did not differ among 5C, 5R, and 10R treatments, which were greater than U paddocks in April and August and less than HS paddocks in June. Mean annual forage productivity was greater in HS, 10R, 5R, and 5C paddocks (6 744 +/- 62 kg ha-1 mean +/- SE) than in the U paddocks (1 872 +/- 255 kg ha-1). Mean P concentration of forage outside exclosures was greatest during the spring (0.21 +/- 0.01%), and lowest during the fall (0.13 +/- 0.01%). Mean annual P uptake by forage followed the same trend as forage production, being greater in the HS, 10R, 5R, and 5C paddocks (13.9 +/- 2.0 kg ha-1) than in the U paddocks (3.7 +/- 0.5 kg ha-1). After 3 years, RLD decreased in the ungrazed paddocks, but was unchanged in the HS, 10R, 5R, and 5C paddocks. Forage production and P uptake by forage is stimulated by forage harvest, either by grazing or hay harvest in smooth bromegrass pastures.