stubble height and utilization measurements · of grazing that might be allowed in order to...

Stubble Height andUtilization Measurements:Uses and Misuses

Agricultural Experiment StationOregon State University

A Western Regional Research Publication

Pc i4qçj

Station Bulletin 682 5_o e(7May 1998

For additional copies of this publication, write

Publications OrdersExtension and Station CommunicationsOregon State University422 Kerr AdministrationCorvallis, OR 97331-2119

A Western Regional Research PublicationParticipating Western Region State Agricultural Experiment Stations

University of AlaskaAmerican Samoa Community CollegeUniversity of ArizonaUniversity of CaliforniaColorado State UniversityUniversity of GuamUniversity of HawaiiUniversity of IdahoCollege of Micronesia

Montana State UniversityUniversity of NevadaNew Mexico State UniversityCommunity College of Northern Mañana IslandsOregon State UniversityUtah State UniversityWashington State UniversityUniversity of Wyoming

A Western Regional Research PublicationUnder the procedure of cooperative publication, this regional report becomes, in effect,an identical publication of each of the participating experiment stations and agenciesand is mailed under the indicia of each.

Agricultural Experiment StationOregon State UniversityStation Bulletin 682May 1988

Stubble Height andUtilization Measurements:Uses and Misuses

Contents

Introduction 1

Utilization Standards: The Quandary Revisited 3

Seasonal Effects on the Measurement and Interpretation of Utilization 9

Variation in Utilization Estimates Caused by Differences among Methods, Years, and Observers 17

Interpretation of Utilization and Long-term Frequency Measurements for Rangeland Management 25

Stubble Height and Function of Riparian Communities 29

Developing and Achieving Management Objectives on National Forest System Lands 47

Impact of Federal Grazing Reductions on Wyoming Ranches 50

Comparison of Economic Impacts from Public-land-based Tourism and Grazing: A Case Study 57

Integrating Utilization Measurements into Monitoring Programs 71

This bulletin contains the papers from a symposiumin February 1997 at the 50th annual meeting of theSociety for Range Management, in Rapid City, SD.The symposium provided a venue for discussingthe pros and cons of using utilization estimates asthe primary source of information for managinggrazing lands, particularly indigenous rangelands.The impetus was the belief of the members of thetwo sponsoring Western Coordinating Committees,WCC-40 (Rangeland Ecological Research andAssessment) and WCC-55 (Rangeland ResourceEconomics and Policy), that utilization estimatesoften are used incorrectly in making rangelandmanagement decisions. Although the committeesbelieve that utilization estimates can serve asimportant information on which rangeland manage-ment strategies and tactics can be based, we believesuch estimates often are either inaccurate or easilymisconstrued and misused.

The symposium examined both the technical andsocial aspects of utilization estimates as they relateto how such estimates should be made, interpreted,and used. The potential impact that utilizationbased management strategies can have on live-stock-dominated grazing economies was alsoexplored. To accomplish this, the two committeesinvited a group of distinguished rangeland scien-tists and agricultural economists to present papers.After the symposium, the authors prepared writtenpapers based on their presentations. The paperswere then peer-reviewed and revised under thedirection of the WCC-40 administrative advisor,Jim Jacobs.

Ken Sanders helps set the stage by reviewing thehistorical use of utilization estimates as a rangelandmanagement tool. He reminds us that the currentemphasis on using utilization estimates as theprimary tool for making grazing management

IntroductionRod Heitschmidt, Chair, WCC-40

USDA Agricultural Research ServiceFt. Keogh Livestock and Range Research Laboratory

Miles City, MT 59301

decisions is not new; history does repeat itself.Lamar Smith's paper examines how, when, andwhere utilization estimates should be made, withemphasis on the inherent risks of improperly usingthe tool. Bill Laycock focuses attention on thepossible errors due to methods, observer difference,and time. These can lead to inappropriate use orinterpretation of utilization estimates. He providesevidence of the risk associated with use of utiliza-tion as the primary variable for making manage-ment decisions.

Allen Rasmussen's paper provides a detailedanalysis of the relationship between utilization andrangeland trend data. He shows that utilizationestimates are not strongly correlated with ecologi-cal trend data which, in turn, emphasizes the needto include a wide array of ecological response datawhen developing rangeland management strategies.Quentin Skinner outlines the relationship betweenstubble height and function of riparian communi-ties. He also reviews some of the fundamentalrelationships between vegetation stubble height andstream-channel dynamics, erosion and sedimentdeposition, plant growth dynamics, and ecologicalsuccession. He provides examples of the ecologicalrisk associated with inappropriate utilization orstubble height standards.

The potential economic impacts of changes ingrazing AUMs from federal land are examined inthe papers by Larry Van Tassel and Bob Fletcher.Van Tassel examines the impact on individualenterprises; Fletcher focuses on regional impacts.Both demonstrate the devastating effect that erro-neous and inappropriate use of utilization estimatescan have on the economic well-being of rangelandagriculturists (i.e., ranchers) and associated busi-ness support enterprises. Fred Hall's paper outlinesmanagement steps to enable the USDA Forest

Stubble Height and Utilization Measurements 1

Service to more effectively set and achieve range-land resource management objectives. He empha-sizes the need to monitor management tactics andassociated response variables using quantitative,sensitive, and repeatable methods. Utilizationestimates are but one of many such measures.

The final paper, by Bill Krueger, summarizes andemphasizes that utilization estimates should be onepart of a many-faceted monitoring program rather

than an objective in themselves. Utilization is a toolto achieve management goals and should never bea management goal per Se. Krueger advocatesadopting management strategies that are bothecologically and economically sound and that lendthemselves to quality monitoring programs de-signed to provide managers with accurate informa-tion to assure that goals and objectives are continu-ally met.

2 Stubble Height and Utilization Measurements

Utilization Standards: The Quandary Revisited

AbstractIt has often been said that history repeats itself.That is certainly the case with the current quandaryover the use of utilization standards by many landmanagement agencies. In the early part of thiscentury, range managers encouraged the use ofrotational grazing to maintain a satisfactory foragecrop, with little concern for degree of use. From1926 through the 1940s, considerable emphasis wasplaced on formulating utilization standards andproper use.

At the same time the concepts of range conditionand trend were being developed. In the early 1950s,scientists began questioning the emphasis onutilization and urged the agencies to monitor rangetrend instead. In the past 10 years, with the height-ened concern over riparian areas, some agencieshave again returned to utilization as their primaryand, many times, their onlymonitoring tool.Once again the use of utilization standards is beingquestioned. Utilization data, in conjunction withrange trend data and information on weather, otheruses, and past management actions, can help landmanagers interpret the cause of range trend. Bututilization data alone do not provide adequateinformation to determine whether managementactions are meeting management objectives.

Historical PerspectiveHistory often repeats itself, especially if we do notoccasionally review it. The current debate on the

This paper is issued as Contribution Number 830 of theIdaho Forest, Wildlife and Range Experiment Station,College of Forestry, Wildlife and Range Sciences, Universityof Idaho, Moscow, ID 83844-1135. It is dedicated to thememory of Dr. Lee A. Sharp, who long advocated usinglong-term trends rather than utilization estimates to monitorrangelands.

Stubble Height and Utilization Measurements

Kenneth D. SandersDepartment of Range Resources

University of IdahoTwin Falls, ID 83303-1827

meaning and use of utilization measurements is acase in point. In an article titled "The Quandary ofUtilization and Preference," Cook and Stoddart(1953) questioned the emphasis land managementagencies were placing on utilization estimates.Current events suggest it is time to revisit thisquandary, hence the title of this article.

Through the first couple of decades of this century,rotational grazing systems were advocated formanaging western rangelands (Smith 1895,Sampson 1913). Jardine and Anderson (1919) alsoadvocated deferred grazing for both cattle andsheep using national forests.

A later report by Sampson and Malmsten (1926)stressed the importance of intensity and frequencyof grazing that might be allowed in order to main-tain or improve plant cover and forage production.This report was interpreted as conflicting withSampson's earlier support of specialized grazingsystems. It led to the U.S. Forest Service's placingmanagement emphasis on grazing intensity, ratherthan on systems, for the next two decades.

Campbell (1937) stated, "When continued produc-tivity or gradual death of a good forage grass maydepend upon a difference in foliage removal of aslittle as 10 percent, a more accurate measurementof utilization is necessary." He reported that theU.S. Forest Service was initiating a major researcheffort on utilization standards and proper use incooperation with many state experiment stations.Proper use became the standard with which currentutilization was compared. Proper-use guidelines forindividual species were prepared by interagencycommittees for particular areas or regions, byseason of use and by kind of livestock. Theseguidelines were based on experienced judgmentand research available at that time and were arrivedat by discussion and compromise. Unfortunately,

over the years these estimates of proper use becamesanctified as absolute numbers.

Campbell (1943) recognized the complexity ofattempting to identify proper use standards. Hestated that proper use of a species depended on"several stages of plant succession, considerabledifferences between species as to the relish withwhich they are eaten by livestock at differentseasons, resistance to grazing, and processes ofgrowth, maintenance and reproduction." But inspite of this complexity, Campbell went on to saythe strategy of the cooperative studies was toformulate the results of previous and contemporarystudies "...into simple, readily applicable facts foruse by busy range administrators and managers."Thus began the syndrome of trying to oversimplifya complex subject, a syndrome that continues today(Sharp et al. 1994).

Regardless of whether one agrees with the empha-sis that scientists and land managers placed onutilization and proper use in the 1930s and 1940s,there is no question that the research effort wasvery fruitful. It provided a stimulus for rangeresearch throughout the West by the U.S. ForestService, land-grant universities, and other agencies(Division of Range Research 1944). Along withpassage of the Taylor Grazing Act and formation ofthe Soil Conservation Service in the early 1930s,this stimulus for range research and the agencies'needs for range-trained personnel had much to dowith the start-up of academic programs in rangemanagement at the various land-grant universities.Also in this period, the concepts of range conditionand range trend were being developed. The rangeresearch literature of the period is a virtual who'swho of the founders and early leaders of the Soci-ety for Range Management.

By the l950s, range scientists began to question theemphasis that management agencies were placingon utilization and proper use. While pointing outthe problems in estimating grazing capacity,Stoddart (1952) declared that "nothing but ecologi-cal knowledge plus range-managing experiencewill suffice to determine a standard utilization. Noaccurate method of grazing capacity determinationhas yet been devised which does not rely uponexperience founded upon comparable range of

proved grazing capacity." Cook and Stoddart(1953) added, "...if management is based upon theecological principles considered in range conditionand range trend analyses, it is not necessary for therancher or land administrator to make precisedeterminations of percent utilization for individualforage species."

Although Hedrick (1958) supported the importanceof proper use, he thoroughly reviewed the problemsin determining what proper use is. One of his moreinteresting points is that grazing is generally not asdamaging to the physiology of plants as clipping;however, most utilization standards are based onclipping studies. Blaisdell (1966) indicated thatpreoccupation with exact measurement of herbageutilization seemed to have retarded progress ingrazing research and management. But even asthese questions were being raised, Sharp (1971)pointed out that "rules of thumb" and simpleguides, such as utilization standards, were stillbeing used as a substitute for management guidedby ecological monitoring.

Despite all of the early writings on the benefits ofrotation grazing, until the 1960s most grazing onpublic lands consisted of season-long use (Sharp1971). Both the Forest Service and later the Graz-ing Service/Bureau of Land Management empha-sized inventorying the forage resource in order tobalance animal numbers with resource capacities.The need for the inventory, together with limitedmanpower and funds for range improvements, wereprobably the principal factors in slowing imple-mentation of grazing systems.

After Hormay and Talbot (1961) published theirreport, "Rest-Rotation GrazingA New Manage-ment System for Perennial Bunchgrass Ranges,"Gus Hormay started conducting schools on rest-rotation grazing, winning many converts as hequestioned the need to be concerned about degreeof utilization. Hormay (1970) questioned theproper-use "philosophy" and characterized asunrealistic the assumption that plants can be grazedto a proper level by regulating stocking. In addi-tion, the existing proper use standards were predi-cated on the premise that the foliage would beremoved annually at some given level, which madethese standards inappropriate for rest-rotation


grazing systems. Interest in developing and imple-menting grazing systems began to grow.

In my opinion, progress in grazing managementand grazing systems has been impeded by legisla-tion that places tremendous time demands onfederal and state range managers. Beginning withthe National Environmental Policy Act in 1969 andcontinuing today, range conservationists mustspend too much of their time at their desks comply-ing with a myriad of conflicting environmentalregulations, rather than out on the ground applyingthe art and science of range management. A resultis that land management agencies again place moreemphasis on utilization standards than on grazingmanagement (Sharp et al. 1994, Burkhardt 1997,McKinney 1997).

In a 1993 analysis of utilization monitoring inNevada and elsewhere, Resource Concepts, Inc.(unpublished report) reported that the large major-ity of allotment evaluations and decisions they hadreviewed relied exclusively on short-term monitor-ing data (i.e., actual use and utilization levels) asthe sole determinant and justification for long-termlivestock management decisions to adjust stockingrates. I, as well as others, have had similar experi-ences. Following a meeting of western range con-sultants in Jackson, WY in 1993 to discuss theproblem, Lee Sharp was encouraged to dust off apaper he had presented in 1971, update it, and pub-lish it in Ran gelands. The resulting article (Sharp etal. 1994) is the basis of much of this paper.

Utilization as Management Tool,Not as an ObjectiveWhen used as just one of many management tools,utilization data can provide useful information.Utilization mapping is a very useful tool to assesslivestock distribution. I feel strongly that todaythere are very few U.S. Forest Service or Bureau ofLand Management allotments where stocking rateis a problem. But there are many allotments wherelivestock distribution, season of use, and/or aninappropriate grazing system are a problem. I alsothink it is safe to say that most of our ripariangrazing problems are related to poor distribution inthe pasture. So, by all means, agency personnel

should do utilization mapping. But it should bedone with the permittee(s) and simply mapped asno use, light, moderate, heavy, or very heavy use.Do not pretend greater accuracy by expressing useas a number or percentage. If there is a distributionproblem, then figure out how to solve the problem,and remember that a reduction in livestock num-bers is not likely to solve a distribution problem.

Utilization data, in conjunction with good rangetrend data and other information on weather,insects, wildlife use, and past management actionscan help range managers interpret the cause ofrange trend. But utilization data alone do notprovide adequate information to determine whethermanagement actions are meeting managementobjectives. As pointed out by Sharp et al. (1994),time spent estimating utilization could be betterspent taking photographs of the range at varioustimes of the year. Photos will not only indicateutilization but also range trend.

With increased emphasis on riparian managementand monitoring, utilization standards are once againreceiving increased attention. And once again,some resource managers are using a managementtool as a management objective.

Allowable use levels make poor and inconclusiveallotment and riparian objectives because theyprovide no information by themselves on whetherdesired long-term conditions are being met. Surelyeveryone can agree that a given level of utilizationon riparian areas is not the real objective but rathera tool to help achieve an objective such as improvedplant vigor, more stable streambanks, or more desir-able plant species composition. Once the objectiveis identified, a monitoring plan can be developed togauge whether the objective is being achieved.

Although I believe that the emphasis on monitoringriparian areas should be on long-term trend, I thinkstubble height is a more useful measure thanpercent utilization. However, some permittees aremore nervous about stubble height than percentutilization standards because permittees fear that"cow cops" will be more likely to monitor stubbleheight than utilization. When used properly,stubble height can be a helpful tool for managinglivestock use (Hall and Bryant 1995).


Concerns aboutUtilization StandardsMuch early research on utilization was to developan accurate method of measurement. But how cananyone "measure" what is not there? All methodsof determining utilization are estimates, some moreaccurate than others. All methods also are time-consuming;, thus, it is very difficult to adequatelysample the heterogeneous mix of species and rangesites found in most pastures or allotments. Shouldimportant management decisions, which may affectthe livelihood of one or more families, be made oninformation so limited and of such questionableaccuracy as utilization data?

Another concern about the accuracy and use ofutilization data is that often the personnel using themethods are inadequately trained. One of the morecommon methods, ocular estimate by plot, requiresintensive clipping and weighing during the trainingperiod and then periodic clipping and weighing inestimated plots to provide a correction factor. It isdoubtful that most field personnel using this methodconduct the time-consuming training and correc-tions necessary to accurately estimate utilization.

Most utilization estimates are based either on peakstanding crop or on current-year production to date,the latter of which results in overestimating utiliza-tion. The Society for Range Management (1989)defines utilization as the proportion of the currentyear's forage production that is consumed or de-stroyed by grazing animals. It may refer to a singleplant species or to the vegetation as a whole.

In an excellent discussion of utilization, Frost et al.(1994) pointed out that a strict interpretation of thisdefinition means that the current annual above-ground net primary production must be known,which it seldom is. They also pointed out that mostutilization studies use peak standing crop as anestimate of current-year production, which isalways less than total production. This results in abuilt-in bias for overestimating utilization. Inreality, many utilization estimates are based oncurrent year's production to date, which usuallyresults in an even greater overestimation.

Sharp et al. (1994) compared three studies toillustrate how utilization may be overestimated.

Utilization of crested wheatgrass in studies in Utahand New Mexico were calculated on the basis ofcaged plants, whereas use in an Idaho study wascalculated on the basis of total annual growth.Fifty-percent use in the Idaho study probably meant15 to 20% more herbage removed than at the sameindicated use level in the two other studies. Thus,when each investigator recommended 65% use oncrested wheatgrass, one actually was recommend-ing a substantially higher grazing intensity than theothers. Frost et al. (1994) cited the example of astudy in Arizona where De Muth (1990) clippedsideoats grama to simulate moderate and heavygrazing. Relative utilization (i.e., utilization ofcurrent year's growth to date) in April was 17% inthe moderate and 49% in the heavy intensity.Actual utilization (i.e., utilization in relation topeak standing crop) was 6 and 17%, nearly one-third less than the relative utilization estimates.Measurements of relative utilization should not becompared with proper use standards derived frommeasurements of actual utilization.

In a 1993 review of utilization monitoring, Re-source Concepts, Inc. (unpublished report) listedseveral other problems with how utilization moni-toring was being conducted and analyzd. In manyinstances, riparian or other areas of animal concen-tration were used as key areas for monitoringutilization on the allotment. By definition, areas ofanimal concentration or otherwise sensitive re-sources are termed critical management areas, notkey areas. Monitoring critical areas may be appro-priate to meet specific management objectives, buta critical area should not be used as a key area todetermine grazing effects across the entire pastureor allotment. Resource Concepts also questionedthe use of visual aids, such as the visibility of golfand tennis balls, to estimate utilization classes onplots rather basing utilization on percent use by dryweight. The visibility of such aids is affected byplant density as much as the vegetation height on aplot. The method also does not account for thevariability in annual forage production.

Heady (1949) discussed various methods of deter-mining utilization and pointed out that the realproblem is not the method used but rather theinterpretation of the data. Hedrick (1958) also


supports this conclusion. Caidwell (1984) stated,"Employment of proper use schemes as an integralcomponent of forage allocation should be donewith considerable reservation. If taken at facevalue, these factors imply a level of precision andunderstanding of plants and community dynamicsthat for the most part d not exist. While thesefactors might provide some guidelines for appropri-ate forage utilization, the numerical values maycreate an impression of more precision than iswarranted."

Conclusionsand RecommendationsRules of thumb and simplistic guides, such asutilization standards, are not an acceptable substi-tute for experienced, on-the-ground managementbased on sound, long-term range trend information.As stated by Sharp et al. (1994), using utilizationdata to adjust management programs, particularlywith a simple mathematical formula, is an oversim-plification of resource management. And asCostello (1957) noted, oversimplification leads topoor interpretation, and poor interpretation leads topoor management.

Instead of relying on utilization standards, I recom-mend range managers make sure their goals andobjectives are written to reflect what they want tohappen to the resource. Do not use utilizationstandards as goals or objectives. Place monitoringemphasis on long-term trend, on both uplands andriparian areas. Permanent trend photo plots aremuch faster and easier to take than trying to esti-mate utilization and will provide a permanentrecord of not only trend but also use. Considerutilization and stubble height information as man-agement tools rather than as the only bases forgrazing decisions. Ranchers should take the initia-tive to suggest management options to correctunsatisfactory conditions that may be due to live-stock grazing, such as poor distribution or inappro-priate season of use.

In closing, I urge every professional range scientistand manager not to allow utilization standards totake precedence over the more important job of on-the-ground range management, based on monitor-ing long-term trend.


ReferencesBlaisdell, J. P. 1966. Range management viewed from

fore and aft or a fabrication of facts, figments, andphilosophy. Abstracts of Papers, 19th AnnualMeeting of the American Society of Range Manage-ment, Feb. 1-4, 1966, New Orleans, LA.

Burkhardt, J. W. 1997. Grazing utilization limits: Anineffective management tool. Rangelands 19:8-9.

Caldwell, M. M. 1984. Plant requirements for prudentgrazing. In: Natural Resources Council/NationalAcademy of Sciences: A Report Prepared by theCommittee on Developing Strategies for RangelandManagement. Boulder, CO: Westview Press.

Campbell, R. S. 1937. Problems of measuring forageutilization on western ranges. Ecology 18:528-532.

Campbell, R. S. 1943. Ecologyprogress in utilizationstandards for western ranges. Journal of the Wash-ington Academy of Science 33:161-169.

Cook, C. W. and L. A. Stoddart. 1953. The quandary ofutilization and preference. Journal of Range Manage-ment 6:329-336.

Costello, D. F. 1957. Application of ecology to rangemanagement. Ecology 38:49-53.

Dc Muth, C. A. R. 1990. Seasonal variation in utiliza-tion estimates on sideoats grama plants. M.S. thesis,University of Arizona, Tucson.

Frost, W. E., E. L. Smith, and P. R. Ogden. 1994.Utilization guidelines. Rangelands 16:256-259.

Hall, F. C. and L. Bryant. 1995. Herbaceous stubbleheight as a warning of impending cattle grazingdamage to riparian areas. USDA Forest ServiceGeneral Technical Report PNW-GTR-362.

Heady, H. F. 1949. Methods of determining utilizationof range forage. Journal of Range Management 2:53-63.

Hedrick, D. W. 1958. Proper utilizationa problem inevaluating the physiological response of plants tograzing use: a review. Journal of Range Management11:34-43.

Hormay, A. L. 1970. Principles of rest-rotation grazingand multiple-use land management. USDA ForestService Training Text 4 (2200).

Hormay, A. L. and M. W. Talbot. 1961. Rest-rotationgrazinga new management system for perennialbunchgrass ranges. USDA Production ResearchReport 51.

Jardine, J. T. and M. Anderson. 1919. Range Manage-ment on the National Forests; Washington, DC:USDA Agricultural Bulletin 790.

McKinney, Earl. 1997. It may be utilization, but is itmanagement? Rangelands 19:4-7.

Sampson, A. W. 1913. Range improvements by deferredand rotation grazing. USDA Bulletin 34.

Sampson, A. W. and H. E. Malmsten. 1926. Grazingperiods and forage production on the national forests.USDA Bulletin 1405.

Sharp, L.A. 1971. How can allotment managementplans be improved? In: Proceedings of the MontanaRange Management School, USD1 Bureau of LandManagement, Bozeman, MT.

Sharp, L. A., K. D. Sanders, and N. Rimbey. 1994.Management decisions based on utilizationis itreally management? Rangelands 16:38-40.

Smith, J.G. 1895. Forage conditions of the prairieregion. In: USDA Yearbook, 1895.

Glossary Revision Special Committee. 1989. A glos-sary of terms used in range management. 3rd ed.Denver: Society for Range Management.

Stoddart, L. A. 1953. Problems in estimating grazingcapacity of ranges. In: Proceedings of the 6th Inter-national Grassland Congress, 1952, PennsylvaniaState University, University Park, PA. Vol. II: 1367-1373.

USDA Forest Service, Division of Range Research.1944. The history of western range research. Agri-cultural History 18:127-143.


Seasonal Effects on the Measurementand Interpretation of Utilization

AbstractSeasonal effects on measuring utilization, interpret-ing "proper use," and estimating carrying capacityare examined. Utilization is defined as the propor-tion of the current year's forage production that isremoved or damaged by grazing animals. Peakstanding crop is the best one-time estimate of cur-rent year's biomass production. Proportion of bio-mass removed to standing crop measured at anyother time is not utilization but instead should becalled relative or seasonal utilization.

Appropriate use levels (such as take half, leavehalf) should be developed for the phenological stagein which grazing takes place. Utilization standards,consistent with the accepted definition of utiliza-tion, should not be applied to relative utilization.Utilization of individual plants has little or no rele-vance to the subsequent growth or reproduction ofthe plant unless the phenological stage when useoccurs is specified. Utilization should not be usedto adjust stocking rates unless combined with othertypes of data because the fundamental assumptionsfor the use of utilization on key species are not metif plant growing conditions change within a grazingperiod.

ObjectivesThe objectives of this paper are to examine threequestions.

How does season of the year affect the mea-surement of utilization?

How does season of the year affect the interpre-tation of "proper use"?

How does season of the year affect the validityof estimating "proper" stocking rates fromutilization data?

E. Lamar SmithRangeland Resources Program

University of ArizonaTucson, AZ 85722

BackgroundCattle and sheep ranchers have always been wellacquainted with the concept of utilization. Theirinformal estimates of the forage usedor, morelikely, the amount remainingtold them when theywould have to move, feed, or adjust the numbers oflivestock. Traditionally, ranchers looked at utiliza-tion from the animals' point of view, not from thestandpoint of the plants' welfare.

Early forest rangers recognized that some plantmaterial must be left on desirable forage species ifthe plants were to be maintained on the range. Theirestimates of "proper use" were usually around 80to 85% removal of the forage available from thebetter plants (Stoddart and Smith 1955). Theconcept of "proper use factors" was developedabout 1910 as part of the ocular reconnaissancemethod of range inventory. Utilization levels werebased on ocular estimates. In the 1 930s, quantita-tive methods were developed to measure utilizationon individual plants (e.g., see Lommasson andJensen 1938, Crafts 1938, Pechanec and Pickford1937) and qualitative methods for inventory ofutilization patterns (e.g., Deming 1939)..

Experience and research during the 1930s, 1940s,and 1950s also changed, to a more conservativelevel, the perception of what constituted "properuse" of individuals forage plants. Crider's classicstudy (1955) and numerous other studies of defolia-tion effects on carbohydrate reserves, root growth,and biomass production resulted in the general"take half, leave half' rule of thumb still prevalentin range management.

The concepts of utilization and proper use, as wellas the methods used to measure utilization, weredeveloped mainly by Forest Service personnel


before 1950. Most Forest Service rangelands,except perhaps in the Southwest Region, are char-acterized by springsummer growing seasons andsummer grazing seasons. Also, most studies ofdefoliation effects are based on clipping plants tovarious degrees and at various frequencies duringthe growing season. Thus, both the methodologyand the interpretations were based mainly onseason-long grazing which coincided more or lesswith the growing season. In this situation, use ofutilization data to adjust livestock stocking ratesworked reasonably well; but on year-long ranges orrotational grazing systems, both measurement andinterpretation became more complex.

What Is Utilization?The Society for Range Management defines utiliza-tion as "the proportion of current year's forage thatis consumed or destroyed by grazing animals"(Glossary Revision Special Committee 1989). Thisdefinition is widely accepted by all range manage-ment agencies (Interagency Technical Reference1996). Utilization (or use) commonly is said toapply to single plant species, groups of species, orto the vegetation as a whole.

The preceding statement may be a point of confu-sion. Stoddart and Smith (1955, p.138) state:

Utilization of a range means the degree towhich animals have consumed the usableforage production expressed in percentage. Thisproduction should be based on animal-monthsconsumed compared to animal-months avail-able when the range is correctly used.

When dealing with an individual plant, how-ever, utilization has a different usage and isdefined as the degree to which animals haveconsumed the total current herbage productionexpressed in percentage. These two usages areconfusing and will require clarification when-ever the term is used. It is suggested that rangeuse might be a better term for the first meaningand percentage utilization better for the secondmeaning.

The current definition is generally applied to bothconcepts presented by Stoddart and Smith withoutthe clarification they recommended. "Range use" isdefined in terms of available forage and therefore is

related to proper grazing use of the range as awhole. "Utilization" relates only to use of indi-vidual plants and has no necessary relation toproper stocking rate or canying capacity.

Some have suggested that it is more important, andmore straightforward, to measure the amount ofresidual vegetation (stubble height or biomass) thanthe percentage removed (e.g., Hyder 1954). Theyargue that it is the amount of residual biomass thatis important to the plant's ability to recover or tothe amount of soil protection provided. Removal ofa certain percentage of annual forage productionwould result in greatly different amounts of bothforage removed and residual vegetation left be-cause production varies greatly from year to year.Emphasis on residual vegetation has increased dueto the interest in leaving residual vegetation forwildlife cover, soil cover, and sediment trapping onfloodplains. However, measurement and interpreta-tion of residual biomass suffers from some of thesame difficulties as utilization measurement.

Purpose of Measuring UtilizationThe new Interagency Technical Manual on Utiliza-tion Studies and Residual Measurements (1996)states, "Residual measurements and utilization datacan be used: (1) to identify use patterns; (2) to helpestablish cause-and-effect interpretations of rangetrend data; and (3) to aid in adjusting stocking rateswhen combined with other monitoring data."

MOst range professionals would agree with thoseuses of utilization data. However, the federal landmanagement agencies increasingly have been usingutilization data alone to estimate carrying capacityand to establish standards for allowable range usewithin single grazing seasons. This is not consistentwith acceptable range management principles. Forexample, Stoddart and Smith (1955) discussed anumber of methods of estimating percentageutilization of individual species and the Demingmethod of evaluating overall level of range use.They concluded the discussion with the followingstatement:

It should be emphasized that, of the methodsdiscussed here, only the latter (the Demingmethod) gives its answer directly in terms ofcorrectness of range use. The other methods all


are aimed at determining percentage utilizationof a single species or a group of species pre-sumed to be the most important forage species.The resulting percentage tells nothing as towhether the range is underused, overused, orcorrectly used. Since for virtually no species dowe know what percentage utilization is correct(i.e., what the plant can endure), the value ofthese percentages is limited to general interpre-tation or to comparisons of one range withanother, or of one year with another. Utilizationdetermination is not an exact science either inmethod or interpretation.

Although that statement was made in 1955 and wehave learned a lot about plants' physiologicalresponses to herbivory since then, the statement isstill true. The "correct" level of utilization dependsnot only on the species (or even the ecotype) ofplant but also on such variables as site, weather,time and environmental conditions since the lastdefoliation, kind and level of utilization of associ-ated species, and the phenological development ofthe plant.

Seasonal Effectson Measurement of UtilizationBy definition, measuring utilization requires know-ing the total production for the year for the speciesin question. This requirement makes a true mea-surement of utilization virtually impossible undermanagement conditions. Total yearly productioncannot be measured at one time. The best that canbe done in a one-time effort is to estimate peakstanding crop of current-year production whichusually occurs near the end of the growing season.Peak standing crop is always less than total produc-tion because herbage is continually lost even duringthe growing season. Measurement before the pointof peak standing crop results man even lowerestimate of total biomass because annual produc-tion has not been completed. Measurement afterpeak standing crop is reached also results in lowervalues because biomass is lost to weathering,insects, and decay. However, peak standing crop isthe best estimate of total production that can bemade in a single measurement, and it is the valuethat most accepted utilization measurement tech-niques take for "total" production. Using ungrazed

peak standing crop to estimate total annual produc-tion assumes that grazing during the growingseason does not increase total production.

Measuring utilization by the SRM definitionrequires measuring production at the end of thegrowing season, while utilization should be mea-sured at the end of the grazing season. It often isnot feasible or practical to do this, as the followingexamples show.

Example 1. Continuous grazing during the growingseason. Both production and utilization would bemeasured at the same time, i.e., at the end of thegrazing season and at the end of the growingseason. In this case, utilization can be estimatedreasonably if regrowth is ignored. The concept ofutilization and most of the utilization standards arebest suited to this situation.

Example 2. Continuous grazing during the dormantseason. In this case, production and utilizationcannot be measured at the same time. Utilizationstandards developed for growing-season use maynot apply for dormant-season use.

Example 3. Continuous, year-long grazing. Utiliza-tion most logically would be measured at the end ofthe dormant season for vegetation, but productioncannot be measured reliably at this time.

Example 4. Rotational grazing. There are manypossible rotational grazing situations. Short grazingperiods during the growing season mean animalsare removed from a pasture before growth iscomplete. Thus, as noted previously, utilization andproduction cannot be measured at the same time. Insome rest-rotation situations, forage utilized mayrepresent growth from two different growingseasons, which makes the concept of utilizationeven more difficult to apply. Some form of rota-tional grazing is practiced on most rangeland today.

Recognition that utilization, as defined by SRM,cannot be measured under these conditions has ledto the use of other terms such as "relative use"(Frost, Smith, and Ogden 1994) or "seasonal use"(Interagency Technical Reference 1996) to describea comparison of grazed versus ungrazed plants atany time of year, recognizing that the differencedoes not represent "utilization." Difficulty in


measuring and interpreting utilization has also ledto more emphasis on residual vegetation measure-ments, e.g., stubble height or residual biomass,which are less affected by seasonal considerations.

"Relative use" (or seasonal use) measured duringthe growing season is always a higher percentagethan "utilization" expressed by the standard defini-tion. The amount of herbage removed is the samein both cases, but the ungrazed amount againstwhich use is measured is always less for relativeuse than for utilization.

To illustrate this point, Table 1 shows data calcu-lated from Ganskopp's clipping study (1988) onThurber needlegrass. Plants were clipped at variousphenological stages during the growth period.Relative use was calculated from the mg/cm2removed at each date as a percentage of standingcrop on that date. Utilization was expressed asherbage removed as a percentage of total unclippedproduction at seed shatter on July 17. Utilization

Relative use is biOmas removed/total production to date.2 is biomass removed/total production at last

clipping date.

Total production is preclip + postclip production for theyear.

and relative use are equal at the time of peakstanding crop (hard seedlseed shatter stage). Re-sidual herbage was estimated as 10 mg/cm2 for allclipping dates.

In 1985, relative use increased from 80% in thevegetative stage to 96% at the time of hard seed!seed shatter. Utilization ranged from 17% in thevegetative stage to 96% at the hard seed stage. In1986, growing conditions later in the season weremore favorable than in 1985. Consequently, bothrelative use and utilization are lower in the earlierphenological stages than in 1985. Relative useincreased from 60 to 96% during the growingseason. Utilization on the earliest clipping date wasonly 7% of total annual production and increased to96% at seed shatter. Some regrowth occurred afterthe final clipping in 1986, but this was ignored inmy calculations.

Table 1 shows the relationship of utilization torelative use during the growing season. Utilization

Table 1. Effect of clipping at various phenological stageson subsequent growth of Thurber needlegrass(Ganskopp 1988).

growth is aboveground production in the year aftertreatment.

Roots is root biomass in year after treatment.


Phenological stageUtilization/ Preboot Early Late Anthesis Soft Hard Seedgrowth 10//0 boot (%) boot (%) (%) dough (%) seed (%) shatter (%)1985

Relative use 80 84 92 93 94 96 96Utilization2 17 27 48 53 71 94 96

Total production3 51 37 71 71 73 97 100Top growth4 78 53 53 80 89 95 100Roots5 70 55 70 93 94 95 100

1986

Relative use 60 70 82 92 94 96 96Utilization 7 10 20 51 69 96 96

Total production 61 54 71 86 79 I 00 88

Top growth 63 50 72 91 93 100 98

Roots 68 61 68 87 84 100 93

cannot be predicted from relative use measuredduring the growing season because we cannotpredict the amount of growth that will occur afterrelative use is measured. For this reason, it is notfeasible to measure relative use during the growingseason and adjust the resulting value to somestandard utilization target.

Seasonal Effectson Interpretation of Proper UseProper use is "a degree and time of use of currentyear's growth which, if continued, will eithermaintain or improve the range condition consistentwith conservation of other natural resources"(Glossary Revision Committee 1989). This defini-tion refers to the range as a whole.

A proper use factor (PUF) is "an index to the graz-ing use that may be made of forage species basedon a system of range management that will main-tain the economically important forage species, orachieve other management objectives...." The PUFfor a key species typically represents the amount ofutilization a plant can receive and still maintain orimprove its productivity and reproduction. It is,therefore, related to the physiological and morpho-logical ability of the species to withstand grazing.

PUFs for associated species are a measure ofrelative preference of those species compared tothe key species. PUFs are influenced by kind ofgrazing animal, season of the year, frequency ofgrazing, vegetation type, site conditions, andmanagement objectives.

Utilization standards (or PUFs) for key speciestypically are derived from studies of effects ofclipping or grazing during mid to late growingseason. Such studies (e.g., Crider 1955) are mainlyresponsible for the "take half, leave half" standardwidely applied in range management. However,such studies cannot be applied for utilizationstandards during other seasons of the year, nor dothey usually account for frequency of grazing. Inaddition, an average utilization of 50% on a keyspecies may result in widely varying amounts ofutilization on each individual plant due to the waymany animals graze. Finally, PUFs based on thephysiological and/or morphological tolerance to

TIME

Figure 1. Theoretical relationship of utilization onkey species as a function of time (from Smith 1965).

grazing of key species and relative preference ofother species have no direct relevance to othermanagement concerns such as adequate soil cover,residual cover for nesting birds, or stubble heightrequirements for sediment capture.

Ganskopp's study (1988) again can be used as anexample of the effects of season, or phenologicalstage, "proper use" levels. Table 1 shows thatclipping Thurber needlegrass to a 2.5-cm stubbleheight in anthesis or a later growth stage had littleeffect on total annual production, nor did it havemuch effect on total root weight or top growth thenext year. Clipping during early or late boot stagedid reduce annual production and root weight andtop growth the next year. Clipping in the prebootstage was somewhat less prejudicial. These resultswere fairly consistent in 2 consecutive years eventhough the growth pattern between the years wasquite different.

Clipping at anthesis resulted in about 50% utiliza-tion by weight. Clipping at later phenologicalstages ranged up to 96% by weight. Thus, clippingto 50% or more had little or no effect, while utiliza-tion of 7 to 50% earlier in the year did reducesubsequent plant growth. (These are short-termeffects and do not necessarily indicate that late-season clipping rates could be maintained if clippedfor several years in succession.) Relative use was60% or more at every stage of growth, but theamount of relative use was not correlated witheffects on subsequent plant growth. Thus, relativeuse appears to have little value as a "utilization"standard, and because subsequent growth varies


from year to year, relative use cannot be usedreliably to predict utilization values.

It may be argued that residual measurements(stubble height) would avoid many problems inmeasuring utilization. From the standpoint of theindividual plant, that may not be true. In Gan-skopp's (1988) study, residual weight was the samefor all clipping treatments, but effects varied con-siderably among phenological stages. Thus, in thiscase, a standard residual stubble height or biomasscould not be applied without considering pheno-logical stage.

Although residual stubble height or biomass isimportant for the welfare of the individual plant,most current uses of this approach seem to be basedon other factors, e.g., residual cover for wildlife orresidual height for sediment accumulation. Such"standards" need to be justified by showing rela-tionships of residual vegetation to the factor ofinterest. They cannot be justified based on clipping

Table 2. Cattle diet composition percentages by phenological seasonsat the Santa Rita Experimental Range, AZ (Smith, Ogden, and Gomes 1993).

studies which show effects of residual amounts onthe physiological response of the individual plant.

Seasonal Effects on Estimationof Proper Stocking RatesRange scientists and the new Interagency TechnicalReference (1996) agree that utilization should notbe used to establish proper stocking rates unlesssupported by other monitoring data. However, boththe Bureau of Land Management and the ForestService currently use utilization data as a basis foradjusting livestock numbers and for removinglivestock when certain utilization or residual levelsare reached. Therefore, it is important to considerseasonal effects on the estimation of livestock'sproper range use. The distinction between rangeuse and utilization, mentioned earlier, is importantin this regard.

Using utilization on key species to estimate properstocking levels implies there is a known relation

'Seasons, assigned according to plant phenology, roughly correspond to: summer = July-October; winter = November-January; spring = February-March; and dry spring = April-June.


Plant species

Season1

Summer WinterEarlyspring

Dryspring

Grasses 78.2 40.0 38.5 58.3

Aristida spp. 9.8 7.0 3.8 1.1

Bouteloua spp. 14.2 7.6 16.5 13.2

Botriochloa barbinodes 4.3 1.3 0.3 0.2

Era grostis lehmanniana 5.1 2.2 9.4 3.5

Heteropogon con tortus 27.2 2.5 0.2 0.6

Muhienbergia porteri 4.6 15.3 4.8 26.5

Sporobolus spp. 7.6 0.6 1.3 1.4

Others 5,4 3.5 2.2 11.8

Shrubs 7.9 58.7 28.3 41.4

Atriplex spp. 1.3

Opuntia spp. 4.5 42.4 23.6 25.5

Prosopis juliflora 2.3 10.9 1.6 14.0

Others 1.1 5.4 3.1 0.6

Forbs 13.9 1.3 33.2 0.3

between such utilization and the total AUMs offorage removed from a pasture. Smith (1965)discussed the use of PUFs on key species to esti-mate stocking rates. He stated that the key-speciesconcept implies a unique relationship between thepercentage utilization of the key species and theutilization of the other important forage species.Further, the key species must be used gradually andcontinually throughout the grazing season with nosudden or marked changes in utilization.

Figure 1 shows acceptable and unacceptable pat-terns of seasonal use on key species. Smith (1965)concluded that, under most levels of utilization,these conditions probably were met sufficiently toallow carrying capacity estimates to be made. Thatconclusion may be justified under some conditions,e.g., if grazing use is confined to only a short timeor entirely to one growth period. However, if lon-ger grazing seasons are used, or if grazing periodsoverlap two or more growth phases of plants, theconclusion seems highly doubtful.

I did not find any studies showing changes in utili-zation on different species as a function of time ofgrazing. But, seasonal diet studies are fairly com-mon. For example, Table 2 shows diet compositionof cattle in four seasons of the year at the SantaRita Experimental Range in southern Arizona.Percentages of shrubs, grasses, and forbs varymarkedly by season. Percentages of the diet madeup by different major forage species also vary byseason. The assumption that use of key species iscontinuous throughout the grazing period and has aconsistent relation to use of other species clearly isnot met if grazing is through more than one season.

ConclusionsUtilization by accepted definitions cannot be

measured under most practical grazing manage-ment situations, especially when grazing is notcoincident with the growing season.

Relative or seasonal use can be measured when-ever livestock are removed from a pasture, bututilization standards developed from studies usingthe standard definition cannot be applied.

Utilization of individual species has little or norelevance to the subsequent growth or reproduction

of the plant unless the phenological stage of growthwhen use occurs is specified. Timing of use hasmore impact than amount of use as far as thephysiology of the plant is concerned.

Utilization should not be used to adjust stockingunless combined with other data. The fundamentalassumptions of the use of utilization on key speciesto estimate total forage removed are not met if graz-ing periods extend into different growing conditions.

Utilization standards for key species that arebased on the grazing tolerance of the plant have nodirect relevance to standards of utilization orresidual vegetation aimed at wildlife or soil cover,sediment capture, or other nongrazing effects.

ReferencesCrafts, E. C. 1938. Height-volume distribution in range

grasses. Journal of Forestry 36:1182-1185.

Crider, F. J. 1955. Root growth stoppage resulting fromdefoliation of grass. USDA Soil ConservationService Technical Bulletin 1102.

Deming, M. H. 1939. A field method of judging rangeutilization. (Mimeo.) USD1 Division of Grazing.

Frost, W. E., E. L. Smith, and P. R. Ogden. 1994.Utilization guidelines. Rangelands 16:256-259.

Ganskopp, David. 1988. Defoliation of Thurberneedlegrass: herbage and root responses. Journal ofRange Management 41:472-476.

Glossary Revision Special Committee. 1989. A glos-sary of terms used in range management. 3rd ed.Denver: Society for Range Management.

Hyder, D. N. 1954. Forage utilization. Journal ofForestry 52:603-604.

Interagency Technical Reference. 1996. Utilizationstudies and residual measurements. BLM/RS/ST/-96/004+1730.

Lommasson, T. and C. Jensen. 1938. Grass volumetables for determining range utilization. Science87:444

Pechanec, J. F. and G. D. Pickford. 1937. A comparisonof some methods used in determining percentageutilization of range grasses. Journal of AgriculturalResearch 54:753-765.

Smith, Arthur D. 1965. Determining common usegrazing capacities by application of the key speciesconcept. Journal of Range Management 18:196-201.


Smith, E. L., P. R. Ogden, and Hilton de Souza Gomes.1993. Forage preference and grazing behavior ofHereford and Barzona cattle on a Southern Arizonarange. In: Deborrah D. Young, ed. VegetationManagement of Hot Desert Rangeland Ecosystems.Bureau of Land Management symposium, July 28-30, Phoenix, AZ.

Stoddart, L. A. and A. D. Smith. 1955. Range manage-ment. 2nd ed. New York: McGraw-Hill.


Variation in Utilization EstimatesCaused by Differences

among Methods, Years, and Observers

AbstractA number of characteristics make utilization esti-mates inexact and unreliable. Errors and differ-ences in estimates of utilization can be significantdepending on methods, areas measured, observers,and years. Rather large differences in utilizationestimates can be obtained from different methods.Both the ocular estimate by plot and the caged!open clipped plot methods appear to overestimateutilization. In a given area, year-to-year variation inutilization can be quite large. Grazing by herbi-vores is never uniform which leads to a great dealof variation in utilization from one plot to anotheror from one area to another within a given grazingseason. Utilization estimates can vary considerablyamong observers, even those receiving intensivetraining. Vegetation under cages has been shown toproduce as much as 30% more than uncaged areas,mainly because of environmental conditions underthe cage. Utilization is a tool, not a land manage-ment objective, and should never be used as anobjective or used to set or adjust stocking rateswithout measuring trends. Utilization should bemeasured only at the end of the growing season.

IntroductionA number of characteristics make utilization aninexact and often unreliable indicator of the amountof actual use and, more important, the significanceof a given level of measured utilization to thehealth of a plant or of the rangeland. I will describebriefly the differences that occur in the measure-ment of utilization among methods, plots or areas,years, and observers. I also will describe some ofthe environmental and other factors that may

W.A. LaycockRangeland Ecology and Watershed Management

University of WyomingLaramie, WY 82071

influence the difference between caged anduncaged areas other than removal of foliage bygrazing. Because of these and other factors pointedout by other authors in this symposium, utilizationshould be used as only one tool in managingrangelands, and it should not be the primary factorin determining stocking rates.

Differences among MethodsThe earliest comparison of methods of determiningutilization was that of Pechanec and Pickford(1937). The trial used bluebunch wheatgrass(Agropyron spicatum) on a sagebrushgrass range-land at Dubois, ID. The authors experimentallyremoved a prescribed amount of foliage frombluebunch wheatgrass plants on 100 plots 5 x 5 feetin size. Three observers then independently esti-mated amount of utilization using four differentmethods. Two of the observers were experienced inthe methods compared, and one was not.

The ocular estimate by plot, ocular estimate ofplants by plot, and the leaf length methods yieldedsomewhat similar results; however, all but the ocu-lar estimate of plants by plot overestimated actualutilization (Table 1). Average utilization was esti-mated at 44%, 38%, and 43% respectively for thesethree methods (actual clipped utilization was 37%).

For the plant-count method, Pechanec and Pickford(1937) counted grazed and ungrazed plants, and thepercentage of plants grazed (54%) was consideredequivalent to percentage removal by weight. Thisresulted in an inherent bias because the methodassumed utilization at 54% instead of the actual37%. In addition, the average of the three observers


yielded estimates of 68% utilization, a considerableoverestimate of the actual 37% removal. The differ-ences among observers will be discussed below.

In contrast to Pechanec and Pickford's (1937)

results, Springfield (1961) found that the grazed-plant method gave similar estimates of utilizationas the "difference" method for crested wheatgrass(Agropyron desertorum) in northern New Mexico.Grazed and ungrazed plants were counted on plots,and the difference in weight was measured betweencaged and uncaged areas.

On a tall forb community grazed by sheep in south-west Montana, Laycock, Buchanan, and Krueger(1972) tested three methods of determining utiliza-tion: caged and open plots clipped after grazing,ocular estimate by plot, and botanical compositionof esophageal fistula samples converted to percent-age of each species in the plant community thesheep grazed.

Twenty pairs of caged/uncaged plots, each 4.8square feet, were clipped immediately after sheep

Table 1. Comparison among utilization methods using three observers on bluebunch wheatgrassat Dubois, ID (Pechanec and Pickford 1937).

grazing, 10 pairs in each of two pastures grazed inearly, middle, and late summer. Only data from theearly summer trial are presented.

Before clipping the plots, two observers estimatedthe percentage utilization by weight for eachspecies. The observers had been trained intensivelyin the ocular estimate by plot method (Pechanecand Pickford 1937). Each observer estimated halfthe 40 plots in each pasture (a total of 80 plots,each 4.8 square feet). The percentage compositionof species in the fistula samples of seven sheep,collected for 3 days during the grazing trial, wasconverted to percent utilization by multiplying thecomposition by the assumed consumption and thencomparing that to the amount of each species in thecaged plots.

Even though the fistula method was considered notvery accurate because of rather complicated calcu-lations, the fistula method and the ocular estimateby plot method yielded similar results (Table 2).Pechanec and Pickford (1937) indicated that theocular estimate by plot method probably overesti-

* Percentage of plants grazed (54%) overestimated by 46% the actual percentage of foliage removed by weight (37%).


Method ObserverActual %removed

Estimated %removed Error (%)

Ocular 37 41 +11estimateby plot

2 37 44 +19

Al ±22Avg. 37 44 +19

Ocular 37 35 -5estimateby plant

2 37 40 +831

Avg. 37 38 +2

Leaf 37 42 +14length 2 37 35 -5measure-ments 3 37

+16Avg. 37 43

Plantcount(% plants

Averageof

threegrazed) observers 54* 68 +26

mates utilization. The estimated utilization of forbsand all vegetation from the paired, clipped plotswas more than twice that from the two other meth-ods and 75% higher for grasses. This unexpectedlylarge difference was attributed partly to "tram-pling" damage and other "invisible" utilization.Entire leaves of some forbs, such as yarrow (Achil-lea millefolium), may be broken off easily byactivities other than actual consumption. This isdifficult to recognize when making ocular esti-mates. Other unknown 'factors may have contrib-uted to the differences. Whatever the cause, two ofthe most widely used methods of determiningutilization yielded very different results.

Differences among YearsBecause of journals' space restrictions, most pub-lished grazing studies document only mean utiliza-tion figures. Thus, the magnitude of variation inutilization among years is seldom reported. A studyon sandhill rangeland in eastern Colorado reportedyearly utilization for pastures that steers grazedlightly, moderately, and heavily from May 1 toOctober 1 for 10 years (Sims et al. 1976). Utiliza-tion was determined using the difference between60 to 80 caged and uncaged plots in each pasture.Utilization was reported for all major species, butonly those for needle-and-thread (Stipa comata) arepresented here as an example of yearly variationsin utilization. Other species had similar variation.

Average use of needle-and-thread over the 10-yearperiod was 26%, 60%, and 77% for the light,moderate, and heavy grazing respectively (Table3). Under moderate grazing rate, the utilization rateover 10 years (average 60%) ranged from a low of27% to a high of 85%. The confidence interval atthe 95% level was 14%, meaning that the truemean utilization for the 10-year period was be-tween 46 and 74%. This relatively wide confidenceinterval was in spite of 10 years of data. Each year,60 to 80 pairs of cagedluncaged plots were clippedin each pasture. Smaller sample sizes of 10 orfewer plots, such as are used for utilization samplesin most management situations, probably wouldyield much wider confidence intervals.

Many other examples of year-to-year variationexist. On cattle range in north-central Oregon,

Table 2. Early-summer percentage utilization bysheep, comparing three methods in the tall forb type,Centennial Mountains, MT (Laycock et al. 1972).

20 pairs of caged/uncaged plots clipped after grazing (10pairs per pasture)2 80 plots using ocular-estimate by plot method (40 plots!pasture) after intensive training

Elliott (1976) reported wide variation betweenutilization over 2 years. Utilization in Areas I andV and for the pasture on average differed consider-ably between years (Table 4). In the Blue Moun-tains of Oregon, Clark (1996) reported similarly

Table 3. Percentage utilization of Stipa comata onsandhill rangeland in eastern Colorado, 1957-1 966(Sims et al. 1976).

Grazing season was May 1-October 1 every year.

Stocking rates were:light = 10 A/steer (2.8 A/AUM or 0.35 AUM/A)moderate = 5 A/steer (1.4 A/AUM or 0.7 AUMJA)heavy = 3.3 A/steer (0.9 A/AUM or 1.06 AUM/A)


Stocking rate1

Light Moderate Heavy

Utilization 8 10 82 36 68 59for each of 9 52 50 73 93 7110 years

44 4 70 85 93 96

34 32 60 71 87 29

43 27 45 75 77

CategoryPairedplots1

Estimatedplots2

Fistulasamples

All grass 49 28 30

Yarrow 57 1 4

Stickygeranium 23 2

Knotweed 43 2 25

Northwestcinquefoil 37 42 26

All forbs 39 16 16

All vegetation 40 17 17

Average 26 60 77

Range 4-52 27-85 29-96

Years aboveaverage 5 5 4

Stipa comataresponse Increase Slight increase Decrease

Table 4. Cattle utilization on rockpile improvedpasture in north-central Oregon, 1974-1 975(Elliott 1976).

wide variations in utilization over 2 years on bothelk sedge (Carex geyeri) and Idaho fescue (Festucaidahoensis) on ranges grazed by sheep (Table 5).

Burkhardt (1996) reviewed the evolutionary historyof grazing in the intermountain region and foundno evidence that conservative, uniform utilizationevery year ever naturally existed on rangelands. Heconcluded that "conservative utilization limits donot appear to be part of natural herbivories such asin Africa today, the plains bison of the 1800s, orthe Pleistocene megafauna. Utilization limitsappear to be a human-made concept. The fossilrecord gives no indication of prehistoric forestrangers attempting to enforce use limits on mega-fauna.... Managing grazing by utilization standardsor guidelines reduces range management from anapplied science and an art to a policing action."

Burkhardt (1997) stated that utilizationlimits weredeveloped to manage season-long grazing duringthe growing season every year. Burkhardt (1997)said further, "The current agency approach to graz-ing management is in reality a non-management

Table 5. Sheep utilization percentages onbluebunch wheatgrass rangeland in the BlueMountains of Oregon, 1993-1 994 (Clark 1996).

10/ \Utilization io)

scheme. By rigorous and subjective application ofutilization standards, livestock grazing will bereduced to a token activity which no longer causesadministrative or political headaches."

Variation among Plots or AreasAny herbivore's utilization pattern is never uni-form, unless use levels are very high. Grazing,especially light to moderate grazing, naturally is inpatches (Kellner and Bosch 1992). This leads to agreat deal of variation from one plot to another,especially for estimation methods using plots. This,coupled with the fact that relatively few plots usu-ally are estimated or measured, leads to relativelyhigh standard errors and wide confidence intervals.

The same variation caused by uneven grazing oc-curs in various parts of a pasture or allotment whengrazed. Tables 4 and 5 show, in addition to greatdifferences in utilization between years, that utili-zation among areas in the same pasture or grazingarea can vary considerably within a given year.

The assumption usually is that utilization levels(usually measured on key areas) reflect the level ofuse onthe pasture or area as a whole (see the latersection on "tacky tricks"). This may or may not betrue depending upon the location(s) where utiliza-tion is measured and the particular pattern ofgrazing in a particular year.

Thetford (1975) measured sheep utilization in theCoast Range of Oregon for 2 years in areas grazedat moderate, heavy, and "overstocked" rates. In1973, results were as most would have predictedthe lightest utilization (64%) was in the moderatelygrazed area, and the heaviest utilization (86%) wasin the overstocked area (Table 6). However, in1974, the exact opposite occurredheaviest utili-zation (67%) was in the moderate pasture, and thelightest utilization (37%) was in the overstockedpasture even though the same areas apparentlywere sampled for utilization in both years. Theseresults also illustrate differences that occur betweenyears, especially in the overstocked pasture;

Differences among ObserversIn Pechanec and Pickford's (1937) trial, threeobservers tested four methods of determining


Elk sedge Idaho fescue

1993 1994 1993 1994

18 50 11 5.3

7 20 11 31

2 31 5 33

1 16

4 29 9 39

Area

Utilization (%)

1974 1975

I 8 57

II 4 7

III 5 9

Iv 5 5

V 24 55

Avg. 9 27

Table 6. Sheep utilization percentages in white oaklDouglas-fir vegetation in the Coast Range ofOregon (Thetford 11975).

utilization of bluebunch wheatgrass. For the ocularestimate by plot method; the three observers esti-mated utilization of 44%, 47%, and 41% (Table 1).The average was 44% which overestimated themeasured foliage removal (which was 37%) by19%. The authors concluded that the method wassubject to personal error, and the estimated percent-age removed differed appreciably from that actu-ally removed.

In a variation of the ocular estimate by plotmethod, percent removal of every plant was esti-mated and averaged over the plot. This was consid-erably more accurate. There was less variationamong observers, and the three observers overesti-mated by an average of only 3%. However, themethod was very time consuming.

In the method that measured each plant's leaves,the percent of length removed was assumed toequal percent of weight removed. The three observ-ers estimated 42%, 35%, and 52% removal (Table1), an average overestimate of 16% more than theactual removed, which was 37%. Pechanec andPickford (1937) concluded that this method yielded

Table 7. Differences among teams in samplingalder utilization in the Pacific Northwest(Fred Hall, personal communication).1'2

Average difference (%) in browsedvs. unbrowsed twig length

A negative percentage means the average length of thebrowsed leaders was greater than the average length of theunbrowsed leaders.

2 Each value is the mean for one team of two observers.

large and significant difference among observers,was slow, and was not recommended.

Using the method that counted plants grazed, theestimated weight removed was 54%, which overes-timated actual percentage foliage removed byweight (37%) by 46%. The three observers esti-mated that 68% of the plants had been grazed,which further exaggerated the overestimate to 84%greater than the measured weight removal. Littledifference was noted among observers, but themethod was the least accurate one tested becauseestimated percentages of plants grazed were uni-formly greater than the actual. Pechanec andPickford (1937) stated that this method was notsuited for bunchgrasses.

In the Pacific Northwest, experienced teams mea-sured browsed and unbrowsed twigs to determineutilization alder (Frederick C. Hall, personal com-munication). Each team read at least two transects,with erratic results (Table 7). The negative figuresindicate that browsed leaders were longer thanunbrowsed leaders. All but three team estimates onthe three transects were negative, and the estimatesvaried from -40% to 27%, which indicates a greatdifference among individuals and teams.

Effects of Cageson Utilization EstimatesIn Britain, Cowlishaw (1951) found that yieldsfrom areas under cages were significantly greaterthan from unprotected areas due to reduced windand increased humidity inside the cages. Cook andStoddart (1953) found rather large errors in inter-preting utilization of crested wheatgrass using thepaired caged/uncaged method during differentperiods in the growing season. Cook andStubbendieck (1986) stated, "A common objection(to use of cages) is that differences in growth onthe protected and grazed areas may distort utiliza-tion. The greater the period of time between cagingand clipping the larger this becomes."

The distortion caused by cages was quantified byHeady (1957) on California annual grasslandvegetation. He located 25 to 38 cages, each 2.5 x3.5 feet, in ungrazed pastures. The uncaged andungrazed plots produced 33% less than the caged


Transect 1 -1 -44 -40 -18 -8

Transect 2 -21 +7 +27 -5 0

Transect 3 -6 +4

Grazingtreatment Intensity

Utilization (%)

1973 1974

Moderate 3.7 AU/A 64 67

Heavy 4.9 AU/A 81 59

Overstocked 6.2 AU/A 86 37(6.2 AU/A)

plots in the open grass and 21% less in a grasslandunder thin tree cover (Table 8). Thus, apparentmeasured utilization would have been 33% and21% when actual utilization was zero.

In a taligrass prairie area, Owensby (1969) used 10cages of 1 square meter in one 60-acre pasture and30 cages in another 44-acre pasture. Both pastureswere ungrazed. The uncaged and ungrazed plotsproduced 33% less than the caged plots (1,518versus 2,158 lb/A respectively). This would indi-cate 33% utilization in the uncaged plot when, infact, no utilization occurred. When Owensby(1969) calculated the utilization using cages innearby pastures, he found that the cage effectwould account for 47% of the apparent utilizationunder moderate stocking, 45% under light stocking,and 31% under heavy stocking. Of what value areutilization figures that are wrong by that magni-tude?

Another error that can affect caged/uncaged com-parisons is the fact that grazing, especially heavygrazing, can reduce production compared toungrazed areas. This difference is more likely onareas grazed heavily for a number of years, buteven a single year could reduce production andinfluence the cagediuncaged comparison.

In a long-term grazing trial at the Streeter Station inNorth Dakota (Patton et al. 1996), production wasseverely reduced on heavier than normal grazingtreatments compared to the long-term ungrazedpasture (Table 9). This effect certainly is not whatwould be expected from grazing for 1 year in acaged/uncaged comparison, but it does point outanother potential for error. Based on production ofthe grazed pastures compared to the ungrazed pas-ture in this study, apparent utilization before anygrazing would have been 18%, 36%, and 44% forthe normal grazing, one and one-half times normal

Table 9. Effect of long-term grazing on total forageproduction at Central Grasslands Research Center,Streeter, ND (Patton et al. 1996).

'Means followed by the same letter are not significantlydifferent at p = .05.

2 Reduction in production compared to no grazing.

grazing, and twice nonnal grazing treatments,respectively. This effect, added to any environmen-tal effect of the cage, could increase the error of thecomputation even more. That is why any long-termexclosures must not be used for the ungrazedcomparison in making utilization estimates.

"Tacky Tricks" with UtilizationMcKinney (1997) presented nine "Tacky Tricks"that illustrate how management agencies some-times use utilization inappropriately to reducestocking rates on an allotment. The tricks are:

Measure utilization during the growingseason (when animals are in pasture)notafter the growing season (as required bydefinition).

Monitor only most favored plant speciesfor the season grazed. Ignore all otherspecies even if they become most favored inother seasons.

Produce use-pattern maps from drivingalong roads. Never get out of the pickup to

Table 8. Effect of cages (2.5 x 3.5 ft.) on production on California annual grassland (Heady 1957).


Vegetation type Pairs (#)Average production (lb/A)

Caged Uncaged Diff. (lb.) Duff. (%)

Open grass

Grass underthin treecover

38

25

332

228

224

180

108

48

-33

-21

Grazingtreatment

Production(lb/A)1

Reduction inproduction (%)2

No grazing 2,608 a

Half normal 2,438 a 7

Normal 2,149ab 18

1.5 times normal 1,668b 36

Twice normal 1,462b 44

get away from the travel route for animalsprovided by the road.

Manage for a Utilization Standard at aKey Area and get this written up in a LandUse Plan. This allows you to giveit theforce of law and pretend that it is manage-ment.

To get even greater reductions in animalnumbers, use #4 with afloating Key Areamoved to coincide with heavy use areaseach year.

Obtain major reductions in livestocknumbers by picking riparian areas fordoing #4.

To completely rid the range of livestock,combine #6 with #1.

When calculating pasture averages,throw out any utilization lower than moder-ate. You can average moderate (50%) withheavy (70%) and always show the range isoverstocked, no matter how few animals arepresent and how good the management is.This technique sounds reasonable andconservative when explained in a sincerevoice.

To dodge those obnoxious unwantedcomments, leave the utilization standardsout of the draft Land Use Plan, but slide'em into the final.

McKinney (1997) summarized, "And that's thestory on utilization: a fine old range managementtool with valuable but limited application is nowbeing used less as a management tool and more asa political tool for removing livestock and wildhorses from rangelands. In the resulting brouhaha,everyone involved soon forgets all the reallyinteresting stuff we have learned about manage-ment over the past 40 years."

SummaryAccuracy and precision of utilization estimatesgenerally are not very high because:


1) Patterns of utilization are highly variable inboth space and time. Herbivory, by nature, isnot uniform across the landscape nor is ituniform from year to year.

2) Different methods of determining utilizationwill yield different results.

3) Different observers get different results usingthe same method to estimate utilization.

4) An average utilization figure is, at best an indexto amount of use and is not an exact figure.

5) Using the paired cage/uncaged plot techniqueoverestimates utilization by 30% or morebecause:

The cage environment enhances forageproduction.

Grazing can decrease production outsidecages.

6) Based on early research, the ocular-estimate byplot method probably also overestimates utili-zation.

How should utilization be used?

Utilization, by definition, must be measured atthe end of the growing season, not earlier.

Utilization is only one tool to achieve a landmanagement objective (such as a Desired PlantCommunity). It never should be the objective ofmanagement.

Without a measured trend over time, utilizationalone is not an accurate indicator of the effectof grazing on a pasture and never should beused as the sole factor to adjust stocking rates."Tacky tricks" to adjust stocking rates are beingused in some areas by both the Forest Serviceand Bureau of Land Management. The papersby Van Tassell and Richardson and by Fletcherin this publication address the economic im-pacts of reducing AUMs caused by imposingutilization standards or other managementactions.

Utilization standards should be applied as anaverage over years (such as the number of yearsfor a complete cycle in a grazing system), notimposed every year.

23

ReferencesBurkhardt, J. W. 1996. Herbivory in the Intermountain

West: an overview of evolutionary history, historiccultural impacts and lessons from the past. IdahoForest, Wildlife, and Range Experiment StationBulletin 58. Moscow: University of Idaho.

Burkhardt, J. W. 1997. Grazing utilization limits: anineffective management tool. Rangelands 19:8-9.

Clark, P. E. Use of livestock to improve the quality ofelk winter range forage in northeastern Oregon.Ph.D. dissertation, 1996. Oregon State University,Corvallis.

Cook, C. W. and L. A. Stoddart. 1953. Some growthresponses of crested wheatgrass following herbageremoval. Journal of Range Management 6:267-270.

Cook, C. W. and J. Stubbendieck. 1986. Range Re-search: Basic Problems and Techniques. Denver:Society for Range Management.

Cowlishaw, S. J. 1951. The effect of sampling cages onthe yields of herbage. Journal of the British Grass-land Society 6:179-182.

Elliott, J. C. Diet and feeding behavior of cattle at riskto dietary acute bovine pulmonary emphysema(DABPE). M.S.thesis, 1976, Oregon State Univer-sity, Corvallis.

Heady, H. F. 1957. Effect of cages on yield and compo-sition in the California annual type. Journal of RangeManagement 10: 175-177.

Kellner, K. and 0. J. H. Bosch. 1992. Influence ofpatch formation in determining the stocking rate forsouthern African grasslands. Journal of Arid Envi-ronments 22:99-105.

Laycock, W. A., H. Buchanan, and W. C. Krueger.1972. Three methods of determining diet, utilizationand trampling damage on sheep ranges. Journal ofRange Management 25:352-356.

McKinney, E. 1997. It may be utilization, but is itmanagement? Rangelands 19:4-7.

Owensby, C. E. 1969. Effect of cages on herbage yieldin true prairie vegetation. Journal of Range Manage-ment 22:13 1-132.

Patton, B., P. Nyren, B. Kreft, J. Caton, and A. Nyren.1996. Grazing intensity research on Coteau range-lands. In: 1996 Grass and Beef Research Review.Central Grasslands Research Center. Fargo: NorthDakota State University.

Pechanec, J. F. and G. D. Pickford. 1937. A comparisonof some methods used in determining percentageutilization of range grasses. Journal of AgriculturalResearch 54:753-765.

Sims, P. L., B. E. Dahi, and A. H. Denham. 1976.Vegetation and livestock response at three grazingintensities on sandhill rangeland in eastern Colorado.Experiment Station Technical Bulletin 130. FortCollins: Colorado State University.

Springfield, H. W. 1961. The grazed plant method forjudging the utilization of crested wheatgrass. Journalof Forestry 59:666-670.

Thetford, F. 0. Jr. Responses of vegetation and sheep tothree grazing pressures. M.S. thesis, 1975, OregonState University, Corvallis.


Interpretation of Utilizationand Long-term Frequency Measurements

for Rangeland Management

AbstractAs a tool, vegetation utilization has been used tohelp make management decisions for many years.It is recommended that utilization should be com-bined with long-term trend measurements to helpinterpret these measurements. It has been assumedthat yearly utilization measurements taken overseveral years would relate to long-term trend ofrangeland vegetation. However, long-term data setsin western Utah (1982-1995) show no significantcorrelations between yearly utilization and long-term trend measured (by frequency of key species).

It has been assumed that degree of use relates to thephysiological needs of the plant, but recent studieshave shown that phenological stage of plant growthis more important than degree of use. Numerousmanagement strategies have adapted this knowl-edge where use levels as high as 70% are acceptedon riparian areas and big game winter range whenthese areas are grazed early in the growing season.However, recent data have shown that for someshrubs (e.g., Artemisa tridentata), degree of use isvery important for the subsequent year's produc-tion and for long-term survival.

These data suggest utilization is not very useful indetermining the relationship between managementand long-term trend of rangelands in western Utah.It is an important tool to help determine livestockdistribution and is suggested where managementcould be intensified in the future.

The range management profession has had a longand sometimes arduous debate on how to integrateannual utilization measurements with long-term

G. Allen RasmussenDepartment of Rangeland Resources

Utah State UniversityLogan, UT 84322-5230

trend data to manage rangelands. Basic rangemanagement courses have taught that these twomeasurements should be used together in makinglong-term management changes. Under the as-sumption that if changes in long-term trend wereundesirable and the utilization level exceeded adesired objective, then the utilization level could beused to help establish the cause of the undesirabletrend. Reasoning for this argument is based on theplants' physiological needs, which assumes there isa proper use of a key species. Excessive use (be-yond the estimated proper use level; see Sanders,this volume) will lead to changes in the plantcommunity. In addition, the excessive use of aplant would lead to the decline of that particularplant.

Crider (1955) found that root growth was reducedas aboveground phytomass was removed. Rootgrowth was virtually stopped when 50% of thecurrent growth was removed, and root growth wasdramatically reduced when use exceeded 50% onmany plants (Table 1). These data have beeninterpreted to mean that if root growth is inhibited,

Table 1. Smooth brome (Bromus inermis) rootgrowth 3 weeks after defoliation (adapted fromCrider 1955).


Percentremoved

(by weight)Percent

root growth

0 127

50 2

60 -20

70 -60

then the ability of the plant to compete with itsneighbors will be reduced; therefore, over a periodof time, use relates to changes in the plant commu-nity. This supports the widely accepted generalitythat range managers should allow only 50% use ofcurrent-year growth on key species to maintainthem in the plant community. Using this informa-tion, many land managers have a hard time under-standing why utilization data cannot be used tomake long-term decisions.

Data from Cook (1966) and Olsen and Richards(1989) suggested that it is not utilization level thatprimarily drives response to grazing but rather thephenological stage of plant growth when useoccurs. These studies have helped lead to confusionand conflict on how utilization data should be usedin interpreting long-term changes in rangelandplant communities and the consequences of range-land management practices.

If land managers are to use utilization data andlong-term trend data together to establish cause,there should be a direct relationship between theaverage yearly utilization levels over a period andthe changes in trend over that same period. Fewdata are published on the relationship betweentrend data and utilization levels.

Bureau of Land Management has provided a dataset containing 33 sampling sites from a grazingallotment in western Utah that I used to determinehow well annual utilization levels and frequency ofthe key species were correlated. The allotmentcontained 153,333 hectares (378,734 acres) consist-ing primarily of two community types, salt desertshrub and Wyoming sagebrush steppe. The averageannual precipitation was 76 millimeters. A total of8,673 AUMs were allocated. However, actual useranged from 867 to 8,673 AUMs depending onyearly weather conditions. The allotment wasgrazed from November to June, using a deferredrotation system. From 1982 to 1994, averageutilization on the allotment was 40% (standarderror = 6.8). The change in the frequency duringthis period was -0.4% (standard error = 2.9). Thekey species on the salt desert shrub communitieswere shadscale (Atriplex confertifolia), galletagrass (Hilaria jamesii), and Indian rice grass (Stipahymenoides). On the sagebrush steppe, the key

Table 2. Regression analysis of the change in long-term trend (plot frequency) and average utilizationlevels for the previous 4, 3, 2, and 1 years.

Averageutilization (%)

Number ofprevious

species were Wyoming sagebrush (Artemisatridentata var. wyomingensis), and bluebunchwheatgrass (Agropyron spicatum).

Regression analysis was used to evaluate the rela-tionship between utilization of and change in thefrequency of the key species. Complete data on allsites were available from 1986 to 1994 for yearlyutilization of key species. Plot frequency of the keyspecies was read in 1986, 1991, and 1994 on allsites. This allowed comparisons between changesin frequency for two periods, 1986 to 1991 and1991 to 1994. A separate regression analysis usedthe average utilization levels on each site for theprevious one, two, three, and four grazing periods(years). No difference was found between thesagebrush steppe and salt desert shrub communi-ties, so they were combined in the analysis.

No significant relationship was found betweenaverage annual utilization levels, regardless of thenumber of years of utilization data used, andchanges in long-term trend using plot frequency(Table 2). Average allotment use was only moder-ate, less than 40% over the entire period of analy-sis. However, the average utilization on the indi-vidual sampling sites for the previous four grazingperiods ranged from 8 to 65%. While land manag-ers have been taught that utilization and long-termtrend should be used together to determine theconsequences of their rangeland management,these data indicate little relationship betweenutilization levels and changes in long-term trend.

Many land managers disagree with this conclusion,often citing both observations from their experi-ence and studies like Crider's (1955). It is easy tounderstand their concerns if one accepts the as-


years R-squared P value Max. Mm.

4 0.0037 0.634 65 12

3 0.0167 0.301 67 8

2 0.0304 0.162 67 2

I 0.0173 0.293 80 3

sumptions of the impact of root growth and plantcompetitiveness behind Crider's (1955) work. Thestudy suggests that when a plant has to reduce itsroot growth, its ability to compete in the commu-nity will be reduced, and over time the plant can beeliminated from the plant community. By thisreasoning, plants that can maintain root growth willhave the greatest success in the community.

However, more recent studies (Richards 1984)compared the root growth of crested wheatgrassand bluebunch wheatgrass (Table 3). He foundcrested wheatgrass after defoliation reduced its rootgrowth significantly compared to control plants,but bluebunch wheatgrass only slightly altered itsroot growth after defoliation. Of these two species,crested wheatgrass is considered the more tolerantof grazing, but it has the greater reduction in rootgrowth after defoliation. Crested wheatgrass maytolerate grazing better because it is able to adjustthe root biomass to a level that can be supported bythe remaining photosynthetic tissue on the plant.

While many argue that experience shows utiliza-tion drives the long-term change in the plant com-munity, other literature indicates that it is not thelevel of utilization but the timing of use that driveschanges in the plant community (Cook 1966, Olsenand Richards 1989). Cook originally related thisplant response to the available carbohydrates in theplant. That work led to the opinion that early-season grazing would be detrimental to plantsbecause their available carbohydrates are decliningor at a low point. This does affect a plant, butCaidwell et al. (1981) found the stored energy wasless important than the photosynthetic capacity.Olsen and Richards (1989) agreed with Cook'sconclusion that the phenological stage at which the

Table 3. Relative root growth of bluebunch wheat-grass (BW) and crested wheatgrass (CW) followingdefoliation (adapted from Richards 1984).

Relative root growthafter defoliation(%)

plant is grazed will have a greater effect on theplant's ability to compete than the utilization level.Their work indicates that removing the apicalmeristem before seed set determined the plant'slong-term production and its ability to compete inthe environment. They found early-season (vegeta-tive growth stage) and late-season (seed ripe) heavyuse had little long-term impact on plant production.Moderate use, because it removed the apical mer-istem auring the stem elongation stage, reducedlong-term production. The findings suggest a plant'sability to withstand grazing is related more to usein a narrow season than to the actual level of use.

The BLM data presented here and the literatureindicate that utilization levels have little to do withchanges in long-term trend. With all this evidence,why is there still debate on using utilization?

I have found that land managers who cite theirexperience with the correlation of utilization andlong-term trend were using season-long grazingsystems in which plants were grazed every yearduring their active growth stages. I suspect heavyuse (>50%) in these grazing systems removesapical meristems during the active growth period,leading to reduced competitiveness of some speciesand species frequency changes in the plant commu-nity. Moderate use levels (50%) left many plantswith their apical meristems intact, and so long-termchanges were minimized. Many of the grazingsystems that have been developed prevent repeateddefoliation of the same plants each year during thecritical phenological stage.

Another possible explanation for the lack of corre-lation between utilization and trend is that thecommunities studied may have been in a stablestate because of grazing history or other factors.

Laycock (1994) stated that lower successionalsteady states are common in the sagebrushgrassvegetation type. When a shrub-dominated commu-nity is in such a state, reducing grazing pressure oreven stopping grazing completely does not result inchanges in vegetation composition, even overdecades (Laycock 1994). Continued heavy usemight push the vegetation over a threshold into alower successional steady state, but often that usehas to be quite heavy and prolonged.


Treatment 40 days 100 days

BW control 15 36

BW defoliation 13 39

CW control 20 43

CW defoliation 10 26

If utilization measurements do not relate to changesin plant communities, what help is this measure-ment in managing rangelands?

Utilization is most helpful in determining animaldistribution; second, it can help to estimate stock-ing rates if the total forage base is known. This canbe very important in helping design the mostappropriate management strategy for an area. Forsome shrubs, utilization levels can be important.Wyoming sagebrush will reduce its reproductiveeffort and production if more than 50% of currentgrowth is removed; however, bitterbrush willincrease its growth the more it is utilized. Theseopposite responses relate to where the meristematictissue is located when the plant is grazed (Bil-brough and Richards 1993).

In light of this information, which indicates utiliza-tion is difficult to interpret and measure, I suggestannual management focus on other techniques thatcan be useful for other objectives as well as live-stock or wildlife use. These would include cover,stubble height, or "other residual" measurements.

ReferencesBilbrough, C. J. and J. H. Richards. 1993. Growth of

sagebrush and bitterbrush following simulated winterbrowsing: mechanisms tolerance. Ecology 74:481-492.

Caidwell, M. M., J. H. Richards, D. A. Johnson, R. S.Nowak, and R. S. Dzurec. 1981. Coping withherbivory: photosynthetic capacity and resourceallocation in two semiarid Agropyron bunchgrasses.Oecologia 50:14-24.

Cook, C. W. 1966. The role of carbohydrate reserves inmanaging range plants. Utah Agricultural Experi-ment Station Series 499.

Crider, F. J. 1955. Root-growth stoppage resulting fromdefoliation of grass. USDA Soil ConservationService Technical Bulletin 1102.

Laycock, W. A. 1994. Implications of grazing vs. nograzing on today's rangelands. In: M.Vavra, W. A.Laycock, and R. D. Pieper, eds. Ecological Implica-tions of Livestock Herbivory in the West. Denver:Society for Range Management.

Olsen, B. E. and J. H. Richards. 1989. Grazing effectson crested wheatgrass growth and replacement incentral Utah. Research Bulletin 516. Logan, UT:Utah Agricultural Experiment Station and Utah StateUniversity.

Richards, J. H. 1984. Root growth response to defolia-tion in two Agropyron bunchgrasses: field observa-tions with an improved root periscope. Oecologia64:21-25.


Stubble Height and Functionof Riparian Communities

Quentin D. SkinnerDepartment of Rangeland Ecology and Watershed Management


AbstractStubble height, a measurement of remaining veg-etation, will be discussed as it influences attributesof channels and floodplains, sediment deposition,plant vigor, physical stability of riparian zones, anduse of woody plant species. Because vegetation inchannels causes sediment deposition, water rela-tionships of plants growing along banks may changewith time and channel succession. Consequently, aschannel banks build, plant species composition ofthe near-bank riparian zone also may change.

Evidence will be presented that plants requiringhigh water tables most of the growing season maybe at a disadvantage in competing with plants thatcan withstand a rapid decline in the water table; and,therefore, the latter may become dominant along amature channel system. Furthermore, stubbleheight required to maintain plant vigor also maydecrease when plant succession occurs. However,higher stubble heights of all plants may be requiredto maintain channel bank integrity and to reducegrazing of other desired woody plants during spe-cific seasons. In conclusion, when stubble height isused as an indicator of when to manipulate herbi-vores' grazing, other important factorsflood fre-quency, soils' draining capacity, plant regrowth,and plant dormancyshould be considered andrelated to how they and stubble height standardsmay alter specific management objectives.

IntroductionLivestock and wildlife grazing on western range-land is essential to maintain the economic andsocial values now realized by the livestock industryand general public. However, livestock grazingimpacts on riparian zones are most often cited as

reason to change how public lands are managed asa natural resource (U.S. GAO 1988). Kauffman andKrueger (1984) review and illustrate the impor-tance of riparian zones and how livestock grazingmay impact this wetland resource.

Because livestock utilize vegetation as forage, it iscommon practice to measure utilization to evaluatehow livestock grazing alters attributes of riparianzones. The impact of grazing on plant communi-ties, however, may be estimated in two ways: byestimating the amount of vegetation that has disap-peared by being consumed (utilization); or, bymeasuring the amount of vegetation that remainsafter herbivores have used an area (stubble height).Clary and Webster (1989) address the latter methodof evaluating utilization and support it as a validway to determine when to stop grazing riparianhabitat. In part, this paper endorses Clary andWebster's school of thought because stubble heightis a measure of remaining vegetation and can beseen by the observer. However, like Hall andBryant (1995), this paper also addresses howstubble height relates to function of riparian zones.

In particular, the overall goal of this paper will beto help illustrate that measuring stubble height ofriparian plants can serve as a viable plant commu-nity attribute for helping manage riparian zones.

Specific objectives toward reaching this goal willbe to show:

How sediment deposition alters stream channeland plant succession and, then, how stubbleheight relates to sediment deposition; and

How plant vigor, stream channel characteris-tics, and plant community stubble heightrelationships relate to each other so that land


managers can further evaluate a riparian zone'scapacity to function while being grazed by allclasses of herbivores.

BackgroundChaney et al. (1990) generally address function ofriparian zones. Their work indirectly suggests thatstream channels may go through a successionalsequence from unincised to incised. The authorssuggest that unincised is functioning properly as ariparian zone but the incised is not. This school ofthought has been extended by the Bureau of LandManagement (BLM) to the point that a rating offunctional condition is a way to evaluate riparianzones within lands they manage (USD1 1993).

The BLM manual "Riparian Area Management.Process for Assessing Proper Functioning Condi-tion" suggests that stream channels may movethrough a successional sequence from bare groundto late seral (USD1 1993). The manual also sug-gests that as stream channels move through succes-sion, associated plant community composition maychange until the late seral stage. According to thisassessment, the late seral channel condition andexpected associated plant communities constituteBLM's definition of properly functioning riparianzones. Unfortunately, this as-sumption does not recognize thatthe basic function of a stream is 1996).to remove water and sedimentfrom its drainage basin. Instead, itaddresses only the deposition halfof the natural and cyclic processof erosion.

This paper, in contrast, followsCooke and Reeves (1976) whoshow that arroyos in the south-western United States form andfill in a cyclic process. Thus, it isassumed that in the erosion cycle,sediment deposited at one timeeventually may be removed andtransported farther down thestream. Also, different stages inthis cyclic process may occuralong the course of any onestream at any given time. Thus,

Figure 1. Stubble height and sediment deposition (after Clary et al.

Tall

Plants

Deposited

Sediment

all stages in channel succession may be functioningproperly when the potential for each successionalstage is considered.

Given these facts, measured stubble height is notjust a way to estimate utilization by grazing ani-mals; it also has great value in evaluating how wellvegetation and grazing management meets anobjective of moving from one stage in channelsuccession to another.

Because it appears that water and sediment trans-port in part may create the successional sequencethe BLM uses (USD1 1993) to evaluate riparianzone condition, this paper builds on that thought byaddressing how sediment deposition and erosionrelate to channel succession and, then, plant succes-sion. The purpose of this approach is to illustratethat available soil water may vary for differentstream channel configurations and therefore maybe responsible for changing riparian zone plantspecies composition.

Stubble heights of individual species may alterfunctions of riparian zones in different ways. Thispaper discusses the influence of stubble height onsediment deposition, plant vigor, physical impactsfrom grazing activity, and animal foraging behavior.

0.5 inches

MoreSediment

LessSediment

Short

Plants

Retained

Sediment

Tall

Plants


Stubble Height andChannel and Plant SuccessionAssuming that vegetation remaining on floodplains (stubble height) can filter and stabilizesediment, then channels should move through asuccessional sequence from incised to unincised(Chaney et al. 1990) or from early seral to late seral(USD1 1993). The relationship between riparianzone function and stubble height is that stubbleheight should alter how channels move from a

degraded or empty state ofsediment storage to a statewhere sediment is being stabi-lized and channel banks reachstability or maturity (Chaney etal. 1990, USD1 1993).

This paper accepts that channelsuccession occurs becauseevidence shows that vegetationcauses sediment deposition andthen stabilizes it. However,evidence also suggests thaterosion removes stored sedi-ment at a point in space andtime when storage is greaterthan the system's capacity tohold it in place. This is not newinformation but follows Cookeand Reeves' (1976) review ofsediment cut and fill cyclingwithin arroyos. The process ofremoving stored sedimentdeposition from channels andflood plains appears to be caused by stream chan-nel meandering and slope adjustment (Leopold etal. 1964, Morisawa 1968, Schumm 1977, andHeede 1980). The implications of cut and fillcycling of stored sediment is that channels systemsdegrade and fill within an erosion cycle. As streamchannels "cut" and "fill," there may be correspond-ing change in plant communities. Because manag-ers of riparian zones recognize and judge conditionof riparian zones, in part by plant species composi-tion, it is imperative to discuss how and whychannel and plant succession relate to each other.

As sediment deposited on flood plains increases theheight of channel banks, the soil surface moves

'VKentucky bluegrass

Kentucky bluegrass

Matiir

higher and farther away from the water table duringlow streamfiows. As distance to groundwaterincreases, water-loving plant species will be re-placed by those more resistant to drier conditions.It also can be assumed that during periods of floodplain saturation, the rate and depth of drainage willvary with the soil attributes of the sediment mate-rial stabilized in the flood plain and channel bank.Therefore, it is likely that plant species compositionalso may vary depending on the riparian zone's

Figure 2. A hypothetical sequence illustrating channel and plant suc-cession assuming erosion is a cyclic process.

Plant and Channel Succession

Degraded

K. bluegrass 7'Tufted hairgrass

Sedge

Water Table

drainage characteristics and not just according tothe plants' relative positions next to stream chan-nels. If differences in plant and water table relation-ships or in drainage and the soil's water-holdingcapacity determine plant species composition, thenthe potential for plant stubble to cause sedimentdeposition during flooding also may vary.

To illustrate this point, Clary et al. (1996) made aflume study in which treatments controlled streamflow, sediment supply, and stubble height of flexibleand rigid herbaceous vegetation (Figure 1). Flexiblevegetation (Kentucky bluegrass and native sod)maintained at 0.5 inch high caused more sedimentdeposition than higher stubble and more than rigid


plants tested. Moresediment was depos-ited using 3-inchrigid vegetation(sedge and cornseedlings) than usinghigher rigid vegeta-tion. When test plotswith deposited sedi-ment were subjectedto flooding trialswith no sediment,the 0.5-inch flexiblevegetation lost itscaptured sedimentrather fast at thebeginning of theflood compared torigid vegetation, butflexible vegetationhad a higher net gainat the end of thetrials. No time wasallowed for vegeta-tion to regrowbetween the periodof deposition andrepeated flooding.

Considering evidence presented by Clary et al.(1996), the function and significance of usingstubble height to evaluate riparian zone manage-ment objectives is that the importance of agencystandards may have to be altered as plant speciescomposition changes from rigid to flexible or fromflexible to rigid in relation to different stages inchannel succession. Figure 2 provides an exampleof stream channel configurations observed in fieldsurveys and dominant plant species that often occurwith them. The drawings illustrate that sedimentdeposition not only alters channel configuration butalso that Kentucky bluegrass may replace sedgeand tufted hairgrass as banks build high above thewater table or as the channel incises during thedegradation part of the erosion cycle.

Henszey (1993) explored the relationship of ripar-ian plant communities to depth-to-groundwaterlevels over an 8-year period (Figure 3). He found

Figure 3. Water table and riparian zone plant community relationships (afterHenszey 1993). Percentages are the amount of time the water table level is abovethe ground surface, at 0.5 m deep, and below 1.0 m deep.

Water Table

Surface

43-59%

K. bluegrass

Duration and Watertable level

#' --I25-46% 21-49%

21-56% 26-39% 3%

Tufted hairgrass Sedge Sedge

K. bluegrass Tufted hairgrass

that the Wet Meadow type supported mostly a tallsedge plant community, and the MoistWetMeadow type supported tall and short growingsedges and tufted hairgrass. Although both commu-nity types were flooded from 21 to 49% of thetime, drainage to a water table deeper than 1 meterduring the summer occurred only 3% of the time inthe Wet Meadow type and 26 to 39% of the time inthe MoistWet Meadow type. Drainage to a watertable deeper than 1 meter increased in the MoistMeadow type to 21 to 56% of the time and to 43 to59% of the time in the Dry Meadow type. TheMoist Meadow type supported mostly tufted hair-grass and Kentucky bluegrass , the Dry Meadowtype mostly Kentucky bluegrass. This study sug-gests that depth to groundwater and rate of drain-age or drainage and plants' use of soil water maycause a change in plant species composition alongstream channels as they move through a succes-sional sequence from degraded to mature, aspresented in Figure 2.


O.5ni

1.Om

To further illustrate that sediment deposition mayfavor change in plant composition, Henszey's(1993) plant communities can be placed on ahypothetical degraded straight reach or a buildingpoint bar of a meandering stream, as in Figure 4.Tall sedges often occupy the area close to the low-flow level of this type of stream channel configura-tion. Based on the concept presented in Figure 3,the other plant communities would fall into placeas the depth of the bank material increased over theestablished water table away from the channelitself. Kentucky bluegrass would occupy the driestsite of the riparian zone. As vegetation deposits andstabilizes sediment on the lower portions of thebank, it is reasonable to assume that the degradedbank could eventually reach the channel profileshown in Figure 5. The resulting depth to watertable may cause sedges and tufted hairgrass todecrease and Kentucky bluegrass to increase.

The plant communities that Henszey (1993) studiedare typical of many found in the montane zones ofthe central Rocky Mountains. It is common to findKentucky bluegrass between the dry uplands andthe wetter areas of riparian zones. The argument ismade often that the presence of the introducedKentucky bluegrass in riparian zones, like Henszey(1993) studied, indicates poor grazing practices,and grazing management can influence the expan-sion or reduction of this bluegrass zone. Commonremarks to support this management philosophysuggest that improper grazing may dry out moistarea of riparian zones favoring Kentucky bluegrassover other, more mesic grass species. Also, because

Figure 4. Hypothetical relationship between channel configura-tion and plant succession.

Water Table

I

Kentucky bluegrass is rhizomatous and tuftedhairgrass is a bunch grass, tufted hairgrass may notcompete well for space while being grazed in thepresence of Kentucky bluegrass. Little thought hasbeen given to the arguments that even thoughKentucky bluegrass may have been introduced toNorth America, it is where it should be, and in factit probably is persistent when riparian zones fillwith sediment and soil water dynamics favor itsexistence. In situations like these, presence ofKentucky bluegrass may not be related in any wayto improper grazing but rather to this species'adaptability in competing for soil water with othernative species within certain stages in channelsuccession of riparian zones.

In this paper, the discussion of ecology and man-agement of Kentucky bluegrass verses the presenceof native grasses is much less important than theneed to recognize that Kentucky bluegrass stubbleheight should be managed so that other riparianzone values can be maintained or altered if needed.

Henszey's (1993) data provide a strong argumentthat depth to water table favors one plant commu-nity over another, including the Dry Meadow typewith Kentucky bluegrass. However, some believethat water's capillary rise from the water tableshould be sufficient in stream channel configura-tions like those illustrated in Figures 2, 4, and 5 tosupport most plants common to the wet and moistareas of riparian zones.

To address this question, Yeager (1996) studiedplant growth in sand-filled columns 10 feet tall by

4 inches wide by 1 inch deep. Re-verse column flow and outlet reser-voir control was used to test responseof Nebraska sedge, tufted hairgrass,and Kentucky bluegrass at watertable decline rates of 1, 2, 4, and (5centimeters per day. Average watertable decline rates for Henszey's(1993) 8-year field site study werebetween 2 and 4 centimeters per day.Plants were established in the col-umns and grew rapidly while waterlevels were maintained at the surfaceof

the columns before testing began.Sand was used to minimize water's


capillary rise. Pertinent results are shown in Tables1 and 2. The length and depth of roots Yeager(1996) could see through the clear plastic front ofcolumns suggest Kentucky bluegrass is moreaggressive in growing roots which then may keepup with a declining water table. It is assumed thatcapillary rise in sand was not sufficient to supplyrequired water for growth because at all declinerates all plants died after 30 days. For all species,the ash free weights (Table 2) confirm observationsand again suggest that Kentucky bluegrass may bemore adapted than the two other species tested forgrowing in soils where ground water is deeper ordrainage is rapid.

In summary, Yeager' s (1996) data suggest thatcapillary rise from water tables that decline rapidlyunder riparian zone plant communities may notalways be sufficient to provide water required byall plant species. These data also suggest that soilparameters of sediment that builds channel banksand that alters water's capillary rise may be ex-tremely important in explaining where plant spe-cies grow and why.

Yeager's (1996) study also helps explain whyHenszey (1993) may have found sedge, tufted hair-grass, and Kentucky bluegrass in different locationswithin riparian zones, and why distinct plant com-munities may exist away from a stream's edge touplands. By these arguments, it does appear onecan use channel succession to predict plant commu-nity. It also appears reasonable to use stubbleheight data to manage grazing so that one can setan objective to move from one channel configura-tion to the next as illustrated in Figures 2, 4, and 5.Accomplishing this likely will depend on manag-ers' ability to determine which stubble heightspromote sediment deposition and retention so thatbanks can build while plant vigor is maintained.

Stubble Heightand Sediment DepositionIn review, streambank vegetation has been shownto increase channel roughness (Beschta and Platts1986, Gregory et al. 1991). increased channelroughness in turn may dissipate stream energy andcause sediment deposition (DeBano and Heede

Table 1. Maximum visible root depth and lengthat a water table decline rate of 4 cm per day(Yeager 1996).

Note: All plants were dead after reaching the maximum rootdepth after 30 days.

1987, Debano and Schmidt 1989). Vegetation maysort sediment by size and retain it along riparianzones (Lowrance et al. 1985, Platts and Rinne1985, Beschta and Platts 1986, Middleton 1993).Established vegetation roots appear to bind soil andstabilize stream channel banks (Gebhardt et al.1989, Lowrance et al. 1985, Elmore and Beschta1987, Chaney et al. 1990). Vegetation also hasbeen shown to increase waterway protection bycovering banks and slowing down erosion (Chow1959). However, tall grass may not provide muchflow resistance when it lies over in the direction offlow as streamfiow velocity and channel depthincrease (Haan and Barfield 1978).

There is little doubt that sediment deposition andvegetation are factors in building and stabilizingstream channels. The question is, how can one bestmanage riparian zones to accomplish the task, andhow can stubble height be used to monitor theeffort? The flume study of Clary et al. (1996)discussed above and illustrated in Figure 1 suggeststhat the vegetation should be short and that flexiblegrass like Kentucky bluegrass is better than longer

Table 2. Root ash free weight for 10-cm incrementsat a water table decline rate of 4 cm per day(Yeager 1996).


Depth

I-- øQ)ZU)

Root ash free weight (g)

0-10 117 43 108

10-20 30 27 10

20-30 5 2 0

30-40 2 0 0

SpeciesRoot

depth (cm)Total root

length (cm)

Kentucky bluegrass 31 330

Tufted hairgrass 20 183

Nebraska sedge 11 189

U)U)

U)cew

a, , - D)

and more rigid grasslike plants forcausing sediment deposition. Herbivory,then, at some level should benefit theprocess of sediment deposition andchannel bank building.

Rumsey's (1996) field study of stubbleheight and sediment deposition alongSpring Creek, WY is a logical step fromClary et al.'s (1996) laboratory flumeexperiments to learn if what they foundapplies to a modified stream system.Rumsey was able to maintain somecontrol over direction of flow because the studywas in an urban stream that had been channelizedin a straight, flumelike configuration, and the low-gradient channel bottom was maintained by perma-nent structures. Flood plains and vegetation hadbecome established within the channelized reachbefore the study began in 1994, but she was notable to control streamfiow and sediment supply.

Rumsey found no significant difference in sedi-ment deposited over 2 years when stubble heightswere maintained unclipped (mean = 24 inches) andat 1-, 3-, and 6-inch heights.

Amount of sediment deposited correlated with anddecreased as floodplain elevation increased abovethe level of low streamfiow. The average sedimentdeposited was 0.2 to 0.4 inch after 2 years. Duringthe third year of the study, sediment depositedincreased to about 1 inch deep (Gray et al. 1997).Very few bank-overflows occurred the first 3 years,and none were considered extreme floods. Duringthe fourth year, sediment deposition increased tobetween 1.2 and 3 inches deep, and several con-secutive and extreme floods occurred throughoutthe summer (Gray 1998, personal communication).The 1-inch stubble height sediment deposition was1.2 inches deep; the three other treatments were inthe range of 3 inches deep. Sediment retained on 1-inch stubble was significantly lower than that depos-ited on unclipped and on 3- and 6-inch heights,which were not significantly different from eachother. In the year of frequent floods, sedimentdeposited again correlated to floodplain elevationabove the level of low streamfiow; but unlike whatRumsey (1996) found, sediment increased withbank elevation above low streamfiow.

Figure 5. Hypothetical change in channel configuration andplant species composition caused by deposited sediment andbank building.

K. bluegrass K. bIuegrss

( ; 'A1l aàdA/.4i4 I//i.

Wter TaIe

In comparing Rumsey and Gray's work with that ofClary et al. (1996), extreme and repeated floods inSpring Creek were required to separate differencesin stubble height's ability to cause sediment deposi-tion. As in Clary et al. (1996), very short stubbleheight along Spring Creek appeared to causesediment deposition, especially in years whenflooding was not frequent and extreme. During thelast year of repeated, strong floods, very shortstubble either did not collect sediment as well aslonger stubble, or it did not retain deposited sedi-ment during repeated floods. Clary et al. (1996)show that short stubble collects sediment, but theymake the strong point that this deposition may belost it if it is not stabilized by growing or longerplant material (Figure 1).

Rumsey and Gray's work supports Clary et al.'s(1996) results and discussion. Spring Creek fieldsite data also suggests that differences betweenlonger stubble heights may not make a significantdifference in the overall process of altering sedi-ment deposition. Again, results from both studiesshow that moderate grazing of vegetation may notbe a significant consideration in managing sedi-ment within riparian zones.

In a naturally meandering and high-sediment-transport stream system (Muddy Creek, WY),Goertler (1992) found that in 6 years the existingvegetation on banks and channel conditions did notcause a reduction in suspended sediment in years ofhigh flow but did in years of low flow. Middleton(1993) found that particle-size classes of sedimentchanged with distance away from Muddy Creek'sactive channel, and he related this sorting to de-creased streamfiow power as flooding receded.


Budd (1994) reported that at least 1 foot of sedi-ment was deposited over 6 years in places likepoint bars during Muddy Creek floods and thatoverall change in sediment deposited was notsignificant among vegetation types, straight-streamstudy segments, or years. Apparently, amount ofsediment deposited on flood plains varies withstream channel configuration, amount ofstreamfiow, and sediment supply. In total, resultsof Muddy Creek studies suggest that amount ofsediment deposited because of the condition offlood plain vegetation may be insignificant com-pared to sediment deposition caused by the channeland streamfiow attributes.

In summary, the examples provided above illustratethat vegetation can filter sediment from stream-flow. It may be that shorter stubble can filter more,or at least as much, sediment than taller or ungrazedvegetation. However, very short stubble also maylose sediment deposited from one flood event to thenext if it is not stabilized by growing or tallerstubble.

There may be differences among plant species'ability to filter sediment depending on their flex-ibility. However, measure of vegetation attributesmay address only a small portion of the forces thatshape stream channels. Depending on sedimentload and flood flows, stubble height and the graz-ing that creates it may play only a small role in thesediment deposition phase of bank building. How-ever, because stubble also is responsible for stabi-lizing any sediment deposited, through the produc-tion of root and aboveground biomass, it is impor-tant to explore how stubble height relates to themaintenance of plant vigor.

Stubble Height and Plant VigorPlant vigor can be measured in many ways. Onecommon measure is the change in the relationshipof underground biomass to that produced aboveground in any one growing season or in consecu-tive growing seasons. Roots are assumed importantin creating stable streambanks, and any stubbleheight that depresses root production may affectbank stability and function of riparian zones. It hasbeen shown also that grazing animals can consumeaboveground biomass to such an extent that under-

ground biomass can decrease compared toungrazed controls (Caldwell et al. 1981, Richardsand Caldwell 1985, Engle 1993). A reason oftenprovided is that in these grazing situations, theplant uses its excessive carbohydrate root reservesto support growth of aboveground plant parts at theexpense of growing roots needed to maintain plantvigor. However, recent literature suggests thatphotosynthetic material left after defoliation, andgreater reallocation of the photosynthates producedfrom stubble left to the aboveground shoot system,is the likely reason root biomass and carbohydratestorage decrease under long-term, intensive grazing(Detling et al. 1978, Nowak and Caldwell 1984,Richards and Caldwell 1985, Olson and Richards1988, Briski 1986, Briski and Richards 1994).

This recent research offers strong support formeasuring stubble height to monitor grazing effectson plant vigor in riparian zones. Instead of measur-ing and estimating the plant parts that have disap-peared (utilization), measuring stubble height givesthe opportunity to ensure that enough photosyn-thetic material is left after grazing to promote short-and long-term plant vigor. As important, sedimentdeposition also may influence the plant's ability toregrow after floods because the photosyntheticmaterial is covered. Therefore, measuring stubbleheight also provides the opportunity to see howdeposited sediment alters plant vigor. In riparianzones, it is reasonable to assume that grazing andflood sediment will alter plants' production poten-tial. However, one must also consider that, inriparian zones, stubble regrowth potential may begreater than in surrounding uplands, If so, theremay be more flexibility in managing grazing inriparian zones than on uplands.

Generalized curves showing annual cumulativeaboveground biomass for grass plants like thosegrowing in Wyoming are illustrated in Figure 6. Inthe uplands, grass plants must grow to the mature,flowering stage using limited soil water providedby winter and spring precipitation. In contrast,grasses in riparian zones may have access to suffi-cient soil water to support growth into late summerand fall. Both upland and riparian grasses mustbuild up photosynthetic leaf material in spring anduse it to replenish any root reserves lost in support-


Figure 6. Generalized cumulative aboveground biomass curves forcomparing upland and riparian cool season grasses and therelative time a plant may have to regrow after being grazed.

Water Supply Limited Water Supply Not UmitedLight Not Limited I Ught Not UmitedTemperature Not Umited (Temperature Not Limited

Time

sQ

Riparian Zone

Time

ing the early growing process and, then, to supportgrowth until the plant reaches maturity (Richardsand Caidwell 1985). However, because soil wateris generally a limited resource for plants growing inuplands, the time they have to reach maturity gen-erally is less than in riparian zones. Most uplandsgrasses mature by late spring or early summerwhen their soil water supply generally is depletedand dormancy ensues. Grasses in riparian zones,without the influence of grazing, also follow thisgeneralized upland growth pattern, but the maturegrass may stay greener longer than in uplandsbecause of it has an extended soil water supply.However, when riparian grasses are subjecte todefoliation or flood sediment, they appear able tocontinue growing and able to mature into late sum-mer and early fall. It is this potential for regrowthover an extended period in riparian zones thatmakes measuring photosynthetic material left aftergrazing and deposition of flood sediment so impor-tant to maintaining plant vigor. It also is a verygood reason managers should consider measuringstubble height, not utilization, to evaluate grazingeffects on riparian plants.

In using stubble height as an indicator of change inplant vigor, managers must determine what amountof plant material is needed to manufacture suffi-

cient photosynthates so whole-plant needs are met after defolia-tion. However, accomplishing thistask is complicated and part of thehistorical mystery that inspires thestudy and profession of rangemanagement. Evidence suggeststhat when plants are defoliatedduring periods of acceleratedgrowth, they may be stressed morethan when grazed during early andlate periods, when growth isslower, because during activegrowth the leaf, sheath, and stemmeristems are being elevated andare more susceptible to grazing(Olson and Richards 1988). How-ever, it is important again to pointout that in riparian zones the time aplant has to recover after beinggrazed should be longer than in

uplands because of the lasting soil water supply.Therefore, the extent of impact on the plant and itsability to re-grow after being defoliated when it israpidly growing likely will depend on the amountand distribution of material left (stubble height) andnot on what has been consumed (Olson andRichards 1988).

Briske and Richards (1994) stress that we still havemuch to learn about how grazing modifies plantregrowth. Most research has focused on answeringregrowth questions using single plants and indi-vidual species. It is clear, from their text, that notall grass species respond in a similar way; growthform and ability to withstand competition may varybetween species and environments studied. Towork within this problem, rangeland managersgenerally have followed the "take half, leave half"rule of thumb (Criddr 1955) after finding thatupland range was ready for grazing. Range readi-ness usually has been defined as the time at whichgrass shoots are just ready to flower or in the earlystage of flowering. This means grazing on uplandsgrass at the middle to end of leaf, sheath, and stemmeristem elongation and when soil water supply isabout gone. Potential for managing upland plants'regrowth is, therefore, poor in reality. In riparianareas, if stubble height is sufficient to provide for


photosynthetic activity, then regrowth potentialafter any grazing or sediment deposition is good,and the "take half, leave half' principle historicallyused for uplands may not apply. In fact, dependingon the riparian plant species measured, stubbleheight after taking half by weight may underesti-mate the volume of photosynthetic material left formaintaining plant vigor even when regrowthpotential is ignored. Kinney and Clary's (1994)work illustrates this point (Figure 7).

In Figure 7 are Nebraska sedge, Kentucky blue-grass, and tufted hairgrass, three important plants inriparian zones of the West. Sedge and grass havedifferent growth forms. Herbage weight of leavesand shoots of Nebraska sedge is somewhat evenlydistributed from the plant's crown to its top. Incontrast, most herbage weight of Kentucky blue-grass and tufted hairgrass is near the ground sur-face as leaves; only a small portion of the overallbiomass is elevated as stems and flowers. It appears,then, that grasses with growth forms like Kentuckybluegrass and tufted hairgrass can be grazed muchshorter than plants like Nebraska sedge when "takehalf, leave half' of the total herbage weight is themanagement criterion for maintaining plant vigorafter defoliation. This is because Kentucky blue-grass and tufted hairgrass still have about 50% oftheir biomass weight (mostly as leaves) when theirstubble is between 2 and 3 inches tall, but Nebraskasedge stubble is about 6 inches tall when all matureplants are 18 inches high.

However, this does notmean in either case thatstubble height is equiva-lent to the photosyntheticarea needed to supportregrowth and to maintainplant vigor. In fact, evenif Nebraska sedge weregrazed to 2 inches highand Kentucky bluegrassand tufted hairgrass to 1inch high, it does notmean that stubble wouldnot have enough photo-synthetic area to supportregrowth. One can only

Figure 7. Illustrated height of remaining vegetation when 50% of theaboveground total weight is removed for three 18-inch-high grasslike coolseason plants.

10

assume that if they were grazed this short they mayjust need more time to recover to the point that theymaintained plant vigor. In all cases mentioned,stubble heights presented for the three plants arelikely to cause sediment deposition, but photosyn-thetic requirements for supporting regrowth afterflooding are not known.

There is no reason to believe that less than 50% ofthe weight of a grass plant is not sufficient tomaintain vigor in riparian zones. This may beespecially true if we consider how animals grazegrass under light to moderate use. Grazing animalsusually do not graze plants evenly. They generallyuse only part of individual plants and, thus, leaveparts ungrazed. The combination of defoliated andundefoliated plant parts after grazing may evenincrease light near the crown, promote greaternutrient uptake in roots, increase photosyntheticrates in defoliated leaves, sheathes, and shoots, andinitiate activity of leaf primordial meristems andtillering from auxiliary buds (Olson and Richards1988, Briske and Richards 1994). This may beespecially true if light, temperature, water, andnutrients are not limited. Of all watershed habitat,then, the wet and moist zones of riparian areasshould best promote regrowth under light andmoderate grazing and maybe even under occasionalheavy grazing.

To address the issue of stubble height and plantvigor, Rumsey (1996) used total aboveground andunderground biomass as indicators of change when

0 Percent Weight RemovedAfter Kinney and Clary 1994

0 Percent Weight RemovedAfter Kinney and Clary 1994


Height of Plant Stubble at 50% of Total Blomass

K. bluegrass 3' Nebraska sedge 6"

18 inches Tufted hairgrass 2' 18 inches

plant heights were maintained at 1, 3, and 6 inchesalong Spring Creek, WY over a 2-year period.Each treatment was clipped weekly when plantswere rapidly growing so that stubble heights didnot increase much between clippings. When plantsgrew slowly toward the end of the growing season,they were clipped biweekly. Each treatment wascompared to unclipped controls.

A purpose of the study was to see whether veryshort stubble heights would cause a decrease inroot production and, thus, potentially increasestreambank instability; also, whether stubble heightdifferences produced differences in abovegroundproduction compared to unclipped controls.

Rumsey (1996) reported no significant differencesin underground biomass between the control andclipping treatments after 2 years. Total abovegroundbiomass was not significantly different between theunclipped control (mean = 24 inches high) and the1-inch stubble. However, both of these producedsignificantly more aboveground biomass than the3- and 6-inch treatments. The 3- and 6-inchstubble's aboveground biomass was not signifi-cantly different from each other. Gray (1998,personal communication) has analyzed Rumsey(1995-96) and Gray et al. (1997-98) data. He foundthat after 4 years, the unclipped control plots pro-duced significantly more total aboveground biomassthan each of individual clipping treatments. How-ever, the 1-inch treatment produced significantlymore aboveground biomass than the 3- and 6-inchtreatments, which were not different from eachother. Gray has not completed analysis of under-ground biomass production after 4 years.

These data suggest that maintaining all plants at 1,3, and 6 inches over 4 years will reduce totalaboveground production of plants growing alongSpring Creek. The data also suggests that grassmaintained at shorter stubble heights may producemore total aboveground biomass than the longerstubble when subjected to these extreme clippingtreatments. However, caution should be taken inapplying Rumsey (1996) and Gray et al. (1998)data. Although production of total abovegroundbiomass did decrease for all clipping treatments,clipping was extreme compared to how animalsgenerally graze riparian plants under light or

moderate use. Generally, under light and moderateuse they take only a part of an individual plant,leaving the rest to help provide the photosyntheticneeds for regrowth. Also, the response of indi-vidual plant species was not studied. Therefore,there is always the possibility that plant speciescould have been lost and that less abovegroundbiomass could have been produced by one or a fewspecies more tolerant to clipping. An argument alsocan be made that aboveground production would beeven less if plots used for data collection weresubjected to grazing, because simulating grazing byclipping does not account for any physical impactto plants caused by animal activity.

Rumsey's (1996) data do suggest, however, that inwet and moist riparian areas like Spring Creek,sporadic grazing that reduces stubble height to lessthan 50% by weight of the ungrazed plants may notpermanently damage plant vigor as measured bytotal above- and belowground production. Gray's4-year data, however, suggest that extreme grazingmay reduce production of total abovegroundbiomass. The answer to the question of how these4-year clipping treatments alter total belowgroundbiomass and, thus, potential for maintainingstreambank stability is nearly complete.

To summarize the discussion of plant vigor, stubbleheight does represent plant photosynthetic area.Stubble photosynthetic area then can be maintainedat levels capable of producing the photosynthatesnecessary to maintain plant vigor under grazing. Ifmoderate grazing is defined as taking half andleaving half of the plant's capability to produce atotal amount of aboveground biomass, then stubbleheights after grazing will vary because plants indifferent riparian plant associations may have dif-ferent growth forms. Stubble height does not repre-sent the plant's photosynthetic ability to supportregrowth and plant vigor. Generally, using 50% ofthe aboveground biomass as an indicator of main-taining plant vigor likely will underestimate thephotosynthetic area remaining to support regrowth.

Much has been said about differences in stubbleheight of Kentucky bluegrass, tufted hairgrass, andNebraska sedge. Because of growth form, 50% ofthe total aboveground production results in a muchshorter stubble for Kentucky bluegrass and tufted


hairgrass than for Nebraska sedge (Figure 7).Taking it one step farther, the differences in stubbleheight between these plants and potential forregrowth after grazing or a flood also may beassociated with recognized stages in channel andplant succession. Figure 8 combines the conceptsdiscussed earlier for soil water storage potential(Figure 6), total accumulated biomass (Figure 6),and channel and plant succession (Figures 2, 4, and5). The combination of these concepts may be usedto help predict where short stubble height is likelyunder moderate grazing and how regrowth poten-tial may vary because of available soil water.

Under the protocol of leaving 50% of abovegroundplants in place as stubble, the expectation may bethat in the Kentucky bluegrass zone abovegroundbiomass will be short and the potential for regrowthmay be limited. This is because it generally occu-pies the driest zone of the riparian area, and there-fore it does not have much time to go through itsgrowth cycle before its water supply is depleted.The expectation may be that tufted hairgrassstubble, like Kentucky bluegrass, also will be shortbecause these plants have like growth forms.

However, regrowth potential for tufted hairgrassshould increase because it occupies a wetter area ofthe riparian zone and therefore, it has more timeand soil water to grow. To leave an equivalentamount of biomass weight for Nebraska sedge, thestubble of the tall sedge zone will have to be higherthan Kentucky bluegrass and tufted hairgrass.However, expectations should be that the sedgeplants of this zone should have the greatest amountof time and soil water to recover losses in photo-synthetic area and regrow to maturity.

Evidence presented generally suggests that moder-ate grazing does leave sufficient photosyntheticarea for maintaining plant vigor. Also of interest ishow stubble height can be used to monitor animalforaging behavior and the associated impacts to thephysical stability of riparian zones or the use ofother desirable woody plant species as suggestedby Hall and Bryant (1995). Using stubble height tomonitor grazing impact on these riparian zoneattributes may be more important to the managerthan it is for maintaining stubble heights thatsupport plant vigor and sediment deposition.

Stubble Heightand Physical Attributesof Riparian ZonesGrasslike vegetation has been shown to be impor-tant for maintaining the physical integrity of stream-banks (Zonge et al. 1996, Zonge and Swanson1996). Grazing by large herbivores may cause hoofdamage by using these resources (Buckhouse et al.1981, Kauffman et al. 1983, Marlow and Pogacnik1985, and Myers and Swanson 1992). Physicaldamage may vary depending on season of use andsoil moisture. More physical damage occurs whensoils are wet (Marlow and Pogacnik 1985), butamount of animal use may be less in wet areaswhen surrounding upland vegetation is green inspring and early summer (Kauffman et al. 1983,Clary and Webster 1989, Clary and Booth 1993,Hall and Bryant 1995). To reduce impact to stream-banks and channel characteristics during grazing,Hall and Bryant (1995) recommend managers:

Pay attention to the stubble height of the mostpalatable species as height approaches 3 inches;

Note when stubble height moves from 3 inchesto less than 1 inch; and

Keep track of the greenness of the most palat-able species and, when greenness diminishesand the plants appear to dry, look for animals toseek greener vegetation.

Their advice is excellent and illustrates the wisdomand field experience of the authors.

This paper provides evidence that there are twodistinct areas in riparian zones where stubble heightmay differ assuming that a grazing managementphilosophy of using 50% of available productionwill be enough to maintain plant vigor. These areasare the tall sedge and rush zone in wet areas and thetufted hairgrass and Kentucky bluegrass zones inthe moist to drier areas. It appears that stubbleheights also can be used to predict the extent ofgrazing effects on the physical integrity of riparianzone attributes and the potential for the animal'sforage preference to switch to woody plants.

Excluding vegetationsoil surface compaction,physical damage to riparian zones is likely whensoils are wet and soft and animals' hoofprints are


heavy enough to break through thevegetationsoil surface. Physicalimpacts also include animals' breakingoff or shearing soils of overhangingbank edges along stream channels.Area surrounding ponds, bogs,springs, and lake shores are examplesof where soils may be wet and soft allor parts of a year. The greenlinethatspecific area where a more or lesscontinuous cover of vegetation isencountered when moving away fromthe center of an observable channel(Cagney 1 993)represents the gen-eral area of concern along streams.Figures 2, 4, and 5 illustrate that wetand soft greenline extents vary withchannel succession.

The zones occupied by tall sedges andrushes present a clear and concise image of wherephysical damage may be more than likely. Also,tall sedges and rushes usually are the last grasslikeplants to be defoliated to any great extent whenlivestock graze riparian zones.

Physical impact and grazing vulnerability of thesewet areas, however, may fluctuate with season ofuse and drought in any single season of the yearbecause wet area soils may dry and become firm.Plants like Kentucky bluegrass and tufted hairgrassshould be using their soil water supply as uplandplants do, which should aid drying and increase thestrength of riparian soils above declining watertables at the interface between uplands and wetzones (Figure 8). Therefore, stubble height stan-dards that predict when damage begins may have tobe flexible depending on whether soils are wet andsoft or dry and firm. By monitoring establishedstandards within the transition zone between thetaller grasslike plants of wet soils and those in therest of the riparian zones, information gathered canprovide a first level of protection against physicaldamage to streambanks and other sources of water.Also, these data can be correlated to hoof imprint-ing and bank damage that exceed acceptable levels.

The greenlines of stream channels that support tallsedges and rushes generally are easy to define inthe early stages of channel succession; therefore,

Figure 8. A generalized illustration of where Kentucky bluegrassand tufted hairgrass usually grow. Where they grow, one mayexpect short stubble height after grazing.

Plant Growth and Channel and Plant Succession

Time

Water TableI I I

20(U

a-

Wetlime

2'following the stubble height monitoring protocoldescribed above is appropriate. However, whenchannels reach maturity, the greenline often isdifficult to differentiate from the remaining plantcommunities of the riparian zone, and tall sedgesand rushes may be absent along the channel edgefor reasons discussed earlier in this paper. Conse-quently, the lack of a tall sedge and rush zoneeliminates the opportunity to use stubble heightcharacteristics of plant species that have contrast-ing growth forms to achieve an early warning ofbank damage.

Also, in smaller and mature stream channels, wherethe annual flow regime does not include a lot ofcontribution from other tributary streams duringspring runoff, banks may overhang. In contrast, inlarger streams the majority of flow is contributedby tributaries during spring runoff. The width anddepth of these larger channels may adjust to con-tain high flows. In situations like these, well-vegetated banks often round off into the largerchannels where low flow can be isolated anywherebetween banks, and banks generally do not over-hang except on the outside curve of meander bends.

A case can be made that a small mature headwaterstream with overhanging banks is at great risk ofbeing impacted by animal use and streamfiowdynamics. Assuming erosion is cyclic, this channel


configuration may be at a point when the streamsystem is ready to begin unloading its storedsediment. In part, this is because most roots gener-ally are concentrated in the top 6 inches of soilsurface because grasslike plants may have taken onmore of the growth form of upland plants. As aresult, the strength of the bank soil matrix can varyfrom weak to relatively strong, and the supportingsoil columns for overhanging banks generally areexposed to streamfiow even in drier months.

These factors appear to set the stage for bank failureunder relatively light use even though many man-agers of riparian zones have identified this channelcondition as a standard for what they believe theymust achieve (USD1 1993). By grazing livestockalong channels like these, one assumes a high riskof bank damage because these same banks can bebroken by several classes of animals, streamfiowdynamics, and ice. Therefore, measuring stubbleheight can help minimize risk associated with live-stock grazing when measurements are correlated touse of vegetation and bank damage and subsequentmeasurements are recorded to compare the amountof change caused by livestock with that caused byother animals, streamfiow dynamics, and ice.

Because grasslike plants in the riparian zones ofmany mature streams may have growth formssimilar to Kentucky bluegrass and tufted hairgrass,stubble height after grazing likely will be between3 inches and 0.75 inch at the channel edge (Figures5 and 7). To minimize risk of physical damage, it isimportant to follow Hall and Bryant's (1995) warn-ings. Three inches of stubble can disappear ratherquickly, and if tall sedges and rushes are not presentthere is no early warning as to when animals startusing the vegetation along the banks' edges.

In summary, it appears that there are two distinctareas within riparian zones where stubble heightmay differ assuming that a grazing managementphilosophy of "take half, leave half" the plant'saboveground production is enough to maintainplant vigor. These are the tall sedge and rush zoneoccupying wet areas and the tufted hairgrass andKentucky bluegrass areas of drier sites. Measuringstubble height at the transition zone between theseand correlating data to the amount of hoof imprint-ing in wet areas and bank damage can serve to

monitor animal impacts to riparian zones. Becausewet riparian areas can dry during summer and indrought, stubble height standards should be flexiblewhen used to predict physical and grazing impacts.Where differences in grasslike growth forms do notexist between uplands and water sources, particularattention should be given to Hall and Bryant's(1995) warning that one must closely monitorstubble height as it moves from about 3 inches to0.75 inch. This is especially pertinent when moni-toring the risk associated with grazing small maturestreams and bank damage because: a) overhangingbanks are rather easy to break off; b) a tall sedgeand rush zone may not be present to give an earlywarning of grazing pressure increases along banks;and c) bank damage by other large grazing animals,streamfiow dynamics, and ice can be extensive.

Stubble Height and Grazingof Woody Plant SpeciesShrubs, especially willow, in riparian zones appearto be important for wildlife habitat (Thomas et al.1978, Platts et al. 1983, Loft et al. 1991, Finch andMarshall 1993) and for stabilization of streambanks(Groeneveld and Griepentrog 1985). Also, willowcommunities can be altered or even eliminated afterextended periods of browsing (Chadde and Kay1991, Kovalchik and Elmore 1992, Singer et al.1994). Increases in livestock's willow grazing havebeen observed when grasslike vegetation maturesand turns from green to yellow in summer, andwhen green grasslike plants in riparian zones areconsumed extensively (Kauffman et al. 1983, Claryand Webster 1989, Hall and Bryant 1995).

However, willow use also can be extensive whenwildlife graze riparian zones (Gaffney 1941,Chadde and Kay 1988, Chadde and Kay 1991,Singer et al. 1994). Meiman (1996) illustrates thatwildlife used willow in all seasons in all willowcommunities studied. Yet, results from this studyindicate that livestock use generally was confinedto specific landscapes for allotted periods. Use byboth classes of animals generally was low duringspring to midsummer.

Stubble height resulting from livestock grazinggenerally is used to predict when grazing animals


shift from mostly herbaceous to woody plants.However, as discussed above, grazing animals alsomay shift from grasslike plants of dry and moistcommunities to the tall sedges and rushes of wetriparian zones. The shift may be before woodyplants are used extensively; therefore, measuringany animal use of the tall sedge and rush commu-nity also may serve as an early warning of whenconsumption of browse species begins. Hall andBryant (1995) and Claiy and Webster (1989)provide sound reasons for shifts like these. Theyalso provide stubble heights that appear to givewarning that grazing animals will use shrubs ifgrazing continues.

Clary and Webster (1989) suggest that vegetationon riparian zones should be 4 to 6 inches high aftergrazing. Hall and Bryant (1995) warn managers towatch use of shrubs as grasses' stubble heightdecreases from 3 inches to 0.75 inch. As stubblemoves below 3 inches, chances are shrub use willincrease. The authors give sound advice, but theirdifferences illustrate that the height of remainingvegetation and increased use on woody shrubs maydepend on: a) the growth form of riparian zonegrasslike plants; b) maturity of stream channels;and c) landscape position within individual riparianzones.

The discussion above about stubble height, streamchannel stability, and foraging behavior suggestsflexibility in management is important. If soils ofstreambanks or edges of ponds, bogs, springs, orlakeshores are soft, stubble may need to be higherto prevent physical damage. If soils are hard, lowerstubble may be acceptable. In either case, animals'shift from using grass to areas supporting growth oftall sedges, rushes, and woody plants probablyshould be monitored at the same time. Overall, Halland Bryant's (1995) warning that foraging behaviorchanges rather rapidly when stubble height goesfrom 3 inches to 0.75 inch is pertinent to monitor-ing the dry and stable soil conditions of tall sedgeand rush communities as well as in the wet transi-tion area between these and areas where Kentuckybluegrass and tufted hairgrass may dominate grass-like plants.

All large grazing animals may cause physicaldamage to riparian zones and consume both grass-

like and woody plants. Managers of riparian zonestherefore should use stubble height to predictpotential impacts by animal class. Woody plants inriparian zones provide browse and habitat for elk,moose, and deer (Gaffney 1941, Chadde and Kay1988, Chadde and Kay 1991, Singer et al. 1994),and use may extend through all seasons (Meiman1996). To document overall grazing impacts onriparian zones, livestock producers can monitor useof woody plants and stubble height just before andimmediately after using riparian areas. Other graz-ing impacts can be monitored similarly by appro-priate management agencies and concerned interestgroups. The result of this complete monitoringeffort will be an evaluation of management objec-tives focused towards managing riparian zones inthe western United States.

SummaryRiparian zones and their function along streamchannels must be related to the cycle of erosion. Ithas been demonstrated in this paper that as sedi-ment is deposited, channel succession occurs andthat this process of bank building may be used topredict change in plant succession. As plant speciescomposition is altered, growth form of grasslikeplants may occur, and stubble height after grazingmay then function differently in relationship to: a)how sediment is deposited along stream channels;b) how plant vigor responds to grazing action; c)when physical damage occurs to streambanks andwet and soft soils; and d) when and to what degreegrazing use occurs on other plants of interest.

Stubble height measurements should predict howmanagers can best alter desired changes betweenthe different stages of bank building recognized asmajor components in channel succession. Shortstubble may increase sediment deposition andproduction of aboveground biomass and not dam-age health attributes of grasslike plants. However,in the process of grazing to short stubble, bank andhoof imprinting on soft and wet soils may occur,and forced use of other desirable plants may in-crease. Contrasts in the use of stubble height toalter function of riparian zones therefore exist,which require that managers be flexible whenapplying this monitoring protocol.


Evidence has been presented to suggest that whenmanagers use stubble height to monitor success ofmoving from one stage in channel succession to thenext, they recognize that each stage is independentwithin a point in space and time. Also, differentstages in channel succession can exist along anysingle drainage net. Therefore, the function of eachstage should be evaluated independently basedupon its pertinent characteristics for trappingsediment, building banks, supporting plant species,and maintaining a desired channel configuration.This action can provide the data necessary topredict and determine trend toward meeting spe-cific management objectives within all stages of thecycle of erosion and along the full course of adrainage system.

The desirable attributes of using stubble height topredict change in management of riparian zonessupport this approach. Examples of these are: a)stubble height estimates remaining vegetation andnot the amount that has disappeared as consumedforage; b) stubble height monitoring is relativelyeasy to explain to a varied audience; c) stubbleheight is relatively easy to measure and thereforeserves the wide range of expertise associated withmanaging riparian zones; and d) a large number ofmeasurements can be taken in a short time whichprovides for adequate sampling intensity and moreprecise estimates than conventional utilizationmethodology.

ReferencesBeschta, R. L. and W.S. Platts. 1986. Morphological

features of small streams: significance and function.Water Resources Bulletin, American Water ResourceAssociation 22:369-379.

Briske, D. D. 1986. Plant response to defoliation:morphological considerations and allocation priori-ties. In: Joss, P. J., P. W. Lynch, and 0. B.Williams, eds. Proceedings of the Second Inter-national Rangeland Congress. Rangelands: AResource under Siege. Australian Academy ofScience, Cambridge University Press.

Briske, D. D. and J. H.Richards. 1994. Physiologicalresponses of individual plants to grazing: currentstatus and ecological significance. In: Vavra, M., W.A. Laycock, and R. D. Pieper, eds. Ecological

implications of livestock herbivory in the West.Denver: Society for Range Management.

Buckhouse, J. C., J. M. Skovlin, and R. W. Knight. 1981.Streambank erosion and ungulate grazing relation-ships. Journal of Range Management 34:339-340.

Budd, R. W. A Wyoming stream's riparian zoneresponse to instream structures. M.S. thesis, 1994,University of Wyoming, Laramie.

Cagney, J. 1993. Riparian Area Management:Greenline RiparianWetland Monitoring. USD1Bureau of Land Management TR 1737-8.

Caldwell, M. M., J. H. Richards, D. A. Johnson, R. S.Nowak, and R. S. Dzurec. 1981. Coping withherbivory: photosynthetic capacity and resourceallocation in two semiarid Agropyron bunchgrasses.Oecologia 50:14-24.

Chadde, S. and C. Kay. 1988. Willows and moose: astudy of grazing pressure, Slough Creek Exciosure,Montana, 196 1-1986. Montana Forest and Conserva-tion Experiment Station Research Note No. 24.School of Forestry, University of Montana, Missoula.

Chadde, S. and C. Kay. 1991. Tall willow communitieson Yellowstone's northern range: a test of the"natural-regulation" paradigm. In: Keiter, R. B. andM. S. Boyce, eds. The Greater Yellowstone Ecosys-tem: Redefining America's Wilderness Heritage.New Haven, CN: Yale University Press.

Chaney, E., W. Elmore, and W. S. Platts. 1990. Live-stock grazing on western riparian areas. Produced forthe U.S. Environmental Protection Agency by North-westResource Information Center, Inc., Eagle, ID.

Chow, Ven Te. 1959. Open-channel Hydraulics. NewYork: McGraw-Hill.

Clary, W. P. and B. F. Webster. 1989. Managinggrazing of riparian areas in the intermountain region.USDA Forest Service, Intermountain ResearchStation, General Technical Report INT-263.

Clary, W. P. and G. D. Booth. 1993. Early seasonutilization of mountain meadow riparian pastures.Journal of Range Management 46:493-497.

Clary, W. P., C. I. Thornton, and S. R. Abt. 1996.Riparian stubble height and recovery of degradedstreambanks. Range lands 18:137-140.

Cooke, R. U. and R. W. Reeves. 1976. Arroyos andEnvironmental Change in the American South-west.Oxford, England: Clarendon Press.

Crider, F. J. 1955. Root growth stoppage resulting fromdefoliation of grass. USDA Soil ConservationService, Technical Bulletin 1102.


DeBano, L. F. and B. R. Heede. 1987. Enhancement ofriparian ecosystems with channel structures. WaterResources Bulletin, American Water ResourceAssociation 23:463-470.

DeBano, L. F. and L. J. Schmidt. 1989. Interrelation-ship between watershed condition and health ofriparian areas in southwestern United States. In:Gresswell, R. E., B. A. Barton, and J. L. Kershner,eds. Practical approaches to riparian resource man-agement: an educational workshop. Billings, MT:USD1 Bureau of Land Management.

Detling, J. K., M. I. Dyer, and D. T. Winn. 1979. Netphotosynthesis, root respiration, and regrowth ofBouteloua gracilis following simulated grazing.Oecologia 41:127-134.

Elmore, W. and R. L. Beschta. 1987. Riparian areas:perceptions in management. Rangelands 9:260-265.

Engel, R. K. Root and shoot responses of sand bluestemto defoliation. Ph.D. dissertation, 1993, University ofWyoming, Laramie.

Finch, D. M. and R. M. Marshall. 1993. Bird use ofriparian habitats in north-central Arizona during fallmigration: results and recommendations. In:Teliman, B., G. J. Cortner, M. G. Wallace, and R. H.Hamre, technical coordinators. Riparian manage-ment: common threads and shared interests. USDAForest Service, General Technical Report RM-226.

Gaffney, W. S. 1941. The effects of winter elk brows-ing, South Fork of the Flathead River, Montana.Journal of Wildlife Management 5:427-453.

Gebhardt, K. A., C. Bohn, S. Jensen, and W. S. Platts.1989. Use of hydrology in riparian classification. In:Gresswell, R. E., B. A. Barton, and J. L. Kershner,eds. Practical approaches to riparian resource man-agement: an educational workshop. Billings, MT:USD1 Bureau of Land Management.

Goertler, C. M. Sediment dynamics within channel andriparian zone types along a cold desert Wyomingstream. M.S. thesis, 1992, University of Wyoming,Laramie.

Gray, C. L., Q. D. Skinner, and R. J. Henszey. 1997.The effect of three residual-vegetation heights onstreambank sediment deposition and vegetationproduction. Abstracts, 50th Annual Meeting, Societyfor Range Management, Feb. 16-21, Rapid City, SD.

Gregory, S. V., F. J. Swanson, W. A. McKee, and K.W. Cummins. 1991. An ecosystem perspective ofriparian zones. Bioscience 41:540-55.

Groeneveld, D. P. andT. E. Griepentrog. 1985. Interde-pendence of groundwater, riparian vegetation, andstreambank stability: a case study. In: Johnson,R. R., C. D. Ziebell, D. R. Patton, P. F. Ffolliott, andR. H. Hamre, technical coordinators. Riparianecosystems and their management: reconcilingconflicting uses. First North American RiparianConference, April 16-18, Tucson, AZ. GeneralTechnical Report RM- 120, USDA Forest Service.

Haan, C. T. and B. J. Barfield. 1978. Hydrology andsedimentology of surfaced mined lands. Office ofContinuing Education and Extension, College ofEngineering, University of Kentucky, Lexington.

Hall, C. F. and L. Bryant. 1995. Herbaceous stubbleheight as a warning of impending cattle grazingdamage to riparian areas. General Technical ReportPNW-GTR-362, USDA Forest Service, PacificNorthwest Research Station, Portland, OR.

Heede, B. H. 1980. Stream dynamics: an overview forland managers. USDA Forest Service, GeneralTechnical Report RM-72. Rocky Mountain Forestand Range Experiment Station, Fort Collins, CO.

Henszey, R. J. Riparian zone changes caused bystreamfiow augmentation. Ph.D. dissertation, 1993,University of Wyoming, Laramie.

Kauffman, J. B. and W. C. Krueger. 1984. Livestockimpacts on riparian ecosystems and streamsidemanagement implications: a review. Journal ofRange Management 37:430-438.

Kauffman, J. B., W. C. Krueger, and M. Vavra. 1983.Impacts of cattle on streambanks in northeasternOregon. Journal of Range Management 36:683-685.

Kinney, J. W. and W. P. Clary. 1994. A photographicutilization guide for key riparian graminoids. USDAForest Service General Technical Report INT-GTR-308.

Kovalchik, B. L. and W. Elmore. 1992. Effects of cattlegrazing systems on willow-dominated plant associa-tions in central Oregon. In: Clary, W. P., E. D.McArthur, D. Bedunah, and C. L. Wambolt, compil-ers. Proceedings: symposium on ecology and man-agement of riparian shrub communities, May 29-31,Sun Valley, ID. USDA Forest Service GeneralTechnical Report INT-289.

Leopold, L. B., M. G. Wolman, and J. P. Miller. 1964.Fluvial Processes in Geomorphology. San Francisco:W. H. Freeman.

Loft, E. R., J. W. Menke, and J. G. Kie. 1991. Habitatshifts by mule deer: the influence of cattle grazing.Journal of Wildlife Management 55:16-22.


Lowrance, R. R., R. Leonard, and J. Sheridan. 1985.Managing riparian ecosystems to control non-pointsource pollution. Journal of Soil and Water Conser-vation 40:87-91.

Marlow, T. B. and T. M. Pogacnik. 1985. Time ofgrazing and cattle-induced damage to streambanks.In: Riparian ecosystems and their mangement:reconciling conflicting uses. First North AmericanRiparian Conference, April 16-18, Tucson, AZ.USDA Forest Service General Technical ReportRM-l20.

Meiman, P. J. Effects of large ungulate herbivory onwillow (Salix) in mountain rangelands of northernWyoming. M.S. thesis, 1996, University of Wyo-ming, Laramie.

Middleton, M. A. Chemistry and soil texture survey ofsediments in a Wyoming riparian zone. M.S. thesis,1993, University of Wyoming, Laramie.

Morisawa, M. 1968. Streams: their dynamics andmorphology. Earth and Planetary Science Series.New York: McGraw-Hill.

Myers, T. J. and S. Swanson. 1992. Variation of streamstability with stream type and livestock bank damagein northern Nevada. Water Resources Bulletin, Ameri-can Water Resources Association 28(4):743-754.

Nowak, R. S. and M. M. CaIdwell. 1984. A test ofcompensatory photosynthesis in the field: implica-tions for herbivory tolerance. Oecologia 61:311-318.

Olson, B. E. and J. H. Richards. 1988. Tussock re-growth after grazing: intercalary meristem and axil-lary bud activity of tillers of Agropyron desertorum.Oikos 51:374-382.

Platts, W. S., W. F. Megahan, and G. W. Minshall.1983. Methods for evaluating stream, riparian andbiotic conditions. USDA Forest Service GeneralTechnical Report INT-1 38.

Platts, W. S. and J. N. Rinne. 1985. Riparian and streamenhancement management and research in the RockyMountains. North American Journal of FisheriesManagement 5(2A):1 15-125.

Richards, J. H. and M. M. Caldwell. 1985. Solublecarbohydrates, concurrent photosynthesis andeffiency in regrowth following defoliation: a fieldstudy with Agropyron species. Journal of AppliedEcology 22:907-920.

Rumsey, C. J. The effect of three residual vegetationheights on streambank sediment deposition andvegetative production. M.S. thesis, 1996, Universityof Wyoming, Laramie.

Schumm, S. A. 1977. The Fluvial System. New York:John Wiley & Sons.

Singer, F. J., L. C. Mark, and R. C. Cates. 1994.Ungulate herbivory of willow on Yellowstone'snorthern winter range. Journal of Range Manage-ment 47:435-443.

Thomas, J. W., C. Maser, and J. E. Rodiek. 1979.Riparian zones in managed rangelands: their impor-tance to wildlife. In: Cope, 0. B., ed. Proceedings ofthe Forum on Grazing and Riparian/Stream Ecosys-tems, Nov. 3-4, 1978, Denver. Trout, Unlimited.

U.S. Department of Interior. 1993. Riparian areamanagement: process for assessing proper function-ing condition. USD1 Bureau of Land Management,TR-1739-9.

U.S. General Accounting Office. 1988. Public range-lands: some riparian areas restored but widespreadimprovement will be slow. B-230548, USGAO,Resources,Community, and Economic DevelopmentDivision. Washington, DC.

Yeager, T. D. Critical groundwater hydroperiods formaintaining riparian plant species. M.S. thesis, 1996,University of Wyoming, Laramie.

Zonge, K. L., S. Swanson, and T. Myers. 1996.Drought year changes in streambank profiles onincised streams in the Sierra Nevada Mountains.Geomorphology 15:47-56.

Zonge, K. L. and S. Swanson. 1996. Changes onstreambanks in the Sierra Nevada Mountains:perspectives from a dry and a wet year. RestorationEcology 4(2): 192-199.


Developing and AchievingManagement Objectives

on National Forest System LandsFrederick C. Hall and Richard Lindenmuth

Natural Resources Unit, Pacific Northwest Region, USDA Forest ServicePortland, OR 97208

AbstractTo enable the Forest Service to more effectively setand achieve management objectives for the rangeresource on National Forest System lands, achiev-ing the functionality and desired condition of bothaquatic/riparian and upland systems is essential.Setting measurable objectives is a required stepbefore any monitoring can be developed. Monitor-ing protocols must be workable by grazing permit-tees and concerned citizens as well as by agencyemployees. Monitoring for forage utilization, whilenot a substitute for measuring long-term conditionand trend, has its place in the manager's tool box toguide systems on the road to recovery. Monitoringresidual vegetation is preferred over methods thatestimate amount of herbage removed.

IntroductionThe previous papers presented an excellent sum-mary of the determination, interpretation, and useof forage utilization. Two methods to determineutilization were discussed: amount removed, andamount remaining. But most of all the authorsillustrated idiosyncrasies and constraints.

Our purpose is to relate aspects of forage utilizationto National Forest System (NFS) livestock manage-ment. The NFS operates under two of severalimportant congressional laws: the ResourcesPlanning Act (RPA) as amended by the NationalForest Management Act (NFMA), both of whichrequire analysis of livestock impacts on the grazingresource. These laws do not require determinationof forage utilization, only assessment of the effectsof utilization. They also provide for changingForest Plans when new information becomes

available. The NFS has proposed utilization mea-sures as one way to assess livestock impacts. Thispaper discusses use of utilization in land manage-ment and its place in National Forest Plans.

Allotment Management PlansManagement plans dealing with livestock grazingmight focus on three important features.

Define suitable functioning of ecosystems, bothriparian and upland. Those functioning ad-equately are managed to maintain their func-tion, and those that are not functioning ad-equately are managed to attain and maintainsustainable function.

Define management objectives that specifydesired conditions on the ground.

Design a management strategy that will ensurethe desired outcome based on sustainability. Amanagement strategy is the foundation uponwhich a monitoring plan is formulated.

We should ask how and where determination ofutilization fits in this monitoring plan. Five ques-tions come to mind.

Why determine utilization? To appraise plantphysiology or to appraise animal management?To evaluate grazing effects on vegetation andsoil? What management need does it resolve?Does it monitor movement of the vegetationand soil toward or away from the desiredcondition and suitable functioning?

What does utilization monitor? Amount offorage removed or amount remaining? Animaldistribution? Animal preference?


How is utilization determined? Percent offorage removed? How is removed foragedetermined: height/weight curves or percent ofplants grazed? Amount of forage remaining?Measured stubble height? Percent of twigsbrowsed? Measured twig length?

Where is utilization determined? Criticalareas? Key areas? Benchmark areas? What arethe criteria for selecting areas?

When should utilization be determined? Beforeand after grazing to appraise wildlife use? Afterlivestock move? End of plant growth? End ofseason? Every year?

Monitoring ObjectivesA management plan must define what determina-tion of utilization is supposed to accomplish. Howdoes it monitor maintenance or attainment of adesired condition? For example, how does it relateto monitoring trends in vegetation and soil? Howdoes it relate to animal distribution, allotmentcarrying capacity, and prescription of a grazingplan? How might it be used to appraise animaldamage to soil and vegetation? Can it be used tomanipulate livestock (USD1 BLM 196)?

How is determination of utilization used on anallotment without a management plan? Is it used tomanage livestock? Should it be used annually ormore often to evaluate animal impacts?

There are some monitoring priorities for whichdetermining utilization might be useful if themethod is easy to use, quick, and has minimumvariability between observers (USD1 BLM 1996).In other words, the method should be usable bypermittees, the public, and agency employees. Thefollowing attributes are particularly important forallotments without a management plan. Determin-ing utilization might be used:

1) As a warning sign to prevent livestock damageto soil or vegetation (Hall and Bryant 1995).We are looking for an indication that there hasbeen enough use and that livestock should bemoved. We can measure resource damage afterit has happened (U.S. GAO 1988); what weneed is a livestock management tool to preventdamage (Hall and Bryant 1995).

As a means to develop vegetation structure.Limiting use may provide for an increase inshrub height and crown spread, leave enoughstubble to trap sediments in riparian areas, andleave forage for other animals. Vegetationstructure is a key element in ecosystem func-tion. These are ways to meet requirements of abiological opinion rendered under the Endan-gered Species Act (ESA). Annual monitoringtypically is required.

As a means to prevent soil damage by leavingenough stubble to protect the soil from compac-tion, soil displacement, and wind or watererosion (Hall and Bryant 1995). Preventing andcorrecting soil and vegetation damage leads tosustainable function. Annual monitoring mightbe necessary.

As a means to appraise livestock distribution onkey or critical areas, appraise allotment carry-ing capacity, and adjust the grazing manage-ment system to avoid vegetation or soil damage(USD1 BLM 1996).

Utilization CharacteristicsSystems for determinating utilization to meet moni-toring needs have several characteristics. One criti-cal factor is recognition that all we have to workwith is what remainsstubble height, twigsbrowsed, or twig length, all of which are directlymeasurable (USD1 BLM 1996, Hall and Bryant1995). Removed material and total production mustbe estimated to determine percent use (USD1 BLM1996).

The sampling method must be quick, easy, andreliable with little variation among different ob-servers. This means that when several peoplesample the same utilization transect, their resultsshow a minimum of difference. Measurements, asopposed to estimates, usually vary less amongobservers, require less training, and have fewerpresumptions about how much is removed (USD1BLM 1996). Focusing attention on what remainsseems desirable because it contributes to sustain-able function.

What remains directly influences the rate of shrubheight and crown growth and the stubble's ability


to trap sediment, provide soil cover and protection,and supply forage for other animals; it also affectshow animals eat (Hall and Bryant 1995). Cows arean example. A cow reaches her tongue out of theside of her mouth, draws in forage, tastes it, andbites off a mouthful, filling her rumen quicklywhen forage is plentiful (Hall and Bryant 1995).But as stubble height lowers to 3 to 4 inches, theherbage is too short for the cow's tongue to draw itinto her mouth so she starts eating in bites. Bitestake in less forage, so more time is required to fillher rumen (Van Soest 1988). As a result, her pref-erence changes to less palatable species in order tofill her rumen. Or she may concentrate her grazingin riparian areas where forage remains plentiful,then damaging vegetation or soil and breakingdown stream banks (Hall and Bryant 1995).

The amount of forage remaining directly relates tolivestock preference for forage and foraging areasand reveals opportunities to shift livestock distribu-tion from heavily to lightly used locations. Whenlivestock grazing habits change as stubble heightapproaches 3 to 4 inches, the livestock manager iswarned to move animals to avoid damage andmaintain trend toward the desired condition andsustainable function (USD1 BLM 1996, Hall andBryant 1995).

What remains can be measured, whereas what wasremoved must be estimated (USD1 BLM 1996).What remains can be sampled with little training,and criteria can be established that are easy tounderstand, such as a stubble height of 4 inches.What remains can be shown to people: "This iswhat we want to accomplish" or, "This is when theanimals must be moved."

The concept of measuring vegetation remaining ispreferred by the authors, who recommend it beadopted as a tool in Forest Plan Standards andGuidelines for both riparian and upland ecosystemswhile phasing out the estimation of percent use(what was removed).

ReferencesHall, F. C. and L. Bryant. 1995. Herbaceous stubble

height as a warning of impending cattle damage toriparian areas. USDA Forest Service General Tech-nical Report PNW-GTR-362. Pacific NorthwestResearch Station. Portland, OR.

USD1 Bureau of Land Management. 1996. Utilizationstudies and residual measurements. InteragencyTechnical Reference BLM/RS/ST-96/004+ 1730.National Applied Resource Sciences Center, Denver,CO.

U.S. General Accounting Office. 1988. Public range-lands: some riparian areas restored but widespreadimprovements will be slow. B-230548. US GAO,Resources, Community, and Economic DevelopmentDivision. Washington, DC.

Van Soest, P. J. 1982. Nutritional ecology of theruminant. Corvallis, OR: 0 & B Books, Inc.


Impact of Federal Grazing Reductionson Wyoming Ranches

AbstractThe study examines the profitability of a ranchingoperation that adjusted to a reduced stocking ratesresulting from a decrease in public land use. A 300-cow case study ranch was developed using inputfrom ranchers grazing on U.S. Forest Service(USFS) allotments in the Big Horn National Forestand adjoining Bureau of Land Management (BLM)allotments.

A linear programming model of production alterna-tives was developed to assess how the ranch wouldadjust to a reduction in federal AUMs. The ranchwas allowed to adjust cattle numbers and to converthayland to pasture and/or feed hay. Adjustments to25, 50, and 100% decreases in federal AUMs wereexamined. Each scenario then was analyzed toestimate the financial consequences of the reduc-tion using The Farm Level Income Tax and PolicySimulation Model (FLIPSIM).

As total federal AUMs were reduced 25, 50, and100%, numbers of cows were reduced from 300head to 261, 221, and 144 head, respectively.Hayland converted to pasture under each reductionscenario was 16, 32, and 64 acres, respectively.These reductions translated into a decline in aver-age annual net cash income of $14,263, $30,689,and $56,554, respectively. The ending equity ratiodropped from the original 0.88 to 0.80, 0.63, and0.33, respectively, under the 25, 50, and 100%reduction scenarios.

Larry W. Van TassellDepartment of Agricultural Economics


James W. RichardsonDepartment of Agricultural Economics

Texas A&M UniversityCollege Station, TX 77843

introductionAlmost 49% of the total acreage in Wyoming isfederally owned (USD1 BLM 1994). Of this, 60%is under Bureau of Land Management (BLM)jurisdiction (USD1 BLM 1994) and 30% is admin-istered by the U.S. Forest Service (USDA FS 1993).More than 2.1 million animal unit months (AUM5)of authorized grazing are provided to livestockranchers in Wyoming from BLM- and USFS-administered lands.

When federal grazing lands originally were allo-cated, livestock operators who met the commensu-rability and prior-use requirements were givenpreference for receiving available grazing permits.Grazing fees were set below "fair market value,"and permits were allocated to encourage settlementand stability of western communities. Public landswere incorporated quickly into the ranchers' foragerotations and became an integral ingredient tosuccessful ranching in the arid West. Public landranchers contend their economic viability is relatedto the economies of size achieved from, and sea-sonal forage demands provided by, grazing federallands (Torell et al. 1992).

Livestock AUMs on federal lands can be reducedfor a number of reasons. For example, the ProposedAction outlined in the Draft Environmental ImpactStatement for Rangeland Reform '94 (USD1 1994)projected a 21% reduction from 1993 levels inAUMs on federal lands over a 20-year period.Stocking rate adjustments were predicted to result"from monitoring studies that indicate continuing


resource damage and a declining economic feasi-bility of livestock grazing" (p. 4-3 8). While notspecified, monitoring studies often include themeasurement or estimate of utilization. As has beenpointed out in previous papers, exceeding arbitraryutilization levels, sometimes for only 1 year, canlead to temporary or permanent reductions instocking rates on a federal grazing permit.

Several researchers have examined the importanceof federal grazing to the economic success ofpublic land ranchers. Torell et al. (1981) used alinear programming model to examine the impactof BLM allotment reductions on Nevada ranchers.As the percentage of AUMs were reduced, a corre-sponding reduction in ranch net income occurred.Ranchers adjusted by substituting irrigated pas-tures, deeded rangeland, and USFS grazing for theBLM grazing. Additional hay also was fed andlorcattle numbers were reduced.

Results were similar in Wyoming studies by Olsonand Jackson (1975) and Peryam and Olson (1975).Both studies found that reducing the amount ofBLM grazing permitted caused a reduction in cownumbers and labor used, along with an increase inoff-farm hay sales. Depending on the ranch'sdependence on public lands, net income declinedfrom 5.2 to 3 1.6% as available federal grazingdecreased 90%.

This study's objectives were to determine optimalranch production adjustments to reductions in BLMand USFS grazing reductions for a representativenorth-central Wyoming ranch. The economicsuccess of the ranching operation, given the opti-mal ranch production adjustments, was examinedunder stochastic price and production conditions.

MethodsAnalysis was in three steps. First, data for therepresentative ranch were developed through groupinterviews. A linear programming (LP) model thenwas developed to depict the production process ofthe ranch. Optimal production adjustments tospecified reductions in permitted federal grazingwere determined using the LP model. Third, ranchresources and production practices under eachfederal grazing reduction scenario were incorpo-

rated into a stochastic simulation model to deter-mine the financial performance of the ranch over a10-year period.

Representative RanchA representative federal land ranching operationwas developed for north-central Wyoming, specifi-cally, for Washakie and Big Horn counties. Therepresentative ranch was developed through inter-views with a panel comprising four area ranchersselected by the local county agricultural Extensionagent. Through consensus, the panel developed thecharacteristics, resources, costs, and income struc-ture of a representative federal-land ranching opera-tion in the area. The panel also was used to validatethe simulated representative ranch to ensure theinformation gathered was interpreted correctly.

The ranch comprised 2,200 deeded acres and waslocated at the base of the Big Horn National Forest.Characteristics of the ranch are given in Table 1.The ranch was a typical cowcalf operation, with

Table 1. Characteristics of the representativeWyoming ranch.

Unit Value


Land resources ownedAlfalfa hayland acre 80

Native hayland acre 55

Subirrigated pasture acre 300

Rangeland acre 1,765

Land resources leasedGrazing state land AUM 160

Grazing USFS land AUM 940Grazing BLM land AUM 1,001

Livestock resourcesCows head 300Replacements head 38

Bulls head 13

Horses head 6

Financial resourcesAssets $ 838,719Liabilities $ 127,855

Debt-to-asset ratio % 15.2

Efficiency measuresCalf crop weaned % 92

Calf sale weightSteer lb 550Heifer lb 525

300 head of mother cows and associated replace-ments and bulls. A 92% weaned calf crop wasassumed, with weaning weights of 550 pounds forsteers and 525 pounds for heifers. Livestock graz-ing requirements in the spring, summer, and fallwere met by deeded rangeland and subirrigatedpasture, along with state, USFS, and BLM grazingpermits. State and BLM grazing lands were utilizedin the spring (May and June) and fall (October andNovember). Forest Service grazing lands providedmost of the ranch's grazed forage from July throughSeptember.

Productivity measurements for the deeded range-land and subirrigated pasture were obtained fromNatural Resource Conservation Service (NRCS)site guides for the areas, assuming land in goodcondition (Table 2).

Farming operations consisted of an alfalfa growingand haying enterprise, an alfalfa establishmententerprise that used oats harvested as hay for arotation/cover crop, and a native hay growing andhaying enterprise. The alfalfa enterprise yielded 4tons of hay per acre, the oat enterprise yielded 3tons per acre, and the native hay enterprise yielded3 tons per acre (Table 2). The alfalfa and nativehaying enterprises also contributed 1.7 AUM/acreand 0.6 AUMs/acre, respectively, in aftermath tothe livestock operation. All hay crops were floodirrigated or irrigated with side-roll sprinklers.

Farming equipment was minimal and included twotractors (100 and 65 horsepower), a 12-foot hayswather, a round baler, a hay rake, a harrow, a disk,

Table 2. Productivity measures of harvested andgrazed forages.

Alfalfa hayaftermath

Native hayaftermath

Oat hay

Graze alfalfa hayland

Graze native hayland

Graze subirrigated pasture

Graze rangeland

and a drill. Most machinery and equipment were 10to 20 years old and had been purchased from used-equipment auctions around the valley.

Livestock typically were fed hay from Decemberthrough April. Slightly over 1 ton of hay was fedper animal. Concentrate supplements were reservedfor replacements.

Ranch assets were valued at $838,719. Liabilities,mostly long-term loans, were $127,855, leaving adebt-to-asset ratio of 15.2%.

Linear Programming ModelThe LP model was formulated to account for therepresentative ranch's seasonal forage and laborrequirements. The model's objective function wasto maximize profit, subject to the various resourceconstraints previously described. Forage wasaccounted for on a monthly basis, and labor wassummed by activity. Major activities and theirassociated cost or return coefficients are in Table 3.Forage resources previously described were desig-nated as activities and entered the objective func-tion as costs of production (Table 3). Hay yieldsand/or AUMs provided by each activity were madeavailable to livestock enterprises through monthlyforage transfer rows to be used in their designatedseason of use.

Permittees on the Big Horn National Forest weresurveyed concerning adjustments they would makeif USFS grazing permits were reduced by 25, 50,and 100%. The majority of permittees stated theywould adjust their resources at the base ranch andkeep ranching until they could no longer stay inbusiness; then, they would sell out. A few statedthey could improve some grazing lands (e.g., spraysagebrush), but the majority stated there were notmany improvements they cOuld make.

To allow the ranch to adjust to reductions in USFSor BLM grazing, two activities were thereforeadded to the LP model: graze alfalfa land, andgraze native hayland. Stocking rates were figuredassuming 780 pounds of forage per AUM and a90% utilization rate. Costs of grazing the forageincluded buying and operating electric fences.Leasing pasture from other landowners was notincluded as an option in the model, for two reasons.


Unit Value

tons/acre 4.0AUM/acre 1 .7

tons/acre 3.0AUM/acre 0.6

tons/acre 3.0

AUM/acre 11.6

AUM/acre 7.7

AUM/acre 1.5

AUM/acre 0.3

Table 3. Variable costs1 and pricesof major activities in the linear programming model.Activity Unit Value ($)

Labor requirements were identified separately foreach activity in the model. The ranch owner pro-vided 200 hours labor per month. Additional laborrequirements could be met by hiring a full-timeemployee for $18,000 per year (providing 200hours labor per month) and/or by hiring part-timelabor for $5.00 per hour.

Simulation ModelThe mathematical description of the representativeranch and the appropriate resource adjustmentsfrom the LP optimization were entered in the FarmLevel Income Tax and Policy Simulation Model(FLIPSIM). FLIPSIM is a Monte Carlo simulationmodel developed by Richardson and Nixon (1986).The model has been used for numerous farm levelpolicy and technology analysis (e.g., Anderson etal. 1993; Lemieux and Richardson 1989;Richardson and Smith 1985). FLIPSIM allows theoperator to simulate a representative ranch underalternative policy and management scenarios usingstochastic prices and yields.

Weaning weights and livestock prices were sto-chastic in the simulation. The economic activity ofthe representative ranch was simulated over a 10-year planning horizon for 100 iterations. Each year,weaning weights were generated randomly basedon a historical series of weights from panel produc-ers. Livestock and feed prices also were selectedrandomly for each iteration based on the averageannual prices in the 1996 Food and AgriculturalPolicy Research Institute baseline and associatedhistorical probability distributions of each variable.In the simulation, steer calf prices averaged $79.67per hundredweight., alfalfa prices averaged $60.94per ton and intermediate-term debt financingaveraged 9.15%.

FLIPSIM allows the user to track several financialvariables. Variables of interest in this study wereannual net cash income and owner's equity. Resultsof the 100 iteration analyses constitute an estimateof the probability distribution for each variable.

Using the probability distributions, the probabilityof negative cash income and, the probability of theranch owner's obtaining a lower real equity can bedetermined for each federal grazing reduction

Stubb'e Height and Uti'ization Measurements 53

Grow and harvest alfalfa hay acre 69.04

Grow and harvest grass hay acre 27.74

Grow and harvestalfalfa hay establishment acre 85.64

Grow and harvest oat hay acre 72.58

Graze alfalfa forage AUM 3.80

Graze native grass forage AUM 2.44

Graze subirrigated pasture AUM 4.00Graze private rangeland AUM 3.25

Graze BLM land AUM 7.19

Graze USFS land AUM 9.46

Graze state land AUM 8.59

Purchase alfalfa hay ton 70.00

Purchase grass hay ton 60.00

Purchase protein supplement ton 200.00

Purchase price of bulls head 1,700.00

Maintenance cost of cows2'3 head 37.64

Maintenance costof replacement heifers2 head 21.80

Selling price of steer calves cwt 88.75

Selling price of heifer calves cwt 81.50

'Excludes labor2 Excludes cost of feed stuffsIncludes maintenance of bulls and value of cull animals

First, a surplus of private leases does not exist inthe area. Second using private leases to compensatefor a reduction in federal AUMs would displaceother livestock production in the region and inaccu-rately reflect the impact that a decrease in availablefederal grazing would have on the region.

Constraints in the model included the number ofAUMs available on USFS and BLM lands. Asthese numbers were reduced, the model adjusted byreducing the number of cows and rearranging theforage situation in the most profitable manner. Asthe number of cows were reduced, the associatedinputs (e.g., number of bulls required), variablecosts, and outputs were adjusted accordingly.

Fixed costs (e.g., legal fees, land taxes, insurance,and depreciation) remained unchanged as USFSand BLM permits were reduced.

scenario. A 6% discount rate was used to adjustcash flows for the effects of time.

Federal Grazing Reduction ScenariosA baseline simulation of the representative ranchwithout federal grazing reductions was comparedwith nine reduction scenarios. Scenarios 1 through3 reflected reductions of 25, 50, and 100% in USFSgrazing permits, holding BLM grazing numbersconstant. Scenarios 4 through 6 entailed 25, 50, and100% reductions in BLM grazing permits, holdingUSFS permits constant. The last three scenariosrepresented 25, 50, and 100% concurrent reduc-tions in USFS and BLM grazing permits.

ResultsAnnual net cash income for the representativeranch when USFS and BLM permits were at theirfull use averaged $31,556 (Table 4). Four percentof the time, the ranch had a negative annual netcash income. Ending equity was reduced to 88% bythe end of the 10-year planning horizon, assumingno inflation. Ending equity eroded over the 10-yearplanning horizon for 75 out of the 100 iterationssimulated.

To adjust to the reduction in USFS grazing permits,animal units operated by the ranch declined, somehay acreage was converted to summer grazingpasture, and the amount of labor needed wasreduced (Table 4). The panel that developed therepresentative ranch believed the season of use on

Table 4. Adjustments to reductions in U.S. Forest Service AUMs.Percent reduction in USFS AUMs

BLM lands was fairly rigid and could not be grazedduring the summer to make up for lost USFSAUMs. Therefore, adjustments included grazingsome private rangeland during the summer andconverting hayland to pasture to keep production ata maximum. Because of the seasonal rigidity ofgrazing BLM forage, adjustments in cow numbersor in hayland to pasture conversion were not thesame as the USFS permits were reduced from 25through 100%.

The loss of all 942 USFS AUMs amounted to aloss of 136 mother cows along with the associatedbulls and replacement heifers. The loss of the 942USFS AUMs amounted to a greater than one-to-one conversion on the ranch, as the USFS AUMsamounted to a reduction in only 78.5 AUs. Thiswas because the conversion of hay acreage topasture allowed less feed for wintering cows andalso because of underutilized BLM AUMs.

Labor also was reduced as the ranch lost USFSAUMs. The reduction was not constant on a per-cow basis because some labor requirements werefixed on the ranch. Labor requirements per mothercow increased from approximately 15 hours percow to almost 18 hours per cow as the ranchcompletely lost USFS grazing permits.

Economies of size have been noted as partly re-sponsible for the value attached to federal grazingpermits. This was observed in the simulationanalysis as average annual net cash income wasaffected by the rigidity of the BLM grazing season,


Adjustments in 0 25 50 100

USFS AUMs 942 707 471 0

BLM AUMs 1,001 1,001 1,001 1,001

Number of cows 300 267 221 164

Acres hayed 135 117 95 79

Hay acres converted to pasture 0 17 40 56

Hours of labor 4,481 3,901 3,688 2,872

Average annual net cash income ($) 31,556 20,489 15,893 (20,522)

Probability of negative cash income (%) 4 13 18 100

Ending equity ratio (%) 88 83 78 38

Probability of lower real equity (%) 75 99 99 100

Table 5. Adjustments to reductions in Bureau of Land Management AUMs.Percent reduction in USFS AUMs

the fixed nature of labor, and the fixed costs thathad to be covered no matter how many cows wereoperated.

The financial condition of the ranch quickly dete-riorated as USFS permits were removed. Averagenet cash income was reduced a total of $52,078when USFS grazing ceased, and the probability ofnegative cash income increased from 4% when allgrazing permits were intact to 100% when forestgrazing no longer was available. Continued ranch-ing after forest grazing was removed was not a wisedecision, as rancher equity eroded to 38% of theoriginal value over the 10-year planning horizon.

Patterns were similar for BLM and USFS permitreductions (Table 5). Cow numbers were reducedon a greater than one-to-one basis, though thereduction was not quite as great as when USFSgrazing permits were lost. Accordingly, the finan-cial condition of the ranch deteriorated slightlyless, though the ranch quickly turned unprofitablewithout the use of BLM permits.

When USFS and BLM grazing permits were lostconcurrently (Table 6), the effect on the ranchingoperation was not additive because of the unem-ployed federal grazing resources due to the sea-sonal rigidity of both USFS and BLM grazingpermits. The carrying capacity of the deededranchland settled at 144 mother cows when allfederal grazing permits were removed. The ranchwas not able to cover all variable and fixed costs atthis stage, and it rapidly lost equity.

ConclusionsAs previous studies have shown, federal grazingpermits were important to the success of the repre-sentative ranch used in this study. Economies ofsize, obtained through the additional cows theranch was able to maintain because of the federalgrazing permits, were an important aspect of thissuccess. Costs of buildings, fences, corrals, andequipment were all reduced on a per-cow basisbecause of federal grazing permits.

Employment issues also appeared to be importantconsiderations as federal grazing permits wererelinquished. While some labor fixity was apparentin this study, the ranch was not able to maintainenough work to employ a full-time person whenfederal permits were lost.

As mentioned, many permitees of the Big HornNational Forest stated if they lost forest grazingprivileges, they would make adjustments on theirbase properties until they could no longer stay inbusiness. The simulation analysis in this studyshowed that equity rapidly eroded as federal per-mits were removed. The potential exists, therefore,that without federal grazing permits, much of theland around national forests could change owner-ship. Because of the price most land around nationalforests can demand, the danger is those lands wouldbe subdivided into ranchettes or other residencesrather than stay in productive agricultural use.

Results of this study are limited to the area studiedand to the assumptions made. Further study appears



USFS AUMs 942 942 942 942

BLM AUMs 1,001 751 501 0


Acres hayed 135 123 109 80


Hours of labor 4,481 4,211 3,858 3,152

Average annual net cash income ($) 31,556 19,835 6,697 (23,111)




Table 6. Adjustments to reductions in U.S. Forest Service and Bureau of Land Management AUMs.Percent reduction in USFS and BLM AUMs

to be warranted to assess positive and negativeexternalities, such as the impact upon wildlife,associated with viable ranching operations utilizingfederal grazing permits.

ReferencesAnderson, D. P., J. W. Richardson, R. D. Knutson, J. C.

Namken, T. R. Harris, W. 0. Champney, T. R.MacDiarmid, and A. B. Marshall. 1993. Alternativegrazing fee formula impacts on representative publicland ranches. Journal of Range Management 46:548-554.

Food and Agricultural Policy Research Institute(FAPRI). 1996. FAPRI 1996 U.S. AgriculturalOutlook, Staff Report 1-96. University of Missouri,Columbia and Iowa State University.

Lemieux, C. M. and J. W. Richardson. 1989. Economicimpacts of Porcine Somatotropin on midwest hogproducers. North Central Journal of AgriculturalEconomics 11:171-182.

Olson, C. E. and J. W. Jackson. 1975. The impact ofchange in federal grazing policies on south-centralWyoming mountain valley cattle ranches. WyomingAgricultural Experiment Station Research Journal96. Laramie, WY.

Peryam, J. S. and C. E. Olson. 1975. Impact of potentialchanges in BLM grazing policies on west-centralWyoming cattle ranches. Wyoming Agricultural Ex-periment Station Research Journal 87. Laramie, WY.

Richardson, J. W. and C. J. Nixon. 1986. Description ofFLIPSIM V: a general firm-level policy simulation

model. Texas Agricultural Experiment StationBulletin B-1528, Texas A&M University, CollegeStation.

Richardson, J. W. and E. G. Smith. 1985. Impacts offarm policies and technology on the economicviability of Texas southern high plains wheat farms.Texas Agricultural Experiment Station BulletinB-1506, Texas A&M University, College Station.

Torell, L. A,, E. T. Bartlett, and F. W. Obermiller.1992. The value of public land grazing permits andthe grazing fee dilemma. New Mexico State Univer-sity Range Improvement Task Force Report 31, LasCruces.

Torell, L. A., J. R. Garrett, and C. T. K. Ching. 1981.The economic effects of three changes in publiclands grazing policies. Journal of Range Manage-ment 34(5):373-376.

U.S. Department of Agriculture Forest Service. 1993.Report of the Forest Service, Fiscal Year 1992.Washington, DC. February.

U.S. Department of Agriculture Forest Service, RangeManagement. 1995. Grazing Statistical Summary,1994. Washington, DC. March.

U.S. Department of Interior. 1994. Rangeland Reform'94Draft Environmental Impact Statement. Bureauof Land Management in cooperation with the De-partment of Agriculture Forest Service.

U.S. Department of Interior Bureau of Land Manage-ment. 1994. Public Land Statistics, 1993. Vol. 178.September.



USFS AUMs 942 707 471 0

BLM AUMs 1,001 751 501 0


Acres hayed 135 119 103 71


Hours of labor 4,481 4,090 3,687 2,688

Average annual net cash income ($) 31,556 17,293 867 (24,998)




Comparison of Economic Impactsfrom Public-land-based Tourism and Grazing:

A Case Study

AbstractThis paper examines the economic effects ofdecreases in cattle grazing allowed on the Big HornNational Forest (BHNF) in north-central Wyoming.A 25% reduction of the 105,775 AUMs of grazingallotted to cattle would reduce economic activity inthe four-county area surrounding the BHNF by$1.68 million, of which $441,384 would be per-sonal income for local residents. The communitiesalso would lose 30.56 FTE jobs. Eliminating allcattle grazing on the BHNF would reduce eco-nomic activity by $6.74 million, personal incomeby $1.77 million, and employment by 122 FTE.

Tourism often is touted as the economic savior forsmall local communities facing decreased grazingallotments. To offset the decline in economicactivity from eliminating public-land grazing,recreation would have to increase about 104,000tourism visitor days (TVD). Tourism would need to

Robert R. FletcherDepartment of Applied Economics and Statistics

University of Nevada, Reno 89507

George W. BordenCommunity Development SpecialistUniversity of Nevada, Reno 89507

Thomas R. HarrisDepartment of Applied Economics and Statistics

University of Nevada, Reno 89507

David T. TaylorDepartment of Agricultural Economics


Brett R. MolineExtension Specialist, Department of Agricultural Economics


increase about 127,000 TVD to offset the loss ofpersonal income. Depending on the growth rate ofrecreational use, 10 to 53 years could be requiredfor the Big Horn Mountain Area (BHMA)economy to recover the loss in income from thereduction in livestock grazing. To maintain thecurrent employment level, recreation would have toincrease about 113,000 TVD. If personal incomeand employment are important to a local commu-nity, more than one TVD will be required to re-place each AUM of public land grazing. Thenumbers presented in this paper are for analysisand comparative purposes only. There is no evi-dence to indicate tourism recreation would increaseas a result of decreased livestock grazing.

IntroductionEmotional arguments often are at the center ofpublic-land use issues that involve both commodity


users and and individuals advocating a reduction ortotal elimination of livestock, timber, and mineralproduction on public lands. Economic consider-ations have received low priority in the delibera-tions that result in policy decisions that direct theuse of our public lands. As a result, minimal re-sources have been spent to develop sound scientificmethods to estimate the economic impacts ofalternative public-land uses on local communities.

Depressed economic conditions prevailed in Wyo-ming during much of the 1980s. Decline in theenergy industry and low prices for agriculturecommodities created economic hardships in manyWyoming communities. Economic growth anddevelopment opportunities for these rural commu-nities are highly dependent on increasing economicactivity in the basic industries, which rely heavilyon natural resources located on federal lands.

The Big Horn National Forest in north-centralWyoming is a source of summer grazing for live-stock producers and of timber for local sawmills.The Big Horn Mountains and the surrounding area,including forest boundaries, provide visitors theopportunity to enjoy cañping, hunting, fishing,sightseeing, geology sites, historic sites, and vari-ous other activities in an uncrowded setting.

In 1989, the Big Horn Mountain Coalition formedto explore the possibility of improved tourismmarketing for the region. This area had experiencedan eroding tax base, by over 50% in one county,and loss of jobs. The executive board of the Coali-tion was composed of one county commissionerfrom each of the four counties within the Big HornMountain Area: Big Horn, Johnson, Sheridan, andWashakie counties. Theirprimary objective was toincrease economic activity, jobs, and their tax basesfrom tourism and recreation in the BHMA whilemitigating conflicts with other industries and usersof the natural resource base. The Coalition's firstaction was to commission a study through theWyoming Cooperative Extension Service to de-velop baseline information on tourism in the area.

The Coalition developed a climate for cooperationand coordination between commodity users,recreationists, federal land management agencies,and local business people in the area. This provided

an excellent opportunity to study the economicimpact of proposed alternatives for public-landmanagement decisions. Although economics maynot be the most critical issue in determining public-lands policy, it should be considered in the finalanalysis. The consideration should be based on thebest data available, which historically have beenminimal at best.

This paper is presented in three sections: tourismand recreation, grazing, and economic impactcomparisons between tourism and recreation andgrazing. Sections one and twotourism andrecreation, and grazingdiscuss the purpose,procedures, and impact results for current levels ofactivity. The third section compares the economicimpacts as a result of one activity versus the other.

Tourism and RecreationAnalyses reported in this section are based onsurvey data collected from visitors to the areaduring the summers of 1989 and 1990 and thewinter of 1989-90 and on a survey of eating,drinking, and lodging establishments during thesummer and fall of 1990. The sUtvey data weresupplemented with secondary data based on salestax collections and the census of retail trade andservices for the four counties in the study area.

ProcedureTwo critical factors had to be quantified to estimatethe economic contribution of tourism and recre-ation in the Big Horn area: first, the number ofpeople visiting the area; second, the level of visi-tors' expenditures. Two approaches were used toestimate the two factors. An estimate of the numberof people visiting the area was developed throughinterviews with lodging providers in the area todetermine occupancy and average daily rates bymonth. This provided an estimate of the supply oflodging facilities and capacity utilized. Visitorsurveys provided the average daily expenditure foreach party or each person visiting the area fortourism or recreational purposes. This was used toestimate the demand for goods and services bypeople residing outside the area. Census data forlodging and eating and drinking establishments for1987 were analyzed and updated using the latest


state sales tax figures to provide a check on theaccuracy of the survey results.

LodgingThree general categories of lodging were used toestimate tourist numbers. U.S. Forest Service figureswere used to estimate the number of campers usingcamp sites on the Big Horn National Forest. Sur-veys conducted with private campground ownerswere used to estimate the number of spaces rentedper day. An estimate was made of the averagenumber of nights each site was rented per year andthe average daily rate. Information collected frommotel and other lodging providers was used todetermine both the level and type of occupancy.Commercial travelers were separated from touristsbecause they represent different expenditure pat-terns and markets. Meetings and conventions wereincluded in the commercial travel. Although theyhave different expenditure patterns than individualcompany or government representatives, they are inthe area for a specific purpose. Convention centerstarget this market and are very competitive.

TourismMotels and other similar tourist lodging were esti-mated by subtracting the commercial rooms fromthe total number of rooms rented. Table 1 showsthe total number of rooms occupied by commercial

Table 1. Estimated occupancy rates, Big Horn Mountain area, 1989.

and tourist travelers. Table 2 gives the estimatednumber of nights Forest Service campsites wereoccupied. This estimate was based on 429 devel-oped camp sites with an occupancy rate of 62% forapproximately 92 days during the summer season.Private campgrounds were estimated to rent eachsite an average of 48 nights per year. Based on the969 sites shown in the inventory, this gives anestimate of 46,512 party nights per year.

During the summer season, assumed to be Aprilthrough November, approximately 17% of thevisitors surveyed said the primary purpose of theirtrip was to visit friends or relatives while they werein the area. It was assumed they stayed with thefriend or relatives and did not utilize commerciallodging facilities. These people accounted for41,769 party days or 142,015 person days. Thenumber of person days was calculated from theaverage size of party indicated by the surveyrespondents. Table 2 indicates over 892,000 peoplespent a night in the Big Horn area for tourismpurposes during 1989. This does not count thebackcountry campers or day visitors who visitedthe area, but did not spend a night.

Tourism and Recreation ExpendituresThe number of person days is combined with theexpenditures per person per day to estimate thetotal number of dollars tourism brought into the


Month Total roomsoccupied

Commercialrooms

Touristrooms

Summerparties

Winterparties

January 8,410 6,728 1,682 1,682

February 8,760 7,008 1,752 1,752

March 9,790 7,832 1,958 1,958

April 10,216 8,755 1,461 1,461

May 19,059 10,714 8,345 8,345

June 35,995 10,618 25,377 25,377

July 44,871 9,345 35,526 35,526

August 42,847 10,059 32,788 32,788

September 25,287 9,619 15,668 15,668

October 21,786 8,868 12,918 12,918

November 10,408 7,776 2,632 2,632

December 9,076 6,686 2,390 2,390

Totals 246,505 104,008 142,497 134,715 7,782

area. Lodging and eating and drinkingexpenditures are analyzed individuallybecause they represent specific areasthat can be cross-referenced withsecondary data sources.

Estimates of lodging expenditures de-veloped using the 1987 U.S. census ofservices data for Wyoming and thelatest sales tax collection figures indi-cate approximately $7 to $9 millionwere spent on lodging in the four-county area. Our estimate of $9.3million (Table 3) is slightly higher butwe believe it is justified due to thereporting procedures used for tax purposes. Sometax revenue could find its way into a differentindustry classification. Also, a small percentage oflodging is tax exempt.

Estimated expenditures at eating and drinkingestablishments in the area were divided into tour-ism, commercial travelers, area residents, and dayvisitors. Tourism expenditures were based on theestimated number of person days visiting the areaand their expenditures obtained from the visitorsurveys (see Table 3). Using the estimated persondays and the daily expenditures given by surveyrespondents indicates tourists spent $8.6 million ineating and drinking establishments (see Table 4).Commercial travelers are estimated to spend $22

Table 3. Estimated expenditures for private lodgingin the Big Horn Mountain area, 1989.

Table 2. Estimated number of tourists and recreationists visitingthe Big Horn Mountain area, 1989.

Seasons& accommodations

SummerFS Campgrounds

Private campgrounds

Motels, etc.

Friends/relatives

WinterMotels, etc.

Friends/relatives

Totals

per day per party. This is correlated to each night oflodging. Most, but not all, these travelers are alone.The figures include people attending meetings andconventions who tend to spend more per person,per day on food and beverage.

Local residents are the primary source of incomefor prepared food and drink establishments. A 1990study conducted in Park County indicated eachindividual spends approximately $320 per year inhis or her local community for prepared food andbeverages. This was the best estimate available andwas assumed to apply to the BHMA.

xpenditures shown in Table 4g and drinking sales to individu-

als passing through the areabut not staying overnight. Theexpenditures were calculatedas the residual from the $32.1million total expenditures ateating and drinking establish-ments in the area. Secondarydata sources indicate totaleating and drinking revenuesare between $27 and $29million annually. The $32.1million for total eating anddrinking expenditures shownin Table 4 is a gross figure.That is, it contains paymentsto food and beverage busi-nesses plus tips to employees.Data collected from survey

The day visitors' eare estimated eatin


Tourism/recreationPersondays

Expendituresper day

Totalexpenditures

SummerFS Campgrounds 90,540 0.00 0

Private campgrounds 158,141 3.40 537,679

Motels, etc. 458,031 12.43 5,693,325

Friends/relatives 142,015 0.00 0

WinterMotels, etc. 38,132 9.95 379,413

Friends/relatives 5,851 0.00 0

Total tourism 892,710 $6,610,417

Commercial/businessunits/year 104,008 26.00 $2,704,208

Total lodging 996,718 $9,314,625

Partydays

Numberper party

Persondays

24,470 3.7 90,540

46,512 3.4 158,141

134,715 3.4 458,031

41,769 3.4 142,015

7,782 4.9 38,132

1,194 4.9 5,851

256,442 892,710

Table 4. Estimated expenditures at eating and drinking establishmentsin the Big Horn Mountain area, 1989.

Total eating & drinking

respondents are gross estimates. People indicatehow much money they left in the restaurant, not theamount of the check.

Direct expenditures were calculated using totalvisitor days and expenditures per day data collectedfrom the visitor surveys. Table 5 shows the total in-area expenditures by tourists in the BHMA to be$34.6 million in 1989. The dollar expenditures areheavily weighted toward retail purchases andeating, drinking, and lodging. The low level ofexpenditures for the recreational services sectormay provide some indication of the number ofservices available. There is a clear indication thatmany of the businesses in the community benefitfrom tourism in addition to motels and restaurants.

Total Economic Activity of Tourismand Recreation in the BHMATotal economic activity is estimated using an input/output model and adjusted primary data. Two majoradjustments were made to expenditure data beforethey were run through the model. First, expendi-tures made through the trade sector are margin-alized at 25.5%. This is due to the large leakage ofdollars through the retail sector's purchases of

imported items. Only dollarsretained in the state can gener-ate new economic activity.

The second adjustment is inthe eating, drinking, andlodging sector. Expendituresare assumed to include tips of10.5% (see Table 6), 50 thisamount is attributed to house-hold income and not asreceipts to the providers of theservice. This level of tips issubstantiated from twosources. Operators of eatingestablishments estimated tipsbetween 8.5 and 15%. A

17,152,000 small, nonscientific survey ofemployees in the area wasmade to determine tips as apercentage of total employ-ment income. That varieswidely by type of business.

Some have virtually no tips, while others have arelatively high percentage. This adjustment alsowas warranted through use of the 1987 census ofretail trade for eating and drinking establishments.Those data were adjusted to reflect more currentconditions. That approach indicated the area wouldhave between $27 and $29 million in eating anddrinking sales.

Table 6 gives the direct expenditures in columnone. These expenditures were calculated from thedata in Table 5. Total expenditures from retailstores was $17.4 million, the sum purchased forgroceries and liquor through other purchases. Sincea high percentage of the dôllars from the initialpurchases leave the community to replace the itemssold, only the gross profit or $4.4 million is left forfurther expenditures in the area. The remainder,nearly $13 million, is shown as imports in Table 6.Lodging expenditures combined with purchasesfrom eating and drinking establishments gives atotal of $15.2 million from Table 5. Most of thisamount, or $13.6 million, goes to the eating anddrinking establishments while $1.6 million is tipsand shown as household income in Table 6. Themajority of expenditures for licenses and permits

4,000,000

32,077,750


Tourism/recreationPersondays

Expendituresper day

SummerFS campgrounds 90,540 2.78

Private campgrounds 158,141 5.33

Motels, etc. 458,031 11.62

Friends/relatives 142,015 11.62

WinterMotels, etc. 38,132 12.97

Friends/relatives 5,850 12.97

Total tourism 892,709

Commercial/businessunits/year 104,008 22.00

Area residentsperson years 53,600 320.00

Day visitorsexpenditures

Totalexpenditures

251,701

842,892

5,322,320

1,650,214

494,572

75,875

8,637,574

2,288,176

go to the state or federalgovernment and are listedunder other final payments.However, the U.S. ForestService returned $18,667directly to counties in thearea from campground feescollected on the Big HornNational Forest.

The indirect and induced orsecondary effects are shownin Table 6, column 2; thetotal economic impact oftourism is in the third col-umn. Tourism accounted for$56.3 million of total eco-nomic activity in the BigHorn Marketing area in1989. These figures are notto be confused or compared with the impact of totaltravel. It is important to segment travel into theproper components to generate accurate estimatesof economic impacts.

Income and EmploymentContributions to personal income and number ofjobs created are of interest to local leaders inevaluating potential benefits from economic devel-opment projects. Tourism does not provide identifi-able direct personal income through payrolls. Thatis, there are no tourism factories with reportedpersonal income statistics that are used to measureeconomic contribution in many industries. Directpersonal income from the $34.6 million of tourismexpenditures is only $1.6 million in the form ofgratuities. Purchases made from retail outlets andfrom eating, drinking, and lodging establishmentsand service providers creates $10 million in indi-rect personal income in the area through merchantpayrolls. This represents a large portion of theimpact from the original sale. The retail, food, andother services normally are thought of as support-ing the basic sectors in the economy such as agri-culture, mining, and manufacturing. However,these become basic sectors for that portion of theirproduct or service sold to tourists. These salesbring new money into the area which fuels theeconomy in the same manner as exporting oil or

Table 5. Estimated total direct expenditures (daily purchases per person)from tourism in the Big Horn Mountain area, 1989.

Summer WinterCategory Campers Noncampers visitors

Dollarpurchases

agricultural products. The household row in Table6 shows the direct and indirect components thataccount for the $11.6 million of total personalincome that can be attributed to tourism.

Personal income generated from tourism expendi-tures can be translated into jobs. Tourism in theBHMA is highly seasonal in nature, and it followsthat employment also is seasonal. Employment isestimated on a full-time equivalent (FTE) basis.One FTE is defined as one person working 2,000hours (full-time) for 50 weeks during the year. Therelationship between income and employment wascalculated for each sector in the economy and usedto estimate direct, indirect, and total employment.

The trade sector provides 145 FTE of employmentfrom tourism expenditures. Added to the 525 FTEfrom the eating, drinking, and lodging sector and57 FTE from the services sector, tourism accountsfor 727 FTE of employment through direct transac-tions. Most income associated with the directemployment is from merchant payrolls and isincluded in the indirectlinduced income effects inTable 6. The multiplier effect will create an addi-tional 304 FTE of employment annually. Thiswould provide an additional 87 FTE in trade, 200FTE in eating, drinking, and lodging, and 17 FTEin the services sector. Tourism accounts for a totalof 1,031 FTE of employment in the area on an


Lodging

Eating/drinking

6,610,417

8,632,574

Licenses/permits 1.40 0.68 0.42 660,796

Recreational services 0.30 1.57 1.61 1,288,327

Groceries/liquor 4.50 5.34 3.13 4,593,812

Gas/repairs/maint. 5.50 8.10 8.87 7,029,405

Equipment 0.85 0.47 1.15 483,886

Clothing/other retail 0.59 1.84 1,488,483

Gifts/souvenirs 0.65 3.78 1.08 2,972,298

Other purchases 0.19 1.18 911,863

Number of personsmaking purchases 90,540 758,187 43,982

Total purchases 34,631,861

Table 6. Direct, indirect and total economic impact from tourist visitationsto the Big Horn Mountain area, 1989.

annual basis. This represents over 4,000 jobsduring the peak summer season.

Potential for Tourism ExpansionThere has been'considerable discussion on the needto market tourism and recreation in the Big HornMountain area as well as in other parts of Wyo-ming. Some indication of the capacity to handleincreased numbers of travelers is needed to aid inmarketing efforts. Our data year, 1989, was not oneof the better years for tourism in the area. Table 7indicates excess capacity in the lodging sector atleast 9 months of the year. Any efforts to increasethe number of visitors during the winter would bevery beneficial to the entire service and retailbusiness community. Extending the shoulderseasons into April and May in the spring and intoSeptember and October in the fall would be mostbeneficial. Each of these efforts would require avery specific, narrowly targeted marketing effort.

There was excess capacity in the lodging sectorduring the summer in 1989. This may not havebeen the case in 1990, or at least not at the samelevel. Total rooms available would have accommo-dated one-third more travelers in June. This trans-lates into a potential to increase tourism by ap-proximately 71% assuming the same level ofcommercial travel. July had almost 20% excess

capacity with poten-tial for 31% moretourists. These fig-ures can becomesignificant whenconsidering in-creased occupancyrates.

We had uncon-firmed reports of a10 to 20% increasein occupancy ratesduring the summerof 1990 over 1989.If this isthe case, itprobably representsa 15 to 30% in-crease in tourism.Overall, lodging

will not be a long-run deterrent to increased tour-ism in the area. However, lodging may be a deter-rent for specific locations or peak periods, such aswinter lodging on the Big Horn Mountains orsummer lodging during peak weekdays, specialevents, and holidays. Thus, total utilization of thecurrent excess capacity likely will never be fullyattained. The private sector will respond to meetany increase in demand for lodging facilities. Busi-nesses can operate more efficiently with a longerseason and higher annual occupancy.

It also should be noted that commercial travel tendsto be higher in the summer when more field crewsare in the area. Any major construction project orincreased activity in the energy sector could requireup to 20% of the total lodging facilities available.These activities are good for business and the localeconOmy but should be recognized as potentialcompetition with tourism marketing efforts.

GrazingIn January 1992, members of the Coalition re-quested a second study to estimate the economicand fiscal impacts of both commodity and recre-ational uses on federal lands in the BHMA. TheCoalition specifically requested that livestockgrazing, timber harvest, petroleum and mineralextraction, water storage, and tourism and recre-


SectorsTotaldirect

Indirectinduced

Totalimpact

Other businesses 0 3,051,409 3,051,409

Transportation/communications 0 889,291 889,291

Utilities 0 1,185,380 1,185,380

Trade 4,447,135 2,634,321 7,081,456

Eating/drinking/lodging 1 3,642477 275,827 13,918,304

Finance, insurance, real estate 0 1,717,130 1,717,130

Services 1,288,327 1,044,534 2,332,861

Health 0 191,850 191,850

Local government 18,667 698,651 717,318

Households 1,600,514 9,969,505 11,570,019

Other final payments 642,129 0 642,129

Imports 12,992,612 0 12,992,612

Totals 34,631,861 21,657,898 56,289,759

ation activities be evaluated. This study was re-quested about the same time the USFS was consid-ering a 50% reduction in a grazing allotment beingtransferred in the Paintrock Grazing District in BigHorn County. The results of the grazing componentof that study are reported below.

Recently, there has been an ongoing debate overthe economic importance to local communities oflivestock grazing, timber harvest, and recreation onthe BHNF. It has been suggested that jobs lost fromdecreased grazing can be replaced by expandingtourism in the BHMA. The purpose of this sectionis to evaluate the economic impact of reducedlivestock grazing on the BHNF in the four-countyBHMA in terms of economic activity, personalincome, and employment.

ProceduresAn accurate assessment of the economic impact onadjacent communities from reducing livestockgrazing required estimating the structural change inindividual ranching operations that would resultfrom reducing the number of animals allowed tograze on the BHNF. This information was providedthrough preliminary results of a research project by

Table 7. Estimated tourism lodging capacity, Big Horn Mountain area, 1989.

Larry Van Tassell et al., Department of Agricul-tural Economics, University of Wyoming. Studyobjectives for the BHMA grazing study were:

To use Van Tassell's preliminary estimates ofthe structural change in individual ranchingoperations due to a reduction in the number ofanimals allowed to graze on the BHNF.

To estimate the economic impact of an animalunit month (AUM) of cattle grazing on theBHNF in terms of economic activity, personalincome, and employment supported or gener-ated in the local economy.

To estimate the economic impact caused by areduction in range cows in the BHMA due to25, 50, and 100% reductions in AUMs ofgrazing on the BHNF.

To estimate the number of tourism visitor days(TVD) required to offset the economic lossfrom grazing in terms of economic activity,personal income, and employment (see Eco-nomic Impact Comparison between Tourismand Grazing, below).

The impact of one AUM grazing on the BHNFrequired information on the total dollar sales per

* Lodging capacity is not a concern most of the year, so room availability was considered only during peakdemand months of June, July, and August. Some properties in the area are operated seasonally, due to lowoccupancy rates, in order to reduce costs.


Month

Totalrooms

occupiedCommercial

roomsTouristrooms

Roomsopen

Percentrooms

available

Increasedtourismcapacity

January 8,410 6,728 1,682 * na na

February 8,760 7,008 1,752 na na

March 9,790 7,832 1,958 na na

April 10,216 8,755 1,461 * na na

May 19,059 10,714 8,345 * na na

June 35,995 10,618 25,377 54,000 0.333 0.7095

July 44,871 9,345 35,526 55,800 0.196 0. 3 07 6

August 42,847 10,059 32,788 55,800 0.232 0.395 1

September 25,287 9,619 15,668 * na na

October 21,786 8,868 12,918 * na na

November 10,408 7,776 2,632 * na na

December 9,076 6,686 2,390 na na

Total 246,505 104,008 142,497

AUM under typical economicconditions in the BHMA andthe decline in range cow units(CU) the area would experi-ence for each AUM of grazinglost on the BHNF. An input!output model, developed forthe BHMA, was used to esti-mate the loss in total economicactivity, personal income, andemployment (Moline 1991,Moline et al. 1992). Assump-tions were as follows.

There are 120,828 AUMsof allotted grazing forcattle and sheep on theBHNF.

Only the 105,775 AUMsallocated to cattle areconsidered in this paper.

Loss of one AUM ofgrazing on the BHNF willreduce the cattle herd inthe BHMA by 0.069 CU.

Based on 1990 prices, thevalue of total output per range CU was $456.98in the BHMA.

Loss of each AUM of grazing on the BHNFreduces total area output for the range livestocksector by $31.53.

Markets will be available for all other agricul-tural commodities.

Recreational use will continue to rise at 2.8%per year with or without livestock grazing.

Tourism visitor days increase in proportion tothe growth rate of total recreational visitor days.

The bases for assumptions 1 and 2, the number ofAUMs of grazing allotted on the BHNF, were dataobtained directly from the district office of theBHNF. Assumption 3 is based on a linear program-ming model for a typical ranch in the BHMA. Thismodel estimated that the loss of one AUM ofgrazing on the BHNF will reduce the herd by 0.069cows after allowing for reallocation of alternativesources of forage. Assumption 4 was calculated bydividing the model ranch's total receipts by the

Table 8. Economic impact from the loss of one animal unit monthof grazing on the Big Horn National Forest.

Dollar flowsSectors

EmploymentDirect Indirect Total FTE

number of its range cows. The total amount,$456.98, includes each cow's share of calf, year-ling, cull cow, and cull bull sales as well as beefproduced and consumed by range cattle producers.

It was assumed that production per cow for themodel ranch is comparable to the average for theregion. Output per cow was used for several rea-sons. Within the linear programming model, thenumber of cows is the dependent variable; i.e., allactivities depend on the number of cows on theranch. Also, most producers utilize informationbased on per-cow figures. An additional reason forusing output per cow is the general public canunderstand what a cow is but may have difficultyunderstanding the concept of animal units.

Assumption 5 was calculated by multiplying theper-AUM reduction of range cows (0.069) by thetotal output of each range cow ($456.98). Assump-tion 6 allows the other agricultural commoditiessuch as hay and grain consumed by range cattle tomove to other markets.


Range cattle 31.53 0.00 31.53 0.000824Ag services 0.00 0.63 0.63 0.000017Timber 0.00 0.01 0.01 0.000000Oil & gas 0.00 0.16 0.16 0.000000Mining 0.00 0.09 0.09 0.000001.

Construction 0.00 0.38 0.38 0.000004Manufacturing 0.00 1.00 1.00 0.000003Transportation& communications 0.00 1.00 1.00 0.000016Utilities 0.00 1.05 1.05 0.00000kTrade 0.00 4.23 4.23 0.000154Eating,drinking, lodging 0.00 0.54 0.54 0.000021

Finance,insurance, real estate 0.00 2.57 2.57 0.000020Services 0.00 0.79 0.79 0.000025

Health 0.00 0.76 0.76 0.000014Local government 0.00 2.23 2.23 0.000053Households 0.00 16.69 16.69 0.0000o0

Totals 31.53 32.14 63.67 0.001156

Assumption 7 is based on the actual increase inrecreation on the BHNF for the most recent 5-yearperiod. Assumption 8 implies recreational used byresidents also will increase by 2.8%.

Economic impact of Grazing in the BHMAThe $31.53 shown in column 1 of Table 8 is thedirect financial loss to the range livestock sectorfrom the loss of each AUM of grazing on the BHNF.The second column indicates the distribution of theindirect and induced effects; the third column is thetotal economic impact of one AUM of grazing onthe BHMA. Column 4 provides the distributioneffect of changes in employment associated witheach AUM of grazing.

Information in Table 8 shows the total economicactivity, personal income, and employment associ-ated with 1,000 AUMs of grazing. Each 1,000AUMs of grazing would generate $63,667 of totaleconomic activity, $16,689 of which is personalincome. Each 1,000 AUMs also will support 1.16FTE jobs.

Two adjustments weremade in the inputloutputmodel to more accuratelyreflect the impact of areduction in AUMs ofgrazing that would resultin a reduction in totalrange cow units producedin the BHMA. First, thetotal requirements matrixwas adjusted to consider achange in output and toremove the need to esti-mate deliveries to finaldemand. Second, interac-tions between range cattleand other agriculturalsectors were removed toallow products that hadbeen consumed by rangecattle to be sold in othermarkets. A 25% reductionin grazing allowed on theBHNF was arbitrarilyselected to represent the

Table 9. Economic impact of a 25-percent loss of animal unit months (AUMs)of grazing on the Big Horn National Forest.

Dollar flows

SectorsRange cattle

Ag services

Timber

Oil & gas

Mining

Construction

Manufacturing

Transportation& communications

Utilities

Trade

Eating,drinking, lodging

Finance,insurance, real estate

Services

Health

Local government

Households

Totals

total impact that would result if a forestwide graz-ing policy was implemented at that level. A 25%AUM reduction would reduce grazing on theBHNF by 26,444 AUMs which would reduce thecow herd by about 1,825 in the BHMA.

Table 9 shows the direct loss to cattle producers tobe $833,814. This direct effect translates into a lossthroughout the area's economy of $1.68 million intotal economic activity, $441,384 in personalincome, and 30.56 FTE of employment.

A 50% reduction in AUMs permitted on the BHNFwould double the impact of the 25% reduction,resulting in a reduction of 52,888 AUMs grazed onthe BHNF with 3,649 fewer cows in area rangeherds. This would reduce economic activity by$3.37 million, personal income by $882,767, andemployment by 61.13 FTE. The most drastic policywould be to eliminate all cattle grazing from theforest. This action would reduce the cow herd byapproximately 7,298 cows, about $3.34 million ofoutput. This would translate into $6.74 million in

EmploymentDirect

833,814

o

0

0

0

0

0

0

0

0

0

o

0

0

0

0

833,814


Indirect Total FTE

0 833,814 21.78

16,585 16,585 0.46

395 395 0.01

4,257 4,257 0.01

2,416 2,416 0.02

10,059 10,059 0.10

26,548 26,548 0.08

26,367 26,367 0.41

27,655 27,655 0.11

111,929 111,929 4.08

14,414 14,414 0.55

68,064 68,064 0.54

20,826 20,826 0.65

20,143 20,143 0.37

58,912 58,912 1.41

441,384 441,384 0.00

849,952 1,683,767 30.56

economic activity, $1.77million in personal income,and 122 FTE of employment.

Economic ImpactComparison betweenTourism and GrazingThe third objective of thispaper is to estimate the num-ber of tourism visitor daysrequired to replace the eco-nomic activity, personalincome, and employment lostthrough decreased grazing onfederal lands. A tourismvisitor day (TVD) represents avisitor from outside BHMAspending 1 day in the localarea. Table 10 shows theimpact of one tourism visitorday to the BHMA. Directexpenditures of $38.79 perday were reported in an earlierstudy of tourism in the BHMA(Taylor et al. 1991).

For comparison, the impact of

Table 10. Total economic impact of one tourism visitor day (TVD)associated with the Big Horn Mountain area.

1,000 TVDs was estimated at$65,052 total economic activity, $13,852 personalincome, and 1.08 FTE jobs. A tourism sector is notincluded in the input/output model, so nonlocalvisitors' expenditures were specified as changes infinal demand for the appropriate sectors.

Table 4 compares the economic impact of 1,000AUMs of grazing and 1,000 TVDs to the BHMA.In order to replace the economic activity lost froma 1 ,000-AUM reduction in cattle grazing, tourismin the area would have to increase by 979 TVDs.More than 1,200 TVDs would be required toreplace the $16,691 of lost personal income fromgrazing. Replacing the 1.16 FTE of employmentlost with the reduced grazing would require 1,074TVDs. Tourism-related employment tends to payapproximately 11% less than employment gener-ated from grazing (Fletcher and Taylor 1992).

The level of total economic activity generated inthe area also can be compared, as shown in Table

12. If all cattle grazing was eliminated on theBHNF, there would be 105,775 fewer AUMs ofpublic grazing in the area which translates to 7,298fewer cow units of production. Column 2 of Table12 shows the increase in TVDs above the currentgrowth rate required to replace the loss in eco-nomic activity that would result from eliminating25% of cattle grazing on the BHNF.

Column 3 shows the TVDs required to offset theimpact of a 50% reduction of grazing. The fourthcolumn shows the increase in TVDs required toreplace the impact of totally eliminating cattlegrazing on the BHNF. The last column in Table 5shows the impact of TVDs on the BHMA in 1990.

With current levels of livestock grazing, recreationon the BHNF has been increasing at about 2.8%per year from 1986 through 1990 (USDA 1992). Itis assumed that this level of growth can continuewith or without livestock grazing. It is further


SectorsDollar flows Employment

FTEDirect Indirect Total

Agriculture 0.00 0.28 0.28 0.000005

Ag services 0.00 0.04 0.04 0.000001

Timber 0.00 0.02 0.02 0.000000

Oil & gas 0.00 0.23 0.23 0.000000

Mining 0.00 0.13 0.13 0.000001

Construction 0.00 0.39 0.39 0.000004

Manufacturing 0.00 1.63 1.63 0.000005

Transportation& communications 0.00 1.14 1.14 0.000018

Utilities 0.00 1.51 1.51 0.000006

Trade 4.98 3.02 8.00 0.000291

Eating,drinking, lodging 16.06 0.45 16.51 0.000625

Finance,insurance, real estate 0.00 2.05 2.05 0.000016

Services 1.44 1.09 2.53 0.000079

Health 0.00 0.63 0.63 0.000012

Local government 0.02 0.82 0.84 0.000020

Households 1.02 12.83 13.85 0.000000

Other final payments 0.72 0.00 0.72 0.000000

Imports 14.55 0.00 14.55 0.000000

Totals 38.79 26.26 65.05 0.001083

Table 11. Economic impact and comparison of grazing and tourismon the Big Horn National Forest.

Type of impact

Total economic activity

Total personal income

Total employment (HE)

Table 12. Number of tourism visitor days (TVD) required to replace grazinganimal unit months (AUM) on the Big Horn National Forest.

Assumed levels of reduced grazing

nomic losses if grazing on the BHNF is reduced oreliminated. It is not certain that tourism visitor dayscan increase at a rate necessary to offset the loss inemployment and income caused by reductions incattle grazing on the BHNF. As Cordell et al. (1990)note, American leisure activity has changed drasti-cally in recent years. Leisure time has decreased by37% in the last 15 years. In addition, leisure activi-ties are close to home due to an aging population,childbearing in the "baby boom" generation, home-video entertainment, reduced leisure time, andenergy costs. Typical American families now planfor shorter, but more frequent, vacations.

These changes have affected the type of recreationuse at national forest sites. Total recreation use atnational forest sites increased by over 16% from1977 to 1987, an average annual increase of 1.6%.However, the proportion of all trips that were 2hours or less in travel time increased from 43% in1977 to 72% in 1986. By comparison, the propor-tion of all trips that were 8 hours or more in traveltime decreased from 23% in 1977 to 6% in 1986.

Similarly, the proportion of all trips that were 1 dayor less in length increased from 30 to 79%, whilethe proportion of all trips that were more than 1 dayin length decreased from 70 to 21%. Thus, at thenational level there appears to have been a substan-tial shift in recreation activity on national foreststoward greater use by regiohal residents. This shiftin national trends for recreation has important


$63,670 $65,052 979

$16,691 $13,852 1,205

1.16 1.08 1,074

assumed that TVDs willincrease in proportion tothe growth rate in totalrecreational visitor days.Given these assumptions, atourism growth rate above2.8% would be required toreplace the loss in personalincome from reducedgrazing. However, depending on the grazingreduction and recreation use increase, it could takeseveral years in terms of net present value for theincome stream generated by additional tourismvisitor days to offset the lost income stream associ-ated with reduced grazing on the BHNF.

Table 13 shows the timeframe needed for theincreased income generated by additional TVDs tooffset the loss in income associated with reducedgrazing on the BHNF. A 4.0 social discount ratewas used to estimate these breakeven points. Threepotential growth rates are considered: 3.5, 4.2, and5.6%. These growth rates represent a 25, 50, and100% increase in recreational growth beyond thepresent rate of 2.8%. If grazing on the BHNF werereduced 25% and recreation increased to 3.5%, itwould take 10 years for the local economy torecover the loss in income. At a 4.2% increase, itwould take 5 years; at a 5.6% increase, it wouldtake 2 years for income generated by recreation tooffset the income lost by a grazing reduction.

If grazing on the BHNF decreased 50%, 22 yearswould be required to offset lost income if recre-ation increased 3.5% per year, 10 years at a 4.2%growth rate, and 5 years at a 5.6% growth rate. Ifcattle grazing were eliminated on the BHNF, thelocal economy would need 53 years at a 3.5%growth rate for recreation to offset the loss fromgrazing, 22 years at a 4.2% recreation growth rate,and 10 years at a 5.6% recreation growth rate.This paper does not intendto imply any relationshipbetween grazing levelsand increased tourism. Itis the intent of this paper Type of impact

to point out the number of Total economic activity

tourism visitor days Total personal incomerequired to offset eco- Total employment (HE)

25% 50% 100%

25,884 TVD 51,768 TVD 103,537 TVD

31,869 TVD 63,738 TVD 127,476 TVD

28,223 TVD 56,445 TVD 112,891 TVD

TVD equal1,000 Animal 1,000 Tourism 1,000

Unit Months (AUM) Visitor Days (TVD) AUM

implications for increasing the growth rate ofBHNF-related tourism in the BHMA. Inparticular, the shift in demand toward moreregional recreational use at national forest sitesmay increase the time required to generate thefuture increases in visitation to the area neces-sary to expand the local economy sufficientlyto offset the loss from reduced livestockgrazing. For example, based on a profile ofcurrent summer visitors to the area (Taylor et al.1991), at least 80% of these visits involve drivingtime of 8 hours or more, almost the opposite of thenational trend. In light of this, the area may bedoing well to sustain the current average annualgrowth rate of 2.8% in tourism visitor days, and aconsiderably lower growth rate may not be overlypessimistic.

SummaryWyoming continues to depend on federally ownedland for economic activity. Livestock grazing andrecreation are two activities on public lands that areextremely important to Wyoming's economy. Thispaper examines the economic effects of decreasesin cattle grazing allowed on the Bighorn NationalForest located in north-central Wyoming. A 25%reduction of the 105,775 AUMs of grazing allottedto cattle would cause the economy of the four-county area surrounding the BHNF to decrease itsactivity by $1.68 million, $441,384 of which wouldbe personal income. The local economy also wouldlose about 30.56 FTE. A reduction of 50% wouldcause economic activity to decline $3.37 million,personal income to decline $882,767, and employ-ment to decline about 61 FTE. Eliminating allcattle grazing on the BHNF would cause economicactivity to decline $6.74 million, personal incometo decline $1.77 million, and employment todecline 122 FTE.

An increase in the growth rate for recreation couldhelp offset the negatives associated with eliminat-ing cattle grazing on the BHNF. To offset thedecline in economic activity, recreation would haveto increase about 104,000 TVDs. Recreation wouldneed to increase about 127,000 TVDs to offset theloss of personal income. Depending on the growthrate of recreational use, 10 to 53 years could be

Table 13. Years for breakeven incomefrom reduced grazing and increased recreation.

required for the BHMA economy to recover theloss in income from reduced livestock grazing. Tomaintain the current employment level, recreationwould have to increase about 113,000 TVDs.

The shift in demand toward more regional recre-ational use at national forest sites may increase thetime required to generate the future increases invisitation to the area necessary to expand the localeconomy sufficiently to offset the loss from re-duced livestock grazing. At least 80% of the TVDsin the BHNF involve driving time of 8 hours ormore, almost the opposite of the national trend.Therefore, the area may be doing well to sustainthe current average annual growth rate of 2.8% inTVDs, and a considerably lower growth rate maynot be overly pessimistic.

Despite these concerns, increasing tourism remainsa viable strategy for revitalizing the BHMAeconomy. Currently, it is one of the few uses ofpublic lands with the potential to expand. However,it becomes a less attractive strategy if it can beincreased only at the expense of other industries inthe BHMA. Given the potential trade-offs and theuncertainty involved, a balanced approach thatconsiders all alternatives is needed to sustain andrevitalize the area's economy.

ReferencesCordell, H. K., J. C. Bergstrom, L. A. Hartmann, and D.

B. K. English. 1990. An analysis of the outdoorrecreation and wilderness situation in the UnitedStates: 1989-2040. U.S. Forest Service GeneralTechnical Report RM-1 89.

Fletcher, R. R. and D. T. Taylor. Measuring economicimpact of tourism versus travel in north-centralWyoming. Paper presented to Mid-continent Re-gional Science Association meetings, 1992,Stiliwater, OK.


Assumed annual recreationand tourism growth rate

Grazing reduction 3.5% 4.2% 5.6%

25% 10 years 5 years 2 years



Fletcher, R. R., D. T. Taylor, B. R. Moline, 0. W.Borden, G. J. Skidgel. Economic contributions offederal lands within the Big Horn Mountain Area,May 1994. Department of Agricultural Economics,University of Wyoming, Laramie.

Moline, B. R. The impact of increased range cattleproduction on the Wyoming regional economies.M.S. thesis, 1991, University of Wyoming, Laramie.

Moline, B. R., R. R. Fletcher, and D. T. Taylor. Aggre-gated vs. disaggregated input-output models. Paperpresented to Western Regional Science Associationmeetings, 1992, Lake Tahoe, NV.

Taylor, D. T., R. R. Fletcher, and T. Clabaugh. Tourismin the Big Horns: a profile of visitors, attractions,and economic impact, 1991. Department of Agricul-tural Economics, University of Wyoming, Laramie.

Taylor, D. T., R. Ship, and R. R. Fletcher. Park Countyhousehold expenditure study: a report to ParkCounty Commissions, 1991. Department of Agricul-tural Economics, University of Wyoming, Laramie.

USDA Forest Service. 1992. Draft supplementalenvironmental impact statement, Bighorn NationalForest. Lakewood, CO.


Integrating Utilization Measurementsinto Monitoring Programs

AbstractUtilization has been controversial for 50 years. Thesolution is to use the methods properly and in con-text of a total land-use program. Utilization shouldnever be used as an objective but rather in supportof clear ecological objectives. Successful grazingmanagement must be ecologically and economi-cally sound. Utilization measurements can contrib-ute to making proper decisions to help attainobjectives.

IntroductionThis series of papers illustrates the history of theuse of utilization measurements in monitoringrangelands. The papers clearly show the benefitsand liabilities in applying utilization concepts inrange management. The historical review pointsout the concerns of scientists about applyingutilization methodologies in inappropriate ways,either in the context of management objectives orby misapplication of the methods. This dialoguehas been conducted irregularly over the last 50years. The reasons for historical and current de-bates over utilization are imbedded in the scientificand management cultures. Managers need tools thathelp them make defensible decisions. They natu-rally are drawn to a concept as intuitively attractiveas utilization. The concept is simple: A givenamount of forage can tolerate a given amount ofherbage removal; when that point is reached, quitgrazing that area. The scientists are concerned thatwhat is simple in concept is immensely complex indetails. Translation of the concept through rapidlycollected field data or observations is exceedinglydifficult. So the controversy ebbs and flows overtime. Large-scale programs over time have focusedheavily on utilization for monitoring. After consid-

William C. KruegerDepartment of Rangeland Resources

Oregon State UniversityCorvallis, OR 97331-2218

eration, the emphasis declines, and utilization isreduced in importance.

Throughout the presentations of the scientists andmanagers in the symposium, a few central ideaswere accepted generally. The most prominent areaof agreement was that utilization is a land man-agement tool, not a land management objective.The current controvery about utilization on range-lands is based primarily on the perception that, inpractice, a utilization standard frequently is used asa management objective with no context of theecological goal for the site being managed. All thescientists found significant fault with this misuse ofthe utilization concept, as did the Forest Serviceland managers.

The contributors to this report point out in detailthat each technique has its own specific strengthsand weaknesses. Consequently, the land managerneeds a good understanding of how and when touse each utilization technique so an appropriatetechnique that gives reliable and accurate results isimplemented.

The authors explain many of these details in theirpapers. Reliability is the repeatability of measure-ments. The relationship of year and season, as wellas the knowledge of field workers, has great effecton reliability. Accuracy refers to the actual correct-ness of the measurement. This often is affected bythe knowledge of the technique, specific ability ofthe technique to measure the vegetation under themonitoring circumstances present, and the care inimplementing the checks and balances of eachtechnique. Both reliability and accuracy are impor-tant and must be considered if the monitoringprogram is to yield useful background informationfor decision makers.


All the authors indicated the complexity of thisconcept. An effort to reduce the complex to rules ofthumb or to simplistic practice simply will notwork. The effect of seasonal changes in vegetationmakes measurement difficult under the best ofcircumstances. The relationship of timing of graz-ing related to vegetation response suggests a needfor a gradient of standards, not a single site-specificstandard. Most of the techniques do not yield exactmeasurement of forage removed. They are notinherently accurate. The concensus is that anyutilization method is useful when used in referenceto an ecological objective and when integrated withother appropriate data that allow the manager toinfer long-term trend in rangeland condition.

The most universally accepted use of utilizationtechniques is to develop large-scale utilizationmaps. These maps will highlight areas of livestockconcentration and low use. When combined withother pasture information, they can help lead themanager to incorporate management strategies toencourage livestock to graze some areas more andother areas less. Yearly maps are useful wheninterpreted with other yearly information such asweather patterns, grazing system, and herd history.

The Forest Service contributors suggested the landmanagement objectives include uplands and ripar-ian zones and stressed ecological function anddesired land conditions. With these decisions made,the manager can determine what needs to be moni-tored and how utilization fits into a monitoringstrategy to evaluate progress toward objectives. Inaddition, the Forest Service authors say that moni-toring should be based on techniques that federalgrazing permittees or concerned citizens canreadily implement. They agree that utilization has aplace in yearly monitoring programs.

The complex needs for information naturallyrequire a wide variety of approaches to collectacceptable data. If detailed measurements areneeded and if statistical reliability is important,then very detailed and expensive monitoringapproaches are necessary. It is not responsible tosample inadequately and then make importantmanagement decisions based on faulty data. Ifnumbers are collected as part of the monitoring

program, they always should be statistically ana-lyzed so the decision maker knows the quality ofthe information. This at least allows a confidenceinterval to be established. With that, the value ofthe information for the manager is clear. Forexample, utilization of 35%, ±5%, does not meanthe same as utilization of 35%, ±30%. If it is notpossible to collect sufficiently reliable utilizationdata, then monitor with a qualitative technique. Formost management needs, photo points are fullyadequate.

Researchers are testing the use of photographictechniques using geographic information systemstechnology for quantification. These tools showpromise for future objective evaluation of residualstanding crop of shrubs. However, the technologyis several years away from being practical.

What do we in WCC-40 suggest in using utilizationmethods for monitoring grazing use?

Develop clear ecological objectives.

Use site-specific objectives and techniques.

Monitor parameters that relate to the objectives.If you are interested in plant vigor, utilizationcould be a good choice to monitor. If you areinterested in plant community change, soilmoisture and timing of grazing could be goodchoices to monitor.

Understand the quality of your data so you caninterpret them effectively.

Select a monitoring tool you can interpret.

Use feedback to adjust management.

If the toolutilization--is used right, it will work.

Finally, remember that decisions that result frommonitoring programs can influence the number ofanimals grazing an area. Whether the land is publicor private, there is an economic result from thedecision. The stocking rate on rangelands must besustainable. If not, it will decline due to deteriora-tion. Reductions in grazing that are not required tosustain the resource can cause real economichardship to the ranch family and consequently tothe community. Economic impacts are importantand need full consideration in making land-usedecisions.


stubble height and utilization measurements · of grazing that might be allowed in order to...

Documents