Journal of the NACAA
ISSN 2158-9429
Volume 8, Issue 1 - June, 2015

Editor: Lee Stivers

Weed Control in Bermudagrass: Effect on Biomass Production and Quality

Lemus, R., Extension Forage Specialist, Mississippi State University Extension Service
White, J.A., Forage Variety Testing Manager, Mississippi State University


Weeds can reduce forage yield and stand persistence of desirable forage species in pastures and hay fields.  Although some weeds might provide some nutritional value to grazing animals while in the vegetative stage, certain weed species can be poisonous to livestock. They can also affect the quality and palatability of available forage to livestock.  A study was conducted at Mississippi State University to evaluate grass response and weed control with Pastora™ (nicosulfuron + metsulfuron methyl), Plateu® (imazapic), and Eraser (glyphosate) at different rates and mixtures in bermudagrass.  Eight herbicide treatments were implemented as pre- or post-harvest application in bermudagrass to reduce the impact of weeds on forage production.  Herbicide treatment effects on forage yield were significant by harvest. However, total annual yield was not affected by treatment.  At 30 days, all herbicides were effective in controlling weeds, but treatments containing Plateu and the treatment using Pastora plus glyphosate suppressed weeds the greatest. Acid detergent fiber (ADF), neutral detergent fiber (NDF) and crude protein (CP) fractions of the forage were affected by herbicide treatment and harvest time.


Grasssland (pasture and range) in the US covers approximately 614 million acres and represent 27% of the total land area (Nickerson et al., 2011).  Pasture production in the southern USA is a major component of a successful livestock enterprise.  There are approximately 900,000 acres of established bermudagrass (Cynodon dactylon) in the state of Mississippi (Lemus, 2012) and over 33.9 million acres in the southern USA (Lemus and White, 2015).  Forage production in perennial warm-season grasses could be affected by weed competition for space, light and nutrients (Carlisle et al., 1980).  Weeds can impact livestock production by decreasing yield potential, impacting forage quality, producing anti-quality factors, increasing the cost of livestock production and reducing the value of the land. Although production and profit losses attributed to weed infestations in pastures can be difficult to assess, Watson (1976) reported that the production of the desirable species may increase by 400% or more if proper weed control practices are followed. 

There are currently several herbicides labeled for bermudagrass forage production.  One disadvantage is that these herbicides have only a limited broad spectrum control on both broad leaf and grass weeds.  A few herbicides are labeled to control grass weeds commonly found in Mississippi pasture and hay fields such as crabgrass (Digitaria sp.), Johnsongrass (Sorghum halapense), dallisgrass (Paspalum dilatatum), broadleaf signalgrass (Brachiaria platyphylla), vaseygrass (Paspalum urvillei), and foxtail (Alopecurus sp.).  New herbicide chemistry has been introduced to the pasture and range market, and there is a need to test their efficacy alone or in tank mix.   This new chemistry should be used as part of a strategic long-term plan to eradicate weed populations using management practices that could be economical and sustainable to improve competition of desirable species, reduce seed bank and prevent reinvasion by other weed species. 

Pastora™ (56.2% nicosulfuron plus 15.0% metsulfuron methyl; E. I. du Pont de Nemours and Company, Crop Protection, Wilminton, DE) is a product that has a full label for bermudagrass. No other forage grass species is listed on the label.  This product has relatively limited soil activity.  The herbicide recommendations for Pastora is to apply at a rate of 1.0 to 1.5 oz/ac with a nonionic surfactant at 0.25% v/v before bermudagrass green up or three to seven days after bailing.  It is recommended not to exceed applications of 2.5 oz/ac/yr.  According to the label, Pastora is also reported to have herbicide activity on controlling foxtails, seedling Johnsongrass and some broadleaf weeds.  In some instances Pastora may result in temporary yellowing or stunting of bermudagrass. There are no grazing or haying restrictions.  Shoup (2009) reported that midsummer postemergence applications of Pastora provided greater than 85% control of Johnsongrass one month after treatment (MAT) and 100% by 2 MAT.

Plateu® (23.6% ammonium salt of imazapic; BASF Corporation, Research Park, NC) is a product labeled for weed control, growth suppression in pastures and native grass establishment.  Application of Plateu at a rate of 4 to 12 oz/ac could provide seedhead and grass growth suppression along with control of tall growing summer annual broadleaf weeds.  The soil residual activity of imazapic could provide broad spectrum weed control that can last throughout the summer following a single late spring application.  Bermudagrass growth will be suppressed after an application of Plateu for 30 to 45 days with some bermudagrass varieties more susceptible  than others.  Label recommendations indicate not to use this product on drought stressed bermudagrass or during the transition period between dormancy and green-up, within 30 days of aeration or newly established bermudagrass (especially World Feeder).  There are no grazing restrictions following application, but a 7-day haying restriction is labeled.  Studies in the southern USA have shown that Plateu can provide control of crabgrass, signalgrass and nutsedge (Cyperus esculentus) at a 4.0 oz/ac rate (Ducar, 2002).  Vassey grass control reached 85% after two months at 6 to 10 oz/ac rate (Ducar, 2002).

Eraser (41% Glyphosate; Control Solutions, Inc., Pasadena, TX) sometimes can be used over completely dormant bermudagrass for winter-annual weedy grass control. With a broadleaf post-emergent tank-mix herbicide, the tank-mix can often be used for both winter annual grass and broadleaf weed control in dormant bermudagrass.  Glyphosate application should be made when temperatures are above 60 ºF.  Above this temperature, winter annual weeds are generally metabolically active and are more effectively controlled by a glyphosate application.   The application is best made when temperatures are in the 60s but definitely prior to bermudagrass green-up. Data has shown that bermudagrass can be slightly or marginally injured when sprayed with glyphosate during green-up (Yelverton, 2009).

The objective of study was to evaluate grass and weed control with nicosulfuron plus metsulfuron methyl, glyphosate, or imazapic in bermudagrass.


The study was conducted at the Henry H. Level Animal Research Farm at Mississippi State University. The soil type is a Marietta fine sandy loam (Fine-loamy, siliceous, active, thermic Typic Endoaquults).  Eight herbicide treatments (Table 1) were applied to bermudagrass to determine weed control efficacy and effect on biomass and forage quality.  The experimental design was a randomized complete block design replicated four times.  Plot dimensions were  6 ft x 30 ft.  Treatments were applied with a CO2 backpack sprayer at 25 PSI.   Treatments contained pre-emergence (Pre) and post-emergence (Post) applications. 

Table 1. Herbicide treatments applied to bermudagrass in 2013.

Abbreviation Description
UC Untreated Check
E-Pre 6 oz  Eraser (Pre)
P-Pre 1 oz Pastora™  (nicosulfuron + metsulfuron methyl) + 0.25% V/V1 NIS2 (Pre)
PE-Pre 1 oz Pastora™  + 6 oz Eraser (41% glyphosate) + 0.25% V/V NIS (Pre)
PL-Pre 8 oz Plateu®  (ammonium salt of imazopic) + 0.25% V/V NIS  (Pre)
PLE-Pre 8 oz Plateu®  + 6 oz Eraser + 0.25% V/V NIS  (Pre)
P-Post 1 oz Pastora™  + 0.25% V/V NIS (Post)4
PL-Post 8 oz Plateu®  + 0.25% V/V NIS  (Post)

1V/V = Volume to volume; 2NIS = Non-ionic surfactant; 3Pre = herbicide application done before bermudagrass green up; 4Post = herbicide application after the first bermudagrass harvest.


Pre-emergence treatments were applied prior to bermudagrass green up, especially those treatments containing glyphosate.  Post-emergence treatments were applied after the first bermudagrasss harvest.  Weed incidence was evaluated as a whole and not for each specific weed present at 15, 30, 45, 60 and 90 days after treatment applications.  Percent weed incidence was measured using a frequency grid sampling (Vogel and Masters, 2001).  The frequency grid was a Providence square containing 25 (5x5) cells.  The frequency grid was randomly placed four times in each plot. The number of cells containing one or more weeds were counted. Counts were directly converted into a frequency occurrence by dividing the total number of cells from the four squares that contain weeds by 100.  Bermudagrass was harvested for biomass every 28–35 days or when plants had reached 15-inches of biomass growth.  Dates between harvests varied due to environmental conditions.  Plots were harvested using a commercial mower (52 inches width) and removing a 4.3 ft x 30 ft swath from the center of the plot to minimize border effect.   Bermudagrass was fertilized with 50 lb N/ac after each cut and up to 150 lb N/ac for the entire season.  Biomass subsamples from each harvest were collected for dry matter determination and to calculate harvest and seasonal yield.  Subsamples were dried at 50 °C for 3 to 5 days and ground to pass a 2-mm screen.  Ground samples were used to determine forage quality analysis [crude protein (CP), acid detergent fiber (ADF), and neutral detergent fiber (NDF)] using the Foss 6500 NIRS instrument and using the mixed hay equation developed by the NIRS Forage and Feed Testing Consortium (Hillsboro, WI).    Data were analyzed using repeating measures in PROC GLM of SAS (SAS Institute Inc., 2013), and means were separated by the Least Significant Differences (LSD) at α = 0.05. 


Herbicide Application Schedule

Herbicide treatments “Pre” was first applied on March 27th with sunny conditions, wind less than 5 mph and temperature of 62°F.  The second application “Post” was applied on June 17th with sunny conditions, wind speed of 5 mph and temperature of 84°F.   


Plots were first harvested June 14th and again on July 17th and September 4th.  Harvest yields averaged between 1800 and 2200 lbs/ac.  Herbicide treatment effects on forage yield was significant by harvest (Table 2). Treatments applied before green up that included Plateu decreased yields up to 43% in the first harvest (Table 2). However, total annual yield was not affected by treatment (Table 3).   This effect was mitigated by the second and third harvest.  Pastora and glyphosate had no effect on forage yields through the duration of the trial. Though forage yield was generally unaffected by herbicide treatment (data not shown), a visual assessment described suppression of the bermudagrass growth for both Pastora and Plateu in “Pre” treatments, but was consistently more severe in the Plateu treatments.

Table 2. Forage yields for eight herbicide treatments and three harvest dates in 2013.

  Harvest Dates      
Treatment Jun 14 Jul 17 Sep 4 Mean CV, % Pr
  --------------------------------- lbs DM/ac ---------------------------------    
Untreated Check   2369Aa 1672Ba 2098ABb 2046 17  0.0881**
Eraser (Pre)    2088Aab 2000Aa 2049Ab 2046 13 0.8940
Pastora  (Pre)     2069Aab 1721Aa 1980Ab 1924 20 0.6953
Pastora  + Eraser (Pre)     1971Aab 1933Aa 1962Ab 1955  9 0.9520
Plateu  (Pre)   1030Cc 2045Ba  2761Aa 1945 20  0.0020*
Plateu  + Eraser (Pre)   1196Cc 1756Ba   2339Aab 1764 14  0.0016*
Pastora  (Post)    2169Aab 1565Ba 2239Ab 1991 16   0.0503**
Plateu  (Post)   1816ABb 1266Ba 2213Ab 1765 27  
Mean          1839          1745        2205      
CV, %              16 23 15      
Pr <0.0001* 0.1966 0.0616**      

Upper case letters are for comparison or harvest within a herbicide treatment.

Lower case letters are for comparison of herbicides treatments within a harvest date.

*, ** denotes significant difference at the α = 0.05 and α = 0.10, respectively.


Table 3. Seasonal bemudagrass biomass production across eight herbicide treatments in 2013.

  Seasonal Dry Matter Yield
Treatments lbs DM/ac
Untreated Check 6139
Eraser (Pre) 6138
Pastora  (Pre) 5771
Pastora  + Eraser (Pre) 5866
Plateu (Pre) 5836
Plateu  + Eraser (Pre) 5292
Pastora  (Post) 5973
Plateu (Post) 5295
Mean 5789
CV, %    11
LSD0.05* NS

*Least Significant Difference among means and within a forage quality parameter at α = 0.05

†NS: Not significant difference, P>0.4768.


Forage Quality

Crude protein (CP), acid detergent fiber (ADF), and neutral detergent fiber (NDF) fractions of the forage were affected by herbicide treatment and harvest time.   Post application of Plateu had the higher crude protein and lower NDF compared to the rest of the treatments (Table 4).  No other treatments were significant different from each other. The bermudagrass growth suppression by the “post” Plateau treatment likely increased forage quality by keeping it in a more vegetative state, and control plots were cleaned off of winter weeds during the first harvest allowing for fresh yet minimal bermudagrass growth to be achieved. Harvest date indicated higher crude protein in July harvest compared to the September harvest (Fig. 1).  On the other hand, the fiber components (ADF and NDF) increase with harvest dates.  This increase is correlated with maturity.

Table 4. Treatment effect on bermudagrass forage quality parameters: crude protein (CP), acid detergent fiber (ADF) and neutral detergent fiber (NDF).

  Forage Quality Parameters
Treatment CP ADF NDF
  ----------------------------- % DM -----------------------------
Untreated Check 14.2b 35.1a 59.6abc
Eraser (Pre) 12.3c 36.6a 61.7a
Pastora  (Pre) 12.9bc 36.0a 60.9ab
Pastora + Eraser (Pre) 12.5bc 36.4a 61.6a
Plateu (Pre) 12.6bc 36.9a 60.7ab
Plateu + Eraser (Pre) 14.0bc 35.8a 58.2c
Pastora  (Post) 13.1bc 36.0a 60.9ab
Plateu  (Post) 16.2a 33.2b 58.5bc
P Value <0.0018 <0.0035 <0.0344
CV, % 13.7 4.8 4.1

Letters are for comparison of herbicide treatments within each forage quality parameter at α = 0.05.


Figure 1. Harvest date effect on bermudagrass forage quality parameters: crude protein (CP), acid detergent fiber (ADF) and neutral detergent fiber (NDF). Letters are for comparison of harvest dates within each forage quality parameter. P<0.001 CP, ADF, and NDF.

Weed Composition    

Two weed assessments were made to the entire trial area at the time of herbicide application.  Prior to the first herbicide application (“Pre”) the following weeds were present in the plots: Annual ryegrass (Lolium multiflorum), curly dock (Rumex ciprus), buttercup (Ranaculus arvenis), buckhorn plantain (Plantago lanceolata), chickweed (Sterallia media), dandelion (Pyrrhopappus carolinianus), cheatgrass (Bromus secalinus), wild carrot (Daucus carota), bluegrass (Poa annua), shepard’s-purse (Capsella bursa-pastoris), yellow nutsedge (Cyperus esculentus), purple deadnettle (Lamium purpureum) and henbit (Lamium amplexicaule). Prior to the second herbicide application (“Post”) the following weeds were present: annual ryegrass, yellow nutsedge, buckhorn plantain, buttercup, dallisgrass (Paspalum dilatatum), dogfennel (Anthemis cotula), Johnsongrass (Sorghum halapense), curly dock, barnyard grass (Echinochloa crus-galli), blue vervain (Vervena hastate), yellow foxtail (Setaria pumila), horsenettle (Solanum carolinense), carpetweed (Mollugo verticillayta), large crabgrass (Digitaria cillaris), and marestail (Equisetum arvense).  The effect of herbicides on each specific weed was as not quantified due to the transitional period between cool-season and warm-season annual weeds.  That means that any other broadleaf or grass species besides bermudagrass was considered a weed.

Percent weed incidence in the plots was significant in a treatment by time interaction (P<0.0225) for “Pre” treatments (Fig. 2).  At 15, 60, and 90 day evaluations all treatments were similar to the control in respect to weed composition.  At 30 days all herbicides were effective in controlling weeds, but treatments containing Plateau and the treatment using Pastora with glyphosate suppressed weeds the greatest. By 45 days after application, all herbicide treatments had lower weed percentage than the untreated check, but were similar to each other in efficacy.


Figure 2. Evaluation of weed incidence by pre-emergence herbicide application to bermudagrass. Herbicides were applied on March 27, 2013 and treatments effects were evaluated at 15, 30, 45, 60 and 90 days after application. Letters are for comparison of treatments within post-herbicide application days. Bars having the same letter within days after treatment are not significant different from each other.  NS = No significant difference α = 0.05.


Percent weed incidence in the plots was significant in a treatment by time interaction (P<0.0003) for “Post” treatments (Fig. 3).  When plots were evaluated for weed percentage at 15 days after the summer herbicide application, no difference between Pastora and Plateu could be detected. However, both herbicides reduced the percentage of weeds up to 23% compared to the untreated check. The relatively limited effect of the herbicides after 30 days is partially due to the relative increase in warm-season annual grasses along with dallisgrass in the treatment plots. By July, the plots consisted only of grass species with the majority of the broadleaf species already controlled.


Figure 3.  Evaluation of weed incidence by post-emergence herbicide application to bermudagrass. Herbicides were applied on June 17, 2013 and treatments effects were evaluated at 15, 30, 45, 60 and 90 days after application. Letters are for comparison of treatments within post-herbicide application days. Bars having the same letter within days after treatment are not significant different from each other.  NS = No significant difference α = 0.05.


Bermudagrass yield was significantly suppressed by Plateu when applied before green-up and after the first harvest. Plateu had the greatest negative effect on yield when applied at green-up.   Pastora™ did not negatively affect bermudagrass at either application suggesting its possible use throughout the summer to control weeds.  All herbicides were more effective when applied before green-up and weed control lasted up to 45 days.  Plateu had greater efficacy controlling weeds in the spring over Pastora except when Pastora was mixed with glyphosate.  As a result early spraying of Pastora before green-up may benefit from a mixture with glyphosate to eliminate any existing winter weeds actively growing.  Both Pastora and Plateu had little effect on summer annual grasses as well as in dallisgrass, suggesting that other methods of weed control need to be implemented if these are the dominant weeds in a pasture.


Carlisle, R.J., V.H. Watson, and A.W. Cole. 1980. Canopy and chemistry of pasture weeds. Weed Sci. 28:139–141.

Ducar, J.  2002.  A new herbicide for grass weeds in hay fields.  In: Jackson Stockman.  Jackson County Extension Livestock and Forage Newsletter.  University of Florida.  Online at (verified 26 February 2015).

Eraser Weed Killer (Glyphosate) Herbicide Label.  Online at (verified 26 February 2015).

Lemus, R.  2012.  Forage Production Survey in Mississippi.  Mississippi State University Extension Service.  Unpublished data.

Lemus, R. and J. White.  2015.  Harvest Date and N Application Impact in Bermudagrass Forage Production and Quality.  In: Proceedings of the Southern Branch of the American Society of Agronomy.  Atlanta, GA.  February 1–3.

Nickerson, C., R. Ebel, A. Borchers, and F. Carriazo. 2011.  Major Uses of Land in the United States, 2007, EIB-89, U.S. Department of Agriculture, Economic Research Service, 67 pp.

Pastora™ Herbicide Label.  Online at (Verified 22 February 2015).

Plateu® Herbicide Label.  Online at (verified 26 February 2015).

Shoup, D.E. 2009.  Grass control with nicosulfuron and metsulfuron methyl in bermudagrass.  In: Proceedings of the North Central Weed Science Society 64:28.

Watson, V.H. 1976. Weed control and nutritional benefits of Banvel and Weedmaster in warm season grass pastures. Proceedings 29th Annual Meeting Southern Weed Science Society. 142.

SAS Institute Inc.  2013.  SAS Enterprise Guide 6.1 Help and Documentation, Cary, NC.

Vogel, K.P. and R.A. Masters.  Frequency Grid:  A Simple Tool for Measuring Grassland Establishment.  J. Range Manage. 54 (6):653-655.

Yelverton, F.  2009.  Spraying glyphosate on semi-dormant bermudagrass.  Turf Files.  North Carolina State Univ.  Online at (verified 26 February 2015).


This study was supported by DuPont Chemicals.  Thank you student workers Isaac Picket, Dereck McCain, Kyle Munn, and Daniel Moore for helping with data collection and sample processing.


Mention of a trademark, proprietary product, or vendor does not constitute a guarantee or warranty of the product by Mississippi State University and does not imply its approval to the exclusion of other products or vendors that also may be suitable.