Introduction
In the United States, North Carolina ranks as the top-producing state for sweetpotato, with production primarily occurring from the coastal plain to the Sandhills region of the state (Schultheis et al. Reference Schultheis, Sorensen, Monks, Holmes and Thornton2005). In North Carolina, more than 36,000 ha of sweetpotato were harvested in 2017, with a value of $346 million (USDA-NASS 2017). Because of their uniform production and higher percentage of premium grade storage roots (no. 1), both Beauregard and Covington are important commercial clones grown in the state (Schultheis et al. Reference Schultheis, Sorensen, Monks, Holmes and Thornton2005; Smith et al. Reference Smith, Stoddard, Shankle, Schultheis, Loebenstein and Thottappilly2009; Yencho et al. Reference Yencho, Pecota, Schultheis, Zvezdana-Pesic, Holmes, Little, Thornton and Truong2008).
With sweetpotato production representing a large portion of cropland in North Carolina, the need for effective weed management practices is a concern for many producers (A. Thornton, personal communication). Several weed management methods including preplant (PP), PRE, and POST herbicides; tillage; row-middle cultivation; and hand removal of weeds are used to provide acceptable weed control in sweetpotato (Monks et al. Reference Monks, Jennings, Meyers, Smith, Korres, Korres, Burgos and Duke2019; Schultheis et al. Reference Schultheis, Sorensen, Monks, Holmes and Thornton2005). However, the limited number of herbicides registered in sweetpotato contributes to the difficulty in managing the diversity of weeds (Kemble Reference Kemble2019). Compounding the issue is repeated removal, via regulatory processes, of herbicides that have previously been registered in the United States and Canada as well as increased restrictions for legal approval of new products (Fennimore and Doohan Reference Fennimore and Doohan2008). For example, halosulfuron was registered for use in sweetpotato in 2010 by Gowan Company (Anonymous 2010) and S-ethyl dipropylthiocarbamate received an Environmental Protection Agency (EPA) Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) Section 3 registration for sweetpotato in 1999 (US EPA 1999). However, neither of these herbicides is registered for use in sweetpotato. Fluridone was registered in sweetpotato in North Carolina by an EPA FIFRA Section 18 label in 2017; however, the label expired in July 2017 and has not been renewed.
Sweetpotato and many other traditionally grown agriculture row crops rely on herbicides with similar modes of actions (e.g., protoporophyrinogen oxidase [PPO] inhibitors or acetyl CoA carboxylase inhibitors), which can encourage evolution of resistant weed species. Globally, more than 250 reported weed species show resistance to more than 160 different herbicides (Heap Reference Heap2019). PPO-resistant weeds, including Palmer amaranth have been confirmed in North Carolina and PPO-resistant goosegrass [Eleusine indica (L.) Gaertn.], common ragweed (Ambrosia artemisiifolia L.), tall waterhemp [Amaranthus tuberculatus (var. rudis)], and Palmer amaranth have been reported throughout the United States (Heap Reference Heap2019), leading to the need for additional herbicides with alternative modes of action for control of these resistant weeds.
Palmer amaranth in sweetpotato negatively affects sweetpotato growth and reduces marketable yield up to 81% (Meyers et al. Reference Meyers, Jennings, Schultheis and Monks2010b). The current recommendation to control Palmer amaranth is 107 g ha−1 flumioxazin (a PPO inhibitor) PP followed by (fb) 800 g ha−1S-metolachlor post-transplant (PT) 7 to 14 d after transplanting (DAT) (Meyers et al. Reference Meyers, Jennings, Schultheis and Monks2010a). This herbicide program provides initial control and, at times, season-long residual weed management with greater than 90% control of troublesome weeds, including Palmer amaranth (Coleman et al. Reference Coleman, Chaudhari, Jennings, Schultheis, Meyers and Monks2016; Meyers et al. Reference Meyers, Jennings, Schultheis and Monks2010a). Approximately 98% sweetpotato ha in NC is treated with flumioxazin. With the extensive use of flumioxazin in sweetpotato, a need exists for registration of herbicides with alternative modes of action for the weed management programs to remain successful. Several herbicides have been evaluated for their potential use in sweetpotato production systems (Barkley et al. Reference Barkley, Chaudhari, Jennings, Schultheis, Meyers and Monks2016; Beam et al. Reference Beam, Jennings, Chaudhari, Monks, Schultheis and Waldschmidt2018a, Reference Beam, Chaudhari, Jennings, Monks, Meyers, Schultheis, Waldschmidt and Main2018b; Chaudhari et al. Reference Chaudhari, Jennings and Meyers2018; Smith et al. Reference Smith, Jennings, Monks, Schultheis and Reberg-Horton2019). One potential alternative herbicide is bicyclopyrone, a member of the p-hydroxyphenylpyruvate dioxygenase–inhibiting herbicide family (Dunne Reference Dunne2012).
Bicyclopyrone is a Weed Science Society of America Group 27 selective herbicide that has PRE and POST activity on broadleaf weeds and annual grasses (Anonymous 2018). Currently, it is sold as part of a premix of atrazine, mesotrione, and S-metolachlor (tradename, Acuron®; Syngenta Crop Protection, Greensboro, NC) for use in field corn, seed corn, silage corn, sweet corn, and yellow popcorn. Bicyclopyrone has the potential to provide season-long control because of its root absorption when applied PRE and its foliar uptake when applied POST (APVMA 2017).
Bicyclopyrone has been evaluated for weed efficacy as well as vegetable crop tolerance (Accinelli et al. Reference Accinelli, Mencarelli, Balogh, Ulmer and Screpanti2015; Chen et al. Reference Chen, Hu and Doohan2018; Felix and Ishida Reference Felix and Ishida2015; Gage Reference Gage2016). Carrot [Daucus carota subsp. sativus (Hoffm.) Arcang.], dill (Anethum graveolens L.), onion (Allium cepa L.), and radish (Raphanus sativus L.) were tolerant to PRE and POST bicyclopyrone in soil with high organic matter content (muck), but application of bicyclopyrone POST over the crop caused severe crop injury (Chen et al. Reference Chen, Hu and Doohan2018). Felix and Ishida (Reference Felix and Ishida2013, Reference Felix and Ishida2015) reported variable level of tolerance and yield response of direct-seeded onion to PRE and POST (at 2-leaf stage) bicyclopyrone. Bicyclopyrone PRE or POST application controls a wide spectrum of weed species including hairy galinsoga (Galinsoga quadriradiata Cav.), common purslane (Portulaca oleracea L.), common lambsquarters (Chenopodium album L.), redroot pigweed (Amaranthus retroflexus L.), hairy nightshade (Solanum physalifolium Rusby), kochia [Bassia scoparia (L.) A.J. Scott], and spotted ladysthumb (Polygonum persicaria L.) (Chen et al. Reference Chen, Hu and Doohan2018; Felix and Ishida Reference Felix and Ishida2013, Reference Felix and Ishida2015; Peachey Reference Peachey2015).
With the lack of herbicides registered in sweetpotato and the need for alternative herbicides in sweetpotato to better manage herbicide resistance, studies were conducted to determine sweetpotato tolerance to bicyclopyrone PP, PT, and PT-directed, and its efficacy in controlling Palmer amaranth.
Materials and Methods
Greenhouse Studies
Studies were conducted at the Syngenta Crop Protection greenhouse (36.07°N, 79.91°W) in Greensboro, NC, in 2018. Covington and NC04-531 sweetpotato nonrooted cuttings (10 cm with one to three leaves) were planted in 20-cm wide by 20-cm deep polyethylene pots (Nursery Supplies Inc., Chambersburg, PA) containing nontreated Orangeburg sandy loam soil (fine-loamy, kaolinitic, thermic, typic Kandiudults) with pH of 6.2, CEC of 2.8 mEq 100 g−1 and 0.6% organic matter from the Horticultural Crops Research Station, Clinton, NC. Sweetpotato plants were grown during summer months in a greenhouse with natural sunlight and temperatures between 20 and 30 C.
A total of 20 treatments, consisting of combinations of two application timings (PP and PT), five rates of bicyclopyrone (0, 25, 50, 100, and 150 g ha−1), and two sweetpotato clones (Covington and NC04-531) were arranged in a randomized complete block design with four replications. The study was conducted twice, with two runs separated temporally. Bicyclopyrone treatments (PP and PT) were applied on June 18 and July 6, 2018, for run 1 and run 2, respectively. Treatments were applied using a CO2- pressurized spray chamber with a single XR 11002VS nozzle tip (TeeJet Technologies, Springfield, IL) at 280 kPa and calibrated to deliver 187 L ha−1. PP treatments were applied to the soil surface and sweetpotato plants were planted immediately after application. After the application, pots were watered lightly to incorporate the herbicides but not so much as to allow leaching through the pots. PT treatments were applied on the same day after sweetpotato transplanting. Sweetpotato plants were not watered for 72 h after PT applications to avoid potentially washing herbicide from leaves; thereafter, pots were watered as needed.
Data recorded included percent visual injury (i.e., chlorosis, necrosis, and leaf deformation) and stunting on a 0% (no injury) to 100% (crop death) scale determined weekly for 8 or 9 wk. Sweetpotato vine length was measured weekly from the soil surface to the node of the last fully expanded true leaf. Destructive harvest of sweetpotato plants was conducted 64 ± 2 DAT by cutting shoots at the soil surface; shoots were placed in paper bags and air dried at 25 to 45 C for 30 d before determining shoot dry weight.
Sweetpotato shoot dry weight, vine length, injury, and stunting data were subjected to ANOVA using PROC MIXED (SAS, version 9.4; SAS Institute, Inc. Cary, NC) with a separate analysis performed for each rating date. Clone, application timing, application rate, and their interactions were considered fixed effects and replication, replication within run, and run by treatment interactions were considered random effects. All data were checked for variance homogeneity and normality before statistical analysis by plotting residuals. The arcsine transformation was used to improve normality and homogeneity of error variances for stunting, shoot dry weight, and injury data. When ANOVA indicated a significant rate effect, data were subjected to regression analysis against rate by SAS PROC MIXED to determine best-fit models. Least square means were used to estimate coefficients for logistic and Gompertz models via SAS PROC GLM.
The three-parameter logistic model was used to describe the relationship between application rate and vine length or shoot dry weight: Y = a / [(1 + k) × exp(− c × X)], where Y is sweetpotato vine length or shoot dry weight, a is the upper asymptote, k and c are constants, and X is bicyclopyrone application rate in g ai ha−1.
The three-parameter Gompertz equation was used to describe the relationship between application rate and plant stunting or injury: Y = a × exp [−k × exp (− c × X)], where Y is percentage of sweetpotato stunting or injury, a is the upper asymptote for injury or stunting, k and c are constants, and X represents the application rate of bicyclopyrone in g ai ha−1 (Knezevic et al. Reference Knezevic, Evans, Blankenship, Van Acker and Lindquist2002).
Field Studies
Studies were conducted in 2016, 2017, and 2018 at the Horticultural Crops Research Station in Clinton, NC (35.0232°N, 78.2804°W). Soil was a Norfolk sandy loam (fine-loamy, kaolinitic, thermic Typic Kandiudults) with pH of 6.0 and 0.8% organic matter and pH of 6.2 and 1.2% organic matter in 2016 and 2017, respectively. In 2018, soil was an Orangeburg sandy loam (fine-loamy, kaolinitic, thermic Typic Kandiudults) with pH of 6.2 and 0.8% organic matter. Field-grown slips (nonrooted cuttings) from a field propagation bed in 2016 (Beauregard), 2017 (Covington), and 2018 (Covington) were mechanically transplanted 10-cm deep and 30-cm apart using a tractor-pulled, commercial mechanical transplanter. Plots were two rows, each 1-m wide by 6.1-m long. The first row of each plot was a nontreated border row and the second row received a treatment.
The experimental design was a randomized complete block with four (2016 and 2018) and three (2017) replications. Treatments consisted of PP, PT, and PT-directed herbicide applications to the row middle between raised beds (Tables 1 and 2), and nontreated weedy and weed-free checks. PT-directed applications of bicyclopyrone in 2016 and 2018 contained a nonionic surfactant at 0.25% vol/vol. Treatments were applied with a CO2-pressurized backpack sprayer equipped with two TeeJet XR 8003VS nozzles (PP and PT) or one TeeJet Turbo TT11003 nozzle (PT-directed) and calibrated to deliver 187 L ha–1. PP, PT, and PT-directed treatments were applied one d before transplanting, 1 to 2 wk after transplanting (WAT), and 5 ± 2 WAT, respectively. PT-directed treatments were applied to the row middle between raised beds when sweetpotato plants averaged 0.36-m vine length. Sethoxydim (Poast®; BASF Corp., Research Triangle Park, NC) at 0.34 kg ai ha−1 plus 1% vol/vol crop oil (Agri-Dex; Helena Chemical Co., Collierville, TN) was applied as needed to control goosegrass and large crabgrass [Digitaria sanguinalis (L.) Scop.].
Table 1. Herbicide treatments applied to field studies in Clinton, NC in 2016, 2017, and 2018.
a Abbreviations: fb, followed by; PP, preplant; PT, post-transplant; WAT, weeks after transplanting.
b PP, PT, and PT-directed applications were made 1 d PP, 1 to 2 WAT, and 5 ± 2 WAT, respectively.
c Nonionic surfactant at 0.25% vol/vol were added with bicyclopyrone in 2016 and 2018.
Table 2. Sources of herbicide treatments used in field studies from 2016 to 2018.
a Abbreviation: N/A, not applicable.
Sweetpotato injury (i.e., chlorosis, necrosis, leaf deformation, and stunting) was estimated visually on the same visual scale as in the greenhouse studies at 2, 3, 4, and 8 WAT. Palmer amaranth control was estimated visually on a scale from 0% (no control) to 100% (complete control) at 4 and 8 WAT. Sweetpotato storage roots were harvested 113 ± 5 DAT using a tractor-mounted chain digger to place storage roots on the soil surface and then hand graded into jumbo (≥8.9-cm diam), no. 1 (≥4.4 cm but <8.9 cm), and canner (≥2.5 cm but <4.4 cm) grades (USDA 2005). Total marketable yield was calculated as the sum of no. 1, jumbo, and canner grades.
Data were subjected to ANOVA using SAS, version 9.3 (SAS Institute Inc.) PROC MIXED with the fixed effect of herbicide treatments and random effect of replication. Data were analyzed separately for each year because of the limitation of the experimental design; that is, one additional treatment was included in 2017 and one fewer treatment in each of 2016 and 2018. For the same reason, all data were presented by year, because a significant year by treatment interaction was observed when analyzed using the treatments that were repeated during all 3 yr. Data were checked for homogeneity of variance by plotting residuals. Weedy and weed-free check plots were included in the analysis for sweetpotato yield. Because if a lack of variance, however, these treatments were not included in analysis of sweetpotato injury and Palmer amaranth control. Data for sweetpotato injury and Palmer amaranth control were subjected to arcsine transformation to improve normality and homogeneity of error variances; however, back-transformed means are presented. Means were separated using the Fisher LSD test at the 0.05 significance level.
Results and Discussion
Greenhouse Studies
The two- and three-way interactions among rate, application timing, and clone were not significant (P > 0.05) for all measured variables except for shoot dry weight, for which the rate by clone interaction was significant. Therefore, the results are presented with respect to significance of main effects or their interactions.
Sweetpotato injury
Injury was observed on the growing tips as early as 1 WAT and increased with time. Averaged over clones and application timing, a linear trend of increasing injury was observed with increasing rate of bicyclopyrone at 3, 5, and 8 WAT (Figure 1A). Sweetpotato injury was less than 12% from all bicyclopyrone rates at 3 WAT; however, by 8 WAT, injury ranged from 15% to 39%. Plants began to recover from injury at 8 WAT; therefore, a similar injury trend was observed 9 WAT (data not shown). Injury to plant clones only appeared significant at 1 and 9 WAT, when Covington (1% and 33%, respectively) was less affected at earlier rating dates but showed greater injury than NC04-531 (6% and 27%, respectively) at the later rating date. Application timing (PP or PT) did not significantly affect the level of injury.
Figure 1. Influence of bicyclopyrone rate on (A) sweetpotato injury (I) including chlorosis, necrosis, and leaf deformation; (B) stunting (ST); (C) vine length (VL); and (D) shoot dry weight (DW) in 2018 greenhouse studies in Greensboro, NC. Data are combined over application timings, clones, and runs for sweetpotato I, ST, and VL. However, for shoot DW, data are combined over application timings and runs. Points represent observed mean and vertical lines represent the mean ± SE. I3WAT = 12.03 exp [−2.59 × exp (−1.20 × X)]; R 2 = 0.98. I5WAT = 20.53 exp [−2.36 × exp (−1.33 × X)]; R 2 = 0.96. I8WAT = 39.35 exp [−2.64 × exp (−1.53 × X)]; R 2 = 0.98. ST3WAT = 10.50 exp [−3.55 × exp (−3.44 × X)]; R 2 = 0.97. ST5WAT = 29.77 exp [−2.65 × exp (−1.22 × X)]; R 2 = 0.97. ST8WAT = 43.50 × exp [−2.99 × exp (−1.37 × X)]; R 2 = 0.99. VL2WAT = 3.66 / [(1 − 0.07) exp (−0.95 × X)]; R 2 = 0.78. VL5WAT = 4.52 / [(1 − 0.25) exp (−1.02 × X)]; R 2 = 0.96. VL8WAT = 4.71 / [(1 − 0.34) exp (−0.50 × X)]; R 2 = 0.98. DWCOV = 0.33 / [(1 − 0.7816) exp (−0.19 × X)]; R 2 = 1.00. DWNC04-531 = 0.68 / [(1 − 0.77) exp (−1.22 × X)]; R 2 = 1.00. COV, Covington; WAT, weeks after transplanting.
Sweetpotato stunting
Similar to sweetpotato injury, stunting increased as bicyclopyrone rate increased (Figure 1B). Bicyclopyrone application timing (PP or PT) was not a significant factor for crop stunting. Stunting for sweetpotato clone was statistically significant (P < 0.05) at 8 and 9 WAT with Covington (18% and 25%, respectively) being stunted less than NC04-531 (27% and 36%, respectively).
Sweetpotato vine length
At 7 WAT, NC04-531 (15.6 cm) was taller than Covington (13.8 cm) (data not shown). The greater vine length for NC04-531 compared with Covington may be attributed to different growth characteristics for these clones (Harrison and Jackson Reference Harrison and Jackson2011; La Bonte et al. Reference La Bonte, Harrison and Motsenbocker1999) compared with other clones. However, there was no interaction observed for clone by application timing, or clone by application rate; therefore, data are presented averaged over clone and application timing at 2, 5, and 8 WAT (Figure 1C). An effect of bicyclopyrone rate on sweetpotato vine length was not observed 2 WAT, but reduced vine length corresponding to the increasing rates of bicyclopyrone was observed at 5 and 8 WAT (Figure 1C).
Sweetpotato shoot dry weight
A significant interaction between clone and bicyclopyrone application rate was observed for shoot dry weight (Figure 1D). This interaction likely resulted because nontreated NC04-531 (3.0 g plant−1) had higher shoot dry weight than nontreated Covington (1.5 g plant−1). As a result, the shoot biomass reduction (percentage of nontreated check) was higher for NC04-531 compared with Covington for all rates of bicyclopyrone (data not shown). The difference in biomass with respect to clone is likely due to the inherent differences in clone biomass accumulation (Harrison and Jackson Reference Harrison and Jackson2011; La Bonte et al. Reference La Bonte, Harrison and Motsenbocker1999). The shoot dry weight of both sweetpotato clones decreased with increasing rates of bicyclopyrone (Figure 1D).
Field Studies
Palmer amaranth control
In 2016 and 2018, although minor differences occurred among treatments for Palmer amaranth control, all herbicide treatments provided at least 95% control at 4 WAT (Table 3). At 8 WAT, no differences in Palmer amaranth control were observed among the treatments, with control ranging from 67% to 95% and 97% to 100% in 2016 and 2018, respectively. In 2017, Palmer amaranth control was lower compared with the other 2 yr from the treatments that had bicyclopyrone PP and clomazone PP fb S-metolachlor PT. However, flumioxazin PP fb S-metolachlor PT and flumioxazin PP fb S-metolachlor PT fb bicyclopyrone PT- directed controlled Palmer amaranth 92% or greater at 4 and 8 WAT.
Table 3. Effects of herbicide treatments on Palmer amaranth control in field studies at Clinton, NC, in 2016, 2017, and 2018.
a Abbreviations: —, treatment not included in the study; fb, followed by; PP, preplant; PT, post-transplant; WAT, weeks after transplanting.
b Means within columns fb the same letter are not significantly different according to Fisher Protected LSD (α = 0.05).
c Treatments conducted in 2016 and 2017.
d Treatments conducted in 2017 and 2018.
Sweetpotato injury and stunting
In 2016, at 3 WAT, injury was 23%, 14%, and less than 2% after treatment with bicyclopyrone PP fb S-metolachlor PT, bicyclopyrone PP, and all other treatments, respectively (Table 4). By 5 and 8 WAT, sweetpotato had mostly recovered; injury was 0% to 9% from all herbicide treatments except for flumioxazin PP fb S-metolachlor PT fb bicyclopyrone PT-directed, which was 28%. At 3 WAT, sweetpotato stunting was 0% to 5% with no differences in stunting among herbicide treatments. Plant stunting at 5 WAT was 33% for bicyclopyrone (100 g ha−1) PP fb S-metolachlor PT, 7% for bicyclopyrone PP, and no stunting for all other treatments.
Table 4. Effect of herbicide treatments on Beauregard sweetpotato injury, stunting, and yield in field studies at Clinton, NC, in 2016.
a Abbreviations: —, nontreated weedy and weed-free checks were not included in the statistical analysis of sweetpotato injury and stunting; fb, followed by; PP, preplant; PT, post-transplant; WAT, weeks after transplanting.
b Means within columns fb the same letter are not significantly different according to Fisher protected LSD (α = 0.05). Means with no letter in a column do not differ according to Fisher protected LSD (α = 0.05). The nontreated weedy and weed-free checks were not included in the statistical analysis of sweetpotato injury and stunting.
c Marketable is the aggregate of jumbo, no. 1, and canner grades of sweetpotato roots.
In 2017, less than 1% injury or stunting was reported from flumioxazin PP, flumioxazin PP fb S-metolachlor PT, flumioxazin PP fb S-metolachlor PT fb bicyclopyrone PT-directed, and clomazone PP fb S-metolachlor PT, regardless of rating timings (Table 5). However, more injury was observed from treatments with bicyclopyrone PP, and injury ranged from 12% to 85%, 16% to 94%, and 88% to 96% at 2, 4, and 8 WAT, respectively. Likewise stunting from these treatments was high and ranged from 72% to 89% at 4 WAT.
Table 5. Effect of herbicide treatments on Covington sweetpotato injury, stunting, and yield in field studies at Clinton, NC, in 2017.
a Abbreviation: —, nontreated weedy and weed-free checks were not included in the statistical analysis of sweetpotato injury and stunting; fb, followed by; PP, preplant; PT, post-transplant; WAT, weeks after transplanting.
b Means within columns fb the same letter are not significantly different according to Fisher protected LSD (α = 0.05). Means with no letter in a column do not differ according to Fisher protected LSD (α = 0.05). The nontreated weedy and weed-free checks were not included in the statistical analysis of sweetpotato injury and stunting.
c Marketable is the aggregate of jumbo, no. 1, and canner grades of sweetpotato roots.
In 2018, higher level of crop injury (10% to 12%) and crop stunting (8% to 40%) were observed from the treatments with bicyclopyrone PP regardless of evaluation timing (Table 6). All other treatments showed 4% or less injury or crop stunting.
Table 6. Effect of herbicide treatments on Covington sweetpotato injury, stunting, and yield in field studies at Clinton, NC, in 2018.
a Abbreviation: —, nontreated weedy and weed-free checks were not included in the statistical analysis of sweetpotato injury and stunting; fb, followed by; PT, post-transplant; WAT, weeks after transplanting.
b Means within columns fb the same letter are not significantly different according to Fisher protected LSD (α = 0.05). Means with no letter in a column do not differ according to Fisher protected LSD (α = 0.05). The nontreated weedy and weed-free check were not included in the statistical analysis of sweetpotato injury and stunting.
c Marketable is the aggregate of jumbo, no. 1, and canner grades of sweetpotato roots.
Sweetpotato yield
In 2016, no statistical differences among treatments were observed for canner yield (Table 4). No. 1, jumbo, and marketable yields were greatest for the weed-free check, 50 g ha−1 bicyclopyrone PP fb S-metolachlor PT, and flumioxazin PP fb S-metolachlor PT. The increase in bicyclopyrone rate to 100 g ha−1 PP fb S-metolachlor PT negatively affected sweetpotato yield regardless of grade as a consequence of greater crop injury and stunting. All other treatments were lower in no. 1 and marketable yields than the weed-free check, likely due to higher level of crop injury or lower Palmer amaranth control.
In 2017, jumbo and marketable yields were lower from the treatments with bicyclopyrone PP compared with the weed-free check, flumioxazin PP, flumioxazin PP fb S-metolachlor PT, and flumioxazin PP fb S-metolachlor PT fb bicyclopyrone PT-directed (Table 5). No. 1 yield between the weed-free check and flumioxazin PP fb S-metolachlor PT were not different and was higher than with all other treatments.
In 2018, canner, no. 1, or jumbo root yields did not differ across treatments (Table 6). Similarly, marketable yields for all treatments except bicyclopyrone PP fb S-metolachlor PT were statistically similar and ranged from 39.8 to 49.2 × 103 kg ha−1. This decrease in yield from bicyclopyrone PP fb S-metolachlor PT compared with other treatments was possibly due to greater crop injury and stunting.
Weather had a substantial effect on Palmer amaranth control as well as on sweetpotato response to treatments applied in field studies from 2016 to 2018. In 2017, plots received 3.1 cm of rain 1 DAT, and a total of 4.7 cm within the first week (Figure 2). Bicyclopyrone is highly mobile in soil (APVMA 2017; Hand et al. Reference Hand, Nichols, Kuet, Oliver, Harbourt and El‐Naggar2015) and high levels of rain immediately after PP application likely moved bicyclopyrone from the upper level of the soil (where most of the germinating weed seeds are present) to below the soil surface near the sweetpotato root zone, causing a reduced level of Palmer amaranth control, higher crop injury and stunting, and lower yields.
Figure 2. Rainfall from 0 to 10 d after sweetpotato planting at Clinton, NC, in 2016, 2017, and 2018. Rainfall data were collected from regional weather stations and provided by the State Climate Office in North Carolina.
Combined results from the greenhouse and field studies suggest bicyclopyrone has potential to be used in sweetpotato. However, the crop injury and yield reduction associated with higher rates of bicyclopyrone are concerning. At rates corresponding to twice and three times the desired (50 g ha−1) registration rate, crop injury ranged from 34% to 37% at 8 WAT in greenhouse trials, which is an unacceptable level to growers (K.M. Jennings, personal communication). Similarly, in field studies, bicyclopyrone PP 100 g ha−1 fb S-metolachlor PT caused 33% or greater crop stunting and 44% or greater marketable yield reduction compared with the weed-free check during 2016 (Beauregard) and 2017 (Covington). Sweetpotato treated with PP 50 g ha−1 bicyclopyrone alone or fb S-metolachlor PT had minimal to moderate injury and similar no. 1 and jumbo yield as the weed-free check in 2 of the 3 yr. Bicyclopyrone PT-directed application was safe in 2 (2017 and 2018) of the 3 yr. Bicyclopyrone PP should not be applied immediately before or after rainfall events greater than 2.5 cm, because considerable plant injury and yield loss were observed by possible movement of the herbicide into the sweetpotato root zone. Additional care with application, including cultivation to cover the leaf and vine tips, is necessary for late-season PT-directed applications of bicyclopyrone to be safe to sweetpotato. If bicyclopyrone comes into contact with the mature plant leaves and vines, substantial injury, including leaf chlorosis, necrosis, and stunting, can occur, leading to reduced yields.
The addition of a new herbicide mechanism of action in a weed management program for sweetpotato would help reduce the selection pressure in fields where flumioxazin typically has been applied, especially in areas where herbicide-resistant weeds have already evolved. Bicyclopyrone has higher biological activity than other herbicides in the triketone family (i.e., mesotrione, tembotrione, and suclotrione) (Hartzler Reference Hartzler2015). Therefore, registration of bicyclopyrone either PP or PT-directed for sweetpotato at lower rates (≤50 g h−1) would provide a new option to control weeds. However, on the basis of our results, additional research is needed to evaluate lower rates of bicyclopyrone PP or PT-directed on additional soils and sweetpotato clones to ensure crop safety, and more research should be done to evaluate the potential of lower rates of bicyclopyrone in combination with other herbicides to design a season-long weed management program.
Acknowledgements
The authors thank Syngenta Crop Protection, the NC Agricultural Research Service, NC State Extension, and NC Department of Agriculture and Consumer Services for supporting this research. They thank fellow graduate students and undergraduate students in the Department of Horticultural Science for assisting with this research and the staff at the Horticultural Crops Research Station in Clinton, NC, for management of these studies. No conflicts of interest have been declared.
