Switchgrass and Big Bluestem Response to Spring Burning and Glyphosate Application: Hay, Biomass, and Seed Yields

by M.A. Sanderson, R.R. Schnabel, W.L. Stout, and D. Genito, USDA-ARS Pasture Systems and Watershed Management Research Unit, University Park, PA; W.S. Curran, Dep. Crop and Soil Sciences, The Pennsylvania State University, University Park, PA; B.F. Tracy, Dep. Crop Science, University of Illinois, Champaign-Urbana, IL

printable version of this article with tables


Cool-season weeds frequently invade stands of warm-season grasses, such as switchgrass (Panicum virgatum L.) and big bluestem (Andropogon gerardii Vitman) in the northeast USA. Spring burning or glyphosate [N-(phosphonomethyl) glycine] application may be useful in controlling cool-season weeds. Our objective was to determine when burning or glyphosate application in the spring affected forage or seed yield in switchgrass and big bluestem. We burned or applied glyphosate to switchgrass and big bluestem in mid-April, early-May, and late-May in 1998 and 1999 at Rock Springs, PA. Dry matter yield was measured in July (hay harvest) or September (bioenergy feedstock harvest). Switchgrass seed yield was measured in September. Burning in mid-May 1998 reduced hay yields of switchgrass and big bluestem by 40 to 44% compared with burning in mid-April. Glyphosate reduced yields of both warm-season grasses when applied later than mid-April. Applying glyphosate to switchgrass in late May reduced hay yields by 80% and biomass feedstock yields by 30%. Late application of glyphosate to big bluestem reduced hay yields by 90% and biomass feedstock yields by 40%. A late burn reduced yields less than a late application of glyphosate. Switchgrass seed yields were not affected by burning or glyphosate in 1998, whereas burning increased seed yields in 1999. Switchgrass and big bluestem can be burned any time through the first week of May in central Pennsylvania. Glyphosate application, however, should be done before mid April if switchgrass or big bluestem are to be cut for hay in summer or by the first of May if the grasses are to be harvested in September or later.


Cool-season weeds often invade stands of warm-season grasses during early spring before the warm-season grasses are competitive. These weeds reduce both warmseason forage yield and stand life. Glyphosate can be used in early spring before green up of the warm-season grass to control cool-season grasses and broadleaf weeds. Burning the grass residue in spring may also control some weeds. Timing of these weed control techniques is critical to both control weeds and prevent damage to newly emerging warm-season grasses. Sometimes weeds may not be at the proper developmental stage for effective control with herbicides before the warm-season grass has greened up. Thus, waiting to apply a nonselective herbicide, such as glyphosate, or burning too late may damage the warm-season grass. Applying herbicide or burning too early may allow for subsequent weed emergence that can compete with the warm-season grasses.

Perennial warm-season grasses, such as switchgrass and big bluestem, have multiple uses including livestock forage, soil conservation, and wildlife habitat. Switchgrass has also been studied extensively as a biomass energy crop (Sanderson et al., 1996). Early season weed control is critical in each of these uses. Our objective was to determine how timing of burning or glyphosate application in the spring affects forage, biomass, and seed yield of switchgrass and big bluestem.

Materials and Methods

The field experiment was conducted in 1998 and 1999 at the Russell E. Larson Agricultural Research Center near Rock Springs PA. Soil at the site is a Hagerstown silt loam (fine, mixed, semiactive, mesic Typic Hapludalfs). Soil tests indicated no need for lime, P, or K. Established (10 to 20 years old) plantings of three switchgrass cultivars (Pathfinder, Trailblazer, and NJ50) and two big bluestem cultivars (Pawnee and Niagara) were burned or treated with glyphosate at 1 lb a.i./ac on three dates each year. Plots were burned on 15 April, 6 May, and 22 May 1998. Glyphosate was applied on 21 April, 6 May, and 25 May 1998. In 1999, plots were burned on 28 April, 6 May, and 14 May. Glyphosate was applied on 28 April, 6 May, and 18 May. At the April treatment date, switchgrass and big bluestem were just greening up and had about 1 to 2 inches of visible growth. At the early May treatment date, both grasses were 5 to 7 inches tall and had about two fully expanded leaves. At the late May treatment date, both grasses were 10 to 14 inches tall and had three to four fully expanded leaves but had no elongated internodes.

There was one planting of each grass cultivar. Treatments were randomly assigned to two replicate plots (20 ft by 30 ft) within each grass cultivar. Thus, there were six plots total of each herbicide and burning treatment and six control (not burned or treated with herbicide) plots for switchgrass and four treated and control plots for big bluestem. In early April of each year, standing residue was clipped to a 4-inch stubble and removed from the plots designated for herbicide treatment. Standing residue and litter were left in place on the control plots, whereas most above ground residue was removed by fire in the burned treatments. A second set of plots not used in 1998 was used for treatments in 1999.

The plots were split in half and one half harvested in July and the other half harvested in September. The July harvest simulated the use for hay (harvest at the late boot to early head stage), whereas the September harvest simulated the use for bioenergy feedstock (harvested after seed maturation). At each harvest, four 2 ft by 2 ft quadrats were cut to a 6-inch stubble by hand. Each quadrat sample was bagged separately and dried at 135F for 48 h to determine dry matter yield. In September of each year, the panicles in each quadrat were removed and threshed to determine seed yield for switchgrass only. Panicles were threshed by hand and the seed cleaned by hand- rubbing and separation in an air-column seed blower. Seeds were air dried and seed yield (total caryopsis weight) per unit area was calculated. Weeds are not problematic in this study, so no weed control data were collected.

The design of the experiment was a randomized complete block. Data were analyzed separately for each grass species, year, and harvest within years. The individual grass cultivars were not replicated, thus the cultivars were treated as blocks with two replicate plots nested within each block. Comparisons of treatment means included 1)control vs the average of burning and glyphosate treatments, 2) early vs late burn, 3) early vs late glyphosate, and 4) late burn vs late glyphosate.

Results and Discussion

In 1998, switchgrass hay yields were reduced by both burning and glyphosate when applied in May (Table 1). A late application of glyphosate, however, reduced switchgrass hay yields by 77% compared with the April application date, whereas a late burn reduced switchgrass hay yields by 44% when compared with early burning. Switchgrass harvested in September to simulate a biomass feedstock harvest was not affected by the date of burning; however, a late May glyphosate application reduced biomass yields by 35% compared with a mid-April treatment. A similar pattern of responses was observed in big bluestem (Table 2). Late burning reduced hay yields in big bluestem by 40% compared with a mid-April burn, whereas late glyphosate application reduced hay yields by 84% and biomass feedstock yields by 48% compared with a midApril treatment.

Switchgrass hay or biomass yields were not affected by the date of burning in
1999 (Table 1). Similar to 1998, however, a late glyphosate application reduced hay yields by 80% and biomass feedstock yields by 29% compared with the mid-April treatment. In 1999, big bluestem hay yields were not affected by burning, whereas glyphosate application reduced hay yields by 26 to 91% and biomass feedstock yields by 41% (Table 2). Biomass feedstock yields of big bluestem, however, increased by 80% with later burning. Mitchell et al. (1994) reported that a mid-spring (late April) burn increased big bluestem forage yield on native tallgrass prairie by 18 to 36% compared with an early spring (March 31) burn and increased yields by 52% compared to not burning. Burning young (5 year) stands of switchgrass and big bluestem, however, reduced yields compared to unburned plots during 3 yr in Nebraska (Cuomo et al., 1996). Delayed burning from mid-March to mid-May progressively decreased yields of both grasses. Burning in midMay reduced yields by half compared to unburned plots each year. Burning in consecutive years reduced yields.

Seed yields of switchgrass were not statistically different among treatments in 1998 (Table 3). Burning increased switchgrass seed yields in 1999 compared to the glyphosate and control treatments. Later burning treatments increased seed yields in 1999, whereas later glyphosate treatments reduced seed yields.

Although weed control was not a component of this study, previous research indicates that cool-season weed control generally increases as application is delayed in the spring. Therefore, timing of glyphosate for effective control of cool-season weeds while maintaining warm-season grass tolerance may be a difficult compromise. Both weed and grass stage of development should be assessed before glyphosate application to help determine the effect of spring application.


Generally, the timing of burning treatments did not have a large effect on hay or biomass yields of switchgrass and big bluestem. When yields were affected, the late burning treatment did not reduce yields as much as a late application of glyphosate. Delaying glyphosate application into May frequently reduced warm-season grass yields. Thus, mid April seems to be the threshold date for glyphosate application to switchgrass and big bluestem sods in climates similar to central Pennsylvania. Frequently, the herbicide effect on hay yields persisted into the later biomass harvest; however, burning effects on yield were most evident in the first (hay) harvest. Thus, burning or glyphosate application may be more flexible for warm-season grasses managed for a single late-season harvest, such as for biomass energy feedstock production.

Literature Cited

Cuomo, G.J., B.E. Anderson, L.J. Young, and W.W. Wilhelm. 1996. Harvest frequency and burning effects on monocultures of three warm-season grasses. J. Range Manage. 49:157-162.

Mitchell, R.B., R.A. Masters, S.S. Waller, K.J. Moore, and L.E. Moser. 1994. Big bluestem production and forage quality responses to burning date and fertilizer in tallgrass prairies. J. Prod. Agric. 7:355-359.

Sanderson, M.A., R.L. Reed, S.B. McLaughlin, S.D. Wullschleger, B.V. Conger, D.J. Parrish et al. 1996. Switchgrass as a sustainable bioenergy crop. Bioresource Technol. 56:83-93.