Cover crops can provide many benefits to a peanut and cotton rotation including suppressing weeds, conserving soil moisture after termination, increasing soil organic matter, and reducing soil erosion (Clark Reference Clark2007; Dabney et al. Reference Dabney, Delgado and Reeves2001; Kasper and Singer Reference Kasper, Singer, Hatfield and Sauer2011; Lu et al. Reference Lu, Watkins, Teasdale and Abdul-Baki2000). High-residue cover crops have been shown to suppress weeds in no-till or strip-till cropping systems through resource competition, alleopathic affects, physical impediment, and light suppression (Aulakh et al. Reference Aulakh, Price and Balkcom2011; Dabney et al. Reference Dabney, Delgado and Reeves2001; Price and Norsworthy Reference Price and Norsworthy2013; Reberg-Horton et al. Reference Reberg-Horton, Grossman, Kornecki, Meijer, Price, Place and Webster2011; Reeves et al. Reference Reeves, Price and Patterson2005). In recent years, throughout the Southeastern United States there has been an increasing practice of using cover crops and conservation tillage (Claassen et al. Reference Claassen, Bowman, Mcfadden, Smith and Wallander2018; SARE CTIC 2017). Producers often use residual herbicides during the growing season to extend the period of weed control and provide another control method to herbicide programs especially to manage herbicide-resistant weeds. However, residual herbicides can prevent the successful establishment of fall-seeded cover crops, thus reducing biomass and subsequent weed suppression and achieving longer-term benefits provided by cover crops (Curran et al. Reference Curran, Lingenfelter, Garling and Wagoner2006; Rogers et al. Reference Rogers, Talbert and Frans1986; Yu et al. Reference Yu, Van Eerd, O’Halloran, Sikkema and Robinson2015).
Previous studies have evaluated soybean and corn herbicide carryover onto fall-seeded cover crops with varied results. One study observed pyroxasulfone caused a 12%, 16%, and 11% reduction in plant density for cereal rye, hairy vetch, and wheat, respectively; however, the reduction in plant density did not lead to significant biomass reductions (Palhano et al. Reference Palhano, Norsworthy and Barber2018). The same study found crimson clover biomass reductions of 13%, 12%, and 11% for atrazine, pyroxasulfone, and S-metolachlor, respectively, when applied during the growing season; however, there were no significant reductions in plant density (Palhano et al. Reference Palhano, Norsworthy and Barber2018). Yu et al. (Reference Yu, Van Eerd, O’Halloran, Sikkema and Robinson2015) found imazethapyr, S-metolachlor + atrazine + mesotrione, and saflufenacil + dimethenamid-p did not cause any significant injury or biomass reductions to oat, hairy vetch, and cereal rye when planted 3 mo after application at the labeled rates. One study evaluating the carryover effects of cotton herbicides (fluometuron, MSMA, trifluralin, linuron) on hairy vetch and wheat found ground cover reductions varied greatly by soil type, with more injury found in Dundee silty clay than the silt loam soils (Rogers et al. Reference Rogers, Talbert and Frans1986). Cornelius and Bradley (Reference Cornelius and Bradley2017) found cereal rye to be the most tolerant cover crop to all of the corn and soybean herbicides evaluated, with cloransulam-methyl, flumioxazin, fomesafen + S-metolachlor, and metribuzin causing reductions in cover crop stands or biomass. The study also found that crimson clover and Austrian winter pea (Pisum sativum L.) were the most sensitive to herbicide carryover (Cornelius and Bradley Reference Cornelius and Bradley2017). In addition, pyroxasulfone, imazethapyr, fomesafen, and flumetsulam carryover reduced stand and cover crop biomass more than other herbicides evaluated (Cornelius and Bradley Reference Cornelius and Bradley2017). A greenhouse study in Iowa showed radish was the most sensitive to corn and soybean herbicide carryover, whereas cereal rye was the most tolerant (Hartzler and Anderson Reference Hartzler and Anderson2015). All of these previous studies have shown there are cases in which herbicide carryover can affect the establishment and biomass of fall-seeded cover crops; however, there is not a comprehensive understanding of the effects of cotton and peanut herbicides on Southeastern and mid-South cover crops.
Herbicide carryover can reduce cover crop biomass and subsequent weed suppressive qualities. It can increase expenses associated with cover crop establishment if replanting is needed in the fall or more herbicide applications the following season are needed for weeds. Herbicide chemistry and soil properties including pH, texture, organic matter, clay content, temperature, and moisture determine herbicide persistence in the soil. Overall, few of the previous studies evaluated commonly used peanut and cotton residual herbicides and limited research has been conducted in the Southeast, which has different environmental conditions and soil compared with other regions. Therefore, the objective of this study was to investigate the responses of 6 cover crops (daikon radish, cereal rye, oat, crimson clover, winter wheat, and common vetch) to simulated carryover from 12 common soil residual herbicides used in peanut and cotton.
Materials and Methods
Field trials were conducted in Macon County (32.4939°N, 85.8903°W)), and Henry County (31.3512°N, 85.3146°W), Alabama, in 2016–17 and 2017–18. The Macon County trial had a Kalmia sandy loam soil (fine-loamy over sandy or sandy-skeletal, siliceous, semiactive, thermic Typic Hapludults) and the Henry County trial had a Dothan fine sandy loam soil (fine-loamy, kaolinitic, thermic Plinthic Kandiudults). Soil composition and pH for each location are listed in Table 1. At each location, soil was conventionally tilled 1 wk prior to herbicide application to provide ideal soil seeding conditions and prevent previous crop residue interference with herbicide application. The study was set up as a completely randomized block design with herbicide treatments in each block applied in strips, and the cover crops planted in perpendicular strips across the herbicide treatment. Plots were 1.8 m by 3.7 m with four replications in Henry County and three replications in Macon County each year. Herbicide rates were set at 10% of full-labeled rate and all treatments were applied prior to cover crop planting (Table 2). The treatment rate of 10% of the full-labeled rate was selected to simulate high concentrations of herbicide residue carryover beyond cotton or peanut harvest. Herbicide treatments were applied November 18, 2016 and October 30, 2017 in Macon County and November 3, 2016 and October 30, 2017 in Henry County. Herbicide treatments were applied using a backpack sprayer with a six-nozzle boom (Teejet TT110025 flat-fan nozzles in Henry County and Teejet XR11002VS extended-range flat-fan spray tips in Macon County; Teejet®, Spraying Systems Co., Wheaton, IL 60187) propelled by compressed CO2 at a spray volume of 187 L ha−1. Plots were immediately irrigated with 1.3 cm after herbicide applications to ensure activation. Six cover crops were planted: Daikon radish, Wrens Abruzzi cereal rye, Coker 227 oat, crimson clover, Pembroke 2017 winter wheat, and AU Olympic vetch. Cover crops were drill-seeded with a Hege plot grain drill November 21, 2016 and November 7, 2017 in Macon County, and November 7, 2016 and November 3, 2017 in Henry County with a Great Plains 1205 no-till drill. Rye, oat, and wheat were planted at 100 kg ha−1. Clover and vetch were planted at 22 kg ha−1. Radish was planted at 11 kg ha−1. No rain was received for 7 d after planting (DAP) either year at either location.
Table 1. Locations and soil information of field trials conducted in 2016–17 and 2017–18.a
a Soil information was provided by Auburn University Soil Testing Laboratory (Auburn AL).
b Abbreviation: OM, organic matter.
c Fine-loamy, kaolinitic, thermic Plinthic Kandiudults.
d Fine-loamy over sandy or sandy-skeletal, siliceous, semiactive, thermic Typic Hapludults.
Table 2. Herbicide treatments and rates.
a All treatments are 10% of the full-labeled rate to simulate carryover.
Plant stands and heights for 10 random plants were collected for each herbicide-by-cover crop treatment at 42 to 52 DAP and 145 to 149 DAP. Stand counts were taken in two linear 1-m rows in broadleaf cover crops and three 30-cm linear row stand counts were collected from the cereal grain cover crops. Heights were measured from the base of the plants at the soil to the highest growing point. A fresh weight biomass was recorded for each plot at 146 to 149 and 148 to 150 DAP in 2017 and 2018, respectively. In Henry County, cover crops were harvested with a hay cutter then two 1-m2 quadrats were raked and weighed onsite. In Macon County, a Carter® flail forage harvester was utilized to harvest and weigh the center 1.5-m by 3-m of the plot.
All data were converted to a percentage of the nontreated (NT) prior to statistical analysis for each individual cover crop. Then, converted data were processed with the PROC GLIMMIX procedure in SAS® 9.4 (SAS Institute Inc., Cary, NC 27513). Year, treatment, location, and block were subjected to ANOVA for a randomized complete block design. Each cover crop species was analyzed separately because the objective of the study was not to compare the different cover crops to each other but to evaluate the effects of different residual herbicides on each species individually. Treatment and location were considered fixed effects, while block was a random effect. If treatment by location was not significant, then location was used as a random effect and data were combined over location for analysis. If the interaction was significant, data were analyzed and presented by location. All means were separated using Tukey’s honestly significant difference test with P < 0.10 to reveal statistical difference. This significance level was used because differences in cover crop injury and growth are difficult to distinguish, and P < 0.10 will, without question, reveal biologically significant differences among treatments.
Results and Discussion
There was a year-by-treatment interaction (P < 0.10) for each cover crop species; therefore, 2016–17 and 2017–18 were analyzed separately for stand counts and plant heights. Data were combined over both locations for stand counts because there were no location-by-treatment differences for each year. Stand counts were evaluated at 48 to 52 DAP and 145 to 148 DAP at each location in 2016–17 (Table 3). Stand reductions of 43%–52% in rye and 44%–75% in wheat, respectively, were observed at 48 to 52 DAP for S-metolachlor, acetochlor, pyroxasulfone, imazapic, and bentazon plus acifluorfen over both locations. Wheat also had stand reductions of 36% with chlorimuron-ethyl use and 52% with diclosulam use at both locations. Vetch had significant stand reductions for all herbicide treatments, except for prometryn, at 48 to 52 DAP, ranging from 14% to 80% over both locations. The sensitivity of vetch to residual herbicides was not observed in two previous studies that evaluated corn and soybean herbicides (Bradley et al. Reference Bradley, Hoorman, Ellis, Shannon and Broz2016; Yu et al. Reference Yu, Van Eerd, O’Halloran, Sikkema and Robinson2015). However, two other studies found significant vetch injury or biomass reductions, indicating that environmental or soil composition factors are likely playing a role in herbicide carryover effecting vetch establishment (Bryan Reference Bryan2014; Palhano et al. Reference Palhano, Norsworthy and Barber2018). S-metolachlor, pyroxasulfone, and acetochlor had the largest negative impacts on stand counts for rye (52%, 45%, 44%), wheat (75%, 59%, 67%), and vetch (80%, 66%, 74%) at 48 to 52 DAP. Similarly, Palhano et al. (Reference Palhano, Norsworthy and Barber2018) saw stand reductions 14 DAP of 12% and 11% with pyroxasulfone used with rye and wheat, respectively. Clover, radish, and oat were not affected by any herbicide treatment at 48 to 52 DAP. By 145 to 148 DAP, there were no stand reductions for any of the cover crops evaluated in 2016–17. The stand recovery was due to late-season germination of the affected cover crops. In 2017–18, stand counts were evaluated at 42 to 45 DAP and 147 to 149 DAP at each location (Table 4). At 147 to 149 DAP, diclosulam and trioxysulfuron-sodium reduced clover stand by 29% and 38%, respectively. Diclosulam reduced radish stand by 64% at 147 to 149 DAP. Chlorosis and stunting were observed for clover and radish plants following herbicide treatments that had stand reductions at 147 to 149 DAP. Oat, rye, and vetch did not have any stand reductions at either 42 to 45 DAP or 147 to 149 DAP. Overall, more stand reductions were observed at 147 to 149 DAP than at 42 to 45 DAP, which was different from 2016–17, when by 145 to 148 DAP there were no observed stand reductions. Conditions in 2016–17 favored more stand reductions by herbicides than in 2017–18. In 2016–17 there was more rainfall and lower soil temperatures than 2017–18 (Table 5). These environmental factors may have slowed cover crop germination and emergence allowing the seedling to be exposed to herbicide for a longer period compared with 2017–18, especially for herbicides that are not very water-soluble.
Table 3. Plant stand response to residual herbicides in peanut and cotton rotation in 2016–2017.a
a Abbreviations: NT, nontreated; DAP, days after planting.
b Data collected in Henry County December 12, 2016 and March 20, 2017. Collected in Macon County January 12, 2017 and April 18, 2017.
c Clover, radish, and oat did not have any significant stand reductions, and therefore were not included in this table.
d Means followed by the same letter in the same column do not differ based on a mixed model analysis of variance of a randomized complete block (P = 0.1). Data are expressed as percentage of nontreated.
Table 4. Plant stand response to residual herbicides in peanut and cotton rotation in 2017–2018.a
a Abbreviations: NT, nontreated; DAP, days after planting.
b Data collected in Henry County on December 18, 2017 and March 29, 2018. Collected in Macon County on December 19, 2017 and April 5, 2018.
c Rye, vetch, and oat did not have any significant stand reductions, and therefore were not included in this table.
d Means followed by the same letter in the same column do not differ based on a mixed model analysis of variance of a randomized complete block (P = 0.01). Data are expressed as percentage of nontreated.
Table 5. Average monthly rainfall, temperature, and soil temperatures
a Soil temperatures were taken at 10-cm depth.
b Soil temperature data was missing from November 25, 2017 to December 14, 2017 because the Macon County weather station soil sensor was down for repairs. Soil temperature data for the missing days came from Natural Resources Conservation Service Weather Station on Morris Farms, an Alabama SCAN site, approximately 10 miles away.
Location by treatment was different for radish height and was analyzed by each location in 2016–17 (Table 6). Treatments applied in Henry County did not result in any height reductions for radish at 48 or 145 DAP. In Macon County, height reductions of radish were reduced by 31% at 52 DAP when imazapic was used. This was the only time and location to show radish plant height reduction. Radish sensitivity was also observed in other studies in which radish had more injury from herbicide carryover than any other cover crop evaluated (Anderson Reference Anderson2014; Bradley et al. Reference Bradley, Hoorman, Ellis, Shannon and Broz2016; Bryan Reference Bryan2014; Hartzler and Anderson Reference Hartzler and Anderson2015). By 147 DAP, radish had recovered and there was no height differences compared with NT plants. Radish did not have height reductions at either timing in 2017–18. Again, no other cover crop had height reductions in either year or location of this study. Previous studies have not considered height as a potential growth parameter to evaluate for herbicide carryover onto fall-seeded cover crops. Based on the results of this study, plant height reductions are not a good visual indicator of herbicide carryover. Because radish was the only cover crop to have a plant height reduction, this was likely due to radish being sensitive to environmental factors or, possibly, to higher clay content in the soil in Macon County in 2016–17. Overall, based on these data, stand reductions are a better indicator of herbicide carryover compared to height reductions. Although stand losses were observed for some cover crops each year, this did not lead to biomass reduction at the end of the growing season either year. Also, the average biomass of all treated plots was not different from the average NT plots for each cover crop. Oat had the largest average biomass of the evaluated grass species in both treated and NT plots both years (Table 7). Although vetch had stand reductions for all but one herbicide in 2016–17 at the beginning of the season, it had the greatest average amount of biomass of the broadleaf species evaluated in the treated and NT plots. Even though clover did not exhibit any reductions, it did not have the largest amount of biomass of the broadleaf plants in 2016–17. Therefore, if a producer is growing a cover crop for biomass, then species selection and herbicide carryover need to be considered. Clover did not show any stand reductions but did not have the greatest amount of biomass at the end of the growing season in 2016–17 of the broadleaf plants evaluated. In 2017–18 radish exhibited the largest average biomass of the broadleaf species in the treated and NT plots. Overall, some cover crops had stand reductions; however, the new plants that did emerge were able to compensate for the reduced population and produce a biomass similar to that of the NT plots with a full growing season for each cover crop evaluated in this study. Cover crop response to herbicides varied from year to year, even though the same amount of herbicides were applied each year they did not vary between locations, suggesting environmental factors favored certain herbicide persistence each year and not soil composition. Although this study applied low rates of herbicides prior to planting to simulate carryover, in field settings, these herbicides would be applied weeks to months before planting fall-seeded cover crops. Environmental factors including soil pH, soil composition, microbial activity, soil temperatures, air temperatures, and other conditions can extend soil herbicide residual persistence, thus increasing carryover chances on to fall-seeded cover crops (Curran 2016). In addition to herbicide chemistry, how it degrades, and its half-life can affect how long an herbicide will be persistent in a soil. Overall in this study, diclosulam caused more stand reductions than any other herbicide by affecting all cover crops with the exception of oat and rye. Trifloxysulfuron-sodium, diuron, flumioxazin, pyrithiobac, and prometryn did not affect the establishment of any grass cover crop stands either year. Prometryn was the only herbicide to not affect broadleaf cover crops. Although injury and/or stand reduction is possible with residual herbicide use, producers should likely expect cover crops to recover and produce full biomass potential.
Table 6. Radish plant height response to residual herbicides in peanut and cotton rotation in Macon County in 2016–17.a
a Abbreviations: NT, nontreated; DAP, days after planting.
b Data collected in Macon County on January 12, 2017 and April 18, 2017.
c Means followed by the same letter in the same column do not differ based on a mixed model analysis of variance of a randomized complete block (P = 0.1). Data are expressed as percentage of nontreated.
Table 7. Average wet weight cover crop biomass in nontreated and all treated plots.a
a Data collected in Henry County March 31, 2017 and March 30, 2018. Collected in Macon County April 19, 2017 and April 6, 2018.
b Treated plots are the average of all plots treated with herbicides across both locations.
c Abbreviation: NT, nontreated.
Overall, oat showed the most herbicide tolerance with no treatments reducing stands, heights, or biomass in either year. This aligns with other studies that found oat to be tolerant to many corn and soybean herbicides, including S-metolachlor, imazethapyr, atrazine, and mesotrione (Hartzler and Anderson Reference Hartzler and Anderson2015; Yu et al. Reference Yu, Van Eerd, O’Halloran, Sikkema and Robinson2015). Previous studies have shown rye to be more tolerant than oat to residual herbicides (Bryan Reference Bryan2014; Cornelius and Bradley Reference Cornelius and Bradley2017). One study found cereal rye had the most tolerance out of all the cover crops tested, with stand and biomass reductions caused by only a few herbicides, including by flumioxazin, cloransulam, sulfentrazone, metribuzin, and fomesafen + S-metolachlor (Cornelius and Bradley Reference Cornelius and Bradley2017). Another study also found rye to be the most tolerant to commonly used corn herbicides including S-metolachlor (Bryan Reference Bryan2014). However, this study did observe rye stand losses with a number of herbicides including S-metolachlor in 2016–17. Wheat exhibited the most sensitivity to herbicide carryover out of the grasses evaluated as more stands were reduced compared with other grasses. Rogers et al. (Reference Rogers, Talbert and Frans1986) observed significant cover reductions of wheat to cotton herbicides (fluometuron, MSMA, trifluralin) in three different soil types. Broadleaf cover crops showed more sensitivity to herbicide carryover compared to grass species. Vetch response was variable in that it had the most sensitivity to herbicide carryover of all the broadleaf plants with all but one herbicide effecting stand establishment in 2016–17; however, it did not exhibit a stand reduction the following year. Some studies have shown hairy vetch to be the most tolerant cover crop to herbicide carryover, whereas other studies showed it to be the most sensitive (Bryan Reference Bryan2014; Hartzler and Anderson Reference Hartzler and Anderson2015; Rogers et al. Reference Rogers, Talbert and Frans1986; Stahl Reference Stahl2016; Yu et al. Reference Yu, Van Eerd, O’Halloran, Sikkema and Robinson2015). Palhano et al. (Reference Palhano, Norsworthy and Barber2018) found clover had reduced biomass in a field study from residual herbicides but emergence reductions were not observed, whereas the opposite occurred in this study. Another study found acetochlor and S-metolachlor caused biomass reductions of clover during one year of the study; however, neither herbicide caused injury in this study (Cornelius and Bradley Reference Cornelius and Bradley2017). Previous studies indicated radish, similar to clover and vetch, had both tolerance and susceptibility to carryover (Cornelius and Bradley Reference Cornelius and Bradley2017; Yu et al. Reference Yu, Van Eerd, O’Halloran, Sikkema and Robinson2015). One study evaluating oilseed radish tolerance found it to be sensitive to a number of residual herbicides, including fomesafen, S-metolachlor/fomesafen, and imazethapyr but was not affected by them the following year, likely due to increased rainfall (Cornelius and Bradley Reference Cornelius and Bradley2017). Another study that also had varying results from year to year did not recommend planting oilseed radish within 3 mo of an imazethapyr application but did not report injury with S-metolachlaor + atrazine and saflufienacil + dimethenamid-P (Yu et al. Reference Yu, Van Eerd, O’Halloran, Sikkema and Robinson2015). The results of this study and all previous studies indicate that residual herbicides have the potential to reduce fall-seeded cover crop establishment; however, weather conditions, soil textures, application timings, and other environmental factors affect the severity of damage observed.
Overall, cover crop stand establishment varied over the years but not locations, similar to previous studies with other row crop residual herbicides, likely due to environmental factors affecting herbicide persistence (Cornelius and Bradley Reference Cornelius and Bradley2017; Tharp and Kells Reference Tharp and Kells2000; Yu et al. Reference Yu, Van Eerd, O’Halloran, Sikkema and Robinson2015). Fall-seeded cover crop should be planted based on the residual herbicides applied to row crops the previous season, when the last application of residuals occurred and based on the biomass goal and nutrient needs of the field. Although initial injury and stunting may occur, cover crop biomass was not affected by the residual herbicides evaluated in this study and producers can still expect the full benefits offered by cover crops. Further research needs to be conducted to determine the minimum plant-back interval needed for fall-seeded cover crops after herbicide applications in the previous crop, especially as the utilization of cover crops increases in the Southeast.
Acknowledgments
We thank the National Peanut Board for funding support and the Alabama Agricultural Experimental Stations for trial initiation and management. No conflicts of interest have been declared.
