Introduction
Interest in the production of species of bunch grape [winegrape (Vitis vinifera L.), fox grape (Vitis labrusca L.)], and muscadine (Vitis rotundifolia Michx.) across the southern United States continues to grow (USDA 2017). Management of weeds is an important part of vineyard maintenance. If left unmanaged, vineyard weeds can reduce airflow, impede labor efficiency, harbor unwanted pests, reduce yield, and alter fruit quality (Agusti-Brisach et al. Reference Agusti-Brisach, Gramaje, León, Garcia-Jiménez and Armengol2011; Basinger et al. Reference Basinger, Jennings, Monks, Mitchem, Chaudhari, Heitman, Havlin, Howard and Spayd2018; Wistrom and Purcell Reference Wistrom and Purcell2005). Weed control in vineyards is an important factor in the establishment of young grape vines, and weeds can continue to compete with established vines for water and nutrients (Celette et al. Reference Celette, Findeling and Gary2009; Holland et al. Reference Holland, Howard, Heitman, Sauer, Giese, Sutton, Agam and Havlin2014). Concerns over the development of resistant weed biotypes have pushed growers to seek ways to minimize factors contributing to herbicide resistance (Alcorta et al. Reference Alcorta, Fidelibus, Steenwerth and Shrestha2011; Beckie and Harker Reference Beckie and Harker2017). Although no resistant weed biotypes have been identified in southern US vineyards, resistant biotypes of horseweed [Conyza canadensis (L.) Cronq.] and junglerice [Echinochloa colona (L.) Link] to EPSP synthase inhibitors, and multiple resistance of hairy fleabane [Conyza bonariensis (L.) Cronq.] to EPSP synthase inhibitors and PSI electron diverters have been reported in other regions (Heap Reference Heap2018). Additionally, Italian ryegrass [Lolium perenne ssp. multiflorum (Lam.) Husn.] with resistance to glutamine synthase inhibitors and biotypes with multiple resistance to ACCase inhibitors, PSI electron diverters, and EPSP synthase inhibitors have been confirmed. To mitigate selection pressure and limit the development of resistance in vineyards, herbicides with additional modes of action must be integrated into vineyard weed management programs.
Vineyard weed management programs often rely on tank mixes of PRE and POST herbicides (Nita et al. Reference Nita, Brannen, Burrack, Pfeiffer, Mitchem, Lockwood and Sial2017). However, many PRE herbicides registered for use in grape are limited in weed control spectrum or have a short residual that does not provide season-long weed control (Abit et al. Reference Abit, Rainbolt, Krutz, Shaner and Hanson2012; Kadir and Al-Khatib Reference Kadir and Al-Khatib2006). Indaziflam, a Group 27 cellulose biosynthesis inhibitor, provides an additional mode of action that is not widely utilized in vineyards in the southern United States. Indaziflam has broad-spectrum residual activity on annual grasses such as crabgrass species [Digitaria sanguinalis (L.) Scop. and D. ischaemum (Schreb.) Shreb. ex Muhl.], goosegrass [Eleusine indica (L.) Gaertn.], Italian ryegrass, foxtail species [Setaria faberi Herrm., S. viridis (L.) P. Beauv., S. pumila (Poir.) Roem. & Schult.], and many broadleaf species including pigweed species (Amaranthus blitoides S. Wats., A. retroflexus L., A. spinosus L., A. hybridus L.), buckhorn plantain (Plantago lanceolata L.), spurge species (Euphorbia maculata L., synonym E. supina Raf.), and morningglory species (Ipomoea hederacea Jacq., I. lacunosa L.) (Anonymous 2017).
Thus, interest exists in integrating indaziflam in current vineyard weed management programs to provide season-long residual control of weeds and an additional mode of action in vineyard weed management programs. However, there has been some concern about crop safety due to indaziflam label restrictions (Anonymous 2017). Label restrictions include the placement of adequate soil barriers between the soil surface and the major portion of the grape root system. At the time of study establishment, registration restrictions also required that grape vines be at least 5 yr old and be planted 30 cm deep, or have a 30-cm barrier between the soil surface and the major portion of the root system (Anonymous 2012). Additionally, muscadine grape, which is grown widely in the southern United States, and a subgenus of Vitis rotundifolia grown in the region, are not mentioned on the indaziflam label. The authors know of no research that has been conducted to determine the effect of indaziflam on muscadine grape. Therefore, a study was conducted to determine the effect of indaziflam rate and application timing on non-bunching muscadine and bunch grape (tolerance, yield, fruit quality).
Materials and methods
The study was conducted from 2015 to 2016 at commercial vineyards across North Carolina. Grape types included fresh-market muscadine (FM) ‘Triumph’ and ‘Summit’, processing muscadine (PM) ‘Carlos’, and a processing/table bunch grape (SUN) ‘Sunbelt’. Each grape type was grown at two locations, for a total of six locations (Table 1). All vines were healthy and at least 5 yr old. FM, PM, and SUN vines were spur pruned and trained to a bilateral cordon on a single high-wire trellis system (Carpio et al. Reference Carpio, Safley and Poling2008). Between-row spacing was 3 m for all locations. In-row vine spacing was 6.1 m except for SUN, which was 2.4 m. Plot size varied according to grape type and was 1.2 m by 6.1 m for SUN; 1.2 m by 12.2 m for FM, and 1.2 m by 18.3 m for PM. Studies were a randomized complete block design with three replications for FM (‘Summit’) or four replications for all other cultivars.
Location, grape type, and soil information for study sites investigating indaziflam rate and timing in muscadine and bunch grape in 2015 and 2016
a Grape types included a processing/table bunch grape (SUN), fresh-market muscadine (FM), and a processing muscadine (PM).
All herbicide treatments were applied to the vegetation-free strip beneath the planted vine row using a CO2-pressurized backpack sprayer with one 8003EVS nozzle (Teejet® Technologies, 200 W. North Ave., Glendale Heights, IL) calibrated to deliver 187 L ha–1 at 206 kPa. Each treatment was applied to both sides of the vegetation-free strip and included an inert blue dye to minimize overlap. Treatments included indaziflam at 0, 50, 73 g ai ha–1 (Alion®; Bayer Crop Science, P.O. Box 12014, 2 T.W. Alexander Dr., Research Triangle Park, NC) or flumioxazin (213 g ai ha–1) (Chateau Herbicide SW®; Valent USA Corp., Walnut Creek, CA) applied alone in April, and sequential applications of indaziflam (36, 50, 73 g ai ha–1) or flumioxazin at 213 g ai ha–1 applied in April followed by the same rate applied in June (Table 2). Early PRE treatments were applied at Eichhorn-Lorenz stage 15 (eight leaves separated, flowers in compact groups) (Coombe Reference Coombe1995). Sequential late PRE applications were made 8 wk after the early PRE application in June when vines were at the Eichhorn-Lorenz stage 32 (beginning of fruit cluster closure). All early PRE indaziflam and flumioxazin treatments included oryzalin PRE at 4,480 g ai ha–1 (Surflan AS®; United Phosphorus, Inc., 630 Freedom Business Center, Suite 402, King of Prussia, PA) for PRE grass control, and glufosinate at 986 g ai ha–1 (ReckonTM 280SL Herbicide; Solero ATO, LLC, P.O. Box 21720, Mesa AZ) was included in all herbicide treatment applications to control emerged weeds. A weed-free check was also included for comparison and was established with an initial application of glufosinate and oryzalin and then maintained weed-free with hand removal.
Growth, yield and yield parameters, and berry quality of ‘Summit’ and ‘Triumph’ muscadine grape (FM), as influenced by indaziflam grown in Vale and Kings Mountain, NC, in 2015 and 2016
a Glufosinate (986 g ai ha–1) was included with all treatments.
b Oryzalin at 4,480 g ai ha–1 was included with all indaziflam and flumioxazin early PRE treatments and the weed-free check.
c Means for vine growth parameters, yield components, and berry quality did not have significant interactions of year, variety, and location. Yield data for ‘Summit’ and ‘Triumph’ were combined and are reported over location, year, and variety. Means within the same column followed by the same letter are not different according to Tukey’s HSD (P ≤ 0.05 level).
d Trunk cross-sectional area (TCA) measured parallel and perpendicular to the vine row 0.3 m above the soil surface, and calculated based on the area of a circle using the average diameter calculated from the parallel and perpendicular measurements.
e Titratable acidity (TA). Samples were titrated to a pH of 8.2 using 0.1 N NaOH to determine g of tartaric acid L–1 of juice.
Crop injury was assessed based on visual estimates on a scale from 0 (no crop injury) to 100% (crop death) (Frans et al. Reference Frans, Talbert, Marx, Crowley and Camper1986). Assessments of crop injury were conducted at 1, 2, 4, and 8 wk after application of the early PRE application and again at 1, 2, 4, and 8 wk after application, after the sequential herbicide application.
Vine growth parameters including cordon length and trunk cross-sectional area (TCA) were measured (Parker and Meyer Reference Parker and Meyer1996; Smart and Robinson Reference Smart and Robinson2001). Cordon length was measured from the dormant end of one arm of the cordon to the dormant end of the corresponding cordon of the same vine. To measure TCA, a measurement parallel and perpendicular to the vine row was made 0.3 m above the soil surface using a digital caliper (Mitutoyo 500 196–30, 965 Corporate Blvd., Aurora, IL). TCA was calculated based on the area of a circle using the average diameter calculated from the parallel and perpendicular measurements.
To calculate yield, berry number and cluster number were counted for two 1-m sections of each cordon for muscadine (FM, PM). Yield in kilograms per hectare was calculated by multiplying the average yield (kg) of the two 1-m sections of cordon by the length of cordon for each vine, resulting in yield per vine. Yield per vine was then multiplied by the number of vines in a hectare based on the vineyard planting density to calculate yield in kilograms per hectare. For SUN, clusters per vine were counted prior to harvest, and 20 clusters per plot were weighed to determine average cluster weight. Yield for SUN was calculated by multiplying average cluster weight by cluster number and then by vines per hectare. For all locations, 100 berries per plot were placed into sealable polypropylene bags and placed in coolers containing gel ice packs for transport between the field site and Raleigh, NC. To determine average berry weight, 100 berry samples were weighed. Fruit samples were stored in a –20 C freezer until analysis. In September 2015 and 2016, frozen grape samples were thawed to room temperature and blended in an Oster® 10-Speed Blender (Sunbeam-Oster Co., Inc., Fort Lauderdale, FL). The pulp was filtered through a coarse fluted filter (Qualitative P8 24 cm; Fisher Scientific, Pittsburgh, PA). Soluble solids concentration (SSC) was measured using a digital refractometer (Refracto 30px; Mettler Toledo, Columbus, OH) and expressed in °Brix. Grape pH was measured using a pH electrode (HI9125; Hannah Instruments, 584 Park East Dr., Woonsocket, RI) and pH meter (PHi 510; Beckman Coulter, Life Science Division Headquarters, 5350 Lakeview Parkway S Dr., Indianapolis IN) standardized to pH 4, 7 and 10. A 5-ml aliquot of juice was placed in a 6-ml centrifuge tube and placed in a freezer at –20 C for analysis of titratable acidity (TA). In January 2016 and 2017, the frozen subsamples were placed in a cooler with ice packs and transported to the Plants for Human Health Institute, Kannapolis, NC, for processing. Subsamples were thawed at room temperature and then shaken for 10 s to re-suspend any settled solids, at which time 2 ml of juice was placed in 60 ml of distilled water. Samples were titrated to a pH of 8.2 using 0.1 N NaOH and an automated titrimeter (Metrohom 862 Compact Titrosampler, 6555 Pelican Creek Circle, Riverview, FL) and reported in grams of tartaric acid per liter of juice (Spayd et al. Reference Spayd, Wample, Evans, Stevens, Seymour and Nagel1994).
Response variables for yield and yield components (berries per cluster, average berry weight, berry number m–1 cordon and cluster number m–1 cordon, and yield ha–1) and berry quality components (SSC, pH, and TA) for SUN and FM were analyzed using the GLM procedure (SAS 9.4; SAS Institute, Cary, NC), with year by treatment as the error term. Location by treatment was used as the error term for response variables in PM as a result of the total crop loss in 2016. Data were analyzed by muscadine type (FM or PM) to account for cultivar differences. For SUN, response yield per hectare had a treatment-by-year effect and was analyzed by year. Cluster weight, berries per cluster, cluster number m–1 cordon, average berry weight, SSC, pH, and TA for SUN did not have a treatment-by-year effect and were combined over years and analyzed using the GLM procedure and using location by treatment as the error term. The yield for FM, PM, and SUN was log transformed to increase variance homogeneity. Means reported for yield are not transformed. Differences for response variables between treatments were determined using Tukey’s HSD at a level of P ≤ 0.05.
Results and discussion
Crop injury
Crop injury was not observed at any location (data not shown), and cordon length or TCA were not affected by herbicide treatments (Tables 2–4). The maximum registered rate per season of indaziflam for grape is 73 g ha–1. Even when applied sequentially at rates above 73 g ha–1, no visible injury was observed on FM, PM, and SUN. Injury from indaziflam has been reported in other crops such as pecan [Carya illinoinensis (Wangenh.) K. Koch)]. However, injury to pecan was due to a combination of production practices such as flood irrigation and high soil pH, resulting in indaziflam being more mobile in the soil profile (Jefferies and Gannon Reference Jefferies and Gannon2016; Gonzalez-Delgado et al. Reference Gonzalez-Delgado, Ashigh, Shukla and Perkins2015; Jhala and Singh Reference Jhala and Singh2012). Soils in southeast vineyards typically have acidic pH (Havlin et al. Reference Havlin, Hardy, Ghel and Spayd2012), and irrigation is typically applied through drip emitters or microsprinklers to reduce the potential for soil mobility of indaziflam. However, vines that are on soils high in gravel content (>20%) and low in organic matter, and vines that are <3 yr old or have much of their root system <15 cm deep may be susceptible to crop injury (Anonymous 2017). Vineyard locations in this study represent varied soil types and growing conditions where grapes are grown in North Carolina. All locations in this study had vines older than 5 yr planted in soil types that would minimize the potential for crop injury.
Growth, yield and yield parameters, and berry quality of ‘Carlos’ muscadine grape (PM), as influenced by indaziflam grown in Vale and Kings Mountain, NC, in 2015a
a Data for PM are only for 2015 as a result of a total crop loss in 2016 due to a late-spring freeze.
b Glufosinate (986 g ai ha–1) was included with all treatments.
c Oryzalin at 4,480 g ai ha–1 was included with all indaziflam and flumioxazin early PRE treatments and the weed-free check.
d Means within the same column followed by the same letter are not different according to Tukey’s HSD (P ≤ 0.05 level).
e Trunk cross-sectional area (TCA) measured parallel and perpendicular to the vine row 0.3 m above the soil surface, and calculated based on the area of a circle using the average diameter calculated from the parallel and perpendicular measurements.
f Titratable acidity (TA). Samples titrated to a pH of 8.2 using 0.1 N NaOH to determine g of tartaric acid L–1 of juice.
Yield and yield components
Treatments did not affect yield, berries per cluster, average berry weight, cluster number m–1 cordon, and yield ha–1 for FM and PM relative to the weed-free check (Tables 2 and 3). Yields for all varieties were within the normal range for FM and SUN in 2015 and 2016, and PM in 2015 (Tables 2–4) (Streigler et al. Reference Streigler, Carter, Morris, Clark, Threlfall and Howard2005). Yield and yield parameters for PM could not be measured in 2016, as a late-spring frost had resulted in a total crop loss.
Treatments did not affect berries per cluster or average berry weight. Cluster number m–1 was highest in the oryzalin early PRE treatment. However, none of the indaziflam or flumioxazin treatments were different from the weed-free treatment (Table 4). SUN yield was not reduced by any herbicide treatments compared to the weed-free check in 2015 and 2016. Yield response to treatments was not consistent across years. In 2015, sequential applications of indaziflam at 73 g ha–1 resulted in the highest yield that was greater than a single application of flumioxazin at 213 g ha–1 or indaziflam at 50 g ha–1 and greater than the weed-free check. In 2016, indaziflam early PRE at 73 g ha–1 had the greatest yield, whereas flumioxazin followed by flumioxazin had the lowest yield. However, flumioxazin followed by flumioxazin treatment was not different from the weed-free check nor the grower standard treatment of a single application of flumioxazin at the early PRE timing. Results indicate that labeled rates of indaziflam at 73 g ha–1 affected yield similarly in each year as the grower standard of a single application of flumioxazin at the early PRE timing and were not different in either year from the weed-free check.
Growth, yield and yield parameters, and berry quality of of ‘Sunbelt’ (SUN) bunch grape, as influenced by indaziflam grown in Vale and Kings Mountain, NC, in 2015 and 2016
a Glufosinate (986 g ai ha–1) was included with all treatments.
b Oryzalin at 4,480 g ai ha–1 was included with all indaziflam and flumioxazin early PRE treatments and the weed-free check.
c Means for vine growth parameters, yield components, and berry quality did not have significant interactions of year and location and are combined over year and location. Means within the same column followed by the same letter are not different according to Tukey’s HSD (P ≤ 0.05 level).
d Trunk cross-sectional area (TCA) measured parallel and perpendicular to the vine row 0.3 m above the soil surface, and calculated based on the area of a circle using the average diameter calculated from the parallel and perpendicular measurements.
e Year-by-treatment interactions were significant for SUN yield grown at Vale and Kings Mountain, NC. Data are reported by year combined over locations and is reported as non-log transformed means.
f Titratable acidity (TA). Samples titrated to a pH of 8.2 using 0.1 N NaOH to determine g of tartaric acid L–1 of juice.
Fruit quality
In both years, herbicide treatments had no effect on muscadine grape or SUN fruit quality parameters (Tables 2–4). Other studies have examined the effect that weeds and in-season groundcover have on fruit quality (Basinger et al. Reference Basinger, Jennings, Monks, Mitchem, Chaudhari, Heitman, Havlin, Howard and Spayd2018; Hickey et al. Reference Hickey, Hatch, Stallings and Wolf2016; Meyers et al. Reference Meyers, Jennings, Monks and Mitchem2014; Sanguankeo et al. Reference Sanguankeo, Leon and Malone2009). In the aforementioned studies, increased weed or groundcover presence resulted in higher concentration of soluble solids. Others have reported changes in anthocyanin concentration due to groundcover management (Wheeler et al. Reference Wheeler, Black and Pickering2005). However, pH is not affected by the presence of groundcover (Hatch et al. Reference Hatch, Hickey and Wolf2011; Wheeler et al. Reference Wheeler, Black and Pickering2005). Lack of treatment influence on fruit quality in our study is probably due to the high level of weed control, as increases in soluble solids in these reported studies were due to weed competition with vines.
These results indicate that indaziflam did not reduce yield or berry quality in muscadine (‘Carlos’, ‘Summit’, and ‘Triumph’) or SUN. Yield for indaziflam treatments was similar to yield from flumioxazin (grower standard) treatments. This research shows that indaziflam was safe to use on bunch grape (SUN) and muscadine (‘Triumph’, ‘Summit’, and ‘Carlos’).
Growers integrating indaziflam into vineyard weed management programs should consider a mid- to late-season POST herbicide application with an additional residual herbicide such as flumioxazin. This additional application would control emerged weeds missed by indaziflam and extend residual control through the remainder of the growing season and past harvest. Furthermore, the weed complex present in the vineyard should be considered carefully, as indaziflam does not control all problematic vineyard weeds. Additional herbicide tank-mix partners or additional applications should be utilized if these weeds are not adequately controlled by indaziflam. During the course of these studies, the restrictions on the indaziflam label were changed. A soil barrier of 15cm between the surface of the soil and the majority of the root system is now acceptable instead of 30cm. Based on the results of this study, indaziflam is safe to use in muscadine and bunch grapes in the Southeast when following the current indaziflam label. Overall, indaziflam is a safe and viable herbicide option for southern US muscadine and bunch grape and could be integrated by growers into vineyard weed management programs in the Southeast.
Author ORCID
Nicholas T. Basinger https://orcid.org/0000-0002-7317-5835
Acknowledgments
Support for this study was provided by the North Carolina Department of Agriculture. The authors would like to thank muscadine and bunch grape growers, Ervin Lineberger, Carlos Mungia, and John Gurganus for their cooperation and collaboration. The authors would also like to thank Dr. Cavell Brownie for her assistance in data analysis and interpretation. No conflicts of interest have been declared.
