Impact
Statement

PDF version
for printing

© 2008 Plant Management Network.
Accepted for publication 18 June 2008. Published 18 August 2008.

Managing Field Bindweed in Sorghum-Wheat-Fallow Rotations

Mark A. Marsalis, Assistant Professor, Agricultural Science Center at Clovis, New Mexico State University, Clovis 88101; Leonard M. Lauriault, College Associate Professor, Agricultural Science Center at Tucumcari, New Mexico State University, Tucumcari 88401; Stan H. Jones, Curry County Agricultural Agent, New Mexico Cooperative Extension Service, New Mexico State University, Clovis 88101; and Mark J. Renz, Assistant Professor, Department of Agronomy, University of Wisconsin, Madison 53706

Corresponding author: Mark A. Marsalis. marsalis@nmsu.edu

Marsalis, M. A., Lauriault, L. M., Jones, S. H., and Renz, M. J. 2008. Managing field bindweed in sorghum-wheat-fallow rotations. Online. Crop Management doi:10.1094/CM-2008-0818-01-RS.

Abstract

Infestation of field bindweed (Convolvulus arvensis L.) on farms in the semiarid, Southern High Plains is widespread and has led to significant reductions in crop productivity throughout the region. A 3-year study was conducted in a wheat (Triticum aestivum L.)-sorghum [Sorghum bicolor (L.) Moench]-fallow rotation at two locations to investigate the long-term suppression of field bindweed with chemical methods. Specifically, the effects of one, two, and three years of fall-applied herbicides on bindweed populations were of interest. Chemical treatments used at each location were: quinclorac (0.28 and 0.43 kg/ha); quinclorac + 2,4-D (0.28 + 0.56 kg/ha); dicamba (1.12 kg/ha); and metsulfuron (0.021 kg/ha). Results indicate that the third year of herbicide application may be critical in effectively suppressing field bindweed for more than one year. There were no differences among herbicide treatments between one and two years of application by the third year. All herbicides containing quinclorac and dicamba significantly reduced bindweed populations the following spring after application; metsulfuron alone was not considered effective in either short- or long-term suppression. Field bindweed control may be reduced when 3-year herbicide program begins following a fallow period without any tillage or crop competition.

Introduction

Field bindweed (Convolvulus arvensis L.) is a perennial noxious weed that infests thousands of acres in New Mexico and adjacent states. It is a substantial problem in cultivated fields where it competes for valuable resources and reduces the value of land. Field bindweed exhausts soils of moisture and nutrients and reduces yields in both irrigated and dryland situations, particularly when soil moisture is limiting in dry years (9). It is estimated that crop losses due to bindweed exceed $350 million per year in the US (2). Field bindweed is hardy and is difficult to completely eradicate due to its extensive, deep, perennial root system that gives rise to long, lateral roots from which new plants arise. Carbohydrates are stored in the root system and the plant utilizes these carbohydrates to regrow after defoliation, chemical burndown, or cultivation (1). Also, root fragments can regrow (8), and mechanical control methods (e.g., tillage) often tend to promote the spread of field bindweed by dragging roots into uninfested areas. New plants can develop from seed as well, and seed can remain viable in the soil for greater than 20 years (12). Combined control of field bindweed with chemical and mechanical methods can be effective and economical but is dependent upon the chemical and timing of tillage (14,15). In one study, the combination of sweep tillage and 2,4-D was shown to be the most economically profitable to the producer (14). High herbicide cost and fuel prices combined with low potential yields often limit growers’ ability to adopt an aggressive weed control program. Maximizing herbicide effectiveness per pass across the field is one step in reducing annual inputs and the negative impact of field bindweed.

Dryland wheat-sorghum-fallow rotations are common in the Southern High Plains. In a typical wheat-sorghum-fallow system, wheat is planted in the late summer to winter period (year 1) when good soil moisture is available and harvested the following late spring (year 2). After this, the field is fallowed until the next spring or early summer to permit soil moisture accumulation and weed control before being planted to grain sorghum, which is harvested that year in autumn (year 3). The field is then fallowed again until the late summer of year 4 when wheat is planted. During the fallow periods, tillage and/or herbicides are often used for weed control. The tillage also improves water infiltration and retention (below the tilled zone). Light tillage (e.g., shallow discing) is usually done immediately prior to planting. Information is limited on the effectiveness of fall applications of chemicals on bindweed suppression in these systems. Several chemicals are labeled for control of field bindweed, but historical evidence suggests that use of these chemicals has not been effective in eradicating the weed from these systems and control rarely extends beyond two years (7,11,13). Previous research suggests that fall applications prior to the first frost are the most effective in managing field bindweed (3,6,11).

The objectives of this project were to evaluate the effectiveness of consecutive years of fall applications of herbicides to field bindweed in addition to typical management practices for a grain sorghum/wheat cropping system.

General Methodology

A study was conducted near Broadview, NM (34°49’N, 103°10’W, elevation 1348 m) in a typical dryland wheat-sorghum-fallow rotational system. Soil type was an Olton clay loam (fine, mixed, superactive, thermic Aridic Paleustoll). Fields were selected with patches of bindweed of sufficient size and density (> 50% ground cover) to test if applications in the fall would eliminate populations of field bindweed that were in different phases of the rotation. Two fields were selected, one where wheat was harvested in 2003 (Location 1) and one where grain sorghum was planted and harvested in 2003 (Location 2). Each field was in the same sorghum/wheat/fallow rotation, but at different phases (Table 1). The main plot was herbicide and the subplot was one, two, or three applications of that herbicide over three years. Main plot size was 55.5 m˛ (6.1 × 9.1 m) and each plot was split into three portions as subplots (6.1 × 3.0 m), all of which received an application in the initial year (2003). Two-thirds received an additional application in the second year and one third received another application in the third year (Table 1). Herbicide treatments (Table 2) were applied prior to first frost. Depending on stage in the rotation, wheat had recently been planted or grain sorghum had been harvested when herbicide applications were made. Location 1 (Table 1) received additional main plot herbicide treatments (Table 2) due to a larger area of uniform infestation of field bindweed than that of location 2; however, only treatments common to both locations will be discussed statistically. Treatments were applied to field bindweed shoots with a CO₂ powered backpack sprayer at 140 liters/ha (15 GPA). A nonionic surfactant was used at 0.5% v:v. with all treatments. Methylated seed oil (MSO, 1.2 liters/ha) was added to all treatments containing quinclorac (Table 2). No other herbicides or fertilizers were applied during the study period as this is considered common practice for the region. Only light tillage was used within one month before planting of both grain sorghum and wheat.

Table 1. Stages of wheat-sorghum-fallow rotation during each year of fall-applied herbicide study at two locations near Broadview, NM, 2003-2006.

 ^x Light tillage (discing) was prior to planting sorghum in the spring (S) and prior to planting wheat in the fall (F).

 ^y Applications occurred either after sorghum and wheat harvest over residue or after wheat planting when wheat was tillering. Application dates are for both locations.

Table 2. Treatments used in 3-year, fall-applied herbicide study at two locations near Broadview, NM, described in Table 1, 2003-2006.

 ^x Additives were: MSO, methylated seed oil (1.2 liters/ha); NIS, non-ionic surfactant (0.5% v:v).

Experimental Design and Data Analysis

Experimental design was a randomized complete block with a strip-split plot arrangement of herbicide, annual application regime (one, two, or three consecutive years of applications), and year after the first application (2004, 2005, 2006) as a time factor. There were four replications of each main plot at each location. Bindweed evaluations were taken from each subplot in the spring of each year following herbicide application after wheat was harvested or before grain sorghum was planted, depending on the stage of the rotation (Table 1). Percentage bindweed cover was estimated based on visual rankings of 0 to 100% of ground cover in each subplot division. Visual ratings occurred on: 27 May 2004, 16 June 2005 and 17 May 2006, before sorghum planting.

Ground cover data were subjected to PROC MIXED ANOVA (SAS Institute Inc., Cary, NC) procedures for tests of main effects of herbicide, application regime (subplot), and year and all possible interactions. Means were separated using LSD (Fisher’s Protected Least Significant Difference) procedures where differences (P < 0.05) occurred (10). The two locations were analyzed separately due to a herbicide by location effect likely due to differences in stage of rotation at the beginning of the experiment between the two locations. Results are presented for each location. Significant herbicide by year, herbicide by application regime, and herbicide by year by application regime interactions led to interpretation of results by year within each application regime and associated separate LSD for each year within a location.

Herbicide Efficacy

Several herbicides applied once in the fall of 2003 were effective in reducing field bindweed cover the following spring (Figs. 1A and 2A). Effectiveness of all treatments containing quinclorac were lower (P < 0.05) than the control at location 1, except for the lowest rate (P = 0.07; Figs. 1A and 2A); however, no reductions in bindweed were visible by the following spring (Figs. 1A and 2A). The effect of herbicide control was not evident in 2005 or 2006 (P > 0.05), suggesting an application regime effect and implying that one fall application will only provide suppression of field bindweed through the next summer with plants recovering by the fall even with the most effective treatments. Similar results were seen for the same herbicides at location 2; however, dicamba was effective at reducing bindweed cover significantly (P < 0.05) the following year after application, whereas only a trend occurred at location 1 (Table 1; Figs. 1A and 2A).

Fig. 1. Field bindweed ground cover over three years with applications of herbicide in the autumns of 2003 (A); 2003 and 2004 (B); or 2003, 2004, and 2005 (C) at location 1. Ratings were taken in the spring of each year shown. * = significantly different (P < 0.05) from the control. Data are the means of four replications.

Fig. 2. Field bindweed ground cover over three years with applications of herbicide in the autumns of 2003 (A); 2003 and 2004 (B); or 2003, 2004, and 2005 (C) at location 2. Ratings were taken in the spring of each year shown. * = significantly different (P < 0.05) from the control. Data are the means of four replications.

While consecutive applications of herbicides common to both locations in the fall of 2003 and 2004 resulted in considerable reductions in populations at both sites in the spring of 2005 (Figs. 1B and 2B), bindweed recovered in 2006 relative to the control and no significant reductions in populations resulted with treatments. Although no live plants were observed at location 2 in 2005 with quinclorac treatments, bindweed roots were likely still viable and recovery by 2006 indicated that it was not eradicated from the location (Fig. 2B). Quinclorac and quinclorac (low rate) + 2,4-D combinations exhibited reductions (P < 0.05) in cover two years after treatment at both sites. However, significant differences between quinclorac at the high rate and low rate were observed only in 2005 after two years of treatment at location 1 (Fig. 1B). Again, dicamba reduced bindweed populations significantly only at location 2 (Figs. 1B and 2B).

Consecutive fall applications over three years were effective in reducing cover of field bindweed compared to the control with treatments including quinclorac and dicamba (Figs. 1C and 2C). Neither quinclorac rate nor the addition of 2,4-D, appeared to have much of an effect on control the following spring. In fact, addition of 2,4-D to quinclorac added no improvement to bindweed control except at location 1 in 2005 with the high rate of quinclorac (Fig. 1C, not tested at location 2). While not testable statistically across locations, application of 2,4-D alone at 1.12 kg/ha did not provide any control even after three consecutive fall treatments (applied at location 1 only, Tables 1 and 2). Clearly, three fall applications of quinclorac, quinclorac + 2,4-D, or dicamba alone reduced overall bindweed populations. Effectiveness of quinclorac may be enhanced by continued soil activity after application and during regrowth (4). Metsulfuron (low and high), fluroxypyr + 2,4-D, and 2,4-D alone had no observable effect on bindweed at any of the ratings in any year at location 1 (Fig. 1). Metsulfuron at the low rate did not control bindweed at any of the ratings at either location (Figs. 1 and 2). Limited control of bindweed by metsulfuron one year after application in this study supports previous research (5).

Bindweed cover in general was lower in spring 2006 (Figs. 1 and 2), but recovered by late summer in plots sprayed in either one, two, and three years (by observation only). The significant year effect was caused by much lower overall bindweed coverage across all herbicide treatments in 2006, particularly at location 1. The proposed reason for this discrepancy is lower March-May rainfall in 2006 (36 mm) compared with 2004 (113 mm) and 2005 (91 mm). Though not tested, it appears that differences among treatments occurred between sites; this is likely due to the phase of rotation (Table 1) and the impacts on bindweed cover at the time of herbicide application. The level of crop competition or tillage may have some impact on bindweed competitiveness and the effectiveness of treatments. At location 1, where the first year spraying followed a fallow period (no tillage for 1 year; Table 1), the effect of dicamba was not as great as when the treatment followed a pre-plant tillage (spring 2003) and a sorghum crop (Table 1; Figs. 1B and 2B) or when small wheat plants were growing during application (fall 2005 spraying; Fig. 1C). Tillage prior to wheat planting may have also added to this effect. Similarly, dicamba was more effective at location 2 in 2004 and 2005 (Fig. 2B) where application followed a sorghum crop and when wheat was actively growing, 2003 and 2004, respectively (Table 1). Although dicamba was lower than the control (P < 0.05) at location 2 throughout the study and numerically lower at location 1 with one and two sprays, it required three consecutive years of application to significantly reduce bindweed populations when the system began following a fallow period (Fig. 1C). Therefore, greater control may be likely when fields are treated that had either recently been tilled (15) or presently have active crop competition at time of herbicide application. This effect may or may not be due to moisture levels (13) or could be a result of growing season light and nutrient competition between bindweed and sorghum or bindweed and wheat (9). Having a dense canopy during summer to shade bindweed and/or growing wheat plants at time of application may provide stress to the bindweed plant that allows for greater effectiveness of the herbicide. Also, an initial stress on bindweed plants from tillage may exacerbate the effect of herbicides applied soon after.

Conclusions

Results indicate that one, two and three consecutive fall treatments only provide suppression the following year. Three years of fall treatments may provide further suppression, and potentially local eradication with some treatments; fourth-year data were not obtainable as grower management altered the system. However, observations at location 2 in late 2006 indicated that quinclorac (high rate) and dicamba plots with 0% cover ratings have remained as such for at least 1˝ years after herbicide application. Amount of fall/winter moisture may also enhance the activity of herbicides as it could increase absorption and translocation. Wiese et al. (15) reported improved herbicide control when bindweed plant growth was considered "good". Although typical management practices were utilized, these data also suggest that additional management of bindweed when it is actively growing during fallow periods, such as tillage and possibly additional herbicide applications, in conjunction with fall applications, could further improve long-term suppression. Of all of the herbicides tested, quinclorac and dicamba were the most effective at suppressing bindweed the following growing season. The herbicide selection (quinclorac or dicamba) or rate of quinclorac did not appear to affect control if consecutive treatments were applied for three years.

Acknowledgements

This research was supported by the New Mexico Agricultural Experiment Station. The authors gratefully acknowledge the technical and field assistance of Aaron Scott, Christie Werner and Justin Norsworthy for helping with chemical applications and bindweed observations during the project. Sincere gratitude is extended to Max and Pat Kralicek for allowing the use of their land for this research.

Literature Cited

1. Barr, C. G. 1936. Preliminary studies on the carbohydrates in the roots of bindweed. Agron. J. 28:787-798.

2. Boldt, P. E., Rosenthal, S. S., and Srinivasan, R. 1998. Distribution of field bindweed and hedge bindweed in the USA. J. Prod. Agric. 11:377-381.

3. Enloe, S. F., Westra, P., Nissen, S. J., Miller, S. D., and Stahlman, P. W. 1999. Use of quinclorac plus 2,4-D for controlling field bindweed (Convolvulus arvensis) in fallow. Weed Technol. 13:731-736.

4. Enloe, S. F., Nissen, S. J., and Westra, P. 1999. Absorption, fate, and soil activity of quinclorac in field bindweed (Convolvulus arvensis). Weed Sci. 47:136-142.

5. Heering, D. C., and Peeper, T. F. 1991. Field bindweed (Convolvulus arvensis) control in winter wheat (Triticum aestivum) with herbicides. Weed Technol. 5:411-415.

6. Parker, D. S., and Krall, J. M. 1989. Desiccation of field bindweed (Convolvulus arvensis) in alfalfa (Medicago sativa) seed fields with glyphosate. Weed Technol. 3:99-101.

7. Schoenhals, M. G., Wiese, A. F., and Wood, M. L. 1990. Field bindweed (Convolvulus arvensis) control with Imazapyr. Weed Technol. 4:771-775.

8. Sherwood, L. V. 1945. Field bindweed, Convolvulus arvensis L., root fragments may grow. Agron. J. 37:307-313.

9. Stahler, L. M. 1948. Shade and soil moisture as factors in the competition between selected crops and field bindweed, Convolvulus arvensis. Agron. J. 40:490-502.

10. Steel, R. G. D., and Torrie, J. H. 1981. Principles and Procedures of Statistics: A Biometrical Approach, 2nd Edn. McGraw-Hill, New York, NY.

11. Stone, A. E., Peeper, T. F., Kelley, J. P. 2005. Efficacy and acceptance of herbicides applied for field bindweed (Convolvulus arvensis) control. Weed Technol. 19:148-153.

12. Timmons, F. L. 1949. Duration of viability of bindweed seed under field conditions and experimental results in the control of bindweed seedlings. Agron. J. 41:130-133.

13. Westra, P., Chapman, P., Stahlman, P. W., Miller, S. D., and Fay, P. K. 1992. Field bindweed (Convolvulus arvensis) control with various herbicide combinations. Weed Technol. 6:949-955.

14. Wiese, A. F., Salisbury, D. C., Bean, B. W., Schoenhals, M. G., and Amosson, S. 1996. Economic evaluation of field bindweed (Convolvulus arvensis) control in a winter wheat-fallow rotation. Weed Sci. 44:622-628.

15. Wiese, A. F., Schoenhals, M. G., Bean, B. W., and Salisbury, C. D. 1997. Effect of tillage timing on herbicide toxicity to field bindweed. J. Prod. Agric. 10:459-461.