Effects of dormant season
grazing on herbage production and plant growth
from
the North Dakota Agricultural Experiment Station 2002 Unified Beef
Cattle and Range Research Report
(for complete report go to http://www.ext.nodak.edu/extpubs/ansci/beef/2002/beef01.htm)
Mitch Faulkner, Kevin Sedivec, Jackie Olson, Tim
Faller, Jack Dahl, and Greg Lardy
In May through November, seasonal forage availability declined considerably
between the time of peak production and the beginning of the winter
grazing season. Considering these losses, stockpiling of forage throughout
the growing season for use in late fall or winter resulted in lost
herbage production potential. Furthermore, AUMs/ha for winter-only
grazing areas were severely reduced relative to season-long grazing
use. Incorporating a brief early-summer grazing period on winter pastures
could increase land use and reduce economic loses by increasing stocking
rates (AUMS/ha).
From an ecological and land-use efficiency perspective, a dormant
season grazing system that incorporates moderate early summer use
combined with winter stocking rates utilizing 50% of the standing
plant biomass is a preferable, and moreover, a beneficial management
alternative. This method yielded greater herbage production than other
treatments and resulted in greater needle-and thread and thread-leaf
sedge leaf heights than the season-long or DS 30 grazing treatments.
This method, however, reduced western wheatgrass leaf heights late
in the growing season. If dormant-season defoliation has little effect
on these grasses, limiting litter accumulation on stockpiled pastures
by ensuring at least moderate utilization (50%) of standing plant
biomass may positively affect subsequent herbage production. Furthermore,
season-long grazing may have a more negative effect on needle-and-thread
and thread-leaf sedge growth than winter use at higher (50%) utilization
levels. The direct effects of dormant-season grazing on individual
plant species versus conventional season-long use, at present, are
undistinguished in relevant literature. This research indicates that
the four species examined were generally unaffected by dormant season
grazing.
Preliminary data regarding dormant-season grazing of native rangeland
in the western Dakotas indicated that brief early summer use of dormant-season
pastures and winter stocking rates intended to achieve 50% utilization
of standing aboveground biomass is the preferred management option
relative to grazing treatments of 30 or 50% winter utilization with
no summer use. This method was beneficial from both a land-use and
ecological standpoint. Subsequent data are necessary; however, to
evaluate the long-term ecological and economic sustainability of this
management.
Introduction
Many North and South Dakota livestock producers practice winter or
dormant-season grazing in an effort to lower feed costs. Dormant season
grazing, while not an exclusive winter-grazing period, is defined
as grazing during that time period between plant quiescence in late
fall and green up in early spring. Although adequate information exists
regarding nutritional management of winter grazing cattle, little
is known about the ecological effects of these practices "on
range or pasture land in the upper Midwest and northern Great Plains.
Furthermore, research-emphasizing inferences for specific winter-grazing
management is lacking. Various aspects of dormant season grazing have
been examined in a variety of ecosystem types, and conventional wisdom
dictates that defoliation during winter months while plants are dormant
has little to no effect on plant vigor (Riesterer et al. 2000).
Winter grazing is an appealing management option to many ranchers.
Producing hay or purchasing winter feeds is labor and capital intensive,
while winter grazing offers the potential for flexibility in making
management decisions. Furthermore, this practice allows for more efficient
utilization of range resources. The objectives of this study were
to determine the impacts of winter grazing on herbage production,
growth rate of dominant grass species (short-term), and changes in
plant species composition using various levels and combinations of
winter and summer use (long-term subsequent research).
Study area
This study was located in Adams County, North Dakota and Perkins
County, South Dakota. The Adams County study site was approximately
153 acres and located 5 miles southwest of Hettinger, North Dakota
(El. 817m) on sections 16, T129N, R96W and 25, R97W, T129N. The Perkins
County study site was approximately 143 acres and located 16 miles
south of Lodgepole, South Dakota (El. 803m) on sections 13, T19N,
R12E, and 18, T19N, R13E.
Climate
Growing-season precipitation was 11.3 inches in 2000, which was 4.7
inches below the annual average, with all months except May and July
below average. The 2001 growing season was characterized as a dry
spring and wet July, with average precipitation 1.6 inches below the
30-year average. The fall and winter of 2000-01 received above average
precipitation; however, the fall and winter of 2001-02 received considerably
less precipitation, particularly in November and December.
Monthly average temperatures were generally above the 30-year average
in 2000, with the exception of June, November, and December. Warmer-than-average
temperatures characterized the winter of 2001-2002, as November and
December 2001 and January and February 2002 were substantially warmer
than the 30-year average. Spring and summer temperatures were near
average in both years.
Vegetation
The study areas were found in the northern mixed-grass prairie and
described as the Missouri Slope Vegetation Zone (USDA-SCS 1984). The
plant communities were described as a wheatgrass-needlegrass vegetation
type (Barker and Whitman 1994). Dominant midgrass species were western
wheatgrass (Pascopyrum smithii) and needle-and-thread (Stipa
comata), and dominant short graminoid species were thread-leaf
sedge (Carex filifolia) and blue grama (Bouteloua gracilis)
(Barker and Whitman 1994, Shiflet 1994). Plant names were referenced
from McGregor et al. (1986) and USDA-USFS (2002).
Methods and design
Treatments
A total of two study areas (blocks) were selected in North and South
Dakota based on similar range condition and composition of native
plant species. Each study area was blocked and divided into four paddocks
with one of four treatments 1) season-long summer grazing at 50% utilization
(SL), 2) 25% summer use for 2 weeks in early and mid June and 50%
dormant season utilization [flash grazing (Hart 2001)] (FL), 3) 30%
dormant season utilization (DS 30), and 4) 50% dormant season utilization
(DS 50) assigned randomly to a paddock. The SL treatment was an 80-acre
paddock and the dormant season use treatments each 23-acre paddocks
in North Dakota. The DS 30 and SL treatment paddocks were each 30
acres in size, the FL treatment 37 acres, and the DS 50 treatment
48 acres at the South Dakota site.
Stocking rates
Stocking rates for the summer use treatments were determined using
the United States Department of Agriculture (USDA) Soil Conservation
Service (SCS) Technical Guidelines (1984) for the Missouri Slope Vegetation
Zone. Summer use paddocks were surveyed for ecological site composition
using the USDA SCS soil surveys for Adams County, North Dakota (Ulmer
1987) and Perkins County, South Dakota (Wiesner 1980). The stocking
rate for the SL was calculated for a 4-month grazing period beginning
June 1 and ending October 1. The North Dakota site was stocked at
1.9 ac/AUM with ten 1150 lb cows and their calves. The South Dakota
site was stocked at 1.6 ac/AUM with seven 620 lb spayed heifers.
Summer grazing use of the flash grazing treatments (FL) was targeted
for 25% utilization. The FL treatment carrying capacity was calculated
by stocking for 50% use of the total available AUMs in June while
considering that 50% of the total annual production occurred by mid
June, thus achieving a 25% utilization of total annual biomass. The
North and South Dakota sites were stocked with ten and sixteen 1150
lb cows and their calves or 4.4 ac/AUM and 4.1 ac/AUM; respectively,
for two weeks.
Stocking rates for the winter grazing treatments were calculated
after determining dry-standing plant biomass on Nov. 15, 2000. Ten
randomly placed 0.25m2 frames were clipped for each ecological
site (n=2) existing within a given replicate (n=20). The USDA SCS
(Wiesner 1980, Ulmer 1987) soil survey maps and technical guides were
used to estimate ecological site composition within each paddock to
calculate total standing biomass. Final stocking rates for each treatment
were computed by calculating 25% grazing-use efficiency with 30 or
50% disappearance, depending on treatment (Laycock et al. 1972, Pearson
1975) and a dry matter intake for an 1150 lb non-lactating cow using
the National Research Council (1996) for beef cattle.
The North Dakota DS 50 and FL grazing treatment paddocks were each
stocked with four 1,150 lb cows, or 3.1 ac/AUM; and the DS 30 treatment
paddock was stocked with three 1,150 lb cows, or 4.1 ac/AUM. The South
Dakota DS 50 treatment was stocked with 11 cows or 2.5 ac/AUM, the
FL treatment stocked with 8 cows or 2.4 ac/AUM, and the DS 30 treatment
stocked with 6 cows, or 2.5 ac/AUM. All South Dakota paddocks were
stocked with cows weighing an average of 1150 lb.
Winter grazing cattle were allowed ad libitum access to white salt
and trace minerals and were supplemented with 3 lb/day on an as-fed
basis of 30% crude protein all-natural cake. During the winter grazing
period of 2000-2001, cattle grazed as snow cover allowed for 53 days
beginning November 15 on both the North and South Dakota study sites.
During the dormant-season grazing period of 2001-2002, cattle grazed
on the North Dakota site for 53 days beginning November 15. The cattle
on the South Dakota site grazed for 35 days and animal numbers were
increased to meet set stocking rate guidelines, as turn out was
delayed until January 12 due to mechanical failures affecting
the watering system.
Table 1 shows ac/AUM comparisons of treatments and percent change
in carrying capacities compared to the SL treatment (control). From
a perspective of utilized AUMs, the dormant season only grazing treatments
reduced carrying capacities relative to season-long use; however,
the FL treatment numerically increased carrying capacities slightly
relative to season-long use (3.2 to 5.3%).
Table 1. Stocking rate comparisons among grazing treatments
in North and South Dakota.
----------------------------------------------------------------
SL FL DS 30 DS 50
----------------------------------------------------------------
N.D.
Ac/AUM 1.98 1.93 3.95 2.97
% Difference from SL 0.0 +5.3 -115.8 -61.8
----------------------------------------------------------------
S.D.
Ac/AUM 1.48 1.44 2.47 2.22
% Difference from SL 0.0 +3.2 -61.9 -38.1
----------------------------------------------------------------
SL = season-long summer grazing, FL = 25% summer use for 2 weeks
in early and mid June and 50% dormant season utilization,
DS 30 = 30% dormant season utilization, DS 50 = 50% dormant
season utilization
Herbage production
Herbage production of graminoids and forbs for each treatment was
determined using a paired-plot clipping technique (Milner and Hughs
1968). Twenty cages were distributed in each pasture during the treatment
period. One plot within and outside each cage was clipped using a
0.25m2 quadrat. Clipped herbage was separated into grasses
and forbs, dry matter weights were recorded, and lb/ac plant biomass
and standard error of the mean were calculated for each ecological
site.
In the summer through winter periods of 2000-2001, five cages were
systematically placed on each of the two shallow ecological sites
and two loamy ecological sites before grazing began on each treatment
(n=20), with the exception of the South Dakota 30% treatment where
only five cages were placed on a shallow ecological site since this
site made up only 10% of the study area on the treatment. During the
winter of 2001 on the North Dakota sites, five plots were clipped
for both the loamy and shallow ecological sites on the 30% treatment,
five shallow plots were clipped on the 50% treatment, and no plots
were clipped on the FL treatment due to ice and snow cover. On the
summer treatments of 2001, the 20 sites within each pasture selected
for the tiller study were used to determine production. In 2001-2002,
all plots from the winter grazing treatments were clipped since ice
and snow cover did not prevent clipping as it had in 2000-2001.
Leaf Heights
A study to examine leaf heights throughout the growing season was
initiated in May of 2001, to determine the growth patterns of western
wheatgrass, needle-and-thread, thread-leaf sedge, and blue grama within
each treatment. The species were selected as they were described as
the predominant forage base of the study region (Barker and Whitman
1994, Shiflet 1994). Furthermore, these species were described as
commonly existing together in various successional stages of rangeland
in western North Dakota (Hansen and Hoffman 1988). Goetz (1963) monitored
the growth and development of native range plants in western North
Dakota and used leaf height as a main indicator of plant growth. Furthermore,
researchers have correlated leaf and plant height with plant vigor,
forage yield, competition, range condition and trend, and defoliation
levels (Short and Woolfolk 1956, Buwai and Trlica 1977).
Twenty locations indicative of the dominant forage base were selected
randomly within each treatment in May 2001. On each location, a 0.25
m2 quadrat was selected containing at least 10 western
wheatgrass tillers, five needle-and-thread tillers, 10 thread-leaf
sedge tillers, and 10 blue grama tillers. Cool-season tillers were
marked with uniquely colored rings upon the selection of each site
in mid-May and each tiller was measured monthly until senescence was
observed for each species. Western wheatgrass and needle-and-thread
tillers were measured mid-month for leaf height (height of tallest
leaf) from May to August. Thread-leaf sedge was measured mid-month
for leaf height from May to July. Blue grama was the only warm-season
grass investigated for growth; thus, leaf heights were measured mid-month
during its growth period as described by Goetz (1963), from June to
September.
Statistics
A general linear model (GLM) was used to test for between-subject
effects for treatment-by-date interactions of leaf heights for each
species and herbage production. When interactions were detected (P#0.05),
treatments by date comparisons were made using a GLM model to determine
differences between treatments and date. When interactions were not
detected, data from all periods and replicates were combined and a
GLM model was used to determine differences among treatments (P#0.05).
Mean separations were performed at P#0.05 using Tukey's Honesty Significant
Difference (HSD) procedure (Steele and Torrie 1980, SPSS 1990).
Results and discussions
Herbage production
No differences in herbage production were found between locations
(P=0.296, F=1.097) in 2000. Following one year of treatment, peak
primary production on the winter-only treatments did not differ (P>0.05)
from the SL control treatment (Figure 1). Furthermore, herbage production
was higher (P#0.01) on FL than SL, DS 30, and DS 50 after one year
of treatment. No differences or positive effects of moderate dormant
season grazing treatments, similar to data reported by Coughenour
(1991) who found increased nitrogen in live and dead grasses and fringed
sagebrush on winter grazed areas, were found. Likewise, Schacht et
al. (1998) observed that mowing dormant range of switchgrass, little
bluestem, and big bluestem resulted in a higher yield of annual growth
than a non-mowed control. Engle et al. (1998) also reported that grazing
strategies emphasizing defoliation during the dormant season that
decrease probability of multiple defoliations during the growing season
are less detrimental than those that increase the probability of multiple
defoliations, such as the FL treatment in this study. Relevant research
by Auen and Owensby (1988), Coughenour (1991), Engle et al. (1998),
Schacht et al. (1998) and Reisterer et al. (2000) indicate dormant-season
harvesting of grasses has little or no negative effect on subsequent
herbage production.
Figure 1. Peak herbage production on the summer grazed season
long (SL), June flash + 50% dormant-season use (FL), 30% dormant-season
use (DS 30), and 50% dormant-season use (DS 50) in 2001. Treatments
with the same letter are not significantly different (P>0.05).
(Click here for a 24KB black and white graph.)
Leaf Heights
No differences (P<0.05) in western wheatgrass leaf heights were
detected between treatments for the months of May and June 2002. In
July, western wheatgrass leaf heights in the DS 30 treatment were
shorter (P<0.05) than the SL treatment and in August both the FL
and DS 30 treatment leaf heights were shorter (P<0.05) than the
SL treatment (Figure 2). Negative effects from grazing treatment on
late growing season plant production were also observed by Trent et
al. (1988). Fall grazed winter wheat plants relied more heavily on
photosynthesis later in the growing season than did the non-grazed
wheat plants as they were unable to draw from carbohydrate reserves
during grain filling. Similarly, Buwai and Trlica (1977) found heavy
quiescent defoliation of western wheatgrass reduced TNC relative to
a non-defoliated control. Furthermore, moderate and heavy dormant
defoliation of western wheatgrass reduced both herbage yield and plant
height when compared to the control.
Figure 2. Western wheatgrass leaf heights on the summer grazed
season long (SL), June flash + 50% dormant-season use (FL), 30%
dormant-season use (DS 30), and 50% dormant-season use (DS 50)
in 2001. Treatments with the same letter within each month are
not significantly different (P>0.05). (Click here for a
38KB black and white graph.)
Light winter use (DS 30) resulted in shorter leaf heights than heavy
winter use for needle-and-thread, thread-leaf sedge, and blue grama.
Light use also resulted in lower needle-and-thread and blue grama
leaf heights than SL. These data suggest increased utilization during
the dormant period results in increased herbage yield the following
year. Treatments by date interactions were not detected (P<0.05)
for needle-and-thread, thread-leaf sedge, and blue grama; thus, monthly
leaf height data were combined.
Needle-and-thread leaf heights throughout the growing season did
not differ between the SL, FL, and DS 50 treatments (P< 0.01);
however, the DS 30 treatment had lower leaf heights (P<0.05) than
the SL treatment (Figure 3). Thread-leaf sedge leaf height was also
greater in the FL and DS 50 treatments (P<0.01) than the DS 30
and SL treatments. The DS 30 and SL treatments did not differ in leaf
height (P<0.05) throughout the growing season (Figure 3). Blue
grama leaf heights did not differ (P<0.05) between the SL, FL,
and 50% treatments; however, the SL and FL treatments were higher
(P<0.01) than the DS 30 treatment (Figure 3).
Figure 3. Needle-and-thread, thread-leaf sedge, and blue grama
leaf heights on the summer grazed season long (SL), June flash
+ 50% dormant-season use (FL), 30% dormant-season use (DS 30),
and 50% dormant-season use (DS 50) in 2001. Treatments with the
same letter within each grass species (a,b,c, for needle-and-thread,
k,l for thread-leaf sedge, and x,y for blue grama) are not significantly
different (P>0.05). (Click here for a 23KB black and white
graph.)
These findings are consistent with the peak herbage production observations
and studies by Coughenour (1991) and Manley et al. (1995) who reported
positive effects on herbage production with increased levels of herbage
removal during the dormant season. If dormant-season defoliation is
not detrimental to needle-and-thread, blue grama, and thread-leaf
sedge, removal of standing-dead plant material and the corresponding
reduction in litter on the soil surface may be important to subsequent
herbage production and plant growth. Removal of standing dead plant
material has been noted to elevate soil temperatures; thus, accelerating
decomposition and mineralization in the spring. Furthermore, nutrient
turnover rates are accelerated under grazed systems by returning mineral
nitrogen to the soil in a readily decomposable form, thereby bypassing
slower plant litter decomposition pathways (Coughenour 1991).
Literature cited
Auen, L.M. and C.E. Owensby. 1988. Effects of dormant-season herbage
removal on Flint hills rangeland. J. Range Manage. 41:481-482.
Barker, W.T. and W.C. Whitman. 1994. Rangeland cover types of the
northern Great Plains. In: Shiflets, T.N., Rangeland cover
types of the United States. Soc. for Range Manage., Denver, CO.
154pp.
Buwai, M. and M.J. Trlica. 1977. Multiple defoliation effects on
herbage yield, vigor, and total nonstructural carbohydrates of five
range species. J. Range Manage. 30:164-171.
Coughenour, M.B. 1991. Biomass and nitrogen response to grazing
of upland steppe on Yellowstone's northern winter range. J. Appl.
Ecol. 28:71-82.
Engel, R.K., J.T. Nichols, J.L. Dodd, and J.E. Brummer. 1998. Root
and shoot responses of sand bluestem to defoliation. J. Range Manage.
51:42-46.
Goetz, H. 1963. Growth and development of native range plants in
the mixed grass prairie of western north Dakota. M.S. Thesis. North
Dakota State Univ., Fargo. 165pp.
Hansen, P.L. and G.R. Hoffman. 1988. The vegetation of the Grand
River/Cedar River, Sioux, and Ashland Districts of the Custer National
Forest: a habitat type classification. Gen. Tech. Rep. RM-157. Fort
Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment Station. 68p.
Hart, R.H. 2001. Where the buffalo roamed-or did they? Great Plains
Res. 11:83-102.
Laycock, W.A., H. Buchanan, and W.C. Krueger. 1972. Three methods
for determining diet, utilization, and trampling damage on sheep
ranges. J. Range Manage. 25:352-356.
Manley, J.T., G.E. Schuman, J.D. Reeder, and R.H. Hart. 1995. Rangeland
soil carbon and nitrogen responses to grazing. J. Soil and Water
Cons. 50:294-298.
McGregor, R.L., T.M. Barkley, R.E. Brooks, and E.K. Schofield 1986.
Flora of the Great Plains. University Press of Kansas. Lawrence,
K.S..
Milner, C. and R.E. Hughes. 1968. Methods of the measurement of
primary production of grassland. Blackwell Sci. Pub. Oxford, England.
National Research Council. 1996. Nutrient requirements for beef
cattle (7th ed.) Nat. Acad. of Sciences-NRC, Washington,
D.C.
Pearson, H.A. 1975. Herbage disappearance and grazing capacity
determinations of southern pine bluestem ranges. J. Range Manage.
28:71-73.
Riesterer, J.L. M.D. Casler, D.J. Undersander, and D.K. Combs.
2000. Seasonal yield distribution of cool-season grasses following
winter defoliation. Agron. J. 92:974-980.
Schacht, W.H., A.J. Smart, B.E. Anderson, L.E. Moser, and R. Rasby.
1998. Growth responses of warm-season tallgrasses to dormant-season
management. J. Range Manage. 51:442-446.
Shiflet, T.N. 1994. Rangeland cover types of the United States.
Soc. Range Manage. Denver, CO. 152 pp.
Short, L.R. and E.J. Woolfolk. 1956. Plant vigor as a criterion
for range condition. J. Range Manage. 9:66-69.
SPSS. 2000. Statistical procedures for social science. Marketing
Dept. SPSS. Inc. Chicago, Ill.
Steel, R.G.D and J.H. Torrie. 1980. Principles and Procedures of
Statistics. (2nd Edition), McGraw-Hill, New York Company.
New York, NY.
Trent, J.D., L.L. Wallace, T.J. Svejcar, and S. Christiansen. 1988.
Effects of grazing on growth, carbohydrate pools, and mycorrhizae
in winter wheat. Can. J. Plant Sci. 68:115-120.
Ulmer, M.G. 1987. Soil survey of Adams County, North Dakota. U.S.
Gov. Print Off. Washington, D.C. 174p.
United States Department of Agriculture Soil Conservation Service.
1984. Range Technical Guidelines. Bismarck, ND.
United States Department of Agriculture Forest Service, Rocky Mountain
Research Station, Fire Sciences Laboratory 2002. Fire Effects Information
System, [Online]. Available: http://www.fs.fed.us/database/feis/
[2002, February].
Wiesner, H.C. 1980. Soil survey, Perkins County, South Dakota.
U.S. Gov. Print Off. Washington, D.C. 195p.
Mitch Faulkner: Range Specialist, Bureau of Land
Management, Lander WY
Kevin Sedivec, Jackie Olson, Greg Lardy: Extension State Rangeland
Specialist, Graduate Student and Extension State Beef Specialist,
Animal and Range Sciences Department, NDSU, Fargo
Tim Faller: Director, Hettinger Research and Extension Center, NDSU,
Hettinger, ND
Jack Dahl: Range Specialist, US Forest Service, Dickinson, ND.
