Introduction
The acquisition of detailed knowledge of weed response to various management approaches will enhance our understanding of the multifunctional interactions that influence weed distribution and will improve our efforts for effective management (Korres et al. Reference Korres, Burgos and Duke2019). The cosmopolitan nature of Palmer amaranth, barnyardgrass, broadleaf signalgrass, johnsongrass, large crabgrass, morningglory, and prickly sida, along with yield reductions caused by their competition, are well documented (EPPO 2018; Korres et al. Reference Korres, Norsworthy, Brye, Vaughn and Mauromoustakos2017; Van Wychen Reference Van Wychen2019, Reference Van Wychen2020; Webster and Nichols Reference Webster and Nichols2012).
Palmer amaranth is one of the most problematic broadleaf weeds in North America, especially in the midsouthern United States, owing to its rapid growth rate, prolific seed production, adaptability, and propensity to evolve resistance to several herbicide sites of action (SOAs) (Heap Reference Heap2021; Korres and Norsworthy Reference Korres and Norsworthy2017; Schwartz-Lazaro et al. Reference Schwartz-Lazaro, Norsworthy, Steckel, Stephenson, Bish, Bradley and Bond2018; Van Wychen Reference Van Wychen2016). Currently Palmer amaranth has developed resistance to eight different SOAs, including acetolactate synthase inhibitors (Weed Science Society of America SOA Group 2), microtubule inhibitors (Group 3), synthetic auxins (Group 4), photosystem II inhibitors (Group 5), 5-enolpyruvylshikimate-3-phosphate synthase inhibitors (Group 9), protoporphyrinogen oxidase inhibitors (Group 14), long-chain fatty-acid elongase inhibitors (Group 15), and 4-hydroxyphenylpyruvate dioxygenase (HPPD) inhibitors (Group 27) (Heap Reference Heap2021; Jhala et al. Reference Jhala, Sandell, Rana, Kruger and Knezevic2014). Therefore there is a need to evaluate herbicide options to manage Palmer amaranth and other problematic Amaranthus species (Tranel et al. Reference Tranel, Riggins, Bell and Hager2011).
Non-new herbicide SOAs have been commercialized in recent years (Duke Reference Duke2012). HPPD-inhibiting herbicides are one of the most recently developed commercialized SOAs developed in the 1980s (Lee et al. Reference Lee, Prisbylla, Cromartie, Dagarin, Howard, Provan and Mutter1997; Van Almsick Reference Van Almsick2009). HPPD catalyzes the oxidative decarboxylation of 4-hydroxyphenylpyruvate to form homogentisate (Dreesen et al. Reference Dreesen, Capt, Oberdoerfer, Coats and Pallett2018). HPPD is a key enzyme for the catabolism of tyrosine, hence the formation of fumarate and acetoacetate (Dreesen et al. Reference Dreesen, Capt, Oberdoerfer, Coats and Pallett2018). It also holds an important role for the biosynthesis of plastoquinone, an essential cofactor for phytoene desaturase, a key enzyme in the biosynthesis of carotenoids (Norris et al. Reference Norris, Barrette and Della Penna1995) and an essential component of the photosynthetic electron transport chain in photosystem II. Prevention of carotenoid synthesis has direct effects on chloroplast development and photosynthesis (Pallett et al. Reference Pallett, Little, Sheekey and Veerasakaran1998). These herbicides are also known as bleachers because they prevent pigment biosynthesis and impair chloroplast development, leaving the tissue white on susceptible plants (Dreesen et al. Reference Dreesen, Capt, Oberdoerfer, Coats and Pallett2018; Grossmann and Ehrhardt Reference Grossmann and Ehrhardt2007). HPPD inhibitors like mesotrione and tembotrione are the most commonly used herbicides for weed control in corn owing to their broad-spectrum weed control; mixing compatibility with other herbicides, such as atrazine; preemergence and postemergence activity; and crop safety (Bollman et al. Reference Bollman, Boerboom, Becker and Fritz2008; Fleming et al. Reference Fleming, Banks and Legg1988; Gitsopoulos et al. Reference Gitsopoulos, Melidis and Evgenidis2010; Mitchell et al. Reference Mitchell, Bartlett, Fraser, Hawkes, Holt, Townson and Wichert2001; Nurse et al. Reference Nurse, Hamill, Swanton, Tardif and Sikkema2010; Swanton et al. Reference Swanton, Gulden and Chandler2007). Globally, only two species, Palmer amaranth and waterhemp (Amaranthus tuberculatus [Moq.] Sauer), have known resistance to this SOA, and both are found in the United States (McMullan and Green Reference McMullan and Green2011).
The overreliance on a single herbicide SOA, reduced herbicide rates, and lack of crop rotation can lead to added selection for herbicide resistance evolution (Norsworthy et al. Reference Norsworthy, Ward, Shaw, Llewellyn, Nichols, Webster, Bradley, Frisvold, Powles and Burgos2012). Effective control of Palmer amaranth in corn, for example, requires the application of preemergence followed by postemergence herbicides with distinct SOAs, herbicide rotation, and crop rotation (Chahal et al. Reference Chahal, Varanasi, Jugulam and Jhala2017; Chahal et al. Reference Chahal, Ganie and Jhala2018a).
Crop rotation is a weed management approach that can facilitate the control of problematic weeds by exploiting the morphophysiological differences and production practices of the crop grown and by allowing rotation of the herbicides used in the cropping sequence (Harper Reference Harper1956), resulting in reduced selection pressure for herbicide resistance through stress modifications and mortality factors, which directly affect weed population dynamics (Weisberger et al. Reference Weisberger, Nichols and Liebman2019). Problem weeds in one crop are often effectively controlled in other crops (Johnson and Coble Reference Johnson and Coble1986). Soil and crop management, such as crop rotation, tillage systems, application of herbicides, and other agricultural practices, affect weed seedbanks and weed emergence (Barberi et al. Reference Barberi, Cozzani, Macchia and Bonari1998; Buhler et al. Reference Buhler, Kohler and Thompson2001; Marshall et al. Reference Marshall, Brown, Boatman, Lutman, Squire and Ward2003; Sjursen Reference Sjursen2001). In addition, two or more years of crop rotation where seed production of the target weed is prevented greatly reduced weed populations (Schwartz-Lazaro and Copes Reference Schwartz-Lazaro and Copes2019). Nevertheless, crop rotations have systematically allowed the removal of easily controlled weeds while allowing other species to become well established (Johnson and Coble Reference Johnson and Coble1986), hence the importance of population dynamics in understanding the evolution of herbicide resistance, and the ability to accurately predict how management strategies would impact resistance is paramount.
Knowledge of weed biology and ecology related to weed biomass production, weed seedbank, and weed population dynamics, such as the information provided in this article, can be used to develop and potentially predict how management strategies may affect the population (Korres et al. Reference Korres, Burgos and Duke2019). Therefore the objective of this study was to understand the population dynamics of the naturally occurring weed flora, dominated by Palmer amaranth, under various combinations of HPPD-based herbicide treatments and crop rotation sequences.
Materials and Methods
Experimental Site
A 3-yr trial was conducted at the Northeast Research and Extension Center in Keiser, AR (35.66927°N, 90.08105°W), from 2015 to 2017. The experiment had a randomized complete block design with three replications, with some of the treatments being nested, whereas repeated measures were conducted for some of the variables mentioned in the following paragraphs. Experiments were established on a Sharkey clay (very-fine, smectite, thermic Chromic Epiaquerts) with a pH of 6.7 and 1.7% organic matter. Plot size was 7.7 × 183 m in all years on 96.5-cm-wide rows. The field, prior to the experiment, was under soybean production for 2 yr.
Crop Management
In 2015, corn was planted into all plots with varying herbicide programs, with the subsequent years including soybean and cotton rotation for selected herbicide programs (Table 1). Each treatment consisted of varying herbicide programs (Table 2) and crop rotation sequences: two treatments with continuous corn monoculture; two treatments with corn and soybean rotation, the latter shifting between year 2 and year 3; one corn and cotton occurring in the second year; and one with corn, soybean, and cotton, in that sequence, for each of the years in the 3-yr rotation sequence. Each crop was irrigated and fertilized as needed to optimize yield potential. Corn, soybean, and cotton crop planting dates, seeding rates, and harvest dates were all within recommended time frames for Arkansas (Table 3). The herbicide programs were designed to represent a range of efficacies and selection pressures for resistance, with the weakest being continuous corn with reduced rates of HPPD-inhibiting herbicides (Program 6).
a An explanation of the herbicide programs for each crop can be found in Table 2.
b Abbreviation: HPPD, 4-hydroxyphenylpyruvate dioxygenase.
Table 2. Herbicide treatment programs for corn, soybean, and cotton from 2015 to 2017.a
a Abbreviations: /, herbicide premix; +, herbicide tank mix; fb, followed by; COC, crop oil concentrate (1% v/v; Agridex®, Helena Chemical Co., Collierville, TN, USA); NIS, nonionic surfactant (0.25% v/v; Induce®, Helena Chemical Co.).
b Corn early postemergence (EPOST) treatments were 15 d after planting (DAP), and all mid-postemergence (MPOST) treatments were 30 DAP. Soybean EPOST treatments were applied at the V2 growth stage and MPOST at the V4/V5 growth stage. Cotton EPOST was applied at two- to three-leaf cotton, and MPOST was applied at six- to seven-leaf cotton.
Table 3. Planting date, seeding rate, application dates, and harvest dates for all crops from 2015 to 2017.a, b
a Layby and directed applications were not applicable (NA) for corn and soybean crops.
b Abbreviations: NA, not applicable; PRE, preemergence; EPOST, early postemergence; MPOST, mid-postemergence.
c Applied and incorporated with or applied after the last cultivation of a crop.
d Precise application to a specific area or plant organ, such as to a row or bed or to the leaves or stems of plants.
e The MPOST application for corn in 2016 was not applied due to an accumulated 12 cm of rain over a 10-day period.
Data Collection
In fall 2015, 30 soil cores, 10.8 cm wide and 15.2 cm deep, were collected in a W pattern across each plot to determine a baseline of the soil seedbank. Soil cores were also taken in fall 2016, 2017, and 2018. Soil samples were stored at −20 C for at least 6 wk before each sample was thawed, deaggregated, stirred, and distributed over commercial potting mix in a 28 × 55 cm plastic tray. The trays containing the soil samples were placed on a greenhouse bench and irrigated daily to field capacity. Seedlings were identified and counted over 4 wk, after which soils were then allowed to dry and the irrigation and germination cycle repeated. After the second germination cycle, soils were stored at 5 C or frozen prior to a third germination cycle to aid in breaking the dormancy cycle (Korres Reference Korres2005).
Subsequently, each spring, one permanent covered and one permanent uncovered 1-m2 subplot were placed 1 m apart into each plot at approximately 30, 91, and 152 m between crop rows 2 and 3 and 6 and 7 for a total of six subplots per plot (two at each distance down the row). At each application, the covered subplots were protected with a 2-m2 tarp to act as a control for that specific herbicide program. Once the application was made, the tarp was removed. Weeds in each subplot were counted by species every 2 wk and removed for weed density determination from planting until harvest. Prior to harvest, total weed biomass from additional predetermined 1-m2 areas was collected.
Statistical Analysis
A nested or hierarchical analysis of variance (ANOVA) was used for the analysis of emerged weeds, soil seedbank assessments, weed biomass, and crop yields. More particularly, herbicide programs were nested in the corresponding crop rotation sequences throughout the entire experimental period for each weed species separately that occurred for >2% of the weed population throughout the experimental period. Prior to data analysis, grassweed species (i.e., barnyardgrass, broadleaf signalgrass, johnsongrass, and large crabgrass) were pooled together to overcome the variability between species but also the absence of johnsongrass in 2015 and barnyardgrass in 2016 emergence counts. Similarly, emergence counts for morningglory species (i.e., pitted [Ipomoea lacunosa L.] and ivyleaf [Ipomoea hederacea Jacq.] morningglory) were also pooled together. Comparisons of weed emergence were made between weed counts in the covered and uncovered plots, which were crossed with herbicide program and rotation sequence. This approach enabled the conjoint analysis of herbicide program with its corresponding crop rotation sequence (henceforth treatment). Therefore evaluation of the dependent variable (e.g., weed density, total weed biomass, and soil seedbank) variability between the experimental treatments in time was possible. Nested ANOVA was used on both log-transformed (when assumptions of normality were not satisfied) and untransformed counts of weed emergence and soil seedbank. Results presented here are based on back-transformed values to facilitate a better understanding of the research outcome. JMP 16.0 Pro software (SAS Institute Inc., Cary, NC, USA) was used for data analysis.
Results and Discussion
Weed Occurrence
Fourteen weed species were recorded throughout the experimental period, but only those with occurrence > 2% of the total weed population averaged across treatments (crop rotation × herbicide program) and years of experimentation were analyzed and are presented here (Table 4). The dominant weed species, in descending order, were Palmer amaranth (40%), prickly sida (15%), and pitted morningglory (12%). Additional weed occurrence included broadleaf signalgrass (8%), hophornbeam copperleaf (Acalypha ostryifolia Riddell) (7%), large crabgrass (6%), ivyleaf morningglory (4%), barnyardgrass (3%), and johnsongrass (2%) (Table 4). Barnyardgrass was recorded in 2015 and 2017 only, whereas johnsongrass was recorded in 2016 and 2017 only.
Table 4. Weed species recorded as percentage of the total weed population averaged across treatments and years of experimentation.
Treatment Effects on the Emergence of Naturally Occurring Weeds
All herbicide programs associated with Treatments 1, 2, 3, and 4 were found relatively effective in controlling the emergence of naturally occurring weed species, whereas, in most cases, herbicide programs associated with Treatments 5 and 6 were less effective. HPPD-inhibiting herbicide-based programs applied as preemergence and mid-postemergence (i.e., mesotrione fb tembotrione or as premix, i.e., mesotrione with S-matolachlor and atrazine) and/or herbicides recommended for soybean or cotton in the subsequent years were effective on grassweed species. In general, treatments involving herbicides with six SOAs, such as Treatments 1 and 4 (i.e., inhibition of hydroxyphenyl pyruvate dioxygenase, inhibition of very-long-chain fatty-acid synthesis, inhibition of photosynthesis at photosystem II [PSll], inhibition of acetolactate synthase, inhibition of glutamine synthetase, inhibition of protoporphyrinogen oxidase), or herbicides with seven SOAs, such as Treatments 2 and 3 (i.e., inhibition of very-long-chain fatty-acid synthesis, inhibition of photosynthesis at PSll, inhibition of hydroxyphenyl pyruvate dioxygenase, inhibition of acetolactate synthase, inhibition of glutamine synthetase, inhibition of protoporphyrinogen oxidase, and inhibition of enolpyruvyl shikimate phosphate synthase), reduced weed emergence. On the contrary, treatments involving herbicides with fewer SOAs, such as Treatment 5 (five SOAs) and Treatment 6 (one SOA), were less effective. Only the top three emerged weeds are discussed in what follows.
Palmer Amaranth
Selected herbicide programs associated with corresponding crop rotation sequences across years affected the emergence (i.e., weed counts in uncovered vs. covered plots) of Palmer amaranth (P < 0.0008). Herbicides used in corn–soybean–cotton rotation (i.e., Treatment 2, involving herbicides with seven SOAs) or corn–soybean–corn rotation (i.e., Treatment 4, involving herbicides with six SOAs) (Table 2) reduced the emergence of Palmer amaranth by 89.8% and 88.5%, respectively. The herbicide program for Treatment 2 consisted of preemergence, early, and mid-postemergence herbicide applications fb directed and layby applications (in cotton), and that for Treatment 4 consisted of preemergence, early, and mid-postemergence applications.
The herbicide program associated with continuous corn, a combination of preemergence and mid-postemergence applications at the recommended dose (i.e., Treatment 5, which involved herbicides with five SOAs, namely, inhibition of very-long-chain fatty-acid synthesis, inhibition of photosynthesis at PSll, inhibition of hydroxyphenyl pyruvate dioxygenase, inhibition of acetolactate synthase, and inhibition of glutamine synthetase), reduced the emergence of Palmer amaranth by 76.2% between counts in uncovered compared to covered plots (Figure 1). Similarly, the herbicide program associated with corn–corn–soybean rotation, a combination of preemergence, early, and mid-postemergence herbicide applications (i.e., Treatment 1, which involved herbicides with six SOAs), reduced Palmer amaranth emergence between uncovered and covered plots by 70.1% (Figure 1). Similar results were reported by Oliveira et al. (Reference Oliveira, Jhala, Gaines, Irmak, Amundsen, Scott and Knezevic2017) when preemergence fb postemergence herbicide programs including glyphosate and glufosinate as pre- or tank mix provided 60% to 90% control of waterhemp under conventional corn. In addition, the preemergence application of mesotrione + S-metolachlor + atrazine provided 83% Palmer amaranth control in conventional corn (Chahal et al. Reference Chahal, Irmak, Gaines, Amundsen, Jugulam, Jha, Travlos and Jhala2018b). Herbicide programs associated with corn–cotton–corn rotation (i.e., Treatment 3, involving herbicides with seven SOAs) or those in continuous corn, that is, at half of the recommended dose of mesotrione fb tembotrione (i.e., Treatment 6, involving herbicides with one SOA, i.e., inhibition of hydroxyphenyl pyruvate dioxygenase), had a lesser effect on Palmer amaranth emergence (54.5% and 31.2%, respectively) (Figure 1).
Figure 1. Effects of herbicide program (i.e., herbicide program × crop rotation) and covered plots (i.e., covered vs. uncovered plot) on Palmer amaranth emergence throughout the experimental period. Vertical bars represent the standard error of the mean at significance level a = 0.1. Treatments are as follows: 1, corn [PRE+MPOST] followed by (fb) corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST]; 2, corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST] fb cotton [PRE+EPOST+MPOST+Directed+Layby]; 3, corn [PRE+MPOST] fb cotton [PRE+EPOST+MPOST+Directed+Layby] fb corn [PRE+EPOST]; 4, corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST] fb corn [PRE+MPOST]; 5, corn [PRE+MPOST] fb corn [PRE+MPOST] fb corn [PRE+MPOST]; 6, corn [EPOST+MPOST] fb corn [EPOST+MPOST] fb corn [EPOST+MPOST]. See Table 1 for a detailed description of the field treatments throughout the experimental period (from 2015 to 2017).
Crop rotation leads to the diversification of individual cropping practices, thereby causing changes in weed population and species composition (Cardina et al. Reference Cardina, Herms and Doohan2002; Norsworthy et al. Reference Norsworthy, Ward, Shaw, Llewellyn, Nichols, Webster, Bradley, Frisvold, Powles and Burgos2012; Ross and Lembi Reference Ross and Lembi1985). On the basis of our results, the effective control of Palmer amaranth was achieved when corn was planted in a rotation with soybean and cotton, which enabled the use of preemergence fb early and mid-postemergence herbicide applications with distinct SOAs. The adoption of this approach, that is, crop rotation and vigorous HPPD-based herbicide programs applied preemergence and/or early to mid-postemergence, can also achieve long-term Palmer amaranth control in corn production systems. This is because the inclusion of the early and mid-postemergence herbicide applications in the weed control program enables the control of emerged weeds, including Palmer amaranth, at the time of application (Bollman et al. Reference Bollman, Boerboom, Becker and Fritz2008; Chahal et al. Reference Chahal, Varanasi, Jugulam and Jhala2017; Neve et al. Reference Neve, Norsworthy, Smith and Zelaya2011).
The long-term effects of treatment choice (herbicide program × crop rotation) were revealed when Palmer amaranth emergence counts were compared in uncovered plots only, in particular, when emergence counts in 2017 (final assessment) were compared to emergence counts in 2015 (initial assessment). Treatments 1, 2, 3, 4, and 5 (involving herbicides with six, seven, seven, six, and five SOAs, respectively), in that order, reduced Palmer amaranth emergence by 28-fold, 5.2-fold, 4.5-fold, 1.5-fold, and 1.1-fold, respectively. In contrast, Treatment 6 (i.e., weak HPPD herbicide program with one SOA in continuous corn) caused a 4-fold increase of Palmer amaranth emergence in 2017 compared to 2015 (Figure 1). Cultural practices for weed management that include crop rotations between soybean and corn accompanied with suitable herbicide rotations can be used to control weeds effectively (Liebman and Dyck Reference Liebman and Dyck1993; Riar et al. Reference Riar, Norsworthy, Steckel, Stephenson, Eubank and Scott2013). On the basis of our results, this is true for Palmer amaranth emergence counts in corn–soybean rotation systems, as 88.7% control was obtained under Treatment 4 (a 3-yr corn–soybean–corn rotation) (Figure 1).
Prickly Sida
Herbicide programs, averaged across years, in uncovered (treated plots) compared to covered (untreated controls) reduced the emergence of prickly sida (P < 0.0001) (Figure 1). The most effective herbicide program was Treatment 2, which involved herbicides with seven SOAs (59% emergence reduction) fb, in descending order, Treatment 1 (herbicides with six SOAs) (56% emergence decrease); Treatment 4 (herbicides with six SOAs) (38.3% decrease); Treatment 6 (herbicides with one SOA) (35% emergence decrease); Treatment 3 (herbicides with seven SOAs) (17.4% emergence decrease); and, finally, Treatment 5 (herbicides with five SOAs) (15% emergence decrease). Previous studies have reported high control of S. spinosa with residual herbicides applied prior to or at planting and followed by single or multiple POST herbicide applications (Beyers et al. Reference Beyers, Smeda and Johnson2002; Culpepper et al. Reference Culpepper, York, Batts and Jennings2000; Ellis and Griffin Reference Ellis and Griffin2002), as in Treatments 1, 2, and 3. In another study, Burke et al. (Reference Burke, Askew and Wilcut2002) reported high prickly sida control when PRE application of flumioxazin was included in the herbicide program, as in Treatments 1, 2, and 4.
Morningglory Species
The two morningglory species evaluated have been grouped together owing to the similarity in their trends. Herbicide programs significantly reduced the emergence of both morningglory species in uncovered plots compared to covered plots (P < 0.05) and between years (P < 0.001). Treatment 5 was the least effective on the emergence of ivyleaf morningglory and pitted morningglory in relation to other treatments. More particularly, Treatment 1 (77% emergence reduction) fb Treatment 2 (70% emergence reduction) and Treatment 4 (64% emergence reduction) were the most effective in controlling the emergence of ivyelaf morningglory. In Treatments 3 and 6, the reduction of the weed was approximately 45% (Figure 2). Likewise, Treatment 5 was the least effective (P < 0.001) on pitted morningglory emergence reduction (12% reduction), followed by Treatments 3, 2 6, 1, and 4, which reduced morningglory emergence by 16%, 21%, 33%, 39%, and 43%, respectively (Figure 2).
Figure 2. Effects of herbicide program (i.e., herbicide program × crop rotation) and covered plots (i.e., covered vs. uncovered plot) on broadleaf weeds emergence throughout the experimental period. Vertical bars represent the standard error of the mean at significance level a = 0.1, 0.1, and 5 for hophornbeam, morningglory, and prickly sida, respectively. Treatments are as follows: 1, corn [PRE+MPOST] followed by (fb) corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST]; 2, corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST] fb cotton [PRE+EPOST+MPOST+Directed+Layby]; 3, corn [PRE+MPOST] fb cotton [PRE+EPOST+MPOST+Directed+Layby] fb corn [PRE+EPOST]; 4, corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST] fb corn [PRE+MPOST]; 5, corn [PRE+MPOST] fb corn [PRE+MPOST] fb corn [PRE+MPOST]; 6, corn [EPOST+MPOST] fb corn [EPOST+MPOST] fb corn [EPOST+MPOST]. See Table 1 for detailed description of the field treatments throughout the experimental period (from 2015 to 2017).
The preemergence fb mid-postemergence application of mesotrione and tembotrione, respectively, in the first year, accompanied with premix applications of mesotrioane with atrazine and S-metolachlor and/or tank-mix applications of tembotrione + atrazine + glufosinate, as in Treatments 1, 2, and 4, along with treatments that contained flumioxazin and fluometuron in the following years as fitted to crop rotational system, improved the control of morningglory species compared to other treatments exclusively based on corn monoculture. The addition of atrazine to mesotrione applied postemergence improved Ipomoea coccinea L. control (Armel et al. Reference Armel, Rardon, McComrick and Ferry2007). Furthermore, the addition of atrazine to tembotrione resulted in a synergistic response in Palmer amaranth (Kohrt and Sprague Reference Kohrt and Sprague2017), whereas atrazine added to PRE applications of isoxaflutole also increased control of ivyelaf morningglory (Stephenson and Bond Reference Stephenson and Bond2012). Johnson et al. (Reference Johnson, Young and Matthews2002) and Breeden et al. (Reference Breeden, Rhodes and Mueller2001) reported increased control of ivyelaf morningglory and pitted morningglory by the addition of atrazine to mesotrione. Therefore the co-application of HPPD-inhibiting with PSII-inhibiting herbicides is well documented as a practice that can increase overall herbicidal activity and broaden the weed control spectrum.
Grass Species
Crop rotation allowed the use of different herbicides that improved the control of various grassweed species, such as broadleaf signalgrass, large crabgrass, and barnyardgrass (P < 0.0001) (Figure 3). Johnson and Coble (Reference Johnson and Coble1986), for example, reported that broadleaf signalgrass has been observed in areas where corn, soybean, and peanut (Arachis hypogaea L.) are grown continually or in rotation. Additionally, Johnson and Coble found that broadleaf signalgrass density was highest in plots where corn dominated rotations. Our findings show a similar trend, where Treatment 6, that is, continuous corn monoculture associated with a weak HPPD herbicide program involving one herbicide SOA, had the highest grassy weed density, including broadleaf signalgrass, large crabgrass, and barnyardgrass (Figure 3; Supplementary Figure 1). Like broadleaf signalgrass, barnyardgrass emergence in Treatment 6, averaged across experimental years, was increased by 96.8% in the uncovered compared to covered plots, whereas little differences showed in the emergence of large crabgrass (Supplementary Figure 1). Differential response to herbicides could reflect differences in selection pressure caused by years of cropping system–related herbicide usage. De Cauwer et al. (Reference De Cauwer, Rombaut, Bulcke and Reheul2011) stated that one of the possible reasons for the increase of barnyardgrass emergence is the lack of crop rotation and the lower sensitivity of the weed to postemergence herbicides acting against panicoid grasses, such as HPPD-inhibiting herbicides, as in Treatment 6. Treatment 6 (involving herbicides with one SOA) was the least effective treatment for the emergence of johnsongrass compared to the most effective treatments, Treatments 2, 3, 4, and 5 (i.e., involving seven, seven, six, and five SOAs, respectively), with 95%, 95%, 90%, and 90% johnsongrass emergence reductions, respectively (Supplementary Figure 1). Findings for Treatment 6, a poor HPPD herbicide program in continuous corn, confirm reports by Frans et al. (Reference Frans, McClelland, Horton, Corbin and Talbert1991), who stated that a high level of herbicide control is necessary to control johnsongrass in various crop rotation systems.
Figure 3. Effects of herbicide program (i.e., herbicide program × crop rotation) and covered plots (i.e., covered vs. uncovered plot) on grassweed emergence throughout the experimental period. Vertical bars represent the standard error of the mean at significance level a = 0.1. Grassweed species consist of large crabgrass, broadleaf signalgrass, barnyardgrass, and johnsongrass. Treatments are as follows: 1, corn [PRE+MPOST] followed by (fb) corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST]; 2, corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST] fb cotton [PRE+EPOST+MPOST+Directed+Layby]; 3, corn [PRE+MPOST] fb cotton [PRE+EPOST+MPOST+Directed+Layby] fb corn [PRE+EPOST]; 4, corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST] fb corn [PRE+MPOST]; 5, corn [PRE+MPOST] fb corn [PRE+MPOST] fb corn [PRE+MPOST]; 6, corn [EPOST+MPOST] fb corn [EPOST+MPOST] fb corn [EPOST+MPOST]. See Table 1 for detailed description of the field treatments throughout the experimental period (from 2015 to 2017).
Treatment Effects on Total Weed Biomass
Treatment (herbicide program × crop rotation) significantly affected the total weed biomass (P = 0.003) between 2016 and 2017 (Figure 4). Biomass production in 2016 showed no significant decreases. However, comparisons between biomass in covered and uncovered plots for each herbicide × crop rotation treatment in 2017 revealed that Treatments 2, 1, 5, 4, and 6, in that order, were the most effective treatments in reducing weed total weight (Figure 4). Total weed dry weight between covered and uncovered plots increased by 1.2-fold under Treatment 2 in 2017.
Figure 4. Effects of treatment (i.e., herbicide program × crop rotation) on total weed biomass in 2016 and 2017. Vertical bars represent the standard error of the mean at significance level a = 0.01. Treatments are as follows: 1, corn [PRE+MPOST] followed by (fb) corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST]; 2, corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST] fb cotton [PRE+EPOST+MPOST+Directed+Layby]; 3, corn [PRE+MPOST] fb cotton [PRE+EPOST+MPOST+Directed+Layby] fb corn [PRE+EPOST]; 4, corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST] fb corn [PRE+MPOST]; 5, corn [PRE+MPOST] fb corn [PRE+MPOST] fb corn [PRE+MPOST]; 6, corn [EPOST+MPOST] fb corn [EPOST+MPOST] fb corn [EPOST+MPOST]. See Table 1 for detailed description of the field treatments throughout the experimental period (from 2015 to 2017).
Johnson et al. (Reference Johnson, Young and Matthews2002) reported increased control of common cocklebur (Xanthium strumarium L.), ivyleaf morningglory, and yellow nutsedge (Cyperus esculentus L.) by the addition of atrazine to mesotrione. Mesotrione activity on common ragweed (Ambrosia artemisiifolia L.), horsenettle (Solanum carolinense L.), sicklepod (Cassia obtusifolia L.), and pitted morningglory has also been enhanced by adding atrazine.
Treatment Effects on Soil Seedbank
Weed seedbanks may indicate the effects of crop management on weed population (Buhler et al. Reference Buhler, Kohler and Thompson2001), although the relationship between soil seedbank and weed emergence is not always direct (Sjursen Reference Sjursen2001). However, Palmer amaranth seedbank under Treatments 2, 3 1, 4, and 5, in that order, was progressively reduced (P < 0.001) from 97% to 25% between 2015 and 2018 (Figure 5). These herbicide treatments, involving herbicides with seven, six, and five SOAs, were also more effective on Palmer amaranth emergence. Only under Treatment 6, Palmer amaranth seed counts increased (P < 0.001) 3-fold between 2015 and 2017 (Figure 5). Seed counts of weed species increased progressively toward the end of the experiment, particularly between 2017 and 2018. The soil seedbank of broadleaf weeds prickly sida and morningglory species increased between 2017 and 2018 in all treatments, especially under Treatments 3, 5, and 6. Seed counts of the broadleaf signalgrass and large crabgrass also increased independently of the treatment, whereas johnsongrass increased greatly under Treatment 6 (Figure 5). Increases in soil seedbank could increase the persistence of certain weed species, such as morningglory, which can remain viable for about 40 yr (Toole and Brown Reference Toole and Brown1946). The prolonged period of weed germination, for example, Palmer amaranth (Monks et al. Reference Monks, Jennings, Meyers, Smith and Korres2019), necessitates management systems that ensure long-term reductions of the soil seedbank. Wei et al. (Reference Wei, Qiang, Ma and Wei2005) reported significant decreases, between 27% and 44%, of soil seedbank owing to crop rotation of wheat, corn, and soybean. The use of crop sequences reduces the selection pressure and prevents the proliferation of some weed species that have adapted well to a single cropping system, as in the case of Palmer amaranth. The soil seedbank of Palmer amaranth and morningglory increased under Treatment 6. With a continual increase in herbicide resistance, preservation of currently effective herbicide programs is paramount.
Figure 5. Effects of treatment (i.e., herbicide program × crop rotation) on weed soil seedbank throughout the experimental period at significance level a = 0.1. Treatments are as follows: 1, corn [PRE+MPOST] followed by (fb) corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST]; 2, corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST] fb cotton [PRE+EPOST+MPOST+Directed+Layby]; 3, corn [PRE+MPOST] fb cotton [PRE+EPOST+MPOST+Directed+Layby] fb corn [PRE+EPOST]; 4, corn [PRE+MPOST] fb soybean [PRE+EPOST+MPOST] fb corn [PRE+MPOST]; 5, corn [PRE+MPOST] fb corn [PRE+MPOST] fb corn [PRE+MPOST]; 6, corn [EPOST+MPOST] fb corn [EPOST+MPOST] fb corn [EPOST+MPOST]. See Table 1 for detailed description of the field treatments throughout the experimental period (from 2015 to 2017).
Practical Implications
HPPD-based herbicide programs augmented with crop rotation, especially when soybean and cotton are included in the sequence, reduced the emergence of naturally occurring weed flora and soil seedbank, especially Palmer amaranth. However, corn monocultures and weak HPPD-based herbicide programs with only one SOA resulted in Palmer amaranth and prickly sida emergence and soil seedbank increases. The latter approach cannot sustain a profitable crop production. Therefore integrated weed management that incorporates crop rotation of corn, soybean, and cotton with selected HPPD-based herbicide programs that involve six or seven SOAs can preserve herbicide efficacy in the long term and suppress the presence of weed species in corn fields, hence delaying the evolution of herbicide resistance.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/wet.2022.40
Acknowledgments
The authors thank all the graduate students and research technicians and the Northeast Research and Extension Center farm crew for all their help with this research. No conflicts of interest have been declared. This research received no specific grant from any funding agency or the commercial or not-for-profit sectors.
Open access
