Introduction
Grape is one of the most economically important fruit crops worldwide, cultivated for a wide range of uses including fresh consumption, juice, raisins, and wine. In 2022, global grape production surpassed 78 billion kilograms with a value estimated at US$80 billion. The United States ranks among the top five global producers, with more than 360,000 hectares planted and an annual production exceeding 6 million tons, 90% of which is concentrated in California (FAO 2023; USDA-NASS 2023). Other key states that produce premium grapes for wine and juice include Michigan, New Jersey, New York, Oregon, and Washington. Grape production occurs in at least 30 U.S. states (Vineyards Media Reference Media2025), reflecting a growing interest in local wines and agrotourism ventures. This geographic and economic expansion makes viticulture research nationally relevant.
Weed management is a significant challenge in grape production due to the crop’s perennial nature, which limits opportunities for soil disturbance and crop rotation, its sensitivity to off-target herbicide injury (Haring et al. Reference Haring, Ou, Al-Khatib and Hanson2022), and the high cost of hand labor required for under-vine maintenance (Gagliardi et al. Reference Gagliardi, Fontanelli, Luglio, Frasconi, Peruzzi and Raffaelli2023; Hostetler et al. Reference Hostetler, Merwin, Brown and Padilla-Zakour2007). Uncontrolled weed infestations may cause yield losses of 29% to 45% (Khadatkar et al. Reference Khadatkar, Sawant, Thorat, Gupta, Jadhav, Gawande and Magar2025). In perennial crops like grape, effective weed control is especially important during establishment because early competition can hinder the crop’s growth and reduce vine survival (Mia et al. Reference Mia, Massetani, Murri, Facchi, Monaci, Amadio and Neri2020).
Field bindweed (Convolvulus arvensis L.) is a deep-rooted perennial plant belonging to the morningglory family and a troublesome vineyard weed species (Orloff et al. Reference Orloff, Mangold, Miller and Menalled2018). Native to the Mediterranean region, this species has naturalized widely across North America and is characterized by its vigorous growth and long-term persistence in perennial agroecosystems (Sosnoskie et al. Reference Sosnoskie, Hanson and Steckel2020). Although the economic impacts of field bindweed interference in grape production have not been quantified in the scientific literature, studies have reported substantial yield losses from field bindweed in U.S. crops generally, largely due to direct competition for water and nutrients (Baumgartner et al. Reference Baumgartner, Steenwerth and Veilleux2007). Field bindweed’s ability to climb and overtop grapevines can shade leaves and fruiting zones (Sosnoskie et al. Reference Sosnoskie, Hanson and Steckel2020), reducing photosynthesis, affecting fruit quality, and creating humid microclimates that may favor disease development and reduce fungicide coverage.
Field bindweed is difficult to control due to its deep, energy-storing root system, which allows regrowth from underground rhizomes (Orloff et al. Reference Orloff, Mangold, Miller and Menalled2018; Sosnoskie et al. Reference Sosnoskie, Hanson and Steckel2020). Its drought tolerance further enhances its competitiveness, particularly in semi-arid regions that are common in the western United States (Khadatkar et al. Reference Khadatkar, Sawant, Thorat, Gupta, Jadhav, Gawande and Magar2025). The plant’s prostrate or climbing growth habit allows evasion of mechanical control (Khadatkar et al. Reference Khadatkar, Sawant, Thorat, Gupta, Jadhav, Gawande and Magar2025), and current herbicide options for grape are limited and often ineffective. Nonselective herbicides such as glyphosate are typically applied as directed treatments or with shields to protect the crop but provide only temporary control (Wiese and Lavake Reference Wiese and Lavake1986). Glyphosate is most effective when field bindweed is flowering, typically in mid-summer, although applications during this period can cause injury if the herbicide contacts grapevine suckers or low shoots (Whitesides Reference Whitesides1979). Additionally, growing market and consumer pressure to phase out glyphosate is driving vineyards to seek alternative control measures (Napa Green Reference Green2025). Systemic auxins such as 2,4-D may be more effective against bindweed, but their use on nondormant grape is constrained by the risk of severe vine injury that could result from high volatility associated with these herbicides (Hanson and Miller Reference Hanson and Miller2015; Haring et al. Reference Haring, Ou, Al-Khatib and Hanson2022). Many preemergence herbicides registered for use in grape production can suppress the emergence of field bindweed from seeds, but they typically lack strong activity against established plants. Frequent tillage under the vine may offer some control but is labor-intensive, can damage shallow vine roots or irrigation equipment, negatively affect soil health by reducing soil organic matter content and increasing soil erosion, and may contribute to weed spread through rhizome fragmentation (Dobrei et al. Reference Dobrei, Nistor, Sala and Dobrei2015; Wolff et al. Reference Wolff, Alsina, Stockert, Khalsa and Smart2018; Zumkeller et al. Reference Zumkeller, Torres, Marigliano, Zaccaria, Tanner and Kurtural2023).
A potential option for field bindweed control in grape is quinclorac, a synthetic auxinic herbicide in the quinolinecarboxylic acid family with both foliar and residual soil activity. Quinclorac disrupts normal cell elongation and division, and stimulates ethylene biosynthesis, leading to growth abnormalities and subsequent plant death in susceptible species (Song et al. Reference Song, Jiang, Wang, Fang, Zhou and Kong2022; Wang et al. Reference Wang, Wang, Li, Yu, Lin and Dong2022). Quinclorac has shown strong efficacy against field bindweed and other problematic weeds within the Convolvulaceae family in rice, hazelnut, highbush blueberry, cranberry, asparagus, turfgrass, and corn (Enloe et al. Reference Enloe, Nissen and Westra1999; Grossmann and Kwiatkowski Reference Grossmann and Kwiatkowski2000; Moretti and Peachey Reference Moretti and Peachey2022). However, quinclorac is not currently labeled for use in grape production, and its safety and effectiveness under vineyard conditions are not yet known.
To evaluate quinclorac’s suitability for U.S. vineyard systems, we conducted multistate field studies across diverse climatic and soil conditions representative of grape production regions. This research aimed to 1) evaluate grape tolerance to quinclorac across multiple application timings, rates, and U.S. production regions; 2) assess how rainfall and clay plus organic matter content influence herbicide efficacy and vine injury risk; and 3) evaluate field bindweed control under various rate and timing combinations.
Materials and Methods
Site Description
Field experiments were conducted in 2021 and 2022 in commercial and research vineyards located in Davis, California; Holt, Michigan; Landisville, New Jersey; Portland, New York; and Alpine, Oregon. Grape cultivars and soil types varied by site (Table 1). Site soils ranged from Mediterranean silt loams in California to humid-region loams in Michigan, sandy loams in New Jersy, and silty clay loams in Oregon. Overall, soil textures varied from sandy loams to fine clay loams, with clay content ranging from 8.2% to 32%, organic matter from 1.6% to 5%, and pH between 5.6 and 7.4 (Table 1). To estimate herbicide adsorption capacity, the combined percentage of soil clay and organic matter (COM) was calculated.
Table 1. Quinclorac study information.a
a Abbreviations: COM, combined percentage of clay and organic matter; OM, organic matter.
b Thirty-year climate averages for rain and average temperature were obtained from the National Centers for Environmental Information (ncei.noaa.gov). Annual values for 2021 and 2022 were collected from on-site or nearby weather stations.
Herbicide Treatments and Experimental Layout
At each location, quinclorac (QuinStar 4L; Albaugh LLC, St. Joseph, MO) was applied at 0.42 and 0.84 kg ai ha−1, corresponding to labeled recommendations for field bindweed control and rates previously used in non-crop and perennial systems (Albaugh 2020; Pedroso and Moretti Reference Pedroso and Moretti2022). The herbicide was applied twice in-season, approximately 90 d before harvest (Timing A) and 60 d before harvest (Timing B), with or without an additional post-harvest application before the first killing frost (Timing C) (Table 2). These timings were selected because they coincide with the period when field bindweed is most actively growing (Sosnoskie et al. Reference Sosnoskie, Hanson and Steckel2020). A nontreated weed-free control was included for comparison. Each treatment was applied to the same plots over two consecutive years (2021 and 2022), resulting in a total of four or six quinclorac applications per experimental unit (Table 2), depending on the treatment. This approach enabled assessment of cumulative effects in accordance with the Interregional Research Project number 4 (IR-4) study requirements for perennial cropping systems (IR-4 Project 2021). The herbicide was applied in 0.9-m-wide bands beneath the cordon on both sides of the vine row using CO2-pressurized backpack sprayers fitted with AIXR11003 nozzles (TeeJet Technologies, Glendale Heights, IL [used in California, Michigan, New Jersey, and Oregon]) or with AIXR11002 nozzles (in New York) and calibrated to 280 L ha−1 (in California, Michigan, New Jersey, and Oregon) or 190 L ha−1 (New York) at 275 kPa. Methylated seed-oil (HASTEN-EA; Wilbur-Ellis, Aurora, CO) at 10 ml L−1 was included with each treatment. Treatments were arranged in randomized complete blocks with four replications at every site except in New Jersey, where there were three replications. Plot sizes were 1.8 m wide by 9 m long (in Michigan and Oregon), 1.8 m wide by 7.3 m long (in New Jersey and New York), and 1.8 m wide by 3.6 m long (in California), with two to five vines per plot. Vine spacing ranges from 1.8 to 2.7 m within rows, with rows spaced 2.3 to 2.7 m apart.
Table 2. Quinclorac rates and application timings used in the studies.a
a Abbreviation: NTC, nontreated weed-free control.
b Application timing codes: A, first in-season (90 d preharvest); B, second in-season (60 d preharvest); C, post-harvest (prefrost).
Data Collection
Overall canopy injury, expressed as a composite visual assessment of leaf necrosis, chlorosis, cupping, strapping, and vine stunting, was rated at 7, 14, and 28 d after the first in-season application (Timing A; DAT-A), and at 7, 14, and 28 d after the second in-season application (Timing B; DAT-B). Ratings following the third application (Timing C) coincided with the Timing A assessments of the subsequent season, and thus injury from Application C was captured within those intervals. Injury was evaluated visually as a percentage injury on a scale from 0% (no visible symptoms) to 100% (complete canopy death). Field bindweed control was visually evaluated on the same dates at the Michigan and New Jersey sites, with natural infestations sufficiently dense, using the same scale: 0% represented no observed control and 100% represented complete control. Yield per vine was recorded at physiological maturity for all locations. Dormant pruning weights were measured in winter at the Michigan, New Jersey, and Oregon test locations. Daily rainfall was recorded at each site using on-farm or nearby weather stations.
Statistical Analysis
Data were analyzed using ANOVA within generalized linear mixed models in R software (v.4.3.2; R Core Team 2025). Crop injury and field bindweed control data were converted to values between 0 and 1 (Douma and Weedon Reference Douma and Weedon2019) and analyzed as repeated measures with a beta distribution, using the glmmTMB package (Brooks et al. Reference Brooks, Kristensen, van Benthem, Magnusson, Berg, Nielsen, Skaug, Maechler and Bolker2017). Yield per vine and vine pruning weight were analyzed assuming Gaussian errors. The assumptions of normality and homoscedasticity were assessed using Q–Q plots and residual-versus-predicted plots from the dharma package, and all models met Gaussian assumptions. Fixed effects included treatment, site, year, and their interactions; assessment day was added for crop injury and weed control models. Repetitions nested within site and year was treated as a random effect. When a treatment or interaction was significant (P < 0.05), means were separated using a Tukey HSD test at α = 0.05 with the emmeans package (Lenth Reference Lenth2020).
To identify the factors that are driving quinclorac injury in grape, we regressed the maximum injury recorded for each quinclorac treatment at each site in each year against cumulative rainfall through 56 DAT-A (equivalent to the combined periods of 0 to 28 DAT-A and 0 to 28 DAT-B) and the combined clay + organic-matter soil content (COM), pooling data across sites and years (n = 148). The model was fitted using a linear mixed-effect model with the lme4 package (Bates et al. Reference Bates, Maechler, Bolker, Walker, Christensen, Singmann, Dai, Scheipl, Grothendieck, Green and Fox2020), and treatment significance was evaluated with Type III ANOVA. A heat map was generated from model predictions to visualize injury across rainfall and COM gradients. Principal component analysis (PCA) was conducted on site-year means of maximum injury, rainfall, and COM using the prcomp function (R Core Team 2025) to visualize multivariate relationships. The nontreated control was excluded due to zero injury by design. Graphics were produced with ggplot2 (Wickham Reference Wickham2016).
Results and Discussion
Injury results were significantly influenced by experimental site, year, and treatment, as indicated by the three-way interaction in the mixed-effects model analysis (P < 0.0001). Therefore, additional analyses were conducted separately by experimental site. The effects of year, treatment, and their interaction were significant in some sites and response variables; thus, results are presented and discussed accordingly (Table 3). Repeated measures analysis also showed a significant effect of rating date (P < 0.003); therefore, crop injury and field bindweed control are presented separately for each assessment day.
Table 3. Summary of analysis of variance for injury, field bindweed control, yield, and pruning weights for each site.
a Asterisks represent P-values as follows: * ≤0.05; ** <0.01; *** <0.001; >0.05 NS.
Crop Injury
No canopy injury was observed in California or Oregon at any observation date, both years, and no treatment effects were detected (P > 0.60) (Table 3; Figure 1, A–D; Figure 1, Q–T). In contrast, grapes in Michigan, New Jersey, and New York exhibited significant injury that was affected by treatment, observation date, and the interaction between treatment and observation date (P < 0.003) (Table 3; Figure 1, E–P). Characteristic symptoms included leaf margin chlorosis, mild cupping, and distortion (Figure 2, A–C), consistent with quinclorac’s auxinic mode of action (Grossmann and Kwiatkowski Reference Grossmann and Kwiatkowski2000). In 2021, significant injury (10 to 20%) was observed at 7 and 14 DAT-A in both Michigan and New Jersey (Figure 1, E and I). At 28 DAT-A, observed injury estimates ranged from 5% to 10% across all treatments in New Jersey. Grape canopy injury in Michigan rose to >20% where quinclorac was applied at 0.84 kg ha-1. In New York, 10% to 20% canopy injury was observed at 28 DAT-A (Figure 1M). Between 0 and 14 DAT-B, canopy injury in Michigan ranged from approximately 20% to 30% where quinclorac was applied at 0.84 kg ha−1; injury from the 0.42 kg ha−1 rate was estimated at 10% to 20% (Figure 1F). In New Jersey, injury peaked at approximately 20% at 14 DAT-B, falling to approximately 10% or less at 28 DAT-B (Figure 1J). In New York, all quinclorac treatments injured grape vine canopies from 10% to 20% at 0 DAT-B to 14 DAT-B, falling to approximately 10% or less 28 DAT-B (Figure 1N).
Figure 1. Grapevine injury after quinclorac applications in five U.S. vineyards. Graphs show the time-course of visible grapevine injury recorded every 7 d up to 28 d after the first in-season application (DAT-A) and 28 d after the second in-season application (DAT-B) in 2021 (left) and 2022 (right). Application A occurred ∼90 d before harvest, and Application B ∼60 d before harvest. Data from post-harvest (Application C, prefrost) were not included but are captured in the early 2022 assessment. Results are shown by site: California (panels A–D), Michigan (E–H), New Jersey (I–L), New York (M–P), and Oregon (Q–T). Within each site and rating date, means followed by the same lowercase letter are not significantly different (Tukey’s HSD, α = 0.05); “ns” indicates no significant differences among treatments.
Figure 2. Representative grapevine responses to quinclorac. Symptoms photographed in New York included (A) cupping and interveinal chlorosis on actively growing shoots, (B) leaflet distortion, and (C) strap-leaf formation. Canopy views from Oregon in 2022 show (D) nontreated vines and (E) vines treated three times in 2021 with quinclorac at 0.84 kg ha−1. In all cases, vines remained vigorous, yielded comparably, and displayed no lasting auxinic injury symptoms. Treatments consisted of either two applications (Timings A + B: ∼90 and ∼60 d before harvest) or three applications (Timings A + B + C: two in-season plus one post-harvest before frost).
In New York, early-season symptoms were noted on basal suckers and in new canopy growth prior to quinclorac applications in 2022, indicating that effects from the first year (2021) persisted into the second season. Injury was numerically, but not statistically, higher where post-harvest quinclorac treatments were applied (data not shown). The greatest amounts of injury (10% to 20%) observed at the New York site occurred from 7 DAT-A to 28 DAT-B, where quinclorac was applied at the highest rate for a total of 5.04 kg ai ha−1 (Figure 1, O and P). Injury from all other treatments was less than 10%. In Michigan, injury was significantly influenced by quinclorac treatment across all rating dates. In general, injury was greatest (10% to 30%) where grapes received a total of 2.52 and 5.04 kg ai ha−1 over 2 yr (Figure 1, G and H). Despite minimal observed injury in 2022 (Figure 1, K and L), cordon dieback was observed in spring 2023 on New Jersey vines that had received quinclorac applications for more than 2 yr of the study (data not shown), suggesting latent or cumulative injury.
A linear regression model including cumulative rainfall (0 to 56 DAT-A), COM and treatment effects (with interactions) explained 52% of the variation in injury (R 2 = 0.52, P < 0.001). Rainfall was the dominant factor (P < 0.001), and its interaction with COM was also highly significant (P < 0.001). Regression coefficients indicated that every 10 mm of rainfall increases the risk of injury by approximately 0.3% in 10% COM soils, but by 0.9% in >20% COM soils. These results are visualized in the fitted response surface (Figure 3A) and delineate a “safe zone” of <10% injury under either low rainfall or high COM and an injury threshold surpassed when intense rainfall co-occurs with sandy or low-COM soils, conditions that occurred in Michigan and New Jersey. The cumulative rainfall (Figure 3C) corroborated this pattern. Plots in California and Oregon remained almost rain-free from 0 to 56 d after application A and stayed within the “safe zone” in both years. It is worth noting that the Oregon vineyard was entirely rain-fed, while the California vineyard received drip irrigation at approximately 12 to 36 mm per week during the growing season. Because drip irrigation delivers water slowly and locally to the root zone and does not produce the large soil-moisture pulses typical of rainfall, it likely had minimal impact on the rainfall-injury relationship in our study. The vineyards in New York received 250 mm of rain in 2021 and exhibited moderate injury, whereas in 2022, with 85 mm of rain, injury levels were lower. Grapes in Michigan and New Jersey exhibited the highest injury levels in 2021, corresponding to the greatest accumulated rainfall of 200 mm and 320 mm, respectively. In 2022, they received less rainfall in the first 56 d after treatment—132 mm in Michigan and 95 mm in New Jersey—and exhibited lower injury levels than the previous year.
Figure 3. Factors explaining differences in quinclorac injury across sites. (A) Response surface from a mixed linear regression predicting maximum grapevine injury (%) as a function of cumulative rainfall during 0–56 d after application A (DAT-A; y-axis), which corresponds to the combined periods of 0–28 DAT-A and 0–28 d after application B (DAT-B), and the combined clay + organic-matter (COM) content of the surface soil (x-axis). The color scale shows model predictions, and points indicate observed site-year combinations (green = 2021, red = 2022); warmer colors correspond to greater injury (n = 148). (B) Principal component analysis of maximum injury, rain, and COM. PC1 (67% of the variance) increases with rain and injury and decreases with clay + organic matter, separating the wet/light Midwest–Mid-Atlantic scores (olive triangles, green squares, and blue crosses) from the dry/heavy Pacific-coast scores (purple squares and blue circles). PC2 (24%) captures the remaining spread in soil properties. Arrows show variable loadings, and 95% confidence ellipses group site-years. (n = 148) (C) Cumulative rain curves for each site and year illustrate the contrast between arid California and Oregon and the humid summers of Michigan, New Jersey, and New York.
The mechanistic basis of this relationship aligns with soil adsorption dynamics. As a weak acid herbicide, quinclorac binds readily to clay particles and soil organic matter (Rave et al. Reference Rave, Mendes, Delgadillo, De Paula, De Aguiar, Da Silva and HH2021), reflected by its relatively high soil adsorption coefficient (Koc = 446 mL g−1) (Adams and Lym Reference Adams and Lym2015). In soils with high COM, this strong adsorption limits herbicide mobility, keeping it near the surface and reducing its movement into the root zone, thereby limiting the potential for grapevine uptake (Bailey and White Reference Bailey, White, Gunther and Gunther1970). However, repeated rain can promote desorption and downward movement, increasing root exposure to the herbicide. In contrast, in coarse-textured soils with low COM, quinclorac can leach more rapidly following rainfall, leading to root-zone uptake by grapevines and subsequent auxinic injury symptoms (Lym Reference Lym2016). Soil pH is also a critical factor influencing quinclorac behavior, as its ionization state depends on pH (Corbin et al. Reference Corbin, Upchurch and Selman1971; Curran 2014). Quinclorac has a pKa of 4.34 (Corbin et al. Reference Corbin, Upchurch and Selman1971), meaning it exists primarily in its anionic form under most field conditions. While all soils in this study had pH values above the pKa, limiting the likelihood of pH-driven changes in ionization, variability in pH could have influenced adsorption or mobility indirectly through interactions with soil colloids or cation concentrations. These site-level differences likely contribute to the variation in observed grapevine injury. Similar interactions have been reported with dicamba and 2,4-D, where crop injury is influenced by soil characteristics and rain (Friesen Reference Friesen1965; Jamshidi et al. Reference Jamshidi, Salehian, Babanezhad and Rezvani2022).
The PCA of maximum injury, rainfall from 0 to 28 DAT-B, and COM further supported these patterns (Figure 3B). PC1 captured 67% of the variation and had loadings of 0.64 for rain, 0.51 for injury and −0.57 for COM, separating the wet, low-COM Mid-Atlantic/Midwest site-years from the dry, high-COM Pacific sites. The negative COM loading is not inconsistent with the regression findings; rather, it reflects a contingent pattern in which the heaviest soils (e.g., in Oregon) coincided with minimal rainfall, while the highest rainfall occurred on the lightest soils. PC2 explained about 24% of the variance and captured residual variation in COM. PC3 (not shown) captured unexplained differences (9%) likely related to cultivar or rootstock variation across sites. Differences in scion-rootstock combinations, grapevine age, and training systems are factors known to influence herbicide tolerance (Haring et al. Reference Haring, Ou, Al-Khatib and Hanson2022). Vineyards in humid regions (Michigan, New Jersey, and New York) clustered in the high-injury quadrant, while those where summers are drier (California and Oregon) aligned with minimal injury and higher COM. The consistency observed between the PCA and regression analyses supports the conclusion that rain and COM primarily determine crop safety. Rain likely influences the movement of quinclorac through the soil profile, increasing the depth and speed of leaching and thereby placing more herbicide in proximity to grapevine roots, where uptake is more likely.
Temperature and humidity also differed among sites following quinclorac applications (data not shown) and may have influenced herbicide behavior. More humid post-application periods may have increased volatilization and uptake, compounding rainfall-driven injury (Ouse et al. Reference Ouse, Gifford, Schleier, Simpson, Tank, Jennings, Annangudi, Valverde-Garcia and Masters2018). Conversely, cooler and drier conditions after application likely led to reductions in volatilization and uptake, thereby reducing visible injury. Such interactions between environmental and edaphic factors highlight the complex dynamics underlying grapevine response to quinclorac.
Despite observed injury patterns, quinclorac had no significant effect on grapevine productivity (Figure 4). Yields ranged from 7 to 15 kg per vine across sites, and yields from vines that received quinclorac treatments did not differ from those of the weed-free controls (Figure 4A). Dormant pruning weights echoed the yield response (Figure 4B) in the sites where those data were collected. No significant differences in pruned biomass were observed among treatments in Michigan, New Jersey, or Oregon. Yield and pruning weights varied among years at some sites (Table 3), likely reflecting environmental conditions rather than herbicide effects. For example, less rain in New York during 2022 (Table 1) may have contributed to lower vine productivity and reduced vine growth. Photographs taken at the Oregon site (Figure 2, D and E) confirm unaffected canopy density and leaf area even after three applications, consistent with the zero-injury ratings recorded there.
Figure 4. Grapevine performance following two or three quinclorac applications per year. (A) Yield per vine for each site-year. (B) Dormant-season pruning weight, an indicator of vegetative vigor. Treatments consisted of either two applications (Timings A + B: ∼90 and ∼60 d before harvest) or three applications (Timings A + B + C: two in-season plus one post-harvest before frost) applied annually over two years. “ns” indicates no significant differences among treatments (Tukey’s HSD, α = 0.05).
Overall, these results suggest that rain after treatment is the primary determinant of crop injury, and no differences were noted among the tested rates or the number of applications. In high-COM soils, even three applications of quinclorac caused minimal injury in the years when post-spray rainfall was negligible. Conversely, under low-COM soils combined with high rain levels, injury increased significantly. Prediction models suggested that high-COM soils, while typically protective, present a greater risk of quinclorac injury once cumulative rain amounts exceed 50 mm within 56 d after application, as quinclorac desorbs and moves to the root zone. In low-COM soils, a single heavy rain may be enough to leach the herbicide into the grapevine root zone and be taken up at sufficient levels to cause auxinic symptoms. These results support the use of site-specific management decisions, such as delaying in-season applications or reducing the rate if rain of >50 mm is forecasted on any type of soil.
Field Bindweed
Quinclorac efficacy against field bindweed varied by site and year but was comparable to or better than results reported for other crops (Figure 5). Regardless of rate, in Michigan, quinclorac provided 70% to 75% control at 7 DAT-A in 2021. The control increased above 90% by 14 DAT-B, statistically similar to that of the nontreated weed-free control. In 2022, regardless of application rate or number, quinclorac provided >75% field bindweed control at 7 DAT-A. By 14 DAT-A and until the last evaluation timing, every quinclorac treatment provided statistically similar field bindweed control as the nontreated weed-free control. At 28 DAT-A, every quinclorac treatment provided >95% control and held that level through the final rating.
Figure 5. Field bindweed control after quinclorac applications from the first in-season spray (Day 0) through 28 d after Application A (DAT-A) and 28 d after Application B (DAT-B). Results are shown for the two sites with uniform, quantifiable bindweed infestations: Michigan in 2021 (A–B) and 2022 (C–D), and New Jersey in 2021 (E–F) and 2022 (G–H). Within each site and rating date, means sharing the same lowercase letter are not significantly different (Tukey’s HSD, α = 0.05); “ns” indicates no significant differences among treatments.
In contrast, poor control was observed in New Jersey in the first year, with all treatments providing <20% control, regardless of application rate or timing (Figure 5, E–H). However, the same site showed significant improvement the following year (2022). The post-harvest application (Timing C) improved overall field bindweed control in 2022. The 0.84 kg ha−1 rate with a post-harvest application resulted in >80% control at 28 DAT-A. Similarly, the 0.42 kg ha−1 rate, combined with a post-harvest application, provided comparable control to the 0.84 kg ha−1 rate at 28 DAT-B and outperformed treatments without a post-harvest application.
These results are consistent with those of prior studies. Enloe et al. (Reference Enloe, Nissen and Westra1999), Grossmann and Kwiatkowski (Reference Grossmann and Kwiatkowski2000), and Moretti and Peachey (Reference Moretti and Peachey2022) reported >85% control of field bindweed in rice, highbush blueberry, turfgrass, and corn with quinclorac, particularly when the herbicide was applied during active growth and followed by rain or irrigation. Its effectiveness has been attributed to root uptake and systemic translocation, leading to ethylene overproduction that disrupts meristematic activity and suppresses regrowth (Grossmann and Kwiatkowski Reference Grossmann and Kwiatkowski2000). The significant improvement in field bindweed control observed in 2022 in New Jersey compared to 2021 likely reflects cumulative stress to the plant, particularly the root system, resulting from sequential quinclorac applications. This stress may have reduced field bindweed regrowth ability. Our data suggest that effective field bindweed management with quinclorac requires 1) a post-harvest application and 2) applications over at least two consecutive years. The post-harvest spray coincides with carbohydrate movement to storage organs, which helps the herbicide reach roots, rhizomes, and other underground tissues (Peterson and Stahlman Reference Peterson and Stahlman1989). The second-year application may enhance efficacy by targeting plants with depleted carbohydrate reserves (Willeke et al. Reference Willeke, Krähmer, Claupein and Gerhards2015).
In summary, this study provides multistate field evidence that quinclorac has strong potential for field bindweed management in vineyards. However, quinclorac efficacy on field bindweed and grapevine safety varied by year and site. Variable levels of grapevine injury were observed, particularly under low-COM soils and high precipitation. Although yield reductions were not statistically different from those of the nontreated weed-free controls, the duration of the field studies may not have been sufficient to detect subtle effects. Therefore, under the evaluated rates and timings, quinclorac poses an unacceptable risk of grape injury in many environments, and further research is needed before recommending its registration for use on grape. Future work should evaluate tank-mix compatibility, interactions with irrigation practices, and reduced-rate and alternative application timing strategies to mitigate quinclorac injury in grapevines.
Practical Implications
Quinclorac may offer a useful addition to the under-vine weed management toolbox in grape, particularly where field bindweed is a major problem. However, its safety depends on site-specific conditions, especially soil characteristics and post-application rainfall patterns. In vineyards with heavier soils that contain more clay and organic matter (COM >30%), quinclorac caused no injury in grapevines. In Michigan, two in-season applications at 0.42 kg ai ha−1 consistently provided effective field bindweed control. Efficacy varied across sites and years, as observed in New Jersey. When cumulative rainfall in the first 2 mo after application is less than 50 mm, the risk of crop damage is minimal (<5%), allowing growers to target actively growing field bindweed without significant risks to grapevine health. In contrast, sandy or low-COM soils are more vulnerable to rainfall-driven herbicide mobility and vine uptake, which increases the risk of injury. In irrigated vineyards using sprinkler systems, irrigation can simulate rainfall-driven leaching and increase the risk of root-zone uptake. Although injury symptoms were transient and did not reduce yield or vine vigor in the short term, repeated high-dose exposure under conducive conditions might have cumulative effects. Therefore, quinclorac is not yet a viable herbicide for grapes, and further research is needed to determine whether its use can be optimized.
Funding
This research was supported through the U.S. Department of Agriculture IR-4 project.
Competing Interests
The authors declare they have no competing interests.
Open access
