Weed Control and Soybean ( Glycine max ) Response to Mixtures of a Blended Foliar Fertilizer and Postemergence Herbicides

: Growers commonly mix foliar fertilizers with postemergence (POST) herbicides to reduce application costs and / or to decrease soybean injury. Field studies conducted in 2015 and 2016 in Stoneville, MS, evaluated the impact on weed control and soybean ( Glycine max (L.) Merr) injury, growth, and yield when combining a blended foliar fertilizer with POST herbicide applications. Herbicide treatments included no herbicide and glyphosate alone and in combination with S -metolachlor, fomesafen, or lactofen. The blended foliar fertilizer was applied at 0, 0.39, and 0.78 kg a.i. ha − 1 . In the Weed Control Study, 14 antagonistic e ﬀ ects at various evaluations were detected on Palmer amaranth ( Amaranthus palmeri (S.) Wats) and barnyardgrass ( Echinochloa crus-galli (L.) P. Beauv.) control 7, 14, and 21 d after treatment (DAT) when a blended foliar fertilizer at 0.39 or 0.78 kg a.i. ha − 1 was mixed with glyphosate alone or combined with S -metolachlor, fomesafen, or lactofen. Of the 14 total e ﬀ ects, nine were detected with foliar fertilizer at the higher rate of 0.78 kg a.i. ha − 1 . Seven antagonistic e ﬀ ects were detected for both weed species regardless of herbicide treatment or foliar fertilizer rate. The only treatment combination in which an antagonistic e ﬀ ect was not detected was glyphosate plus lactofen plus foliar fertilizer at 0.78 kg a.i. ha − 1 . Blended foliar fertilizer did not inﬂuence soybean injury. In the Agronomic Study, blended foliar fertilizer did not impact soybean injury, height, dry wt., nutrient conc., or yield. Foliar fertilizer in combination with POST soybean herbicides did not reduce soybean injury and produced inconsistent e ﬀ ects on weed control across herbicide treatments and between weed species.


Introduction
Amaranthus ssp., known collectively as pigweeds, belong to the family Amaranthaceae and have ranked among the top ten most troublesome weeds in southern U.S. soybean (Glycine max (L.) Merr) since the early 1970s [1-3]. Among eight southern U.S. states surveyed in 2013, Palmer amaranth was ranked as the most troublesome weed of cotton Gossypium hirisutum (L.) and soybean in seven and three states, respectively [4]. Palmer amaranth Amaranthus palmeri (S.) Watts is one of thirteen weed species in Mississippi to exhibit herbicide resistance [5]. In 2008, Palmer amaranth was confirmed resistant to glyphosate and acetolactate synthase (ALS) inhibitors in Mississippi [6].
Fomesafen and lactofen are common treatments for Palmer amaranth control in soybean, but soybean injury is often observed following POST applications [5,39,40]. In an effort to reduce the number of applications and decrease soybean injury, growers commonly mix foliar fertilizers with POST herbicides [41,42]. Due to limited research on the interaction between herbicides and foliar fertilizers, field studies were conducted detailing the impact of mixing a blended foliar fertilizer with POST soybean herbicides. The objectives were to (1) evaluate the influence of a blended foliar fertilizer on soybean injury and weed control with POST herbicides and (2) to characterize soybean growth and yield following POST applications of mixtures of herbicides and a blended foliar fertilizer.  Table 1. The experimental sites were known to be infested with barnyardgrass and Palmer amaranth. Each site was conventionally tilled prior to planting to stimulate weed germination and ensure uniform emergence. 'Asgrow 4632' (Monsanto Company, St. Louis, MO, USA) mid maturity group IV soybean was utilized in all siteyears and planted with a John Deere small-plot air planter (John Deere 1730, Deer and Company, Moline, IL, USA).
Visible estimates of soybean injury and weed control were recorded on a scale from 0 to 100% with 0 representing no injury or control and 100 representing soybean death or complete weed control [44]. Square roots of visible injury and control estimates were arcsine transformed. The transformation did not improve the homogeneity of the variance based on visual inspection of the plotted residuals; therefore, nontransformed data were utilized in all analyses. Soybean injury and weed control data were analyzed utilizing the augmented mixed-model methodology described by Blouin et al. [36]. Data for soybean height and yield were subjected to ANOVA using the PROC MIXED procedure in SAS 9.4 (SAS Institute Inc., Cary, NC, USA) with siteyear, replication (nested within siteyear), and treatment-by-rep interactions listed as random variable parameters [45].
Type III Statistics were utilized to test the fixed effects of herbicide and foliar fertilizer for soybean height and yield. Least square means were calculated and mean separation (p ≤ 0.05) was produced using PDMIX800 in SAS v. 9.3 (SAS Institute Inc. 100 SAS Campus Drive Cary, NC 27513-2414, USA), which is a macro for converting mean separation output to letter groupings [46]. When injury and weed control data did not return a significant synergistic or antagonistic effect [36], data were analyzed as described for soybean height and yield.

Agronomic Study
A field study was conducted at the Mississippi State University Delta Research and Extension Center at Stoneville in 2015 and 2016 to evaluate soybean response when mixing a blended foliar fertilizer with POST herbicides. The study was performed at two sites in 2015 (2015-1 and 2015-2) and 2016 (2016-1 and 2016-2). Global position system (GPS) coordinates, series, description, pH, and organic matter (OM) for each siteyear are presented in Table 1. Each site was conventionally tilled, then planted with a John Deere small-plot air planter. 'Pioneer 48T53' (Pioneer Hi-Bred P.O. Johnston, IA, USA) and Asgrow 4632 were planted in 2015 and 2016, respectively.
The treatment structure and experimental design for the Agronomic Study was identical to the Weed Control Study. However, the Agronomic Study was maintained weed-free each siteyear to prevent weed interference with soybean agronomic performance. Plots were hand-weeded or treated with labeled POST and residual herbicides applied with an in-row hooded sprayer (Willmar Fabrication, Benson, MN, USA) to prevent foliar soybean injury.
Visible estimates of soybean injury were recorded 3, 7, 14, 21, and 28 DAT on the previously described scale. Soybean height was recorded 14 DAT and at maturity as previously described. Soybean biomass was collected from 1 m sections of rows 1 and 4 in each plot 14 DAT. Soybean biomass samples were dried at 60 • C for 1 wk and weight converted to g m −2 . Ten trifoliate leaves were collected from the uppermost fully mature nodes of plants in rows 2 and 3 of each plot 14 DAT for tissue analysis. Tissue samples were air-dried in a greenhouse for analysis. Tissue samples were digested with concentrated nitric acid (HNO 3 ) and 30% hydrogen peroxide (H 2 O 2 ) and analyzed by inductively coupled plasma atomic emission spectroscopy (ICP-AES) for nutrient conc. Soybean were harvested using a small-plot combine on Oct. 5, 2015, and Sept. 27 and Oct. 3,2016. Yield data were adjusted to 13% moisture content. Data analyses were identical to the Weed Control Study.

Weed Control Study
No synergistic or antagonistic effects were detected for soybean injury across all evaluation intervals. A main effect of herbicide treatment was detected for soybean injury at 3, 7, and 14 DAT ( Table 2). Pooled across foliar fertilizer rates, glyphosate plus lactofen injured soybean more than other herbicide treatments 3 and 14 DAT. Glyphosate plus S-metolachlor injured soybean more than glyphosate alone, but not as severely as glyphosate plus fomesafen. Differences among treatments for soybean injury 7 DAT were similar as for 3 and 14 DAT (data not presented). Bronzing and necrosis of plant tissue due to lactofen and fomesafen is well-documented [5,40]. By 21 and 28 DAT, no soybean injury was observed (data not presented).
Palmer amaranth control 7 DAT with glyphosate alone was antagonized with 7% reduction by the addition of foliar fertilizer at 0.39 kg a.i. ha −1 and 11 and 13% reduction at 14 and 21 DAT, respectively, with foliar fertilizer at 0.78 kg a.i. ha −1 (Table 3). Control with glyphosate plus S-metolachlor was antagonized ≥11% by adding foliar fertilizer at 0.39 or 0.78 kg a.i. ha −1 at 7 and 14 DAT (Table 3). Antagonism on velvetleaf Abutilon theophrasti (Medik) control has been reported when glyphosate at 0.28 kg a.i. ha −1 was combined with different formulations of manganese (Mn); however, some of the antagonistic effects were overcome by adding ammonium sulfate at 20 g L −1 [42]. Palmer amaranth in the current research was controlled 66% 7 DAT, with reduction attributed to the population contained some glyphosate-susceptible and -resistant individuals.
Palmer amaranth control with glyphosate alone was 56% and less than that with all other herbicide treatments 28 DAT (Table 2). Since glyphosate is a POST herbicide lacking residual control, it should be expected that the residual activity from fomesafen and S-metolachlor would control more Palmer amaranth than glyphosate alone 28 DAT [5, 50,51]. Similar to glyphosate, there is minimal residual control expected with lactofen; however, glyphosate plus lactofen increased control of Palmer amaranth 18 and 21% compared with glyphosate alone 7 and 14 DAT, respectively (data not presented). Palmer amaranth control with glyphosate alone and glyphosate plus S-metolachlor was similar regardless of evaluation interval prior to 28 DAT (data not presented).
Barnyardgrass control was antagonized when foliar fertilizer at 0.78 kg a.i. ha −1 was mixed with glyphosate alone 7 and 21 DAT, glyphosate plus fomesafen 14 and 21 DAT, and glyphosate plus S-metolachlor 14 DAT (Table 4). Differences between the observed and expected levels of control ranged from 6 to 10%. Antagonism for barnyardgrass control was also detected when foliar fertilizer at 0.39 kg a.i. ha −1 was mixed with glyphosate plus fomesafen 14 DAT and glyphosate plus lactofen 21 DAT, and differences between the observed and expected levels of control were 9 and 6% for mixtures of glyphosate plus fomesafen or lactofen, respectively (Table 4).  0.0502 † Evaluation interval and respective herbicide treatment. ‡ Expected values for each rate of foliar fertilizer at each evaluation interval are the same due to a lack of herbicidal activity from the foliar fertilizer; therefore, values are visual estimates of weed control for each herbicide treatment when foliar fertilizer rate was 0 kg ha −1 . † † Asterisks within each evaluation interval denote antagonistic effects between herbicide treatment and foliar fertilizer rate when p ≤ 0.05. ‡ ‡ The p-value nested within each foliar fertilizer rate for each evaluation interval denotes significant differences between observed and expected values within the corresponding rate of foliar fertilizer. Boldness is for ease of identification significant p values. 0.1166 † Evaluation interval and respective herbicide treatment. ‡ Expected values for each rate of foliar fertilizer at each evaluation interval are the same due to a lack of herbicidal activity from the foliar fertilizer; therefore, values are visual estimates of weed control for each herbicide treatment when foliar fertilizer rate was 0 kg ha −1 . † † Asterisks within each evaluation interval denote antagonistic effects between herbicide treatment and foliar fertilizer rate when p ≤ 0.05. ‡ ‡ The p-value nested within each foliar fertilizer rate for each evaluation interval denotes significant differences between observed and expected values within the corresponding rate of foliar fertilizer. Boldness is for ease of identification significant p values.
A main effect of herbicide treatment was detected for barnyardgrass control 28 DAT (Table 2). Barnyardgrass control 28 DAT was 82% with glyphosate plus S-metolachlor due to its residual activity on small-seeded broadleaf and grass species [51]. Other research reported 88% residual control of barnyardgrass 56 DAT with glyphosate plus S-metolachlor in cotton [52]. Residual control with fomesafen primarily targets broadleaf weeds [52] and barnyardgrass control 28 DAT with glyphosate plus fomesafen in the current study was comparable to glyphosate alone or mixed with lactofen ( Table 2). A main effect of foliar fertilizer was detected for barnyardgrass control 28 DAT, and the addition of foliar fertilizer at 0.39 or 0.78 kg a.i. ha −1 reduced barnyardgrass control ≥4% regardless of herbicide treatment (Table 5). Pooled across foliar fertilizer rates, glyphosate plus lactofen reduced soybean height 14 DAT 5 and 4 cm compared with the no herbicide and glyphosate alone treatments, respectively ( Table 2). Similar results have been reported, where lactofen at 0.22 kg a.i. ha −1 alone or mixed with crop oil concentrate (COC) reduced soybean height 4 and 5 cm, respectively, compared with a control and COC alone [40]. Soybean height at maturity was not affected by foliar fertilizer rate and/or herbicide treatment (data not presented). Pooled across foliar fertilizer rates, soybean yield in plots receiving herbicide were similar and greater than yield in the no herbicide treatment ( Table 2).

Agronomic Study
A main effect of herbicide treatment was detected for soybean injury 3, 7, and 14 DAT, soybean dry wt. 14 DAT, and soybean height 14 DAT (Table 6). Glyphosate plus lactofen resulted in the greatest soybean injury followed by glyphosate plus fomesafen. Soybean injury was less with glyphosate plus S-metolachlor compared with glyphosate plus fomesafen, but greater than glyphosate alone at all evaluation intervals.
Treatments containing a PPO inhibitor reduced soybean dry wt. ≥9% and soybean height 14 DAT ≥7.5% compared with the no herbicide treatment. Soybean dry wt. and height 14 DAT with glyphosate plus S-metolachlor were similar to plots receiving no herbicide or glyphosate alone. Foliar fertilizer rate did not affect the measured parameters. Minor differences in tissue nutrient conc. were detected; however, no explanation for these differences was apparent and all values from the analysis were within the nutrient sufficiency range (data not presented; [53]).

Discussion
The injury caused by POST soybean herbicide treatments evaluated in this research was not influenced by the addition of a blended foliar fertilizer; therefore, the blended foliar fertilizer (4-0-0-3-3-3-0.25%; N-P-K-S-Mn-Zn-B) evaluated herein should not be mixed with POST soybean herbicides with the intent to reduce injury.
Mixing the blended foliar fertilizer with POST soybean herbicides influenced weed control. Palmer amaranth and barnyardgrass control were antagonized 7, 14, and 21 DAT by one or more of the herbicide treatments and blended foliar fertilizer rate combinations. Across species and evaluation intervals, 14 total antagonistic effects were detected. Antagonism of glyphosate from foliar fertilizer at 0.78 kg a.i. ha −1 was the most common antagonistic effect across both weed species and all evaluation intervals with four detected effects. Antagonism of glyphosate plus S-metolachlor from foliar fertilizer at 0.78 kg a.i. ha −1 was detected three times. Of the 14 total antagonistic effects, nine were detected with foliar fertilizer at the higher rate of 0.78 kg a.i. ha −1 . Seven antagonistic effects were detected for both weed species regardless of herbicide treatment or foliar fertilizer rate. The only treatment combination in which an antagonistic effect was not detected was glyphosate plus lactofen plus foliar fertilizer at 0.78 kg a.i. ha −1 . A grower with the intention of applying the blended foliar fertilizer (4-0-0-3-3-3-0.25%; N-P-K-S-Mn-Zn-B) at 0.78 kg a.i. ha −1 with a POST soybean herbicide should expect antagonism.
Soybean agronomic performance was not improved by mixing a blended foliar fertilizer with POST herbicide treatments. Since the blended foliar fertilizer (4-0-0-3-3-3-0.25%; N-P-K-S-Mn-Zn-B) did not affect soybean injury, height, dry wt., nutrient conc., or yield, the addition of this blended foliar fertilizer would not be economically beneficial to soybean and would represent an added expense to the grower. Even when the blended foliar fertilizer was applied with no herbicide, the agronomic performance of soybean was not improved.

Conclusions
Foliar fertilizer in combination with POST soybean herbicides did not reduce soybean injury and produced inconsistent effects on weed control across herbicide treatments and between weed species. Foliar fertilizers also did not improve agronomic performance of soybean. Since this research evaluated only one blended foliar fertilizer, growers should be cautious of other foliar fertilizers applied with POST herbicides in soybean. If a soybean herbicide treatment includes glyphosate, no foliar fertilizer should be added.