Introduction
Palmer amaranth (Amaranthus palmeri S. Watson) is an annual summer broadleaf weed native to the desert regions of the southwest United States and northern Mexico (Ward et al. Reference Ward, Webster and Steckel2013). If left uncontrolled, A. palmeri is known to interfere with various cropping systems, causing massive yield losses (Bensch et al. Reference Bensch, Peterson and Horak2003; Burke et al. Reference Burke, Schroeder, Thomas and Wilcut2007; Massinga and Currie Reference Massinga and Currie2002; Massinga et al. Reference Massinga, Currie, Horak and Boyer2001; Moore et al. Reference Moore, Murray and Westerman2004; Smith et al. Reference Smith, Baker and Steele2000) that account for millions of dollars annually (Meyer et al. Reference Meyer, Norsworthy, Young, Steckel, Bradley, Johnson, Loux, Davis, Kruger, Bararpour and Ikley2015). The Weed Science Society of America identified A. palmeri as the most troublesome weed in the United States (Anonymous 2016). Various attributes contribute to A. palmeri becoming such a problem weed, but the most important one is its fast growth rate due to its C4 photosynthetic pathway. Amaranthus palmeri can grow 5 to 7.5 cm d−1 (Legleiter and Johnson Reference Legleiter and Johnson2013; Sfiligoj Reference Sfiligoj2015) and has a growth rate higher than that of other Amaranthus species (Horak and Loughlin Reference Horak and Loughlin2000). Amaranthus palmeri’s photosynthetic rate is three to four times higher than those of cotton (Gossypium hirsutum L.), soybean [Glycine max (L.) Merr.], and corn (Zea mays L.) (Steckel Reference Steckel2007). This allows A. palmeri to compete effectively for resources (Berger et al. Reference Berger, Ferrell, Rowland and Webster2015; Massinga et al. Reference Massinga, Currie, Horak and Boyer2001, Reference Massinga, Currie and Trooien2003; Morgan et al. Reference Morgan, Baumann and Chandler2001).
Amaranthus palmeri is a problem weed in most parts of Kansas, affecting the sustainability of agricultural production. In western Kansas, where the climatic conditions favor its fast growth, A. palmeri is a common weed in the fallow phase of wheat–fallow–wheat and wheat–sorghum–fallow. Because of its fast growth and deep rooting system, this weed can effectively compete for water—a major limiting factor in existing cropping systems (Berger et al. Reference Berger, Ferrell, Rowland and Webster2015; Massinga et al. Reference Massinga, Currie and Trooien2003). Therefore, it is critical that A. palmeri be controlled at early growth stages when the roots are shallow and the leaves are smaller. According to Kirkwood (Reference Kirkwood1999), older plants and or leaves have a relatively more complex and thicker cuticle structure that can reduce the effectiveness of POST herbicides (Menendez et al. Reference Menendez, Rojano-Delgado and De Prado2014). Generally, weed control would be expected to decrease when herbicide application is made to larger weeds (Edmund and York Reference Edmund and York1987; Jordan et al. Reference Jordan, York, Griffin, Clay, Vidrine and Reynolds1997; Kegode and Fronning Reference Kegode and Fronning2005). For most POST herbicides, it is recommended that applications be made when A. palmeri is ≤10-cm tall (Ward et al. Reference Ward, Webster and Steckel2013). Considering its fast growth rate, however, delaying herbicide application by only a day or two can allow A. palmeri to quickly become too large to control easily. This situation is further exacerbated by the fact that A. palmeri has evolved resistance to at least six herbicide modes of action, including microtubule, 5-enolpyruvylshikimate-3-phosphate synthase, acetolactate synthase, photosystem II, hydroxyphenylpyruvate dioxygenase, and more recently, protoporphyrinogen oxidase inhibitors (Heap Reference Heap2019).
Dicamba is a synthetic auxin herbicide that is effective for selective control of broadleaf weeds, including A. palmeri. However, like natural auxins, auxin herbicides can be deactivated by multiple pathways in the plant. According to Grossmann (Reference Grossmann2010), the spectrum of biological activities of auxin herbicides depends on timing, concentration, and tissue sensitivity, which are determined by the type of tissue, plant species, and physiological stage. Currently, no case of A. palmeri resistance to dicamba has been reported (Heap Reference Heap2019); however, if the risk of resistance developing is to be mitigated, understanding how plant height at time of herbicide application affects its performance is crucial. We hypothesized that efficacy of dicamba for A. palmeri control will decline as A. palmeri plant height increases. Therefore, the objective of this study was to evaluate the effect of plant height on dicamba efficacy to control A. palmeri.
Materials and Methods
Field Experiments
Two dose–response studies were conducted under dryland conditions in adjacent fields at Kansas State University Southwest Research–Extension Center, Garden City, KS (37.994°N, 100.818°W), in 2016 and 2018 to evaluate POST efficacy of dicamba on A. palmeri in the absence of crop competition. Soil at the site was a Richfield silt loam (fine, smectitic, mesic Aridic Argiustolls) with a slope ≥1%. The site received a total of 55.9 mm of rainfall during August of 2018, and the 30-yr average is 63.8 mm (Elliott Reference Elliott2017). The average daily minimum and maximum air temperatures during August 2018 were 16.7 and 31.1 C compared with the 30-year average of 24.6 C (Elliott Reference Elliott2017). Fields were disked and field cultivated within the last week of July 2018 to ensure a uniform seed distribution in the top 2.5 to 5 cm of soil, the primary zone of weed seed germination, and to stimulate emergence of naturally occurring A. palmeri at the site. Plots were established on August 5, 2018, when approximately 50% A. palmeri emergence was observed. The studies used a randomized complete block experimental design with four replications and with treatments in a split-plot arrangement. The main plot treatments consisted of three A. palmeri plant heights (≤10-cm tall [Day 0], 15-cm tall [Day 1], and 30-cm tall [Day 4]), and the subplot treatments consisted of dicamba application at seven doses (0, 70, 140, 210, 280, 420, and 560 g ae ha−1). Each subplot was 6 m by 3 m. Dicamba applications were made using a CO2-pressurized backpack boom sprayer on August 10, 11, and 14 when the wind speed at the site was between 3 and 6 km h−1 and approximately 50% of the plants had reached the desired height.
Greenhouse Experiments
For consistency, A. palmeri seed with no known resistance to any commonly used herbicide were collected from the same site (Kansas State University Southwest Research and Extension Center, near Garden City, KS) in 2016. This seed cohort was also chosen because it is an accurate representation of the A. palmeri in the field. Two dose–response studies were conducted at Kansas State University, Manhattan, KS, during summer 2017 and 2018 to further assess the influence of height on dicamba efficacy to control A. palmeri. Seeds of A. palmeri were germinated in small trays (25 cm by 15 cm by 2.5 cm) filled with a commercial potting mixture (Miracle-Gro® Moisture Control Potting Mix, Scotts Miracle-Gro Products Inc. Marysville, OH, USA). Seedlings that were 2- to 3-cm tall were transplanted into 6.5 cm by 6.5 cm by 9 cm plots and allowed to grow. The greenhouse was equipped with sodium vapor lamps supplementing 250 µmol m−2 s−1 of illumination and maintained at 25/20 C day/night temperatures and 15/9 h day/night photoperiods. Amaranthus palmeri seedlings at ≤10-cm tall (Day 0), 15-cm tall (Day 1), and 30-cm tall (Day 4) were treated with varying doses of dicamba (0, 70, 140, 210, 280, and 560 g ae ha−1). All dicamba treatments were applied with a bench-type track sprayer (Generation III, DeVries Manufacturing, RR 1 Box184 Hollandale, MN, USA) equipped with a single, moving, even flat-fan nozzle tip (8002E TeeJet® tip, Spraying Systems, Wheaton, IL, USA) delivering 187 L ha−1 at 207 kPa in a single pass at 4.85 km h−1. Experiments were conducted in a randomized complete block design with four replications (1 plant per pot), and repeated in time. Mortality and biomass measurements were collected at 4 wk after treatment (WAT).
Dicamba Absorption and Translocation Experiments
Two experiments were conducted at Kansas State University, Manhattan, KS, in 2017 and 2018. Amaranthus palmeri seedlings were raised from the same seed that was used in the greenhouse experiment and grown under greenhouse conditions as described earlier. Three days before treatment with [14C]dicamba, seedlings of uniform size were selected and transferred to a growth chamber equipped with fluorescent bulbs capable of delivering 550 µmol m−2 s−1 photon flux at plant canopy level to acclimate. The growth chamber conditions were maintained at 32.5/22.5 C day/night, 15/9 h photoperiod, and 60% to 70% relative humidity. Plants were watered as required until the desired height for treatment was reached. A preliminary study showed no clear differences in absorption and translocation between 10- and 15-cm-tall A. palmeri (data not shown); therefore, 15-cm-tall plants were not included in subsequent studies. The 10- and 30-cm-tall plants were treated with ten 1-µl droplets of dicamba (ring-UL-14C)-ethanol solution (11.4 kBq µl−1, specific activity: 2.87 kBq µg−1; BASF Corp., Florham Park, NJ, USA) totaling 3.3 kBq applied on the adaxial surface of the fourth-youngest leaf using a Wiretrol® capillary syringe (10 µl, Drummond Scientific, Broomall, PA, USA). Non-radiolabeled dicamba was added to the radioactive solution to obtain 560 g ae ha−1 of dicamba in a carrier volume of 187 L. All plants were returned to the growth chamber 30 min following treatment. Each treatment included four replications. Both 10- and 30-cm plants were harvested at 24, 72 and 120 h after treatment (HAT) and were dissected into three parts: treated leaf (TL), tissue above the treated leaf (ATL), and tissue below the treated leaf (BTL). The TL was washed twice in 20-ml scintillation vials with a 5-ml wash solution (10% ethanol aqueous solution and 0.5% Tween-20) for 1 min at each time, and the radioactivity in the rinsate(s) was measured using liquid scintillation spectrometry (LSS; Beckman Coulter LS6500 Multipurpose Scintillation Counter, Beckman Coulter, Brea, CA, USA) after adding 15 ml of Ecolite-(R) (MP Biomedicals, Santa Ana, CA, USA). The dissected plant parts were oven-dried at 60 C for 72 h and then combusted in a biological oxidizer (OXU-501, RJ Harvey Instrument, New York, NY, USA). Radioactivity was then quantified using LSS. Total absorption of [14C]dicamba was determined as:
$$\scale 94%\eqalign{\cr	{\rm{Absorption}}(\% ) \cr	\hskip-5pt={{{\rm{total\, radioactivity\, applied}}-{\rm{radioactivity\, recovered\, in\, wash\, solution}}} \over {{\rm{total\, radioactivity\, applied}}}}$$
([1])Herbicide translocation was determined as:
$${\rm{Translocation}}(\% ) = 100 - \% \,{\rm{radioactivity\, recovered\, in\, TL}}$$
([2])where
$${\rm{Radioactivity\, recovered\, in\, TL}}(\% ) = {{{\rm{radioactivity\, recovered\, in\, TL}}} \over {{\rm{radioactivity\, absorbed}}}}$$
([3])Data Collection and Analysis
Visual injury ratings and mortality of A. palmeri on a 0% (no control) to 100% (complete plant death) scale were assessed at 4 WAT. Aboveground plant biomass was collected and oven-dried (65 C for 3 d) to determine the dry biomass. All data were subjected to ANOVA to determine the significance of the interaction of A. palmeri plant height and dicamba dose on response parameters. If the interaction was not significant at the 5% level, data were pooled over one factor to test significance of the other. Specifically, visual injury ratings were subject to nonparametric kruskal.test and TukeyHSD post-hoc.test in R software for ANOVA and mean separation, respectively. In addition, visual injury ratings and aboveground plant biomass (% of nontreated; Wortman Reference Wortman2014) and mortality were regressed over dicamba doses using a three- and four-parameter log-logistic model, respectively, in R software (Ritz et al. Reference Ritz, Baty, Streibig and Gerhard2015; Seefeldt et al. Reference Seefeldt, Jensen and Fuerst1995). The lack-of-fit test (P > 0.05) indicated that the chosen model accurately described the data. Regression analyses were further used to estimate effective dicamba doses required to achieve 50% control (ED50 for visual injury rating or LD50 for mortality) and shoot biomass reduction (GR50) of A. palmeri and standard errors.
Aboveground plant biomass and [14C]dicamba absorption and translocation data were subjected to ANOVA using PROC MIXED in SAS v. 9.3 (SAS Institute, Cary, NC, USA) to test the significance of the fixed effects (i.e., plant height and dicamba dose) and their interaction. Experimental data were also subjected to ANOVA, and means were separated using Fisher’s protected LSD at α = 0.05 in SAS (SAS 2011).
Results and Discussion
Field Dose Response
No experiment by treatment interaction was detected; therefore, data were combined across experiments. The data from 2-yr field experiments suggest increased control of A. palmeri with an increase in dicamba dose regardless of plant height at time of spraying. Increase in plant height at time of spraying from ≤10- to 15- and 30-cm tall, however, reduced A. palmeri control from 25% to 0% and 99% to 42% and 30%, respectively, with an increase in dicamba dose from 70 to 560 g ha−1 (Figure 1). This reduction in control suggests that dicamba efficacy to control A. palmeri largely depend(s) on plant height at time of spraying (Hedges et al. Reference Hedges, Soltani, Hooker, Robinson and Sikkema2018; Joseph et al. Reference Joseph, Marshall and Sanders2018). For instance, when the label-recommended dose of dicamba (560 g ha−1) was applied, 99% of ≤10-cm-tall A. palmeri control was achieved compared with only 42% and 30% of 15- and 30-cm-tall A. palmeri, respectively, which is less than the level of control (47%) achieved when only one-fourth of the label-recommended dose of dicamba was sprayed on ≤10-cm-tall A. palmeri (Figure 1). Previous studies showed reduced weed control with increasing plant height at the time of herbicide application in other species, including common waterhemp [Amaranthus tuberculatus (Moq.) J. D. Sauer], large crabgrass [Digitaria sanguinalis (L.) Scop.], giant ragweed [Ambrosia trifida L.], kochia [Bassia scoparia (L.) A. J. Scott], and ivyleaf morningglory (Ipomoea hederacea Jacq.) (Chahal et al. Reference Chahal, Aulakh, Rosenbaum and Jhala2015; Cordes et al. Reference Cordes, Johnson, Scharf and Smeda2004; Craigmyle et al. Reference Craigmyle, Ellis and Bradley2013; Hoss et al. Reference Hoss, Al-Khatib, Peterson and Loughin2003). Furthermore, the effective doses of dicamba required to achieve 50% control (ED50) and shoot biomass reduction (GR50) were greater with an increase in A. palmeri plant height at time of dicamba application (Table 1). This suggests that more dicamba is required for a desirable control with later A. palmeri growth stages. The ED50 values increased 2- and 14-fold, when dicamba was applied on (to) 15-cm and 30-cm A. palmeri, respectively, compared with the ED50 value for ≤10-cm-tall plants (Table 1). Similarly, the GR50 values increased almost 5- and 7-fold when dicamba was applied on 15-cm and 30-cm A. palmeri, respectively, compared with the GR50 value for ≤10-cm A. palmeri (Table 1). Chahal et al. (Reference Chahal, Aulakh, Rosenbaum and Jhala2015) reported an increase in both ED50 and GR50 values as A. palmeri height increased from 10 to 20 cm at time of 2,4-D herbicide application.
Figure 1. Amaranthus palmeri control (visual rating, %) at 4 wk after treatment with dicamba as influenced by plant height at time of herbicide application in a field study conducted at Kansas State University Southwest Research–Extension Center, Garden City, KS, in 2016 and 2018.
Table 1. Effective dose of dicamba required to achieve 50% control (visual rating; ED50) and shoot biomass reduction (GR50) of 10-, 15-, and 30-cm-tall Amaranthus palmeri at 4 wk after treatment in a field study conducted at Kansas State University Southwest Research–Extension Center, Garden City, KS, in 2018.a
a ED50 and GR50 values were obtained by regressing visual weed control ratings and A. palmeri dry biomass over dicamba dose using a three-parameter log-logistic model.
Greenhouse Dose Response
No experiment by treatment interaction was detected; therefore, data were combined across experiments. Similar to field observations, A. palmeri control declined on more fully developed plants at time of dicamba application as reported elsewhere (Hedges et al. Reference Hedges, Soltani, Hooker, Robinson and Sikkema2018; Joseph et al. Reference Joseph, Marshall and Sanders2018). For example, increase in plant height from 5 cm to 10 cm and 30 cm at time of dicamba application resulted in a decline in A. palmeri control from 100% to 94% and 84%, respectively (Joseph et al. Reference Joseph, Marshall and Sanders2018). Similarly, when either glyphosate or dicamba was applied on (to) 5-, 15-, and 25-cm-tall glyphosate-resistant horseweed (Erigeron canadensis L.), its control declined from 87% to 76% and 62%, respectively (Hedges et al. Reference Hedges, Soltani, Hooker, Robinson and Sikkema2018). Amaranthus palmeri mortality ranged from 0% to 100% for ≤10-cm-tall A. palmeri compared with 0% to 87.5% for 15- and 30-cm-tall plants when dicamba dose was increased from 0 to 560 g ha−1 (Figure 2). Based on LD50 and GR50 values determined using a fitted log-logistic model, the effective dose of dicamba increased more than 1- and 2-fold to control 15- and 30-cm-tall A. palmeri, respectively, compared with ≤10-cm-tall A. palmeri (Table 2). The level of control of >10-cm-tall A. palmeri was higher in the greenhouse (mortality; Figure 2) compared with the field (visual rating; Figure 1). Additionally, with the exception of ≤10-cm A. palmeri, GR50 values of 15- and 30-cm A. palmeri obtained from field experiments (183 and 284 g ha−1, respectively) were relatively greater than those derived from greenhouse experiments (63 and 120 g ha−1, respectively; Tables 1 and 2). Although our data from the field studies indicate that 15-cm-tall A. palmeri is just as difficult to control as 30-cm-tall plants (Figure 3), our greenhouse data suggest that there is an opportunity to control 15-cm A. palmeri with the label-recommended dose of dicamba (Figure 4). These results, therefore, highlight the need to be aware of the effects of the environment of greenhouse studies on cuticle thickness and its effect on comparisons of weed control in field and greenhouse studies, underscoring the importance of considering environmental impacts when studying herbicide efficacy. Plant cuticle has been shown to interfere with foliar uptake of pesticides (Kirkwood Reference Kirkwood1999). There is a marked difference in cuticle development between field- and greenhouse-grown plants (Hull Reference Hull1958), which may explain the discrepancy in response to dicamba between field- and greenhouse-grown A. palmeri.
Figure 2. Amaranthus palmeri control (mortality, %) at 4 wk after treatment with dicamba as influenced by plant height at time of herbicide application in a greenhouse study at Kansas State University, Manhattan, KS, in 2017 and 2018. Mortality is the percentage of plant death following dicamba application.
Table 2. Effective dose of dicamba required to achieve 50% control (mortality; LD50) and shoot biomass reduction (GR50) of 10-, 15-, and 30-cm-tall Amaranthus palmeri at 4 wk after treatment in a greenhouse in 2017–2018 at Kansas State University, Manhattan, KS.a
a LD50 and GR50 values were obtained by regressing mortality ratings and A. palmeri dry biomass over dicamba dose using four- and three-parameter log-logistic models, respectively.
Figure 3. Dicamba dose response of Amaranthus palmeri at three different heights at 4 wk after treatment in a field study conducted at Kansas State University Southwest Research–Extension Center, Garden City, KS, in 2016 and 2018. Biomass was regressed over dicamba dose using a three-parameter log-logistic model.
Figure 4. Dicamba dose response of Amaranthus palmeri at three different heights at 4 wk after treatment in a greenhouse study at Kansas State University, Manhattan, KS, in 2017 and 2018. Biomass was regressed over dicamba dose using a three-parameter log-logistic model.
[14C]Dicamba Absorption and Translocation Experiments
No experiment by treatment interaction was detected; therefore, data were combined across experiments. Differences in [14C]dicamba absorption were observed in A. palmeri plants at different heights and harvest times (i.e., HAT; Figure 5A), P < 0.05. Amaranthus palmeri plants treated at 10-cm tall absorbed more [14C]dicamba than 30-cm-tall plants, regardless of harvest time (Figure 5A). This reduction in the amount of [14C]dicamba absorbed with increase in plant height may be due to changes in cuticle composition. Older plants or leaves have a relatively more complex and thicker cuticle structure (Kirkwood Reference Kirkwood1999). Thicker cuticles are less permeable to foliar-applied pesticides, thus reducing the effectiveness of POST herbicides (Menendez et al. Reference Menendez, Rojano-Delgado and De Prado2014) by limiting the amount of active ingredient entering the cytoplasm of plant cells. Similarly, at 24 HAT, up to 75% and 60% of [14C]dicamba was absorbed in 10-cm compared with 30-cm A. palmeri, respectively. However, maximum [14C]dicamba absorption occurred at 120 HAT in A. palmeri at both plant heights, 10- and 30-cm tall, with 93.3% and 86.7% absorption, respectively. Because a plant leaf is not a homogeneous substrate (Menendez et al. Reference Menendez, Rojano-Delgado and De Prado2014), the amount of [14C]dicamba absorbed did not always increase with time in 30-cm-tall A. palmeri, unlike in 10-cm-tall plants.
Figure 5. Absorption (A) and translocation (B) of [14C]dicamba in 10- and 30-cm-tall Amaranthus palmeri. Lowercase letters indicate difference between hours after treatment (HAT) at 5% level. P-value indicates significant difference between plant heights. Error bars represent standard error of the mean.
Although more than 50% of [14C]dicamba absorbed was translocated out of the TL over the course of the three sampling times (i.e., 24, 72, and 120 HAT) in both 10- and 30-cm A. palmeri, higher and more rapid translocation of [14C]dicamba was recorded in 10-cm compared with 30-cm A. palmeri (70% and 55.5%, respectively; Figure 5B). As a result, more [14C]dicamba reached ATL (Figure 6A and B) and BTL (Figure 6C) in 10-cm (49.3% and 20.7%, respectively) compared with 30-cm A. palmeri (47.5% and 7.9%, respectively). The acropetal translocation exceeded basipetal translocation of [14C]dicamba regardless of plant height. Dicamba is a systemic herbicide; however, it is translocated mostly via phloem (Chang and Vanden Born Reference Chang and Vanden Born1968; Cox Reference Cox1994; Ou et al. Reference Ou, Thompson, Stahlman, Bloedow and Jugulam2018). Therefore, its movement in the plant is highly dependent on the source-to-sink transport of sugar, also referred to as source-sink strength (Lemoine et al. Reference Lemoine, La Camera, Atanassova, Dédaldéchamp, Allario, Pourtau, Bonnemain, Laloi, Coutos-Thévenot, Maurousset and Faucher2013). Hence, to be effective, dicamba must be distributed throughout the plant and, more importantly, the actively growing shoot tissue. This is the reason an increase in translocation to ATL will improve control of sensitive plants with dicamba.
Figure 6. Distribution of [14C]dicamba to (A) above treated leaf (ATL), (B) treated leaf (TL), and (C) below treated leaf (BTL) in 10- and 30-cm-tall Amaranthus palmeri. Lowercase letters indicate difference between hours after treatment (HAT) at 5% level. P-value indicates significant difference between plant heights. Error bars represent standard error of the mean.
In summary, this study demonstrates that (1) dicamba efficacy to control A. palmeri is greatly influenced by plant height at time of herbicide application and (2) an increased absorption and translocation of dicamba at early growth stages contributes to increased efficacy of dicamba in controlling A. palmeri. The results from this study also suggest that opportunities to delay dicamba application for an effective control of A. palmeri are very limited. Therefore, timely management of this weed is crucial. Moreover, while this research has demonstrated that dicamba can effectively control A. palmeri at its earlier stages of growth, it is important to also consider including other effective herbicide modes of action to broaden the spectrum of weed control and delay evolution of resistance to the few herbicides still effective for A. palmeri management.
Acknowledgments
We thank Kansas State University Southwest Research–Extension Center Farm Crew for their help in field preparation. This article is approved for publication as Kansas Agricultural Experiment Station contribution no. 19-325-J. This research received no specific grant from any funding agency or the commercial or not-for-profit sectors. No conflicts of interest have been declared.
