Introduction
Conservation tillage practices are increasingly threatened and, in some cases, have been abandoned because of lack of control of glyphosate-resistant Palmer amaranth and glyphosate-resistant horseweed (Conyza canadensis L. Cronquist) (Price et al. Reference Price, Balkcom, Culpepper, Kelton, Nichols and Schomberg2011, Reference Price, Monks, Culpepper, Duzy, Kelton, Marshall, Steckel, Sosnoskie and Nichols2016a). First identified in 2005 in Georgia, glyphosate-resistant Palmer amaranth is now reported extensively throughout the High Plains, Midsouth and Southeast cotton-producing states (Culpepper et al. Reference Culpepper, Grey, Vencill, Kichler, Webster, Brown, York, Davis and Hanna2006; Van Wychen Reference Van Wychen2016). Acetolactate synthase–resistant Palmer amaranth is also widely reported (Norsworthy et al. Reference Norsworthy, Griffith, Scott, Smith and Oliver2008; Van Wychen Reference Van Wychen2016; Webster Reference Webster2005; Wise et al. Reference Wise, Grey, Prostko, Vencill and Webster2009). Predicting our current situation, Ball (Reference Ball1992) recognized that reduced tillage accelerates additions to seedbanks of noncontrolled weeds, thus maintaining weed control in reduced-tillage systems is imperative. Ball also stated it is likely development of herbicide-resistant weeds will occur more rapidly in reduced-tillage systems because of potential rapid seedbank increases. In addition to crabgrass (Digitaria spp.), goosegrass [Eleusine indica (L.) Gaertn.], morning-glory (Ipomoea spp.), nutsedge (Cyperus spp.), Palmer amaranth, and sicklepod are the highly ranked troublesome weeds in cotton production (Van Wychen Reference Van Wychen2016). Full-season management strategies for Palmer amaranth, grasses, and nutsedge control are especially necessary because of their season-long germination (Bensch et al. Reference Bensch, Horak and Peterson2003; Webster and Coble Reference Webster and Coble1997; Webster et al. Reference Webster, Cardina and White2003). Glyphosate-based POST-only herbicide systems once provided adequate weed control; however, current challenges necessitate the inclusion of PRE herbicides into cotton weed management systems (Cahoon and York Reference Cahoon and York2019; Culpepper et al. Reference Culpepper, York, MacRae, Kichler, Whitaker and Davis2007, Reference Culpepper, Kichler and Sosnoskie2020; Whitaker et al. Reference Whitaker, York, Jordan and Culpepper2011).
In conservation tillage systems, use of high-biomass cover crops, overlapping residual herbicides, and, in extreme cases, strategic inversion tillage to bury shallow-germinating weed seeds such as Palmer amaranth are being recommended by state cooperative extension systems for control of herbicide-resistant and troublesome weeds to regain adequate weed control (Cahoon and York Reference Cahoon and York2019; Culpepper et al. Reference Culpepper, Kichler and Sosnoskie2020; Price et al. Reference Price, Balkcom, Culpepper, Kelton, Nichols and Schomberg2011, Reference Price, Monks, Culpepper, Duzy, Kelton, Marshall, Steckel, Sosnoskie and Nichols2016a; Smith et al. Reference Smith, Bond, Eubank and Scott2019). High-biomass winter cover crops, especially cereal cover crops preceding broadleaf cash crops, provide early-season weed suppression and are increasingly used with other weed management tactics for integrated weed control (Aulakh et al. Reference Aulakh, Price and Balkcom2011, Reference Aulakh, Price, Enloe, van Santen, Wehtje and Patterson2012, Reference Aulakh, Saini, Price, Faircloth, van Santen, Wehtje and Kelton2015; Hand et al. Reference Hand, Nichols, Webster and Culpepper2019; Loux et al. Reference Loux, Dobbels, Bradley, Johnson, Young, Spaunhorst, Norsworthy, Palhano and Steckel2017; Mirsky et al. Reference Mirsky, Curran, Mortensen, Ryan and Shumway2011; Norsworthy et al. Reference Norsworthy, McClelland, Griffith, Bangarwa and Still2011; Palhano et al. Reference Palhano, Norsworthy and Barber2017; Price et al. Reference Price, Reeves and Patterson2006, Reference Price, Reeves, Patterson, Gamble, Balkcom, Arriaga and Monks2007, Reference Price, Balkcom, Culpepper, Kelton, Nichols and Schomberg2011, Reference Price, Saini, van Santen, Sarunaite and Abdurakhmonov2016b; Reeves et al. Reference Reeves, Price and Patterson2005; Ryan et al. Reference Ryan, Curran, Grantham, Hunsberger, Mirsky, Mortensen, Nord and Wilson2011; Smith et al. Reference Smith, Reberg-Horton, Place, Meijer, Arellano and Mueller2011).
In addition, many producers in areas with coastal plain (including sands and sandy loams) or upland soils (clays or clay loams) use strip tillage to disrupt naturally occurring, annual, root-restricting hardpans. Strip-tillage surface residue disturbance width varies depending on equipment design and function. Thus, producers using a wide (30 cm row-1) strip-tillage implement that disturbs more cover crop residue and soil may find increased weed seed germination. The effect of residue disturbance due to strip-tillage width on subsequent weed management has not been evaluated to our knowledge.
Concerns regarding the efficacy of PRE herbicides used in conjunction with cover crops (due to interception and sorption) have led to the recommendation of, at a minimum, banded applications of PRE herbicides in conservation tillage systems to provide in-row residual weed control (Banks and Robinson Reference Banks and Robinson1982; Hand et al. Reference Hand, Nichols, Webster and Culpepper2019; Potter et al. Reference Potter, Truman, Strickland, Bosch and Webster2008). However, little research exists evaluating banded versus broadcast PRE applications in high-biomass reduced tillage systems. Even in high-residue conservation systems (>4.0 megaton ha−1 biomass) following rolling/crimping, soil in-row middles can be observed (A.J. Price, personal communication).
If PRE herbicides are only banded in strip-tilled rows, exposed soil in row-middle areas are likely sites for weed seed germination. With widespread acetolactate synthase– and glyphosate-resistant Palmer amaranth and other troublesome weeds, conservation tillage hectares will decline without development of effective management strategies (Price et al. Reference Price, Balkcom, Culpepper, Kelton, Nichols and Schomberg2011). Therefore, field studies were conducted to determine weed control, cotton population and yield, and revenue after testing two levels of cover-crop biomass, two residue-disturbance widths, and PRE herbicide treatments, either banded or broadcast.
Material and Methods
Experimental Design and Treatments
Identical field experiments were established at the E.V. Smith Research and Extension Center (EVS) located near Shorter, AL, and at the Wiregrass Research and Extension Center (WGS) near Headland, AL, in fall 2010, 2011, and 2012. The soil types were a Compass loamy sand (coarse-loamy, siliceous, subactive, thermic Plinthic Paleudults) at EVS, and a Dothan fine, sandy loam (fine-loamy, kaolinitic, thermic Plinthic Kandiudults) at WGS. The experimental design was a randomized complete block with 12 treatments in a factorial arrangement replicated four times for three years at each location (6 site-years total). Factor 1 was low or high biomass levels of cereal rye cv. Wrens Abuzzi. Cereal rye was established in the autumn of each year with a no-till drill in 17.8-cm rows at a seeding rate of 100 kg ha−1 and managed through fertilization and termination timing to attain the biomass levels. Cereal rye was planted at EVS on November 3, 2010; November 7, 2011; and November 6, 2012; and at WGS on November 23, 2010; November 23, 2011; and December 6, 2012. To attain a low biomass, plots received no nitrogen and rye was terminated approximately 8 weeks prior to cotton planting; thus, little residue was present at the time of cotton planting. High biomass was attained by fertilizing with a split application of nitrogen and sulfur at 34 and 12 kg ha−1, respectively, applied in the fall at emergence and at the same rate in January, and delaying cover crop termination until approximately 2 weeks prior to cotton planting. Factor 2 was strip-till width treatments of (1) subsoiling with a wide residue disturbance (30 cm row−1) 4-row 3-m KMC® Generation 1, 16 Series rip/strip till implement (Kelly Manufacturing Co., Tifton, GA) equipped with two wavy coulters and rolling baskets (Figure 1); or (2) or a narrow residue disturbance (<5 cm row−1) with a 4-row 3-m KMC® Generation 2 subsoiler/leveler implement equipped with two pneumatic tires and hard rubber rollers (Figure 2). Factor 3 was three herbicide treatments of (1) pendimethalin (Prowl 3.3EC; BASF Crop Protection, Durham, NC) at 0.84 kg ae/ha−1 plus fomesafen (Reflex 2L; Syngenta Crop Protection, Greensboro, NC) at 0.28 kg ai/ha−1 applied PRE broadcast; (2) pendimethalin plus fomesafen applied PRE at the same rates banded on the row; or (3) no PRE. After early-season weed control rating, each of these three herbicide treatments was followed by (fb) glyphosate (Roundup Powermax 4.5EC; Monsanto, St. Louis, MO) at 1.12 kg ae ha−1 applied POST fb a layby applications of diuron (Diuron 4L; Drexel Chemical Co., Memphis, TN) at 1.12 kg ai ha−1 plus monosodium methanearsonate (MSMS 6 Plus 6L; Drexel Chemical Co.) at 2.24 kg ai ha−1 applied POST directed to approximately 40cm cotton.
A wide residue disturbance (30 cm), 4-row 3-m KMC® Generation 1, 16 Series rip/strip till implement (Kelly Manufacturing Co., Tifton, GA) equipped with two wavy coulters and rolling baskets.
A narrow residue disturbance (<5 cm) 4- row 3-m KMC® Generation 2 subsoiler/leveler implement (Kelly Manufacturing Company) equipped with two pneumatic tires and hard rubber rollers.
Cover-Crop Sampling, Cotton Management, and Weed Control Assessment
In the spring prior to termination, cover-crop biomass samples were collected by clipping all aboveground plant parts at the soil surface from one randomly selected 0.25-m2 section in each plot. Plant material was dried at 60 C for 72 h and weighed. All plots were rolled with a custom-designed 4-m wide mechanical roller crimper (Bigham Brothers Inc., Lubbock, TX) prior to glyphosate applied at 1.12 kg ae ha−1 as described by Ashford and Reeves (Reference Ashford and Reeves2003) to aid in termination and to provide a uniform mat of residue on the soil surface in high-residue plots. The cotton cultivar ‘Phytogen 375 WRF’ was planted at both locations in all years. Cotton was planted with a 4-row planter equipped with double-disk openers with row cleaners set to minimize residue disturbance within the row. Plots were four 7.6-m rows spaced 0.97 m apart at EVS and 1.02 m at WGS. Cotton was managed according to Alabama Cooperative Extension recommendations. Cotton at EVS was nonirrigated; at WGS, cotton was irrigated by an overhead irrigation to maintain soil above the permanent wilting point.
Before harvest, plant population and heights were recorded from all cotton plants in randomly selected 3-m sections from the two center rows of each plot. Seed cotton yield was determined by machine harvesting the middle two rows of each plot with a spindle picker. In addition, detailed weed-control ratings by species in early season, prior to the POST and layby applications, where taken in rows and in row middles. In-row and between-row weed control, visually based on biomass reduction as compared with the nontreated control, was estimated in early season on a scale of 0 (native noninjured population) to 100 (complete death of all plants or no plants present) (Frans et al. Reference Frans, Talbert, Marx, Crowley and Camper1986).
Economic Analysis
A partial budgeting approach was used to calculate the net returns of each treatment. Net returns were equal to the revenue from cotton production minus the costs associated with cover crop establishment, management, and termination; PRE herbicide; and herbicide application costs. To calculate revenue, the price of cotton lint (USD $1.37 kg−1) USDA-NASS 2019) was multiplied by the average percentage lint turnout (0.40) times the cotton yield. All variable input costs were calculated using current prices, and total variable cost for each treatment are presented in Table 1. The price of ‘Wrens Abruzzi’ cereal rye was obtained from Agri-AFC (Headland, AL). Fertilizer price was obtained from The Feed Lot, Inc. (Pike Road, AL). All other prices were obtained from the 2018 Mississippi State University Cotton Budgets (MSU 2018).
Total variable cost by treatment, including fuel and labor used, in economic analysis comparing cover-crop biomass, strip-tillage residue disturbance width, PRE-herbicide placement on cotton weed control, and yield.
The costs associated with the management of the high-residue rye cover crop included price of cereal rye seed, fertilizer (33-0-0-12S), burndown herbicide (glyphosate; Roundup Powermax®), machinery (i.e., repairs and maintenance for tractors, implements, self-propelled), fuel, and labor. In addition, the high-biomass plots were mechanically rolled flat before planting cotton in the spring. The costs of the two subsoiling treatment operations were assumed to be the same. Both 4-row implements require the same amount of power per shank for in-row tillage and the shank design of each type of subsoiler is identical. The additional energy needed for coulters, rolling baskets, and drag chains on the wide-disturbance subsoiler would likely be negligible compared with the shank power requirements. The broadcasted PRE herbicide costs were calculated as a blanket application (100% coverage in 91-cm rows) with a John Deere® (Moline, IL) 6700 Hiboy sprayer. The banded PRE herbicide costs were calculated using a band sprayer apparatus being attached to the planter; thus, one operation pass is eliminated. A 20-cm band was applied over the row, resulting in only 22% spray coverage, compared with the broadcast treatment. Thus, 78% less herbicide was used in the banded plots as compared with the broadcast plots.
Statistical Analysis
Data were analyzed using generalized linear mixed-model methodology as implemented in PROC GLIMMIX (SAS, version 9.2; SAS Institute, Inc., Cary, NC). Cover-crop biomass, strip-tillage type, herbicide system, and their interactions were considered fixed effects. Effects and interactions were evaluated at P = 0.05. If location was significant at P ≤ 0.05, results were separated and analyzed by location and presented by location. Economic data were analyzed using PROC MIXED, and treatment means were compared using LSD with the same fixed effects. All economic tests were also evaluated at P = 0.05.
Results and Discussion
Cover Crop
ANOVA revealed significant cover-crop high- versus low-biomass treatment effects on biomass amount for year and location; therefore, biomass results are discussed separately by year and location. At EVS, low-biomass treatments did not differ over years and averaged 92 kg ha−1 (Table 2). High-biomass treatments differed, with 6,929 kg ha−1 in 2013 compared with an average of 3,146 kg ha−1 produced in 2011 and 2012. At WGS, the low-biomass treatments differed, with less biomass produced in 2012 (250 kg ha−1) than in 2011 or 2013, which were similar to each other and averaged 415 kg ha−1. High-biomass treatments differed each year, with 5,328, 4,787, and 4,073 kg ha−1 produced in 2011, 2012, and 2013, respectively. Conservation agriculture specialists consider cover-crop biomass amounts exceeding 4,000 kg ha−1 as a “high-biomass” system that can increase soil quality (Derpsch et al. Reference Derpsch, Roth, Sidiras and Köpke1991). However, for substantial weed suppression, weed scientists recommend greater than 6,000 kg ha−1 (Korres and Norsworthy Reference Korres and Norsworthy2015; Price et al. Reference Price, Saini, van Santen, Sarunaite and Abdurakhmonov2016b, 2018; Teasdale and Mohler Reference Teasdale and Mohler2000). Cover-crop biomass in our experiments once met the 6,000 kg ha−1 threshold in only 1 site-year; thus, weed control in our experiments was likely less and shorter lasting than that attainable in higher-residue systems. Timely planting of cover crops after cotton is difficult because cotton harvesting is often late, thus reducing biomass (Price et al. Reference Price, Saini, van Santen, Sarunaite and Abdurakhmonov2016b).
Cereal rye dry biomass at EVS and WGS, 2011–2013.
a Abbreviations: EVS, E.V. Smith Research and Extension Center; WGS, Wiregrass Research and Extension Center.
b Least-squares means followed by the same letter are not significantly different; comparisons are within column.
Weed Control
ANOVA did not reveal a significant main effect of strip-tillage width on weed control. ANOVA revealed significant PRE application placement and residue amount main effects on weed control for EVS 2013, and WGS 2011 and 2012, with a year-by-location interaction for weeds present at both locations; therefore, weed control results for each site-year are discussed separately. At EVS in 2013, the PRE broadcast application increased row-middle large crabgrass, Palmer amaranth, and tall morning-glory control by 28%, 39%, and 28%, respectively, compared with the PRE banded application (Table 3). In addition, high-residue biomass increased weed control in row middles at least 20% for large crabgrass, Palmer amaranth, sicklepod, and tall morning-glory (Table 4). At WGS 2011, carpetweed, common purslane, large crabgrass, Palmer amaranth, tall morning-glory, and yellow nutsedge control in the row middle was 48%, 55%, 40%, 36%, 38%, and 62% greater, respectively, when PRE herbicides were broadcast, compared with the PRE banded treatment (Table 3). For row-middle weed control of yellow nutsedge at WGS, the PRE herbicide activity appeared to be the leading factor in control, because high residues in row middles appeared to decrease Cyperus control from almost adequate to very inadequate (Table 4), possibly due to PRE application interception or lack of weed suppression by the cereal rye cover crop (Banks and Robinson Reference Banks and Robinson1982; Potter et al. Reference Potter, Truman, Strickland, Bosch and Webster2008). At WGS 2011, when looking at residue-level effects on row-middle weed control, high-residue biomass resulted in increased carpetweed, common purslane, large crabgrass, Palmer amaranth, and tall morning-glory weed control by at least 19%; however, yellow nutsedge control was again slightly decreased (Table 4). At Wiregrass in 2012, similar weed control trends were observed, compared with 2011 for in-row and row middle for large crabgrass and carpetweed (Table 3). However, greater than 90% control of carpetweed, Palmer amaranth, tall morning-glory, and yellow nutsedge was attained only in PRE broadcast plots (Table 3). Large crabgrass control increased with increased residue at WGS in 2011 and 2012; all other weed control comparisons were not different (Table 4).
Weed control after either PRE banded or broadcasted herbicide systems, EVS and WGS, 2011–2013.
aAbbreviations: EVS, E.V. Smith Research and Extension Center; WGS, Wiregrass Research and Extension Center.
bLeast-squares means with the same letter are not significantly different; comparisons are within row between banded and broadcast herbicides for the same weed species within the respective management programs.
Weed control in a high- or low-residue cereal rye cover crop system, EVS and WGS, 2011–2013.
a Abbreviations: EVS, E.V. Smith Research and Extension Center; WGS, Wiregrass Research and Extension Center.
b Least-squares means with the same letter are not significantly different; comparisons are within row between high- and low-cover crop residue for the same weed species within the respective management programs.
PRE banded treatments may provide adequate control in low-weed seedbank situations; however, conservation tillage cotton would likely benefit from broadcast soil-applied herbicide applications regardless of residue amount and tillage width when infested with Palmer amaranth and other troublesome weed species. These results agree with the results reported by Hand et al. (Reference Hand, Nichols, Webster and Culpepper2019). Whereas use of only banded herbicides may perform adequately under low or possibly moderate weed infestations, augmentation of the soil weed-seed bank from seed deposited by escapes could adversely affect weed management in future years.
Cotton Population
ANOVA revealed significant cover-crop biomass treatment effects on the cotton population with a year-by-location interaction; therefore, cotton population results are discussed separately by year and location (Table 5). At EVS in 2011, the low-biomass treatment resulted in 81,724 plants ha−1, more than the 68,967 plants ha−1 population attained in high-biomass plots. In 2012 and 2013, biomass amount did not influence the cotton population at EVS. At WGS, similar to EVS in 2011, the low-biomass treatment in 2011 and 2012 resulted in 74,349 and 58,403 plants ha−1, respectively. These populations were higher than the populations attained in high-biomass plots in 2011 (56,410 plants ha−1) and 2012 (47,839 plants ha−1), respectively. In 2013, biomass amount did not influence the cotton population at WGS. In a study evaluating row spacing, tillage system, and herbicide technology on cotton populations, the interaction of tillage systems and site-years were attributed to environmental conditions when results differed between years (Balkcom et al. Reference Balkcom, Price, van Santen, Delaney, Boykin, Arriaga, Bergtold, Kornecki and Raper2010). In addition, plant population variation between conservation systems may be due to such factors as irregular soil disturbance by the subsoiler that the planter cannot operate successfully in, soil–seed contact issues due to “hair pinning” of residue into the subsoiler slot, precipitation events immediately after planting causing downward seed movement into the subsoiler slot, among numerous other factors (Kornecki et al Reference Kornecki, Raper, Arriaga, Schwab and Bergtold2009).
Cover-crop residue, tillage width, and PRE herbicide application effects on cotton population, EVS and WGS, 2011–2013.
aAbbreviations: EVS, E.V. Smith Research and Extension Center; WGS, Wiregrass Research and Extension Center.
bLeast-squares means with the same letter are not significantly different; comparisons are within each column at each location.
Cotton Yield and Economics
ANOVA revealed significant effects of location and year on seed cotton yields, thus results are presented by location and year, and significant effects are discussed. At EVS in 2011, planting into high biomass resulted in cotton yielding 2,356 kg ha−1, whereas low residue resulted in a cotton yield of 2,575 kg ha−1 (Table 6). However, results reversed with the higher biomass, resulting in increased yield in 2012. Also at EVS in 2012, planting into a narrow-tillage width resulted in 187 kg ha−1 increase in yield. No cotton-yield differences were observed at either location in 2013. Drought occurred at both locations in 2011 and 2012, and yield reflects rainfall deficiency (Figure 3). At WGS, yield was only affected by residue level in 2011 similar to EVS, with low residue resulting in higher yield. Residue amount and tillage width did not affect cotton yields at WGS in 2012 or 2013. Yield was positively influenced by the use of broadcast PRE herbicide at WGS at 2011 and 2012, but not EVS. This result is likely due to the higher population of glyphosate-resistant Palmer amaranth observed at WGS. Thus, non-PRE treatments resulted in less than 80% weed control and lower yield compared with broadcast PRE applications.
Cover crop residue, tillage width, and PRE herbicide application effects on seed cotton yield, EVS and WGS, 2011–2013.
aAbbreviations: EVS, E.V. Smith Research and Extension Center; WGS, Wiregrass Research and Extension Center.
bLeast-squares means with the same letter are not significantly different; comparisons are within each column at each location.
Alabama drought monitor maps generated by the National Oceanic and Atmospheric Administration and University of Nebraska-Lincoln (NOAA-UNL 2020). E.V. Smith, E.V. Smith Research and Extension Center; Wiregrass, Wiregrass Research and Extension Center.
ANOVA revealed significant location-by-year effects on economic returns, thus results are presented by location and year. F tests for yield were also significant for revenue because variable costs among treatments were small compared to yield differences attained under different treatments; thus, revenue differences are the same as yield differences. Revenue ranged from a low of $1,291 ha−1 after highresidue at EVS in 2011 to a high of $3,271 ha−1 after low residue at WGS in 2011 (Table 7). At EVS in 2012, revenue from the high-residue treatment exceeded that of low residue by $141 ha−1, and revenue from narrow tillage treatment exceeded wide tillage by $102 ha−1, again reflecting yield results. And last, PRE herbicide placement influenced revenue, with higher revenue obtained after broadcast PRE at WGS in 2011 ($339 ha−1) and 2012 ($625 ha−1) compared with revenue after banded applications.
Cover crop residue, tillage width, and PRE herbicide application effects on cotton lint revenue, EVS and WGS, 2011–2013.
a U.S. cotton lint price was $1.37 kg−1 as of July 2011 (Index Mundi 2020).
b Least-squares means with the same letter within a column are not significantly different for each location.
c Abbreviations: EVS, E.V. Smith Research and Extension Center; WGS, Wiregrass Research and Extension Center.
In most comparisons, tillage width, residue amount, and PRE herbicide placement did not influence within-row weed control; however, broadcast PRE resulted in increased large crabgrass, Palmer amaranth, and tall morning-glory weed control in row middles compared with plots receiving banded PRE. In addition, high-residue treatment increased large crabgrass, carpetweed, Palmer amaranth, tall morning-glory, sicklepod, and common purslane weed control between rows. Narrow tillage and high levels of cover biomass increased cotton yield and revenue in one of six comparisons, likely due to moisture conservation, whereas banded PRE herbicides resulted in decreased yields and revenue compared with broadcast PRE in two of six comparisons. Our results show that conservation tillage cotton would likely benefit from broadcast soil-applied herbicide applications regardless of residue amount and tillage width when infested with Palmer amaranth and other troublesome weed species. Ball (Reference Ball1992) also hypothesized that conservation tillage systems that leave seed near the soil surface, coupled with effective crop rotation and herbicide regimens, could accelerate depletion of troublesome species from the seedbank. Thus, our efforts to minimize seedbank additions through an integrated approach in conservation tillage systems may help preserve conservation hectarage in the future. Integrated systems that include high-residue cover crops, minimal soil and residue disturbance, and broadcast PRE herbicides that increase weed control to accelerate seedbank depletion are needed to combat troublesome and herbicide-resistant weeds.
Acknowledgements
to the authors thank the station personnel at the E.V. Smith and Wiregrass Research and Extension Centers for their technical assistance in this study. Mention of a trade name, proprietary product, or specific equipment does not constitute a guarantee or warranty by the U.S. Department of Agriculture (USDA) and does not imply endorsement of a product to the exclusion of others that may be suitable. No conflicts of interest have been declared. This work was partially supported by the Alabama Famer’s Federation Cotton Commission, Cotton Incorporated, and USDA agreement no. 50-6048-5-020.
