Ozonated Water Spray Does Not Suppress Grapevine Powdery Mildew or Grape Mealybug

yeast (Modesti et al. 2019). In both these laboratory assays and in-field studies, data on the efficacy of ozonated water for season-long manage - ment of agricultural pests and diseases is lacking. In eastern Washington State (United States), disease pressure from grapevine powdery mildew (caused by the fungus Erysiphe necator) is


Introduction
Ozone (O 3 ), in both gaseous and aqueous forms, is a powerful oxidizing agent (Kim et al. 2003).It is recognized as safe for food treatment, storage, and processing (21 CFR §173.368), as it leaves behind no residue, due to its reactive nature (Ikehata and Li 2018).O 3 is an unstable gas that reacts with organic and inorganic compounds, directly or indirectly, through temporary formation of radicals which occur naturally during the self-reaction of O 3 molecules (Gottschalk et al. 2009).O 3 and radical molecules are antimicrobial, though the efficacy of both depends on numerous variables such as the sensitivity and density of the targeted microbe, presence of interfering compounds on the target or in the delivery solution, O 3 concentration, or method of application (Khadre et al. 2001, Guzel-Seydim et al. 2004, Karaca and Velioglu 2007).
Ozonated water offers a direct application process that delays O 3 degradation (Khadre et al. 2001), allowing for broader uses in agricultural pest management.Ozonated water has been shown to reduce plant pathogens and insects under in vitro conditions.These studies have reported increased insect mortality (Keivanloo et al. 2013) and disease suppression of powdery mildews (Fujiwara and Fujii 2002, Fujiwara et al. 2009, Pierron et al. 2015, Ebihara et al. 2016).Even with commercially available ozonated water generators that retrofit to pesticide sprayers, field studies on ozonated water efficacy for commercial pest management are limited.In grape (Vitis spp.) production, such field studies concluded that ozonated water reduced fungal microflora on grape leaves (Raio et al. 2016) and that ozonated water reduced fungal colony forming units, but not those of bacteria and yeast (Modesti et al. 2019).In both these laboratory assays and in-field studies, data on the efficacy of ozonated water for season-long management of agricultural pests and diseases is lacking.
In eastern Washington State (United States), disease pressure from grapevine powdery mildew (caused by the fungus Erysiphe necator) is relatively low due to the semiarid steppe climate, characterized by hot, dry summers and cold winters (USDA Cold Hardiness Zone 7a; https://planthardiness.ars.usda.gov/).With low disease pressure comes the opportunity to evaluate alternative management strategies, such as ozonated water spray (OWS), that may be of limited efficacy if used in climates with higher disease pressure.Management of E. necator can be broken down into three main periods: early season, the critical window, and late season.The critical window is when the developing fruit is most susceptible to infection, two weeks prebloom or rachis elongation (BBCH 57; Lorenz et al. 1995) to four weeks postbloom (BBCH 73 to 77) (Gadoury et al. 2003, Moyer et al. 2016).OWS could be a solution in the early or late season periods, when the risk of management failure is less.Early season use is particularly favorable, as temperatures are low and humidity is greater, slowing ozonated water decomposition (Staehelin et al. 1982, Khadre et al. 2001) and allowing maximum O 3 concentrations to be applied in-field.
Another important pest in eastern Washington State is the grape mealybug, Pseudococcus maritimus.It is a primary vector of Grapevine leafroll-associated virus 3, the predominate virus of grapevine leafroll disease (Rayapati 2011, Bahder et al. 2013a, Donda et al. 2023).Grape mealybug is typically managed at the nymph stage with contact insecticides applied during the delayed dormant period (wooly bud, BBCH 01 to 03) and/or with a systemic insecticide applied in-season after the first male flight (Hoheisel and Moyer 2023).OWS has potential as a residue-free pesticide to control P. maritimus nymphs at the delayed-dormant period, when spring weather conditions are favorable for OWS application.
To better understand how OWS could help manage these common grape pests in eastern Washington, the potential curative effects of ozonated water were evaluated on nascent E. necator colonies, and the acute lethal effects were evaluated on P. maritimus in the lab.These studies were coupled with field evaluations using a commercial OWS retrofit system to manage E. necator and P. maritimus.

Laboratory-scale OWS
A laboratory-scale OWS system was designed to maximize possible O 3 concentration (Figure 1).Water was split from the central tank into two identical lines.Each line had its own diaphragm pump (Model 8000-543-238, Pentair) which forced water through venturi manifolds (Model 287, Mazzei).O 3 gas from two identical O 3 generators (SP-16G, A2Z Ozone Inc.) was introduced through the venturis.
Only one generator was used in most cases, except when the desired O 3 concentration was not achieved using a single generator.A static mixer was installed immediately following each manifold.Following the static mixers, the lines were rejoined in an off-gassing reservoir.This reservoir had a gas relief valve installed at the top of the unit to release excess O 3 , which would otherwise interrupt normal spray operation.Pressurized (0.3 MPa) spray liquid then exited the off-gassing reservoir to the spray nozzle assembly.The spray assembly included a hollow cone nozzle (TXVS-12, TeeJet Technologies) that generated (DV 0.5 ) fine droplet size at 0.3 MPa to deliver 0.65 L/min to biological targets.O 3 concentrations were measured in-line prior to the nozzle orifice exit with a dissolved O 3 transmitter (Model Q45, Analytical Technology Inc.).Biological targets in conducted experiments were placed 18 cm away from the nozzle tip and typically sprayed with solution until just before run-off.

Figure 1
Schematic of the laboratory-scale ozonated water spray (OWS) system.Ozone (O 3 ) produced from two O 3 generators (SP-16G, A2Z Ozone Inc.) is diffused into the water solution through respective venturi manifolds.Excess O 3 gas is released from the off-gassing reservoir so that ozonated water is sprayed through an interchangeable nozzle to deliver the desired volumes.

Laboratory evaluations on the effect of ozonated water on nascent E. necator colonies
The E. necator inoculum for these experiments was cultured on detached Vitis vinifera Chardonnay leaves as described, with modifications (Moyer et al. 2010).If cultured isolates declined, the fungus was re-sourced from field isolates by tapping infected field-sourced leaves onto new, surface-sterilized leaves to transfer conidia.These cultures were then allowed to develop for seven to 14 days.The inoculum age used to develop the conidial suspensions, as described below for each experiment, ranged from sevento 14-day-old colonies.The germination potential of each inoculation solution was assessed immediately before and after each inoculation by transferring a 10-µL sample of the conidial suspension to a glass microscope slide.The slide was incubated in a closed petri dish with moist filter paper for 24 hrs at 22°C, after which the percent germination (i.e., presence of a germ tube at least one-half the length of the conidium) was assessed microscopically on all spores present.In all experiments, germination rates were between 40 and 60%.
Chardonnay leaves used in this study were collected from a vineyard at Washington State University's Irrigated Agricultural Research and Extension Center (WSU IAREC, Prosser, WA), 14 days postfungicide treatment, to ensure no residual fungicides were present.Young leaves were collected at leaf position three from the distal tip and were ~50 to 75% expanded and with a noticeably shinier cuticle than older leaves (Doster andSchnathorst 1985, Merry et al. 2013).Prior to inoculation, leaves were surface-disinfested with a 0.5% sodium chlorite solution for 90 sec and rinsed twice in distilled water.Leaves were placed individually in a double petri dish with the petiole submerged into deionized water and incubated overnight at 22°C (procedure adapted from Moyer et al. 2010).The leaves were inoculated with 10 μL of a conidial suspension (10 4 to 10 6 conidia/mL) in a 0.05% 80 Tween solution.Each leaf was inoculated 10 times (five droplets on each side of the midvein).Droplets dried at room temperature for one hour and the petri dish was then placed in a plant growth incubator (Model 3765, ThermoFisher) at 22°C with a day/night regime of 16:8 hrs until treatments were applied.
The potential curative effect of ozonated water was evaluated using O 3 concentrations of 1 mg/L (low) or 6 to 7 mg/L (high) delivered through the laboratory OWS system described above.Rates were chosen to simulate maximum concentrations seen in our field-scale OWS (1 mg/L), and the highest concentration that could be reached with the laboratory-scale OWS (6 to 7 mg/L).These were compared to three controls: a non-ozonated water control, a no spray (water or O 3 ) control; and a 2% v/v horticultural oil (PureSpray Green, Intelligro).The water used was chlorinated groundwater that meets National Primary Drinking Water Regulations (NPDWR).The water control was sprayed with the laboratory OWS with the ozonated water generator turned off and the recirculating pump on.OWS treatments and water control were applied to the point of just-before-runoff.The horticultural oil control was also applied to the point of just-before-runoff using a hand-held pump sprayer (CHAPIN 16100 Home and Garden one-gallon sprayer; adjustable cone nozzle, 0.3 to 0.4 MPa), with the delivery nozzle placed 18 cm from the adaxial leaf surface.These five treatments were applied to colonies of E. necator either 24 or 72 hrs postinoculation (hpi).In total, there were 10 experimental treatments (five treatment options × two E. necator developmental stages) with 10 detached leaf bioassays per experimental treatment with 10 inoculated colonies per leaf.The number of colonies that developed on each detached leaf bioassay was recorded eight days after inoculation and expressed as a percentage (number of colonies formed/total colonies inoculated).Conidiophore development for each colony was observed at eight days postinoculation to capture sublethal effects of ozonated water treatment.Development was rated as no hyphae, hyphae development but no conidiophore, or hyphae with conidiophore development.This conidiophore development rating was only completed on experiment 1.

Laboratory evaluation of ozonated water from different water sources on P. maritimus nymphs
A rearing method adapted from Blaisdell et al. (2016) was used to obtain P. maritimus first-instar nymphs.Mated third-instar females were collected from a vineyard in Benton City, WA.Once collected from the field, they were transferred to 90-mm petri dishes, with 90-mm Whatman filter paper.The mealybugs were kept in darkened conditions at 22°C in an incubator until oviposition.Eggs were hatched under the same conditions, but with a day/night regime of 16:8 hrs.Once hatched, ~10 first-instar nymphs were transferred to bioassays containing leaf discs on moistened circle cotton rounds, then left in a darkened incubator overnight to be treated the following day.Leaves were obtained from greenhouse-grown Chardonnay vines and five replicate leaves per treatment were used.
The potential lethal effects of OWS were evaluated on P. maritimus first-instar nymphs, using O 3 concentrations of 1 mg/L (low) or 6 to 7 mg/L (high).Rates were chosen to match the OWS system restraints described above and compared to the three controls described earlier.The experiment was repeated six times: three different water sources, repeated twice.Water source 1 (low sediment) was chlorinated groundwater that meets NPDWR, water source 2 (high sediment) was the same chlorinated groundwater filtered through a dirty sediment filtration (AP110 filter, Aqua-Pure), and water source 3 (DI) was distilled water.Water sources 1 and 2 were chosen due to access availability at the time of the experiments.The DI water source was chosen because sediment in water can decrease O 3 concentrations (Khadre et al. 2001).
Treatment effects were measured as mortality at two and 24 hrs posttreatment for the low sediment water source.For the experiments with the high sediment water source, a 48 hr observation point was added.DI water experiments were only observed at 24 and 48 hrs due to previous experimental results.Under a dissecting microscope, nymphs were determined dead if no movement was observed if prodded.Crawlers with any visible leg or antennae movement were considered alive.

Field-scale OWS system
The field-scale OWS was applied using a commercial O 3 generator (AgriOzien) retrofit to an airblast sprayer (Powerblast Pul-tank, Rears Manufacturing) with an injector and recirculating pump.In-tank O 3 concentration was measured with an Ozone Meter (Model I-2019, CHEMetrics Inc.).Oxidation reduction (redox) potential (ORP) of tank samples was also quantified with a meter (ORP-200, HM Digital) as an indicator of bulk reactive oxygen species present in the spray liquid.Field treatments were applied once the tank water solution reached a measured O 3 concentration of 1 mg/L.Typically, ORP values were >700 when an O 3 concentration of 1 mg/L was reached.The OWS application system was calibrated independently for each vineyard location (Hoheisel et al. 2021) and operated within the confines of ISO specifications (ISO 22522:2007).

Field evaluation of OWS system for E. necator management
Efficacy of OWS applied to vines during the prebloom period (BBCH 13 to 57) was evaluated for its potential to control grapevine powdery mildew (E.necator) throughout the season.OWS field trials occurred in an experimental own-rooted Chardonnay vineyard at the WSU IAREC for two growing seasons (2020 and 2021).The vineyard was planted in 2009.Vine spacing was 1.8 × 3.0 m (vine × row), with north-south row orientation.Vines were trellised on a modified vertical shoot-positioning (VSP) system and trained to a dual-trunk bilateral cordon with spur pruning.The vineyard was dripirrigated with natural vegetation undervine and between rows.The vineyard floor was maintained through routine in-season mowing.Field environmental conditions were collected from a WSU AgWeatherNet station ('Prosser.NE'; weather.wsu.edu)located 1.6 km from the vineyard.
The OWS field sprayer was optimized for the vineyard with three hollow cone nozzles (D5 DC25, TeeJet) per side without air-assist (i.e., axial fan was turned off during all applications).This resulted in an application of 702 L/ha.OWS treatment was evaluated against the following controls: a standard fungicide program, a prebloom unsprayed control that transitioned to the standard fungicide program after the prebloom period, and a full-season unsprayed control where no fungicides were applied all season (Table 1).Treatments started at 10 to 20 cm shoot growth (BBCH 13) and continued at either seven-or 14-day intervals.Treatment plots consisted of six consecutive vines in a row, with data collected from the center four vines.Treatments were replicated four times within a randomized block design.To reduce potential drift between treatment plots, tarps were used to cover individual treatment replicates during treatment application in 2020.In 2021, there were six vines serving as buffers between treatments within a row, and an entire non-experimental row between treated rows.
In 2020, the standard fungicide program was applied using an ATV-mounted tank sprayer (ATV2507 ATV Sprayer, WorkHorse Sprayers) at 468 L/ha until 21 May.From 4 June onward, spray was applied with a Rears Manufacturing Powerblast Pul-Tank airblast sprayer at 702 L/ha with D5DC25 TeeJet nozzles, without air-assist.Air-assist was turned on for the last two sprays of the season.In 2021, the standard fungicide program was applied with a Rears custom-built Table 1 Treament application dates and rates for the field evaluation of ozonated water system (OWS) on Erysiphe necator management.OWS and prebloom unsprayed control treatments were applied up to bloom, then converted to the standard fungicide spray program for the remaining season.Fungicides were applied at max label rates unless noted.

Field evaluation of OWS against P. maritimus
In 2021, OWS was evaluated at two separate commercial vineyard locations for its potential to manage grape mealybug populations before budbreak.The treatments were: 1 mg/L OWS, applied once (OWS 1); 1 mg/L OWS, applied twice at one week intervals (OWS 2); and an unsprayed control.Treatments were replicated four times in both vineyards in a randomized block design, using the OWS field system described above.It was optimized with three disccore D5DC45 TeeJet nozzles per side to apply 936 L/ha with no air-assist as a delayed-dormant (BBCH 01 to 05) spray.To reduce potential drift between treatment plots, tarps were used on neighboring treatment rows.Treatments began once mealybug nymphs were found on the dormant spurs through weekly visual scouting starting 22 March 2021.
The first vineyard was located in Outlook, WA and was planted in the early 1990s to own-rooted V. vinifera Nebbiolo.Vine spacing was 1.8 × 3.0 m (vine × row), with eastwest row orientation.Vines were trellised on a VSP system and trained to a dual-trunk bilateral cordon with spur pruning.The vineyard was drip-irrigated with natural vegetation under the vine and between rows; the vineyard floor was maintained through routine in-season mowing and undervine herbicide applications.In this vineyard, treatments were applied to a plot of 10 consecutive vines.OWS week 1 (OWS 1) was applied on 16 April 2021; OWS week 2 (OWS 2), on 16 and 22 April 2021.The second vineyard was located in West Richland, WA and was planted to own-rooted V. vinifera Gewürztraminer in 2001.Vine spacing was 1.8 × 2.7 m (vine × row), with north-south row orientation.Vines were trellised on a VSP system and trained to a dual-trunk bilateral cordon with spur pruning.The vineyard was drip-irrigated, with natural vegetation under the vine and between rows.The vineyard floor was maintained through routine in-season mowing and herbicide application under-vine.In this vineyard, treatments were applied to a plot of 28 consecutive vines.OWS 1 was applied on 9 April 2021; OWS 2, on 9 and 14 April 2021.Environmental conditions were tracked through WSU AgWeatherNet 'Outlook' and 'Red.Mtn.N' weather stations.The 'Outlook' station was 8 km from the Outlook vineyard and the 'Red.Mtn.N' station was 2.2 km from the West Richland vineyard.
In both vineyards, four consecutive vines with noticeable P. maritimus egg sacks within each treatment plot were selected as data-sampling vines.To assess OWS treatment effects, nymph population after the first male flight was monitored using the double-sided tape method, described below.Nymph populations after the first male flight (i.e., the second generation nymphs) have been reported to have increased population size at this time (Grasswitz and James 2008) and move upward within the canopy (Geiger and Daane 2001).This development and movement is primarily due to increased temperature (>21°C) during this lifestage: as temperatures rise, so does nymph mobility (Cornwell 1958, Geiger andDaane 2001).This increased the probability of capture on sticky tapes.Information obtained from verbal communication with vineyard managers who were monitoring male mealybug populations with pheromone-baited traps (Bahder et al. 2013b) indicated that the first male flight started ~11 June 2020, at both locations.This meant that nymphs hatched and began moving around the beginning of July.Sticky tape traps for both locations were set on 1 and 8 July 2021 then collected on 8 and 15 July 2021, respectively.Sticky tape traps were 20 pieces of double-sided tape (Scotch Brand, Minnesota Mining and Manufacturing Company) per treatment replicate, placed on the base of the previous year's spurs (Figure 2) to capture nymphs.Tape traps were kept on the spur to collect nymphs for seven days.After seven days, tape traps were collected, and new tape was placed on the same spur.Once tape trapes were collected, they were brought back to the lab for assessment.The total mealybug nymphs per tape were counted microscopically.

Statistical analysis
Data were analyzed with the JMP statistical program (v. 6.0.0,SAS Institute, Inc.) using the standard least squares model platform.For the detached leaf bioassays evaluating E. necator colonies in each experiment, leaves that died were removed from analysis and treatment replicates were removed at random to the lowest common denominator until each treatment replicate had an equal n.Lab evaluations of OWS for P. maritimus experiments were pooled by water source, determined using a full factorial analysis of experiment and treatment.Foliar and cluster disease ratings for field evaluation of OWS on E. necator management were evaluated by calculating area under the disease progress curve (AUDPC) (Madden et al. 2017).For the field evaluation of OWS on P. maritimus management, each treatment replicate's mealybug population was summed for the 20 sticky tape traps per treatment replicate.Treatments were then evaluated by location and at each time point of collection.The restricted maximum likelihood method was used for all analysis, where replicates were random and treatments were fixed effects with statistical significance set at α = 0.05.Tukey's honest significant difference was used as a post-hoc significance test.

Laboratory evaluations on the effects of ozonated water on nascent E. necator colonies
The three experimental repetitions were analyzed individually, as there were significant differences within experiments (p = 0.0001), even though response trends were similar.Treatments resulted in significant differences in colony development at 24 hpi (p = 0.0001 for all experiments) and at 72 hpi (p = 0.0001 for all experiments), but those results were driven predominately by the efficacy of the oil control application (Figure 3).OWS treatments were not different than the no spray control at reducing E. necator colony formation, regardless of O 3 concentration or timing of treatment at 24 or 72 hpi.OWS treatments did not suppress conidiophore development.All colonies present in all treatments (1 mg/L OWS, 7 mg/L OWS, no spray, and water) at both timings of treatments (24 or 72 hrs) had conidiophores present.

Laboratory evaluation of ozonated water on P. maritimus nymphs
Ozonated low-sediment water had a delayed response on nymph mortality (Figure 4, left).At two hours postexposure (p = 0.0001, not shown on graph), nymphs exposed to 1 or 7 mg/L OWS had greater mortality than no spray nymphs (p = 0.01 and 0.0003, respectively), but the mortality rate of the 1 mg/L OWS treatment was not greater than that of nymphs treated only with water (p = 0.09).

Figure 2
Pseudococcus maritimus nymphs were collected in the field using a tape-trap method.Double-sided tape was placed at the base of last year's cane (i.e., spur) and nymphs were caught as they walked across the surface on their way from the bark to the shoots.Traps were placed after the peak of the first male flight, as determined through male mealybug monitoring using pheromone lures as described (Bahder et al. 2013b).
Figure 3 Percent of grapevine powdery mildew (Erysiphe necator) colonies formed when treatments were applied at 24 or 72 hrs postinoculation (hpi) and then observed at eight days postinoculation, separated by experimental repeat.Treatments were: 7 or 1 mg/L OWS, ozonated water spray (OWS) with an ozone (O 3 ) concentration of either 7 or 1 mg/L, respectively; No spray, a no spray (water or O 3 ) control; Water control, a non-ozonated water control; and Oil control, a 2% v/v horticultural oil (PureSpray Green, Intelligro).Error bars are standard error (n = 9, 7, or 7, respectively, for each experiment).Different letters denote significant differences among treatment means at α = 0.05 using Tukey's honest significant difference.
By 24 hrs postexposure (p = 0.0001), both 1 and 7 mg/L OWS exposure increased nymph mortality relative to the water and no spray controls.This delayed efficacy was not seen with the ozonated high-sediment water (Figure 4, middle).At two (data not presented), 24, and 48 hrs postexposure, the mortality of nymphs exposed to ozonated water was not different than nymph mortality in the no spray control, and both had significantly less mortality than the 2% oil treatment (p = 0.0001 at all three time points).For experiments using DI water as the water source (Figure 4, right) the two hour observation point was not included, given previous experimental results that showed better mortality observation at 24 and 48 hrs.At 24 and 48 hrs, OWS treatments were not different than the no spray control (p = 0.0001 and 0.0001, respectively), but mortality was significantly less than that of the oil control (Figure 4, right).

Field evaluation of ozonated water system for E. necator management
In 2020 and 2021, in-season weather conditions were hot and dry (Figures 5A and 6A).These weather patterns provided poor developmental conditions for E. necator (Wilcox et al. 2015).In 2020, the end-of-season foliar disease severity for all treatments ranged from 11% (O 3 ) to 43% (full-season unsprayed control).The full-season unsprayed control had significantly more accumulated disease (AUDPC) than all other treatments (Figure 5B, p = 0.001).Cluster disease severity for that year was very low, from <1% (standard program, O 3 , prebloom unsprayed) to 4% (full-season unsprayed).All three treatments that received a synthetic fungicide application during bloom (fungicide standard, prebloom unsprayed, and O 3 ) had significantly less accumulated cluster disease severity (AUDPC) than the full-season unsprayed control (Figure 5C, p = 0.02).In 2021, foliar disease severity was as low as 9% (fungicide standard) and only reached 16% in the full-season unsprayed vines.With such low foliar disease, there was no significant separation between treatments (Figure 6B, p = 0.06).There was 0% cluster disease severity at harvest, even in the untreated vines.Neither year showed a treatment effect on total yield per vine (p = 0.78 in 2020 and p = 0.43 in 2021), cluster weight (p = 0.48 in 2020 and p = 0.88 in 2021), or individual berry weight (p = 0.62 in 2020 and p = 0.2 in 2021).

Field evaluation of ozonated water system for P. maritimus management
Weather conditions were tracked at the Outlook (Figure 7A) and West Richland (Figure 7B) sites.Ozonated water, regardless of treatment frequency, did not reduce mid-summer grape mealybug nymph populations in either vineyard below that of the unsprayed control.There was no difference between treatments at either tape trap collection date at either the Outlook (Figure 7C) or West Richland (Figure 7D) locations.

Discussion
Lab studies confirmed that OWS was not effective at reducing E. necator colony formation or delaying conidiophore development.This contrasts with prior studies where efficacy against cucumber powdery mildew (Sphaerotheca fuliginea Pollacci) was observed (Fujiwara andFujii 2002, Fujiwara et al. 2009).Those studies applied OWS to diseased plants with visual symptoms, i.e., the disease was much more progressed in those studies.Also, fixed volumes of OWS were applied: 150 to 200 mL of 4 to 8 mg/L O 3 every three to four days to potted cucumber (Cucumis sativus L. Sharp 7) plants with two to four true leaves.The spray volume during O 3 ) control.Error bars are standard error (n = 5).Different letters denote significant differences among treatment means at α = 0.05 using Tukey's honest significant difference.
one application was ~10 times greater than the OWS spray volume used per treatment application in our study, which was 20 mL and is more aligned with what would be applied through a standard spray application.Improved efficacy in the other studies could be due to the increased spray volume potentially leading to a longer spray duration, which would increase the O 3 gas released from the OWS.This was acknowledged by Fujiwara and Fujii (2002), but they did not quantify O 3 gas in their experiment.This supposition is supported by a study where intermittent O 3 gas (0.2 mg/L) decreased powdery mildew (Sphaerotheca fuliginea) colonization on cucumber (Cucumis sativa) (Khan and Khan 1999).Unfortunately, if generation of O 3 gas is truly the pesticidal component of OWS application, this would render the use of OWS in commercial field applications very challenging, as generating an O 3 fog with an exposure duration necessary for fungicidal effects is not possible with available application technologies.Advancements in application technology that could help in generating a sustained "O 3 fog" may help improve the applicability of OWS in commercial agricultural pest and disease management.
Our field studies on OWS as a management tool for E. necator were limited by when the field-scale OWS could be used during the season, and the environmental conditions seen during the two years of evaluation.The field-scale OWS  only generated measurable O 3 concentrations during the prebloom period -the cooler environmental temperatures during that time allowed for that maximum O 3 generation (excess heat can hinder O 3 generation).Regardless of what treatment was implemented in the prebloom period, when efficacious fungicides were used during the critical window, end-of-season fruit disease severity was always lower than that of the unsprayed control.Thus, when used during a lowrisk time of year (prebloom), in low-disease pressure years (like 2020 and 2021 in eastern Washington), lower efficacy products like OWS could still play a role in integrated pest management.While we can time such applications to align with lower-risk periods in vine development and machine limitations, we cannot plan for lower-disease-pressure environmental conditions.As such, this field study cannot confirm whether the use of OWS during the prebloom period is truly efficacious for disease management.Additional field evaluations of OWS in climates with greater disease pressure are needed to confirm potential efficacy and to fully understand the potential use of this technology in commercial vineyard disease management.
Our laboratory experiments with P. maritimus highlighted the importance of water quality when generating ozonated water of the required concentration using OWS technology.There were notable differences in how much oxygen was required for generators to produce the 7 mg/L O 3 concentration across each water source.The low-sediment water used 4 L/min oxygen, and high sediment required both O 3 generators at 8 L/min oxygen, but DI water only needed 2 L/min oxygen.These differences in oxygen flow rate for each water type suggests that water quality affected the rate at which ozonated water was formed.This is confirmed by other studies where sediment affected O 3 concentration (Khadre et al. 2001).This may explain the differences seen in our experiments, where the ozonated low-sediment water source, regardless of O 3 concentration, achieved greater nymph mortality than the no spray control.However, the ozonated high-sediment water did not increase the nymph mortality rate over that of the unsprayed control.What is unexplained is why the DI water source, which had the least potential sediment, did not perform better (Figure 4).Future studies with OWS should focus on the water quality used, to better understand its role in OWS pesticidal efficacy.In most commercial agricultural applications, water source and quality for pesticide applications is variable; a product or system that requires specific or narrow ranges in water quality is likely not reliable enough to have wide commercial applicability.
In field studies, OWS treatments of first-generation nymphs at the delayed dormant timing did not reduce the subsequent generation of grape mealybug nymphs.While water quality may have played a role, the water used was chlorinated groundwater that meets NPDWR, which is cleaner than most agricultural water sources.One potential challenge with this field-scale system is that the retrofit sprayer tank used to hold the ozonated water was metal and not airtight.Ozonated water can react with metal (Batakliev et al. 2014), which would eliminate its potential efficacy prior to being sprayed.If this technology is to move forward, additional studies on tank material will be needed to find an adequate material for field-holding the ozonated water.Lastly, OWS is highly reactive with any surface it contacts, including grapevine bark.Paired with the mealybug's behavior of moving between exposed location on leaves and protected areas under loose bark (Geiger andDaane 2001, Grasswitz andJames 2008), this would require OWS to penetrate these crevices without being exhausted of all its dissolved O 3 gas (Khadre et al. 2001), or to optimize timing of application to align with nymph movements when they are more readily exposed.

Conclusion
This study examined whether OWS could be an alternative pest management technique for E. necator and P. maritimus in eastern Washington.Both in the laboratory and in a commercial field application situation, OWS did not provide meaningful control of either pest.Improved OWS application technology is needed before any potential adoption of this technology into current commercial vineyard systems.Aspects of ozonated water generation, including choice of inert tank material, water quality, and prevention of offgassing, must be considered carefully in the design of such a system for field use.
percent surface area infected by E. necator on the upper and lower surfaces of 40 random leaves and 20 random clusters per treatment replicate in 2020, and 40 random leaves and 40 random clusters per treatment replicate in 2021.Ratings were done by trained raters and occurred every seven to 14 days.Yield per vine was collected on 4 Sept 2020 and 31 Aug 2021 for each of the four center vines per treatment replicate.Average cluster weight was calculated by weighing eight random clusters (two clusters from each vine) in 2020, and 12 random clusters (three clusters per vine) in 2021.In both years, 50 berries per treatment replicate were collected from these clusters and weighed collectively to calculate average berry weight.

Figure 4
Figure 4 Grape mealybug (Pseudococcus maritimus) percent mortality observations at 24 hrs posttreatment for experiments with low-sediment water source treatments (left); observations at 24 and 48 hrs for experiments with high-sediment water source treatments (middle), and deionized water source (DI; right).Treatments were: Oil control, a 2% v/v horticultural oil (PureSpray Green, Intelligro); 7 or 1 mg/L OWS, ozonated water spray (OWS) with ozone (O 3 ) concentration of either 7 or 1 mg/L, respectively; Water control, a non-ozonated water control; and No spray, a no spray (water or

Figure 5
Figure 5 Weather and grapevine powdery mildew disease severity, represented as area under disease progress curve (AUDPC), for the 2020 growing season.Treatments were: Full unsprayed, full season long unsprayed control; OWS, ozonated water spray at ~1 mg/L; Prebloom unsprayed; and Standard fungicide, standard fungicide control.A) Daily average air temperature and total daily precipitation were acquired from AgWeatherNet (weather.wsu.edu)'Prosser NE' station from in-season weather months, April to October; B) AUDPC for foliar disease severity; and C) AUDPC for cluster disease severity.Error bars are standard error (n = 4).Different letters denote significant differences among treatment means at α = 0.05 using Tukey's honest significant difference.

Figure 6
Figure 6 Weather and grapevine powdery mildew disease severity, represented as area under disease progress curve (AUDPC), for the 2021 season.Treatments were: Full unsprayed, full season long unsprayed control; OWS, ozonated water spray at ~1 mg/L; Prebloom unsprayed; and Standard fungicide, standard fungicide control.A) Daily average air temperature and total daily precipitation were acquired from AgWeath-erNet (weather.wsu.edu)'Prosser NE' station from in-season weather months, April to October, and B) AUDPC for foliar disease severity.Cluster disease severity was not represented, as there was no disease.Error bars are standard error (n = 4).Different letters denote significant differences among treatment means at α = 0.05 using Tukey's honest significant difference.

Figure 7
Figure 7 Daily average air temperature and total daily precipitation weather data was acquired from AgWeatherNet (weather.wsu.edu)from A) 'Outlook' weather station for the Outlook site, and B) 'Red.Mtn.N' weather station for the West Richland location.Arrows within each graph represent ozonated water spray application dates.The two collection dates, both in 2021 and one week apart, of the Pseudococcus maritimus nymphs after first flight for C) the Outlook location, and D) the West Richland location.Error bars are standard error (n = 4).Different letters denote significant differences among treatment means at α = 0.05 using Tukey's honest significant difference.Treatments were: Unsprayed, unsprayed control; OWS 1, 1 mg/L ozone (O 3 ) applied once a week; OWS 2, 1 mg/L O 3 applied twice at one week intervals.
over-the-row multi-tank sprayer with air shear nozzles.Spray applications from the start of the season until 19 May 2021 were applied at 468 L/ha with three nozzles per side, then 702 L/ha with four nozzles per side for the remainder of the applications.Incidence and severity of E. necator were visually rated from June through August in both years.Visual rating determinations involved observing the approximate