Christopher A. Roider, James L. Griffin, Stephen A. Harrison, Billy J. Williams and Roy P. Vidrine
Glyphosate (sold under numerous formulations including Roundup, Touchdown and others) is a nonselective, postemergence herbicide used to control weeds in reduced tillage systems and in glyphosate-resistant soybeans, cotton and corn marketed in the southern United States. The increased use of glyphosate has increased the incidence of off-target movement to sensitive crops. Herbicide drift is most often the result of improper application. In most cases the herbicide rate in drift situations would be somewhere between 1/10th and 1/100th of the applied rate and would be considered sub-lethal. If a crop is very sensitive to the herbicide, however, the effect on growth and yield could be significant.
In the South, glyphosate is applied before planting from February through April to kill vegetation in fields intended for corn, cotton and soybeans. At that time of the year the growth stage of wheat could range from late tillering to flowering. Wheat would be especially sensitive to glyphosate at jointing when tillering is complete and the first node is visible on the stem because this is when the seedhead is formed. In addition, when wheat is exposed to glyphosate at flowering, a reduction in both the number of spiklets produced per seed head and seed weight could affect yield. This research evaluated how simulated drift of glyphosate from first node through early flowering would affect crop growth, yield and yield components of six wheat varieties.
The wheat variety USG 3209 was planted in early December at the Central Research Station near Baton Rouge; at the Dean Lee Research Station near Alexandria; and at the Northeast Research Station at St. Joseph. Glyphosate was applied to wheat at first node, boot stage or early flowering at rates to simulate drift representing 1/8th (four ounces per acre), 1/16th (two ounces per acre) and 1/64th (0.5 ounces per acre) of the use rate of 32 ounces of Roundup Ultra (one pound active ingredient per acre). Injury from glyphosate appeared within three to five days after application as bleaching of leaf foliage followed by some growth inhibition beginning seven to 10 days after application. At 14 days after treatment, wheat injury was 40 percent to 55 percent for the 2-ounce-per-acre rate applied at first node and the 4-ounce-per-acre rate applied at all growth stages (Table 1).
Wheat height 28 days after treatment was reduced 47 percent when glyphosate was applied at 4 ounces per acre at first node and was reduced around 26 percent when glyphosate was applied at 2 ounces per acre at first node and at 4 ounces per acre at boot stage (Table 1). Wheat height was not reduced with glyphosate at 0.5 ounces per acre applied at first node or boot stage. Wheat height was not reduced for any rate of glyphosate applied at early flowering.
For glyphosate at 4 ounces per acre, wheat yield was reduced 72 percent applied at first node, 45 percent applied at boot stage and 54 percent applied at early flowering (Table 1). At 2 ounces per acre, wheat yield was reduced 25 percent to 30 percent for the three application timings. Wheat yield was not reduced when glyphosate was applied at 0.5 ounce per acre at any growth stage. The reduction in yield was reflected in spike density (number of seedheads per unit area), spikelet number per spike (number of seeds per seedhead) and seed weight (data not shown).
Data clearly show that yield reduction from glyphosate applied at 4 ounces per acre is associated with reduced number of spikes, spikelets per spike and seed weight. Yield reduction associated with the 2-ounce per acre rate, however, could not be explained based on individual yield components but is probably due to the additive effects of all the components.
An additional study was conducted at Baton Rouge in 2001 and 2002 to evaluate wheat variety response to simulated glyphosate drift. Wheat varieties included Coker 9663, Mason, LA 422, AGS 2000, Pioneer/26R61 and USG 3209. This study further defined the response of USG 3209 and included other varieties currently grown in the South. Wheat was planted at the Central Research Station near Baton Rouge in early December of 2000 and 2001 and was harvested in May. The six wheat varieties responded the same to glyphosate applied at four ounces per acre and two ounces per acre.
Wheat height 28 days after treatment was reduced an average of 34 percent for 4 ounces per acre and 17 percent for 2 ounces per acre of glyphosate applied at first node, but height was not reduced when glyphosate was applied at early flowering (data not shown). Yield was reduced an average of 58 percent for 4 ounces per acre and 43 percent for 2 ounces per acre applied at first node and 38 percent for 4 ounces per acre and 19 percent for 2 ounces per acre applied at early flowering. In both studies yield reductions in most cases were reflected in reduced spike density, spikelets per spike and seed weight.
Wheat is more sensitive to glyphosate applied at first node compared with boot stage or early flowering. In previous research conducted in Louisiana, early application of glyphosate to two- to three-leaf rice and six-leaf corn reduced yield more than the later applications at panicle differentiation in rice and at one week before tasseling in corn. At the highest rate evaluated for glyphosate (4 ounces per acre or 1/8th of the labeled use rate), which is typical of what could be expected from herbicide drift, rice yield was reduced in two of three years by 99 percent and 67 percent when glyphosate was applied early. Corn yield at the same rate was reduced an average of 78 percent over three years when glyphosate was applied early.
Yield reduction for wheat from firstnode application at four ounces per acre reduced yield 72 percent for the variety USG 3209 and 58 percent when averaged across six varieties in the second study. The yield reductions are comparable to what has been reported for rice and corn and clearly show the negative effect of off-target drift of glyphosate on grass crops.
Glyphosate is effective in controlling grasses, which explains the greater sensitivity of wheat. Even though wheat yield reduction from sub-lethal rates of glyphosate was similar to that reported for rice and corn when applied early at the same rate, visual injury alone was not a good indicator of rice or corn yield loss. This research, however, shows that visible injury would be an accurate indicator of wheat yield reduction. Bleaching of wheat foliage was clearly evident within seven to 10 days after application, but it was not observed as a common injury response to sub-lethal rates of glyphosate applied to rice and corn.
(This article was published in the winter 2008 issue of Louisiana Agriculture.)