PDF version
for printing

© 2006 Plant Management Network.
Accepted for publication 28 December 2005. Published 8 March 2006.

Effect of Repeated Glyphosate Use on Weed Communities in a Soybean-Corn Rotation

R. G. Hartzler, Department of Agronomy, Iowa State University, Ames, IA 50011; J. W. Singer and K. A. Kohler, USDA-ARS, National Soil Tilth Laboratory, Ames, IA 50011; and D. D. Buhler, Department of Crop and Soil Sciences, Michigan State University, East Lansing, MI 48824

Corresponding author: R. G. Hartzler. hartzler@iastate.edu

Hartzler, R. G., Singer, J. W., Kohler, K. A., and Buhler, D. D. 2006. Effect of repeated glyphosate use on weed communities in a soybean-corn rotation. Online. Crop Management doi:10.1094/CM-2006-0308-01-RS.

Abstract

Five weed management systems, ranging from total reliance on glyphosate to no glyphosate, were evaluated over a four-year period (1998-2001) in a soybean [Glycine max (L.) Merr.]-corn (Zea mays L.) rotation in chisel and no-tillage systems. All treatments provided acceptable levels of weed control during the 4-year period, with a maximum end-of-season weed density of less than 27 weeds per 100 ft² in the glyphosate-only treatment during the fourth year of the study. The dominant weed was dandelion. While occasional differences in control were observed among treatments, they were minor and weed escapes were not sufficient to impact crop yields. The glyphosate-only treatment in soybean outyielded other treatments in 2000, but yields were not affected by weed management system in the other years. The weed management systems imposed during the four years of the study did not influence weed density or composition that emerged during 2002 when no herbicides were applied. Giant foxtail, velvetleaf, and Amaranthus spp. were more prevalent in the chisel-plow system than in no-tillage at the conclusion of the experiment, whereas dandelion was present at higher densities in the no-tillage system. While all systems provided high levels of weed control, the glyphosate-only system did allow nearly a 66% reduction in the total amount of herbicide active ingredient applied compared with the system relying on conventional preemergence and postemergence herbicides.

Introduction

The introduction of glyphosate-resistant (GR) soybean in 1995 resulted in a rapid change in weed management strategies throughout soybean producing areas of the U.S. In 2002 glyphosate was applied to 78% of the U.S. soybean acres (9). Advantages of GR soybean compared to conventional varieties include: (i) broad spectrum weed control, (ii) greater flexibility in application timing, and (iii) large margin of crop safety. While the adoption rate of GR corn has been much slower than soybean, increasing acres are being planted to GR corn hybrids (9).

Weed communities in agronomic fields are influenced by many factors, including tillage, crop rotation, edaphic factors, environment, and management practices (4). Greater weed diversity was found in plots maintained in no-tillage for 25 years than in minimum or conventional tillage plots (3). Researchers in Canada found that herbicide use pattern was the second most important farming practice that influenced weed communities, while crop selection was the most important factor (8). The increased reliance on glyphosate for weed management has raised concerns about the impact this approach may have on weed communities (1). Heavy reliance on any single tactic will result in a shift towards weeds that are able to survive that control tactic (7). Rapid weed shifts frequently result in significant economic losses due to yield losses from uncontrolled weeds and increased expenditures for control tactics. The objectives of this study were to measure weed densities and weed population shifts that occur using conventional and glyphosate weed management systems in a soybean-corn rotation.

Field Study

An experiment was initiated near Boone, IA in 1998 to document changes in a weed community after repeated use of glyphosate. A split-plot experiment with tillage (no-tillage and fall chisel/spring field cultivator) as the main plot and weed management treatment as the split-plot with four replicates was used. A soybean-corn rotation was established with soybean planted during the first year. Both crops were planted in 30-inch rows using recommended practices for corn and soybean production. On 5 May 1998 and 17 May 2000, Pioneer Brand ‘9294RR’ soybean was planted at 180,000 seeds per acre, while ‘92B84’ was planted on 20 May 2002. Dekalb Brand ‘545RR’ corn was planted on 29 April 1999 and 26 April 2001 at 33,000 seeds per acre. Individual plots were 12.5 ft wide by 75 ft long. The soils at the experimental site were Canisteo silty clay loam (fine-loamy, mixed, superactive, calcareous, mesic Typic Endoaquolls) and Clarion loam (fine-loamy, mixed, superactive, mesic Typic Hapludolls).

Five weed management strategies were initiated; two utilized conventional herbicides with no glyphosate except for a burndown treatment in no-tillage (PRE and POST; PRE with cultivation). Three management programs included postemergence applications of glyphosate in GR crops as a component of the system: (i) glyphosate applied when weeds reached 4 inches with a sequential application if needed (Glyphosate), (ii) glyphosate when weeds reached 4 in followed by interrow cultivation (Glyphosate and cultivation), and (iii) a preemergence herbicide followed by postemergence glyphosate when weeds reached 4 inches (PRE and glyphosate). In the no-tillage system glyphosate was applied at planting if weeds were present. The management strategies were maintained for four years. Specific products and rates are provided in Table 1.

Table 1. Herbicides used in the different weed management treatments.^x,y

 ^x In the no-tillage system, glyphosate was applied at 0.75 lb ai/acre at planting to control emerged vegetation.

 ^y Rates listed are on a per acre basis.

Weeds were counted annually prior to harvest in an area between the center three rows (375 ft²). The dominant weeds in the experimental area were foxtail (Setaria spp.), common lambsquarter (Chenopodium album), velvetleaf (Abutilon theophrasti), and common dandelion (Taraxacum officinale). Because there were no major shifts in dominant species during the course of the experiment, weeds were combined into annual grasses (AG), annual broadleaves (AB), and perennials (PEREN) to simplify discussion. In order to quantify changes in weed communities after four years, the tillage treatments were maintained but the management systems were discontinued in 2002. The no-tillage plots were treated with a burndown application of glyphosate at planting, while the chisel plow treatment received no herbicide. Weed counts were collected from six 0.1-m² quadrats in each subplot on 23 July 2002.

Plots were harvested using a small plot combine and yield data were adjusted to 13 and 15.5% moisture content for soybean and corn, respectively. Statistical analysis was conducted by year as a split-plot using analysis of variance. Weed count data were square root transformed prior to analysis. Block and block by tillage interactions were considered random effects while weed management system and tillage by weed management system interactions were considered fixed effects. Tillage by weed management treatment interactions were detected for total weeds in 2000, annual grass and annual broadleaf weeds in 2001, but data were pooled across tillage because no consistent pattern was detected. Treatment means were separated using a protected LSD (P < 0.05).

Influence of Management Systems on Weed Communities

While there were occasional differences in weed densities among the management systems, all treatments maintained weed densities below 30 weeds per 100 ft² throughout the duration of the study (Table 2). The majority of weeds present at harvest appeared to have emerged after the completion of weed control tactics. Significant tillage effects for total weed density were observed in 1999 and 2000 due to greater dandelion densities in no-tillage than in chisel tillage. Few tillage by weed management treatment interactions were observed during the course of the study. In 1999 treatments relying on POST glyphosate or a PRE application followed by glyphosate had higher weed densities than the other treatments. The weed community in these treatments was dominated by perennial weeds, primarily dandelion. The dandelion population consisted of individuals that emerged after glyphosate application. Although the perennial and total weed densities remained higher in 2000 in the glyphosate treatments than in the other weed management systems, the weed densities were reduced compared to 1999. In 2001, perennial and total weed densities were higher in the glyphosate alone and PRE with glyphosate post treatments compared to the other treatments.

Table 2. Weed densities of annual grasses (AG), annual broadleaves (AB), perennials (PEREN) and total weed number at harvest from 1998 through 2001 for different herbicide treatments near Boone, IA.

 ^x Means within a column by year followed by the same letter are not different.

 ^y Total may not be cumulative because of rounding.

 ^z Volunteer corn included in total.

Crop yields were not affected by tillage during the study period. Weed management system only influenced yield in 2000, with the glyphosate alone system providing the highest yield of any treatment (Table 3). Systems relying on conventional herbicides yielded less than those including glyphosate. Total weed densities were greatest in the glyphosate alone treatment (Table 2), thus the yield response did not appear to be related to interaction between soybean and weeds. While postemergence herbicides used in soybean occasionally reduce yield (10), the lack of crop response following application (data not presented) suggests factors unrelated to the weed management may have been responsible for the yield differences.

Table 3. Soybean and corn yield response to different herbicide treatments, averaged across tillage, in alternating years in a soybean-corn rotation near Boone, IA, from 1998 through 2002.

 ^x Means within a column followed by the same letter are not different.

The herbicide programs used the previous four seasons had no impact on the density of any weed species in 2002 when the experimental area was left untreated after planting (data not presented). Total weed density did not differ among the tillage systems; however, giant foxtail, Amaranthus spp., and velvetleaf were more prevalent in the chisel plow system than in no-tillage (Table 4). Dandelion, a perennial, was the only species more prevalent in no-tillage than the tilled system. The relatively low weed densities indicated that a variety of tactics can be used to effectively manage weeds in this cropping system.

Table 4. Weed densities of foxtail (SETSP), lambsquarter (CHEAL), Pennsylvania smartweed (POLPY), pigweed (AMASP), velvetleaf (ABUTH), dandelion (TAROF), other, and total weeds in late July of 2002 in soybean near Boone, IA, in plots that were previously in different herbicide treatments from 1998 through 2001 in a soybean-corn rotation in chisel plow and no-tillage.

 ^x Means within a column followed by the same letter are not different.

During the course of this study no increases in weed densities or shifts in weed communities were observed between a system relying totally on glyphosate and those utilizing conventional herbicides, those utilizing glyphosate in combination with other herbicides, or with mechanical cultivation. In addition to reducing the number of herbicides used, the glyphosate-only treatment averaged 1.12 lb/acre active ingredient per year in the chisel tillage system, compared to the conventional treatment that utilized 3.16 lb/acre of active ingredient per year. Although no evidence of weedy adaptation was observed during the course of this study, we believe it is important to recognize the limitations of short term, small-plot experiments in evaluating risks associated with heavy reliance on glyphosate or any other weed control tactic. The likelihood of a rare resistant gene being present in a field decreases with the size of the weed population (5). The ability of weeds to develop resistance to glyphosate is well documented, with resistance confirmed in eight weed species (6). Thus, while the characteristics of glyphosate result in lower risks of selecting resistant weeds than many other herbicides (2), diversification of management systems to reduce selection pressure placed on weed communities is important to preserve the value of glyphosate.

Disclaimer

Mention of trade names or commercial products is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.

Literature Cited

1. Anonymous. 2004. Wisconsin farmers and agri-business call for glyphosate (Roundup) stewardship. White paper, Univ. of Wis. Ext.

2. Bradshaw, L. D., Padgette, S. R., Kimball, S. L., and Wells, B. H. 1997. Perspectives on glyphosate resistance. Weed Technol. 11:189-198.

3. Cardina, J., Regnier, E., and Harrison, K. 1991. Long-term tillage effects on seed banks in three Ohio USA soils. Weed Sci. 39:186-194.

4. Clements, D. R., Weise, S. F., and Swanton, C. J. Integrated weed management and weed species diversity. Phytoprotection 75:1-18.

5. Diggle, A. J., Neve, P. B., and Smith, F. P. 2003. Herbicides used in combination can reduce the probability of herbicide resistance in finite weed populations. Weed Res. 43:371-382.

6. Heap, I., DiNicola, N., and Glasgow, L. 2005. The international survey of herbicide resistant Weeds. Online. Herbicide Resistance Action Comm., N. Amer. Herbicide Resistance Action Comm., and Weed Sci. Soc. of Amer.

7. Kegode, G. O., Forcella, F. and Clay, S. 1999. Influence of crop rotation, tillage, and management inputs on weed seed production. Weed Sci. 47:175-183.

8. Leeson, J. Y., Sheard, J. W., and Thomas, A. G. 2000. Weed communities associated with Saskatchewan farm management systems. Can. J. Plant Sci. 80:177-185.

9. USDA-NASS. 2003. Agricultural Chemical Usage 2002 Field Crops Summary. Online. U.S. Dep. of Agric., Nat. Agric. Stat. Serv..

10. Young, B. G., Young, J. M., Matthews, J. L., Owen, M. D. K., Zelaya, I. A., Hartzler, R. G., Wax, L. M., Rorem, K. W., and Bollero, G. A. 2003. Soybean development and yield as affected by three postemergence herbicides. Agron. J. 95:1152-1156.