Integrated Livestock Production and Conservation: Managed Goat Grazing in Open Oak Systems

Final Report for GNC12-149

Project Type: Graduate Student
Funds awarded in 2012: $9,973.00
Projected End Date: 12/31/2012
Grant Recipient: University of Wisconsin-Madison
Region: North Central
State: Wisconsin
Graduate Student:
Faculty Advisor:
John Harrington
University of Wisconsin-Madison
Expand All

Project Information


Two different rotational goat browsing regimes and a control were tested in an oak savanna complex with a dense, shrubby midstory in Southwest Wisconsin using three 0.5-hectare paddocks replicated in five blocks. After two seasons under the heavy browse regime goats (1) reduced shrub cover and height though not stem density, (2) increased light at the groundlayer and (3) increased cover and richness of sun-favoring herbaceous species, without increasing soil compaction or cover of invasive species. Weight gain among kids exceeded expectations and levels of gastrointestinal nematode parasites were considered acceptable for meat production and goat health.


Management and restoration of any natural ecosystem requires a toolset that can support the goals of the land manager. In ecosystems dependent upon fire, additional tools are often needed as substitutes in situations where fire cannot be employed as an initial treatment to allow fire in the future, or as a more selective treatment between prescribed burns. Intensive rotational grazing has been explored as a vegetation management tool in the U.S. and globally, often with different species and breeds of livestock as each has its own feeding preferences. Although unmanaged grazing can cause an array of environmental problems, including soil compaction and erosion (Fleischner 1994), these impacts can be controlled when carefully managed (Papanastasis 2009, Teague et al. 2011). Intensive rotational grazing is a highly managed method in which livestock are grazed at high stocking densities for a short period of time and rotated successively through a series of small paddocks (Barnes et al. 2008). Vegetation is then allowed to recover during a resting period before livestock are rotated through the same paddocks again in order to avoid environmental impacts often arising from long-term grazing (Bailey and Brown 2011). This study explored the use of rotational goat browsing as an additional tool to restore oak savanna in the North American Upper Midwest and the viability of raising meat goat on dense midstory forage.

Oak savanna, a once-widespread, fire-dependent ecosystem having an herbaceous understory and partial oak canopy, is now categorized as critically threatened due to post-European settlement conversion to farmland and suppression of fire. In the absence of fire, the understory of surviving savannas have filled in with shade-tolerant shrub and sapling species, decreasing light at the ground layer and changing plant community composition (Nowacki and Abrams 2008). The majority of oak savanna remnants in Wisconsin are located on hillsides marginal for farming, most of which are on private land (Wisconsin Department of Natural Resources 2012). Some poorer quality remnants are restorable, but restoration methods are often time intensive with prohibitive costs. Fire is touted as the most effective restoration tool; however, private landowners often perceive fire to be a high liability with few economic benefits (Henderson 1995). In addition, fire can deter shrub colonization in oak savanna systems but can be ineffective in penetrating very dense vegetation with damp fuels (Nowacki and Abrams 2008). In similar mesic prairie systems, burning often results in the aboveground expansion of shrub and fails to kill well-established shrubs (McCarron and Knapp 2003, Heisler et al. 2004). Other methods, such as mechanical removal and herbicides, are limited by cost, slope, weather and potential damage to the herbaceous layer.

Continuous grazing practices have been destructive to ecosystems in the U.S. and globally regardless of livestock species (Auclair 1976, Fleischner 1994). Intensive, short-duration rotational browsing in which livestock are regularly rotated between paddocks may help reduce shrub cover while enhancing livestock production without the damage associated with continuous grazing (Walker et al. 2006). Rotational grazing of Scottish Highland cattle, a breed with a high browse diet, has been studied as an alternative means of shrub removal in oak savanna in Wisconsin. This breed was effective at reducing aboveground growth for specific shrub species and was observed feeding in areas where shrubs were too dense for fire to penetrate. Despite successes, the logistics of frequent transportation of such large animals required for a short duration rotational grazing regime is an obstacle (Harrington and Kathol 2009).

Meat goats have higher browse diets than the Scottish Highland cattle and are smaller and more mobile. Therefore, they may be better candidates for reducing a dense, shrubby midstory in oak savannas. Goats have been studied as agents for woody fuel reduction, invasive species management, pasture maintenance and ecosystem restoration in structurally similar ecosystems in the U.S. and globally (Batten 1979, Tsiouvaras et al. 1989, Severson and Debano 1991, Perevolotsky and Haimov 1992, Torpy et al. 1993, Popay and Field 1996, Luginbuhl et al. 1999, Valderrábano and Torrano 2000). In oak savanna, rotational goat browsing may have the potential to decrease shrub and sapling layer cover and density. Reduction of the shrub canopy in similar midwestern grassland ecosystems has resulted in more light at the herbaceous groundlayer and enhanced the available habitat for the growth of desirable sun-tolerant herbaceous species, particularly warm-season grasses (Heisler et al. 2004, Nowacki and Abrams 2008, McGranahan 2011). Additional sunlight and herbaceous growth would provide a greater, more continuous, fine fuel matrix providing opportunities for the reintroduction of fire as part of the future management plan.

In addition to the potential conservation advantages, the demand for goat meat in the U.S. is increasing, creating a potential economic incentive to raise goats for meat. The U.S. imported more than 16,000 metric tons of goat meat in 2012, up 6% from 2011 and up 45% from 2007 (Rayer 2013,). U.S. consumption of goat meat is expected to continue to increase as ethnic groups for whom goat meat is dietary staple become more prominant (Solaiman 2007). If meat goats could be raised on the forage in a shrub-infested oak savanna while simultaneously providing an ecological service this would result in a mutually advantageous situation for both livestock producers and land managers.

This study seeks to determine the extent that rotational browsing with goats can be both an effective restoration tool for reducing a dense oak savanna midstory and meet basic standards for goat meat production? If rotational goat browsing can reduce the shrub layer without negative impacts on the soil, herbaceous layer or goats, this approach has the potential to be an effective and more accessible restoration tool for ecologically important but shrub-choked open oak ecosystems.

Project Objectives:
  • Reduce midstory shrub layer cover and height
    Increase light availability at the groundlayer leading to an increase in the herbaceous layer
    No increase in soil compaction or invasive species
    Increase in goat kid weight greater than 2.7kg/month (rate of gain considered to be viable for meat production).
    No significant worsening in Famacha scores


Click linked name(s) to expand
  • Bruce Folley
  • John Harrington


Materials and methods:
Site Background

The study site is located on a 12-hectare piece of land within the Yellowstone Wildlife Area (YWA) in Lafayette County, Wisconsin. Lafayette County is part of the larger Driftless Area of southwest Wisconsin. The Wisconsin DNR categorizes the Driftless Area, the portion of land not glaciated during the Wisconsinan glacial period, into two ecological landscape designations with the study site located in the Southwest Savanna region. This region is characterized by broad, open ridge tops and steep, wooded ravines with predominantly silt loam soils that are occasionally thin and rocky on steep slopes (2012). The context is rural with small family farms that contain oak savanna in a conditoin similar to that of the study site.

Presettlement vegetation maps show the study area to have been a mix of prairie and deciduous hardwood forest with predominant tree species including Quercus macrocarpa and Q. alba with some Carya ovata and Q. rubra. The study site is located on the steep transitional slope between a ridge top and valley. Slope steepness ranges from approximately 12% to 30% with the USDA soil survey showing a band of thin rocky soils in middle of the slope (Natural Resources Conservation Service 2011). Bedrock in the area is part of the St. Peter formation and composed of sandstone with some limestone and shale (Mudrey et al. 1982).

The Wisconsin DNR acquired the parcel of land containing the study site in 1989 which, prior to purchase, had been used to graze beef cattle. At the time of purchase the land was degraded due to severe overgrazing even though grazing activities had ceased some time before the purchase. The vegetation community is similar to many degraded oak savanna and woodland remnants currently being targeting for restoration. In the absence of fire and historical browse pressures the canopy filled in with shade-tolerant mesic species such as Ulmus americana and Tilia americana along with numerous shrub species which then shade the understory. In 2008, the site was selectively logged leaving a canopy of approximately 30% (mostly Quercus alba and Q. macrocarpa) and the understory was forestry mowed (Folley 2012). The sudden increase in sunlight led to rapid shrub colonization, particularly by Zanthoxylum americanum, Cornus racemosa, Lonicera x bella, Populus tremuloides and Rubus spp.

Goat Browsing

In 2011, five replicate blocks were located within the 12-hectare site avoiding curving slopes so that each block faced a single aspect. Each 1.5-hectare block was divided into three 100-meter by 50-meter paddocks (0.5 hectares). Paddocks were positioned in such a way that the steepest part of the slope occurred in the center of every paddock. Each paddock was randomly assigned a goat browse treatment: light browse (two-day) or heavy browse (four-day) or a control (no browse). Paddocks were divided in half in order to fully cross goat browse treatments with interseeding. One half of each paddock was randomly assigned and broadcast with a native seed mix while the other half received no treatment. As interseeding may affect goat browse preference at some time in the future goats were fenced within the each half paddock. All block units were buffered with a 1.5-meter mowed swath. The five replicate blocks were positioned so that treatment and control paddocks were oriented up and down the hill, parallel to the slope.

Through consultation with the goat provider, past studies from Australia, and recommendations from the western U.S. the determined stocking rate was 86 goats per 0.5-hectare unit during the 2011 field season. The same rate was used for the 2012 season of study. Goats were browsed on both sections of a treatment paddock and then rotated to the next. A full rotation through all five replicate blocks took 30 days—five two-day treatment sections plus five four-day treatment sections. Previous studies suggest that brushy plants must be browsed as least twice in a single growing season in order for the impact to be long-term (Davis et al. 1975, Hart 2006).

For this study a two-cycle season was optimal due to the constraints of sampling, classes and funding. The goal was for goats to consume 90 percent of shrub foliage in the heavy browse paddocks during both the first and second rotations. One-day and two-day treatments were used during the first rotation of 2011 but were insufficient to reduce the majority of foliage. Browse treatment length was evaluated and doubled to two-day and four-day treatments for the second 2011 rotation. In 2011, the first rotation began June 8 and ended June 22; the second rotation began July 18 and ended August 17. During 2012 the first rotation began June 7, 2012 and ended July 7, 2012. In both years, each paddock was rested a total of 38 days before beginning the second to allow browsed shrubs to regrow leaves.

The two-day and four-day browse treatments were maintained for the first rotation of the 2012 field season. A drought began near the end of the first rotation resulting in a seven-inch precipitation deficit by the end of the summer (U.S. Drought Monitor 2012). The drought limited the regrowth of leaves and the four-day heavy browse treatment was shortened to three days due to lack of forage for the goats during the second rotation. The second rotation began July 16, 2012 and ended August 10, 2012. Drought is also suspected to have impacted the interseeding treatment as the severe lack of precipitation occurred at a time when small plants were just establishing. No evidence of the treatments was recorded at any sampling time and thus, will not be discussed further in this report.

Data Acquisition

Sampling occurred late May prior to goat browsing treatments and in August following both browsing rotations. Permanent nested quadrats were used to monitor the effects of the treatments: 1-square-meter quadrats for herbaceous species; 5-square-meter circular quadrats for shrub and sapling density and height. Quadrats were selected based on a stratified random design to account for the influence of slope with 18 quadrats per paddock. Precautionary five meter buffers were used around each quadrat to ensure that there was no overlap of shrub cover measurements.

All quadrats were sampled for herbaceous species presence/absence and relative cover during June and August sample times. Cover was visually estimated using six cover classes (0-5%; 5-25%; 25-50%; 50-75%; 75-95%; and 95-100%) in accordance with the Daubenmire method (Daubenmire 1959). Litter depth was measured in all paddocks in August only. Twelve of 18 quadrats per paddock were randomly selected as the center of the circular 5-square meter shrub quadrats and measured for cover, height and stem count. Shrub cover percentage was estimated using a cover board at a distance of five meters taken both directions perpendicular to the hill slope (National Applied Resource Sciences Center 1999). The cover board was 2.5 meters tall by 0.25 meters wide and divided into five 0.5 meter bands of alternating black and white. It is a modification of another visual obstruction method, the Robel pole, which is used to estimate vegetation biomass (NARSC 1999). Cover was estimated for each band using the six Daubenmire classes.

Light availability was assessed through the measurement of leaf area index (LAI) using an AccuPAR LP-80 ceptometer (Decagon Devices 2010). Measurements were taken at six quadrat points within each paddock. LAI uses photosynthetically active radiation (PAR) measurements above and below the shrub canopy to estimate how much leaf area per square meter is require to obstruct the amount of light equal to the difference between above and below canopy PAR measurements. Although PAR is the direct measurement of light intensity, it varies with time of day, cloud conditions, and tilt and direction of the light wand. Therefore, LAI was used as a normalized indicator of light reaching the ground layer.

Rotational browse of any livestock has the potential to cause soil compaction which can constrain plant root growth and therefore nutrient acquisition (Brock 1988). Soil compaction measurements were obtained using a soil compaction tester, also known as a penetrometer, after the last goat rotation and sampled at six random quadrats in each paddock. The penetrometer was pushed straight down into the ground at a steady rate until the meter reached the 300lb pressure threshold at which the soil is too compact for root growth (DICKEY-john Corporation 1987). The length of the rod submerged in the soil was then recorded to approximate the depth of soil available for optimal root growth. Due to the rocky and thin nature of the steep hillside soil many sample depths were limited by hitting rock, not the compaction of the soil. This method, while quicker, does not give estimates of soil bulk density. Estimation of the depth of root growth using this device is prone to fluctuations in accuracy depending on soil moisture content (DICKEY-john Corporation 1987); however, when samples are taken in the same day under identical weather conditions, the results are useful to compare treatments and control without comparing between samples taken under different conditions.

Goat dietary intake was monitored for eight days spaced over the total 30 days required for the full rotation. This was done for each rotation during the growing season. Six random goats (three in each half paddock) were each observed for five-minutes four times throughout the course of the day for forage species and type (herb, shrub/tree foliage or twigs). Goat health was assessed using Famacha test scores and body weights which were assessed by the goat owner before the first rotation in June and after the second rotation in August within 24 hours of removal from paddocks for both years of the study. The Famacha test estimates level of gastrointestinal nematode infection in an individual (causing anemia) by assessing the intensity of tissue color under the eyelid in goats and sheep. Color is assessed on a scale of one to five. A score of one indicates a deep red color suggesting very low levels of infection while a score of five indicates a pale pink color suggesting a near lethal level of infection (Lewandowski 2010). Body weight was used to assess overall goat health as well as weight gain in kids raised for meat.

Research results and discussion:
Vegetation Response

Goats were able to reduce the shrub layer height and cover and increase light levels while meeting health and weight gain standards for meat production. The herbaceous layer richness and cover showed neither damage from the first season of browse in 2011 nor evidence of decline or influx of problematic aggressive species. Although litter declined, there was no evidence of soil compaction. This combination of results strongly suggests that rotational goat browsing could be both an effective and economically viable restoration tool for the removal or containment of dense brush in oak savannas of the Upper Midwest.

In 2012, goats significantly decreased shrub cover and by 23.2% and 25.4% and height by 28.8cm and 30.5cm in the light and heavy goat browse treatments between June before browsing and August following two rotations (p < 0.01 for all). The heavy browse treatment reduced cover the most but there was no significant difference between the two browse treatment. Informally, goats were observed engaging in a variety of feeding behaviors that affected shrub height and cover. Goats ate soft annual twig growth, leaf petioles, and Cornus racemosa and Rubus spp. fruits in addition to leaf blades. Goats also scraped their horns against larger sapling, eating the bark when it began to peel. If this activity completely circled the sapling, it often resulted in the death of the plant above the scraping. In order to reach higher forage goats stood on their back legs and pushed over saplings or pulled down branches with their upper bodies. This behavior often resulted in limbs or the main plant stem cracking.

Some large goats were capable of pushing over saplings up to 5cm in diameter and taller than 2.5 meters in height. However, this activity was not sufficient to significantly impact vegetation greater than two meters in height. This suggests the usefulness of goat browsing as a restoration tool is limited by the height of vegetation targeted for control and the consistency with which the tool can be used. After just four years of growth after forestry mowing, many shrubs and saplings at the study site were too tall to be visibly reduced in height even under the heavy browse treatment. Relatively short gaps in treatments of any tool can be sufficient to either erase progress made by prior treatments or to limit the usefulness of certain tools. In this case, goats must be browsed frequently enough that shrub do not grow above the two-meter reach of the goats for this approach to be effective in constraining height.

Rotational goat browsing is expected to take several season of repeated defoliation to show substantial declines in shrub stem density (Davis et al. 1975, Hart 2006). Overall stem density remained the same over the course of the study. Even if goats snapped the main stem of a shrub or sapling, I observed regrowth of leaves from below the snap. Defoliation, such as is the result of repeated browsing, or other above ground damage can result in the rapid expansion of clonal shrubs following the disturbance (McCarron and Knapp 2003, Heisler et al. 2004). Underground carbohydrate stores and access to deeper resources allow for the growth of seedlings as well as vigorous resprouting from established plants (Ratajczak et al. 2011). Stem density among clonal species increased significantly under the heavy browse treatment after just one season of rotational browse in 2011 but did not show any additional increases during later samplings in 2012.

The reduction in shrub cover corresponded to an increase in light availability at the groundlayer indicated by lower LAI in browse paddocks. For LAI, both treatments and the control were significantly different from each other with control paddocks having the highest LAI (2.7) and heavy browse paddocks having the lowest LAI (0.6; p < 0.01 for all). Although an increase in light at the groundlayer was expected to result in an expansion of the herbaceous layer as observed in similar oak savanna environments (Packard and Balaban 1994, Bowles et al. 2007, Brudvig 2010), no increase in overall cover or richness occurred. Cover and richness of sun-favoring species increased significantly in the heavy browse paddocks in June 2012 following the first season of browsing in 2011. While there was no overall expansion in the herb layer, the increase among sun-favoring species indicates a desired warming and drying of the groundlayer that could favor native savanna species in the future, particularly warm season grasses. Overall herbaceous cover declined in browsed paddocks immediately following browsing rotations in August 2012 (p < 0.01 for both). This is suspected to be a result of trampling rather than consumption as goats only spent 18% of time eating herbaceous material. Of the herbaceous species that were consumed most were observed to be tall erect forbs within easy reach at goat height.

A single moderately sized area in which vegetation had been trampled until there was only bare soil at the ground layer was observed in almost every paddock. I observed goats gathering consistently in a specific area when not feeding that corresponded with a bare patch that appeared by the time goats were moved to the next paddock. The experiment design may have contributed to this result as half of the herd was simultaneously fenced on one side of the paddock and the other half on the other side in order to accommodate the fully crossed interseeding treatment. Goats being social herd animals, the two halves of the herd often congregated at the dividing fence. This could be remedied by fencing the herd in a single paddocks and strategically placing water troughs and mineral feeders. Given the extremely dense midstory and lower quality of the herbaceous plant community, the immediate effects of moderate trampling in browsed paddocks were not particularly concerning. However, the results and observations suggest that use of goat browsing in a high quality remnant will require additional study as trampling may be destruction to threatened or endangered herbaceous species.

Despite the observed trampling, the lack of difference between treatments and controls during June 2012 following the first season of browsing indicates that the herbaceous layer was capable of recovering from any immediate negative impacts sustained during the rotations. Litter depth was also affected by trampling. A lack of litter over top of the soil often results in increased levels of erosion due to direct trampling on the soil and increased desiccation (Brady and Weil 2009). However, a large accumulation of litter can suppress the germination of many herbaceous layer species (Pavlovic and Grundel 2009). Litter depth decreased by 4.1mm to 11.6mm in the heavy browse paddocks in comparison to the control paddocks (p = 0.01). Soil compaction was not observed in any of the browsed paddocks, nor did I find evidence of the spread of invasive species due to browsing.

Goat Health and Weight Gain

The demand for goat meat in the United States is increasing but production is not keeping pace (Rayer 2013). This study did not explore whether or not an individual could profitably raise and sell meat goats raised on shrub forage oak savannas. However, this study does show that under a shrub forage regime adult goats stay healthy enough to remain part of a breeding herd as indicated by weight and level of nematode parasitism and that kids gain weight at a rate at which they are considered profitable to sell for meat.

For young goats, called kids, a weight gain of 2.7 kilogram per month is considered normal and viable for meat production though that rate varies depending on goat breed and forage quality (Luginbuhl and Poore 1998, Hart 2006). Between June 7 and August 15 when goats were on the study site during the 2012 season kids gained an average of 4.4 kilograms per month. This is well above the normally expected rate suggesting that forage in oak savanna areas with dense midstories could be used to profitably raise meat goats.

Differential changes in weight were observed among kids, nursing does, and non-nursing with kids gaining a significant 10.0kg and non-nursing does losing a significant 2.3kg. Goats can reach physical maturity within a year so large weight gains for animals younger than a year are expected. As nursing does are still providing nourishment to their kids, minor weight loss is not considered unhealthy. That weight among nursing does did not change even with the stress of extremely warm weather suggests that forage was sufficiently plentiful and nutritious to sustain their weight. Weight loss among non-nursing does would not be expected except that most of the goats comprising this category were sourced from a different location than the rest of the goats. The producer providing goats for this study year purchased 30 lactating, but no longer nursing, does in order to meet stocking density requirements for the study. Prior to arriving at the study site these animals had been on a diet of grain to boost milk production and many were considered overweight (Robel 2012). Increases in exercise regimen, lower energy content in the diet and gradual reduction of udder size would all contribute to the documented weight loss which the producer considered healthy.

Overall, goats showed no change in the level of nematode parasitism indicated by Famacha score. However, scores did worsen significantly among the kids. This is contrary to literature that states intensive rotational grazing of this nature generally has a decreasing effect on gastrointestinal nematode parasitism (Hart 2006). Other literature found infection levels in goats in temperate climates to decrease as age increases which is also in contrast to these results; however, nematode infection was also found to be less during the winter months which may partially explain the increase over the summer with infection levels being greatest between July and September (Tariq et al. 2010). The drought and subsequent decrease in forage quality may have also increased kid susceptibility to higher levels of parasitism.

No standard Famacha scores for production of goat meat were found in the literature. Nematode parasitism does not directly affect the quality of the meat or milk, but high levels of infection will cause lower production of meat and milk (Lugenbuhl 1998). Dewormers used to treat gastrointestinal nematodes will affect production standards as they have withdrawal times during which meat and milk cannot legally be harvested for consumption (SCSRPC 2012). Conventional use of Famacha scoring on sheep holds that any animal scoring a three, four or five should be treated. Goats, however, usually have higher rates of infection than sheep and are more resistant to treatment (SCSRPC 2012). Therefore, treatment of nematode parasites is often subject to the preferences of the farmer. The owner of the goats used during the 2012 season of the study does not treat his goats unless it scores a five (the worst) in the Famacha test, preferring to keep the product as free of chemicals as possible. He was very happy with the condition of the animals when they were moved from the paddocks at the end of the 2012 season. He was primarily concerned that none of the goats scored a five and none worsened more than two scores (Robel 2012).

Participation Summary

Educational & Outreach Activities

Participation Summary

Education/outreach description:

A field day and pasture walk was hosted at the study site in August 2012 to present the current results of the study and directly demonstrate the impacts of rotational browsing on a dense shrub midstory in oak savanna. The day was attended by 50 people interested in land management (public and private) as well as local goat producers. A field day in 2011 drew an estimated 60 individuals.


The Prairie Enthusiasts annual conference. February 25, 2012. Menomonie, Wisconsin. Title: Integrated livestock grazing and conservation: managed goat grazing and interseeding in oak savanna and woodlands. Poster. Authors: K. J. Baumann, J. M. Ela and J. A. Harrington.

Natural Areas Association Conference. October 9-12, 2012, Norfolk, Virginia. Title: Opportunities for livestock production and conservation: managed goat grazing in oak savanna and woodlands. Oral Presentation. Authors: K. J. Baumann, J. M. Ela and J. A. Harrington.


Baumann, K. J. 2013. Rotational goat browsing as a tool for reducing midstory shrub density in oak savanna. Masters Thesis. University of Wisconsin-Madison, Madison, Wisconsin.

Paper for journal publication in progress.

Project Outcomes

Project outcomes:

Evidence of decline in shrub height and cover, increase in key herbaceous species groups, a lack of soil compaction, increases in kid weight, and no worrisome decline in health all suggests that rotational goat browsing in areas with dense shrub midstories would supplement acreage necessary to raise meat goats and at the same time support the goals of oak savanna restoration. This creates a situation that is potentially a win-win situation for both the goat producer and land owner that could lead to a long-term cooperative relationship between the two.

Intact, surviving savanna remnants are very rare in the Upper Midwest; however, land that was once open oak savanna prior to agricultural conversion (including woodlands and barrens), but now densely populated with midstory shrubs and saplings, is abundant. Positive weight gain and health results in this study suggest that such land would be a great opportunity for goat owners to expand meat goat operations. In addition to potentially expanding available grazing land, the goat owner would produce a meat product that is actively beneficial to the local environment and wildlife. Forage in dense, shrubby midstories would also be cost efficient. This is particularly true given current demand for goat meat and observed kid weight gain that exceeded the threshold for profitable meat production while feeding in a shrub-infested savanna environment. Land owners could provide forage free of charge in the form of overgrown oak savanna, and the rotational browsing of the goats would provide an ecological service for the land owner without the higher costs or logistics of other shrub control methods.

For land managers or those interested in improving the ecological health of savannas, the frequency with which goat browsing can be applied is among its greatest advantages over other tools. Fire is an essential tool for restoring oak savanna, but frequency of use is limited by site and weather conditions, burning regulations and fuel load. Given perfect conditions for the use of fire every year, a savanna can be burned at most once annually. In reality, even under ideal weather conditions most oak savannas can be burned only once every two or three years in order to allow sufficient time to accumulate the necessary fine fuels to carry prescribed burn. Goats, on the other hand, can be rotationally browsed through an area multiple times in a single year, returning to previously browsed areas once vegetation has sufficiently resprouted.

In contrast to most methods of mechanical shrub layer removal, rotational browsing results in gradual shrub reduction and increased light availability. This provides time for the adjustment of the herbaceous layer to warmer and drier conditions, potentially limiting opportunity for the spread of very aggressive species. Sudden and complete removal of the shrub layer at Yellowstone Wildlife Area via forestry mowing in 2008 resulted in vigorous and very dense resprouting in subsequent years that quickly diminished the initial increase in herbaceous layer cover. Post-disturbance response among clonal shrub species, aided by shared below-ground carbohydrate stores, is often particularly strong (Rataczjak 2011). The results of this study suggest that if the initial removal treatment had been immediately followed by a secondary treatment such as browse, the level of resprouting may have been reduced. For lands where an overpopulation of shrubs exist, several years of multiple goat browsing rotations that fully defoliate the shrub layer has the potential to deplete belowground carbohydrate stores, thus limiting the resprouting response.

While land managers already have a number of savanna restoration tools available, rotational goat browsing could help increase the efficacy of existing methods. Goat browsing should not be used to the exclusion of other tools, particularly fire which is the primary disturbance to which these ecosystems are adapted. The use of rotational browsing as a restoration tool would be most appropriate for savannas heavily invaded with shrubs. However, goat browse could be used as a pretreatment for two to three years to increase groundlayer light availability and herbaceous fuel accumulation prior to the reintroduction of fire. In savannas that have sufficient fuel loads to carry a fire, goats could be used as a substitute during years in which conditions do not permit a prescribed burn. Goats could also be used as a secondary treatment after mechanical removal of a dense, shrubby midstory in which goats are employed to repeatedly consume resprouting vegetation.

In summary, the addition of rotational browsing to a land manager’s toolbox not only provides an additional management option but expands the effectiveness of other tools as well. Rotational goat browsing offers new management possibilities to private land owners, who are often dissuaded from restoration attempts by cost and liability concerns associated with other tools. Lower cost and the potential for cash revenue with fewer liability or safety concerns make this a tool that is accessible to both would-be restorationists and professional land managers.

The success of this study to date will lead to a third season of using goats on the site, additional field days, a survey of goat owners on current browse practices, and the potential development of a handbook for both goat owners and land managers on effective methods of goat browsing for increasing ecological quality.

Farmer Adoption

This study leased goats from two goat producers who run businesses using goats for vegetation management and was able to confirm that goats can in fact provide the services that are advertised. Field days in 2011 and 2012 were attended by local farmers and a 2011 presentation at the annual Focus on Goats conference in Platteville, WI was attended by goat producers interested in using their goats to improve land quality in addition to expanding available grazing land. One of the long-term goals of this study is to foster a relationship between goat producers and land managers such that managers benefits from shrub control provided by the goats and producers benefit from access to widely available forage. Future farmer adoption will require some ecological training as rotational browsing in some areas and under certain circumstances could be ecologically detrimental. Quality of area targeted for restoration and soil moisture levels, in particular, should be considered. Higher quality remnants may have threatened or endangered plant species that could be damaged by trampling while trampling under very wet conditions, either permanent or temporary, can lead to exacerbated soil compaction.


Areas needing additional study

  • Economic analysis of the use of goats for vegetation management and simultaneous meat production
    Long-term impact of browsing on presence and cover of invasive species
    Impact of goat browsing on soil erosion
    Impacts of goat browsing in high quality oak savanna remnants
    Response of vegetation after removal of browse pressure

Literature Cited and Consulted

A list of references used in this report can be found below.

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.