ALN logo; link to Arid Lands Newsletter home page No. 58, Winter 2005
Soil management for drylands

Soil carbon and microbial biomass carbon after 40 years of grazing exclusion in semiarid sagebrush steppe of Wyoming

by Gyami Shrestha, Peter D. Stahl, Larry C. Munn, Elise G. Pendall, George F. Vance and Renduo Zhang

"[Results of our study] are consistent with the proposal that well-managed grazing may be an option for sequestering carbon in grasslands."


(Back to top)
The sagebrush steppe of Wyoming is classified by the World Wildlife Fund as a desert and xeric shrubland biome with vulnerable conservation status. It is arid or semiarid and has a surface area of 132 400 km2 (~51 120 mi2). Dominated by sagebrush (Artemisia tridentata), fescue (Festuca spp.) and wheat grasses (Agropyron spp.), it faces disturbances like severe winds, grazing, burning and variations in precipitation and temperature (WWF 2001). Between 1959 and 1965, with the objective of studying the effect of grazing on sagebrush steppe vegetation, approximately 100 exclosures were established in central Wyoming as part of a cooperative agreement between the University of Wyoming's Range Management Section and the Bureau of Land Management. Soil organic carbon (SOC) and soil microbial biomass carbon (SMBC) were studied in four of those grazing exclosures between 2003 and 2005. Our research was conducted with the objective of evaluating the impact of long-term grazing exclusion on SOC and SMBC in Wyoming sagebrush grasslands. SOC, measured as a percentage by weight, provides an estimate of how much organic matter there is in the soil. SMBC generally only represents about 1-3% of total soil carbon, but it is an important indicator of nutrient turnover and storage in a given soil-plant association because it is highly labile (that is, it can undergo rapid changes). Both of these measurements provide an indication of overall soil health as well as a way of determining how much carbon is being sequestered.

Managed grazing on previously degraded grasslands was added to the Kyoto Protocol's list of Land Use, Land Use Change and Forestry (LULUCF) activities for carbon sinks (that is, sequestration) during the Marrakesh Accords in 2001 (UNFCCC 2005). Managing the available vast expanses of grazing lands in the US for enhanced carbon sequestration and storage could prove very valuable both for offsetting present carbon imbalances and for preserving resources by maintaining overall soil health.

Review of past and current literature indicates that soil carbon pool, microbial biomass and soil structure are interlinked during carbon accumulation and storage by the soil (Sparling 1992; Breland and Eltun 1999; Fließbach and Mäder 2000; Nilsson 2004). Milchunas and Lauenroth (1993) did a detailed literature review of 34 studies involving grazed and ungrazed sites around the world and reported both decrease (40%) and increase (60%) in soil carbon as result of grazing exclusion. More recent studies of grazed soils worldwide have also shown both increases (Schuman et al. 1999; Reeder et al. 2004) and decreases (Derner et al. 1997; Yong-Zhong et al. 2005) in carbon storage and accumulation compared to adjacent ungrazed soils.

The null hypothesis (H0) for this study was that long-term grazing exclusion has no effect on the aforementioned soil parameters, while the alternative hypothesis (H1) was that it exerts a significant effect on them.

Study sites

(Back to top)
The four study sites, each comprising an exclosure and some immediately adjacent grazed soil, were located in the semiarid sagebrush steppe region of Fremont County, Wyoming. The vegetation type in all the exclosures is sagebrush steppe grassland (Monger and Martinez-Rios 2001) with Wyoming big sagebrush (Artemisia tridentata Nutt. ssp. Wyomingensis Beettle and Young) as the most dominant shrub. Studied exclosures had minimal external disturbance and fence breaches since time of establishment.

Study sites were established at four exclosures:

  1. Granite Mountain (GM),
  2. Upper Government Draw (UGD),
  3. Shoshoni Ant # 8 (Shon 8) and
  4. Shoshoni Ant # 9 (Shon 9).

Soil textures at the sites ranged from sandy loam to clay loam. Average annual precipitation (years 1990-2003) for all four sites approximated 20-21 cm [~7.9-8.3 in] while altitudes ranged from 1600 m (~5250 ft) (Shon 8 and Shon 9) to 1800 m (~5906 ft) (UGD) and 2100 m (~6890 ft) (GM). Using the standard that 1 Animal Unit Month (AUM) equals the amount of forage required by one mature cow and her calf (or the equivalent, in sheep or horses, for instance) for one month, average stocking rates (years 1993-2003) outside the four exclosures (rotational grazing system) were approximately 0.12, 0.08, 0.013 and 0.013 AUM/acre for GM, UGD, Shon 8 and Shon 9 exclosures respectively (1 acre = 0.4047 hectare).

Research methods

(Back to top)
The basic experimental design was a one-factor experiment with three levels (treatment, microsite, depth). At each study site, soil sampling was conducted separately both inside and outside the grazing exclosures, along three 100 m (~328 ft) long transects at three points 20 m apart from each other. These transects were either entirely inside or outside the exclosures. Soil from inside an exclosure was the "ungrazed soil" treatment while the soil outside was the "grazed soil" treatment. At each sampling point along a transect, samples were collected from under three microsites--sagebrush, grasses and bare interspaces--and at two depths (0-5 cm [~0-2 in] and 5-15 cm [~2-6 in]). The total number of samples per treatment in each study site was 54 (3 transects per study site*3 replicate points on each transect*3 microsites at each point*2 depths at each microsite).

After air-drying and sieving through a 2 mm (~0.078 in) sieve for rock removal, carbon analysis was conducted as follows:

  1. A Carlo Erba Carbon Nitrogen Analyzer was used to measure total carbon and nitrogen concentrations for samples from the GM and UGD exclosure sites. An Elementar Variomax Carbon Nitrogen Analyzer was used for the samples from Shon 8 and Shon 9 exclosure sites.
  2. The modified pressure calcimeter method, as described by Sherrod et al. (2002), was used for analysis of inorganic carbon in fine-ground soil samples.
  3. Soil organic carbon content (SOC) was calculated by subtracting the inorganic carbon content from the total carbon contents in each soil sample.
  4. The SOC/ha was calculated from SOC and soil bulk density data (Madden 2005) from the study sites.
  5. The chloroform fumigation-extraction method described by Vance et al. (1987) and Howarth and Paul (1994) was used for extracting SMBC.
  6. Organic carbon concentration in each SMBC extract was determined using the Phoenix 8000 UV-Persulphate TOC Analyzer.

Statistical analyses, except correlation matrices, were conducted using Minitab 13.1 statistical software (MINITAB 2000). The general linear model (GLM) was employed for analysis of variance between treatments, microsites and soil depths within a study site. Tuckey's post-hoc analysis was used for significant interactions and for significant factor comparisons. The a-value was set at 0.05. For analysis of overall means across all microsites and depths within treatments, a 2-sample T-test was employed.

Summary of results and discussion

(Back to top)
SOC concentration in the studied soils ranged from 3.67 mg/g dry soil (0.37%) to 53.8 mg/g dry soil (5.4%). Statistically significant differences in SOC concentrations were not observed in soils inside and outside any of the four sampled grazing exclosures, although significant interactions of grazing treatment with microsite and depth were observed as follows:

  • At GM exclosure, a significant 3-way interaction of treatment, microsite and depth on SOC and SOC/ha was observed, indicating that long-term grazing exclusion exerted greater influence on soils at 0-5 (~0-2 in) depth under shrub canopies than on soil under other microsites and depths.
  • At UGD exclosure, a significant interaction of microsite and depth on SOC concentration was observed, as would be expected, with greater SOC at upper soil depths and under shrubs and/or grasses compared to lower soil depths and bare interspaces. This occurrence is explained further below. However, no significant difference due to treatment was observed.
  • A significant 2-way interaction of depth and treatment on SOC was observed at Shon 9 exclosure. Grazing exclusion was seen to influence soil at 0-5 cm (~0-2 in) more than soil at 5-15 cm (~2-6 in), as expected.
  • Significant 2-way interactions of microsite and depth were observed on SOC at UGD, Shon 8 and Shon 9 exclosures, as expected.

SMBC concentrations ranged from 99 to 1011 µg/g dry soil in the study sites. Greater SMBC was observed in the ungrazed soils of GM and Shon 9 sites than in their grazed soils. In all of the study sites except for Shon 8, greater SMBC concentrations were generally observed in soil from 0-5 (~0-2 in) depths, under shrubs and in ungrazed treatments than in soil from 5-15 cm (~2-6 in) depths, from other microsites and in grazed treatments.

Overall, our study showed lack of significant difference due to grazing exclusion alone on SOC and SOC/ha between grazed soil and soil not grazed for more than 40 years. In other words, this particular study was not able to disprove the null hypotheses with regard to SOC. Kieft (1994) observed similar results with lack of significant differences in SOC between grazed lands and land not grazed for 11 and 16 years.

Our results disagree with a similar study by Schumann et al. (1999), who found greater SOC in grazed soil than in soil not grazed for 40 years. There are minor similarities with Hiernaux et al. (1999) who observed that after 4 years of controlled grazing in a Sahelian rangeland, SOC decreased slightly compared to ungrazed control but after 9 years of intensive grazing, SOC did not decrease further. Our results may have been different from these prior studies because our ungrazed sites had been grazed prior to exclusion and our grazed sites had been grazed for years. Hiernaux et al. (1999) conducted their grazing studies on previously ungrazed soils. Dormaar and Wilms (1998) observed no change in total soil carbon percentage (11.4% in no grazing and 11% in light grazing of 1.2 AUM/ha) after 42 years of light grazing in Canadian fescue grasslands, although it was seen to decrease with increase in grazing pressure.

Although our study data indicated that less than 40 years of grazing removal did not significantly affect SOC, SMBC did exhibit significant changes because it is a more labile fraction of soil organic matter. Thus, this experiment did provide significant support for our alternate hypothesis with regard to SMBC. Specifically, we observed significantly greater SMBC in the ungrazed soil than in the grazed soil in two of our four study sites (GM and Shon 9). These results are similar to Fliessbach and Maeder (2000) who observed early system-induced changes on SMBC while comparing conventional and organic farming systems, whereas no such changes were observed on total soil organic matter (SOM), of which SOC is a significant component. Contradictory to our results, Kieft (1994) did not observe significant and consistent changes on SMBC due to grazing exclusion in grasslands of New Mexico.

Shon 8 was the only site in our study that demonstrated slightly but insignificantly greater MBC in grazed soil than in ungrazed soil. This could be attributed to greater protection of SOM by aggregates and better environments for microbial proliferation in the Shon 8 soil, resulting in faster microbial assimilation of available SOC in the grazed soil than in the ungrazed soil. It could also be due to differences in soil texture--Shon 8 has much higher proportions of sand than GM and UGD sites. However, this latter explanation may need further investigation since these same parameters at Shon 9 exclosure, with its similar soil texture, were differently influenced by grazing exclusion.

In most cases, soils under shrubs also had the highest SMBC concentration, followed by soil under grasses and bare interspaces. Kieft (1994) observed similar results, attributing them to the canopy effect for both grasses and shrubs and to the "resource island" effect for shrubs as discussed above.

It may be that more than 40 years of grazing exclusion is required for SOC to exhibit significant changes. This could be a reason why significant differences in SOC were not observed between soils inside and outside the grazing exclosure, in spite of significant interactions of treatment, microsite and/or depth. The average bulk densities of soil in the study sites did not show significant difference between grazing or non-grazing treatments. There was no evidence of compaction due to grazing. Since soil porosity and soil bulk density are inversely related and porosity is directly related to SOC concentrations, insignificant differences in bulk density and porosity between grazed and ungrazed treatments might be responsible for insignificant differences in SOC and SOC/ha.

Compared to studies that showed changes due to grazing exclusion, the lack of significant differences in SOC concentration between grazed and ungrazed soils in our study sites may be due to:

  1. historical grazing practices and grazing intensities before and after grazing exclusion; and/or
  2. grazing by small mammals throughout inside the grazing exclosures (Kieft 1994).

As mentioned above, recent grazing history in our study sites indicates very light and well-managed grazing outside the four studied grazing exclosures. Well-managed grazing improves nutrient cycling in grassland ecosystems, stimulating aboveground production as well as root respiration and exudation rates (Schuman and Derner 2004). Our results may have differed from prior results showing grazing-induced SOC increase (Schumann et al. 1999, Reeder et al. 2004) or decrease (Derner et al. 1997, Yong-Zhong et al. 2005) because of differences in stocking rates, climate, soil and/or vegetation types in our study sites. As mentioned by Post and Kwon (2000), the length of time and the rate of carbon accumulation in the soil vary widely from one area to another, depending on the productivity of the vegetation recovering from environmental changes, physical and biological conditions in the soil and the history of soil organic carbon input and physical disturbance. Differences in methodology of sampling and analyses, plant response to grazing, and photosynthesis of grazed plants may also be responsible for different results of grazing on SOC (Schumann et al. 2001).

Within each microsite, greater SOC was often observed in the 0-5 cm (~0-2 in) depths of soil. This trend supports other similar findings like Derner et al. (1997) who found that maximum SOC accumulation was restricted vertically to the 0-5 cm (~0-2 in) depth. The surface layers of soil tend to accumulate more SOC than the deeper layers, due to deposition and decomposition of litter at the surface. When left undisturbed due to grazing exclusion, this process can presumably occur more easily than when grazing occurs, because grazing may be affecting the amount and rate of litter deposition and decomposition on the surface soil.

Greater SOC concentration was mostly observed under shrubs rather than under grasses or in bare interspaces. Soils under grasses, in turn, had higher SOC concentrations than bare interspaces. In all study sites, soil from the interspaces had the lowest SOC concentration. These overall results probably reflect the following:

  • Soil resources tend to accumulate under shrubs, which have more extensive root systems and canopies than do grasses (Bird et al. 2002).
  • Soils beneath plant canopies receive greater organic matter inputs, contain more water, are protected from disturbance (Bird et al. 2002), and hence can accumulate more organic carbon than interspace soil devoid of plants. Canopy effects due to grasses and shrubs were also observed by Kieft (1994) on SOC and MBC in the semiarid grasslands and shrublands of New Mexico, during studies on grazing and plant canopy effects.
  • Grasses tend to accumulate more SOC than bare interspaces, due to shoot and root organic matter input and deposition of plant litter redistributed from surface soils between plants during their life span (Derner et al. 1997).


(Back to top)
Data from this study indicate no significant difference in SOC between grazed soils and soils not grazed for more than 40 years. Significant differences in SMBC, however, indicate that grazing exclusion may have induced some differences in carbon dynamics and nutrient cycling in the soil. Microbial activity and health may also have improved throughout the years of exclusion. Further, managed grazing outside the exclosures in our study sites seems to have had a stabilizing impact on the soil organic carbon accumulation. These results are consistent with the proposal that well-managed grazing may be an option for sequestering carbon in grasslands.

Carbon sequestration, in addition to reducing the amount of carbon dioxide in the atmosphere, increases organic matter in soil leading to greater structural stability, water retention, pollutants fixation and toxicity reduction (FAO 2001). Managing the vast expanses of available grazing lands in the US for enhanced carbon sequestration could prove very valuable for offsetting the present carbon imbalance as well as for resource preservation. The organic carbon content in our study sites ranged from approximately 6 Mg/ha to 16 Mg/ha. Our findings support the idea of managed grazing as a carbon sink as proposed during the Marrakesh Accords in 2001. Further studies in this area should include studies of long term grazing exclusion from sites which have not been grazed previously.


(Back to top)
Special thanks to the Department of Renewable Resources of the University of Wyoming, Dr. Snehalata Huzurbazaar, Dr. Ann Hild, Dr. Lachlan Ingram and all the colleagues and technical help of University of Wyoming. We also thank the Lander Bureau of Land Management staff, particularly Mr. John Likins, for their help in gathering secondary information for the grazing exclosure study. This study was funded by the University of Wyoming Agricultural Experiment Station Competitive Grants Program.


(Back to top)
Bird, S. B., J. E. Herrick, M. M. Wander and S. F. Wright. 2002. Spatial heterogeneity of aggregate stability and soil carbon in semiarid rangeland. Environmental Pollution 116: 445-455.

Breland, T. A. and R. Eltun. 1999. Soil microbial biomass and mineralization of carbon and nitrogen in ecological, integrated and conventional forage and arable cropping systems. Biology and Fertility of Soils 30: 193-201.

Derner, J. D., D.D. Briske and T.W. Boutton. 1997. Does grazing mediate soil carbon and nitrogen accumulation beneath C4 perennial grasses along an environmental gradient? Plant and Soil 191(2): 147-156.

Fließbach, A. and P. Mäder. 2000. Microbial biomass and size-density fractions differ between soils of organic and conventional agricultural systems. Soil Biology and Biochemistry 32(6): 757-768.

Howarth, W. R. and E. A. Paul. 1994. Microbial Biomass. In Methods of Soil Analysis, Part 2. Microbial and Biochemical Properties, 753-773. SSSA Book Series No. 5. Madison, WI: Soil Science Society of America.

Kieft, T. L. 1994. Grazing and plant-canopy effects on semiarid soil microbial biomass and respiration. Biology and Fertility of Soils 18: 155-162.

Lecain, D. R., J. A. Morgan, G. E. Schuman, J. D. Reeder and R. H. Hart. 2000. Carbon exchange rates in grazed and ungrazed pastures of Wyoming. Journal of Range Management 53: 199-206.

Madden, S. Long-term effects of domestic livestock removal from Wyoming big sagebrush dominated rangelands: vegetative diversity and soil stability. (Master's thesis, University of Wyoming, 2005).

Milchunas, D. G. and W. K. Lauenroth. 1993. A quantitative assessment of the effects of grazing on vegetation and soils over a global range of environments. Ecological Monograph 63: 327-366.

Monger, H. C. and J. Martinez-Rios. 2001. Inorganic carbon sequestration in grazing lands. In The potential of U.S. grazing lands to sequester carbon and mitigate greenhouse effect. R. F. Follett, J. M. Kimble, and R. Lal, eds., 87-118. Boca Raton, FL: CRC Press.

Nilsson, K. S. 2004. Modeling soil organic matter turnover. (Ph.D. diss., Swedish University of Agricultural Sciences, 2004). Acta Universitatis Agriculturae Sueciae, Silvestria 326. Online:

Reeder, J. D., G. E. Schuman, J. A. Morgan, and D. R. Lecain. 2004. Response of organic and inorganic carbon and nitrogen to long-term grazing of the shortgrass steppe. Environmental Management 33(4):485-95.

Schuman, G. E. and J. D. Derner. 2004. Carbon sequestration by rangelands: Management and potential. Paper presented at the Western Regional Cooperative Soil Survey Conference. Online:

Schuman, G. E., J. D. Reeder, J. T. Manley, R. H. Hart and W. A. Manley. 1999. Impact of grazing management on the carbon and nitrogen balance of a mixed-grass rangeland. Ecological Applications 9(1): 65-71.

Schuman, G. E., J. E. Herrick and H. H. Jansen. 2000. Soil C dynamics of rangelands. In C Sequestration Potential of US Grazing Land, eds. R. F. Follet, J. M. Kimble and R. Lal, 267-290. Chelsea, MI: Ann Arbor Press.

Sherrod, L. A., G. Dunn, G. A. Peterson, and R. L. Kolberg. 2002. Inorganic carbon analysis by modified pressure-calcimeter method. Soil Science Society of America Journal 66: 299-305.

Sparling, G. P. 1992. Ratio of microbial biomass carbon to soil organic carbon as a sensitive indicator of changes in soil organic matter. Australian Journal of Soil Research 30: 195-207.

U.N. Framework Convention on Climate Change (UNFCCC). 2005. Land use, land use change and forestry (LULUCF). Online:

U.S. Department of Agriculture, Natural Resources Conservation Service (USDA-NRCS). 2005. Soil Survey Geographic (SSURGO) database for Fremont County, Wyoming, East Part and Dubois Area. Online:

Vance, E. D., P. C. Brookes and D. S. Jenkinson. 1987. An extraction method for measuring soil microbial biomass C. Soil Biology and Biochemistry 19: 703-707.

World Wildlife Fund. 2001. Wyoming Basin shrub steppe (NA1313). Online:

Yong-Zhong S., L. Yu-Lin, C. Jian-Yuan and Z. Wen-Zhi. 2005. Influences of continuous grazing and livestock exclusion on soil properties in a degraded sandy grassland, Inner Mongolia, northern China. CATENA 59 (3): 267-278.

bar denoting end of article text

Author information

(Back to top)
Gyami Shrestha ( is presently a Ph.D. student of Environmental Systems at University of California Merced. This study was conducted while she was an M.S. student at University of Wyoming, Laramie. Peter D. Stahl and Elise G. Pendall are Associate Professors at University of Wyoming, Laramie. Larry C. Munn and George F. Vance are professors there. Renduo Zhang, previously professor at the same university, is currently at Sun Yat-sen University, China.

About the Arid Lands Newsletter

Link to ALN home page Link to index page for back web issues Link to index page for pre-web issue archive Link to this issue's table of contents