Spatio-temporal Variation of Spinosad Susceptibility in Drosophila suzukii (Diptera: Drosophilidae), a Three-year Study in California’s Monterey Bay Region

Abstract

Spinosyn insecticides are widely used in conventional berry production, and spinosad is regarded as the most effective insecticide for managing Drosophila suzukii (Matsumura) (Diptera: Drosophilidae), spotted-wing drosophila, in organic berry crops. Following the 2017 identification of spinosad resistance in caneberry fields in the Watsonville area, Santa Cruz Co., California, we conducted a study to examine the seasonal and annual susceptibility of D. suzukii over a three-year period. Adult flies were collected from two conventional and two organic caneberry fields in the Monterey Bay region, California, at ‘early’, ‘middle’, and ‘late’ time points during the 2018–2020 growing seasons, and their susceptibility to spinosad was assessed. Results demonstrated that spinosad susceptibility in the D. suzukii field populations generally decreased during the fruit production season (from June through November), and over consecutive seasons. LC₅₀ values of adults from the conventional sites were determined to be as high as 228.7 mg l⁻¹ in 2018, 665.6 mg l⁻¹ in 2019, and 2700.8 mg l⁻¹ in 2020. For the organically managed fields, LC₅₀s of adults were as great as 300.0 mg l⁻¹ in 2018, 1291.5 mg l⁻¹ in 2019, and 2547.1 mg l⁻¹ in 2020. Resistance ratios based on the LC₅₀ values were as high as 10.7-, 13.2-, and 16.9-fold in 2018, 2019, and 2020, respectively. These results should serve as a caution for growers in other production areas, facilitate informed choice of insecticides used in D. suzukii management, and emphasize the need to develop effective insecticide resistance management strategies for this insect.

Drosophila suzukii, also known as spotted-wing drosophila, is a key pest of soft-skinned fruit in Asia, Europe, and the United States (Hauser 2011, Walsh et al. 2011, Calabria et al. 2012, Cini et al. 2012, Depra et al. 2014, Asplen et al. 2015, Diepenbrock et al. 2016). Since its invasion in 2008, this pest has posed a significant threat to California’s berry production industry (Bolda et al. 2010), which in 2019 valued at more than $2.8 billion (CDFA 2020). Caneberries in particular are a preferred host of D. suzukii (Walton et al. 2019), and California accounts for 89.4% of all production in the United States, with the Monterey Bay region producing about half of the state's raspberries and blackberries (CDFA 2020). This pest has now spread to all major berry and cherry growing areas of the United States.

Several cultural practices, including the timing of harvest, sanitation measures, drip irrigation, and mulching can contribute to the integrated management of D. suzukii (Schöneberg et al. 2021, Tait et al. 2021). However, due to high market value of berry fruit and a zero-tolerance policy for fruit infestation by D. suzukii, control of this pest in conventional production has principally relied on calendar-based applications of pyrethroids, organophosphates, and spinosyns (Beers et al. 2011, Van Timmeren and Isaacs 2013). In organically managed berry production, chemical options are limited, and the Entrust (Corteva Agriscience, Indianapolis, IN) formulation of spinosad has been shown to be the most effective insecticide against D. suzukii that has been certified by the Organic Materials Review Institute (Fanning et al. 2018). Spinosad label restrictions obligate its rotation with a different chemical class after two applications to prevent resistance development. Thus, with limited options, organic berry growers usually rotate spinosad with pyrethrin sprays despite their relative ineffectiveness at controlling D. suzukii, resulting in higher levels of infestation compared to the conventional production (Goodhue et al. 2011, Van Timmeren and Isaacs 2013, Farnsworth et al. 2017).

Selection pressure from frequent treatments, particularly with a single product, has been recognized as a likely route to the development of insecticide resistance (e.g., Georghiou et al. 1983, Vontas et al. 2011, Andreason et al. 2018). Additionally, D. suzukii may be more prone to evolve insecticide resistance due to its short generation time, high reproductive rate, and its ability to disperse (Tait et al. 2021). These concerns motivated the initiation of nationwide insecticide resistance monitoring efforts, which include the development of a glass vial residual bioassay for D. suzukii. This technique was developed for rapid laboratory bioassays of field populations (Van Timmeren et al. 2019) and was previously employed by Gress and Zalom (2019) to assess susceptibility of a D. suzukii population from Watsonville, CA to spinosad. Their study documented the emergence of low to moderate levels of spinosad resistance (RR₅₀s: 4.3 for males and 5.2 for females) in D. suzukii compared to susceptible flies from an untreated mixed-fruit site in California's central valley known as the USDA Wolfskill National Clonal Germplasm Repository.

The results from Gress and Zalom (2019) provided a snapshot of D. suzukii resistance in the Watsonville region at a single point in time. However, understanding the spatio-temporal patterns of resistance—i.e., variation across sites and over time—is critical to inform on-farm decisions to best manage D. suzukii and mitigate emerging spinosad resistance. The purpose of this study was thus to determine the in-season and annual changes in spinosad resistance in D. suzukii over a three-year period at four caneberry field sites in the Monterey Bay region. Monitoring insecticide resistance is essential for sustainable pest control and could be used to inform pesticide use decisions for D. suzukii management.

Materials and Methods

D. suzukii Populations

D. suzukii adults were live-trapped from four commercial caneberry fields in Santa Cruz County, CA. Two conventional fields, ‘Conv 1’ and ‘Conv 2’, and two organically managed fields, ‘Org 1’ and ‘Org 2’, were selected. The Conv 1 field was located near Watsonville, the Org 1 was near Amesti, and the Conv 2 and Org 2 fields were near Johnston Corner. The Conv 1 field was ~1.4 km from the Org 1 field, and the Conv 2 field was ~0.8 km from the Org 2 field. The two fields designated as Conv 1 and Org 1 were ~6.3 km northwest of the two fields labelled as Conv 2 and Org 2. Caneberries in this region are typically grown under hoop tunnels with three rows planted per hoop. At each site, three hoops were selected, and five plastic McPhail-type traps (Great Lakes IPM, Inc., Vestaburg, MI) baited with approximately 20 ml of a mixture of 7 g yeast, 113 g sugar, and 355 ml water were deployed in the middle row of each hoop for live-capturing of adult flies. Traps were modified by fitting a mesh barrier such that D. suzukii adults were able to enter the trap unobstructed but were prevented from reaching the yeast–sugar–water lure and drowning. Traps were collected the next day and transported back to our laboratory at the University of California, Davis, where they were then placed in a walk-in refrigerator (4°C) for 5–10 min to slow down fly activity and allow separation of D. suzukii, by aspiration, from nontarget species. Approximately 100 D. suzukii were randomly selected from larger field collections to establish colonies for bioassays. Of these, 15 females and 15 males were then transferred into Fisherbrand drosophila bottles (Fisher Scientific, Inc., Portsmouth, NH) containing Bloomington standard Drosophila cornmeal diet. Diet bottles were plugged with cotton, labeled with site identification, and maintained in a walk-in growth chamber at 23 ± 1°C, 55–65% RH, and a photoperiod of 14:10 (L:D) h (Percival Scientific Inc., Perry, IA). Flies were transferred to new diet bottles every 6 d, and their progeny were used in bioassays or to expand the colonies. In three instances, fewer than 100 total flies were trapped in a field collection. In these cases, all of the flies were used to establish colonies, and their progeny were reared for an additional generation to produce enough flies for bioassays.

Fly collections were conducted at ‘early’, ‘middle’, and ‘late’ time points during the harvest season for three consecutive years, 2018–2020. The collection dates were June 7, August 8, and October 23 in 2018, June 17, August 16, and October 9 in 2019, and July 9, September 25, and November 11 in 2020. No early-season flies were captured from the Org 1 field in 2018 and 2019 or from the Conv 2 field in 2019 due to later fruit ripening times in these fields and years. Thus, there was a lack of fruit to attract the flies on the early collection dates. Fly numbers collected late in the season from the Conv 2 field in 2018 and early season from the Conv 2 and Org 2 fields in 2019 totaled fewer than 100, and these populations were increased for two generations to obtain sufficient adults for bioassays.

The Wolfskill colony, established in fall 2017 from an untreated mixed fruit orchard near Winters in Solano County, CA (roughly 185 km north of the study site), was previously used as the susceptible population by Gress and Zalom (2019) and was also used as the control in this study. Given that D. suzukii flies in California were not observed to exhibit distinct genetic structure based on whole genome sequencing (Lewald et al. 2021), we deemed the Wolfskill colony to be an appropriate untreated control for flies collected in the study sites.

Bioassays

Susceptibility of D. suzukii adults to spinosad (Entrust 24 SC; Corteva Agriscience, Indianapolis, IN) was evaluated using the glass vial residual bioassay method developed for D. suzukii by Van Timmeren et al. (2019). First, 1 ml of spinosad solution was pipetted into each 20-ml glass scintillation vial (Fisher Scientific, Pittsburgh, PA). Bioassay vials were then capped and gently turned to ensure the solution evenly coated all interior surfaces of the vial. The solution was then poured out into a waste container and the vial was tapped five times on an absorbent paper to remove excess liquid. After 10–60 min, vials were inverted and tapped again to prevent the remaining solution from settling at the bottom of the vial. Treated vials and caps were placed upright in a fume hood and allowed to dry overnight. Treated vials were then stored in a refrigerator at 4°C and were used in bioassays within a week after being prepared.

A concentrated stock solution of spinosad was obtained by diluting Entrust (22.5% spinosad, a mixture of spinosyn A & D, Corteva Agriscience, Indianapolis, IN) in an Induce (Helena Chemical Company, Memphis, TN) – deionized water solution at a rate of 1266 μl Induce per liter of deionized water. Induce, a nonionic surfactant, was used to help spread the solution more uniformly and promote adherence to the vial surfaces. The concentration of 928 mg l⁻¹, previously used as the LC₉₉ × 2 discriminating dose for susceptible D. suzukii (Gress and Zalom 2019), was made from the stock solution and was used to initially assess the susceptibility of the field populations to spinosad. Due to survival of flies at this diagnostic concentration, dose-response bioassays were performed with concentrations of 13, 30, 100, 300, 500, and 928 mg l⁻¹ in 2018, concentrations of 30, 100, 300, 500, 928, 2,000, and 3,000 mg l⁻¹ in 2019, and concentrations of 30, 100, 300, 928, 3,000, 6,000, 10,000, 12,000, 15,000, and 20,000 mg l⁻¹ in 2020. For the Wolfskill population, concentrations of 3–928 mg l⁻¹, 30–928 mg l⁻¹, and 30–3,000 mg l⁻¹ spinosad were tested in the 2018, 2019, and 2020 bioassays, respectively. Control vials were treated with only the Induce and deionized water solution.

For bioassays, five male and five female adults (3–5 d after emergence) from a given site were aspirated into each treated vial, which were then resealed with treated caps. The treated vials were kept on their side in a tray and were maintained in the previously described walk-in chamber for 8 h, after which mortality was assessed and number of dead females and males recorded separately.

Data Analyses

All data analyses were performed using R version 4.1.1 for windows (R Core Team 2021) and were evaluated for significance at P < 0.05. Control mortality was zero or negligible, therefore, no mortality correction was performed. For each field population, proportions of dead individuals at the discriminating dose among the three seasonal time points (‘early,’ ‘middle’, and ‘late’ season) were compared using a generalized linear model (GLM) with a binomial distribution (glm function) followed by LSMeans with a Tukey adjustment (lsmeans function) when significant differences were found.

For dose–response analysis, a two-parameter log-logistic model with lower limit at 0 and upper limit at 1 (drc package) was used to fit the mortality data for males, females, and combined sexes (Ritz et al. 2015). In the drc package, the function ‘ED’ was used to calculate estimated effective doses (LC₅₀ and LC₉₀ values) and their corresponding standard errors. Pairwise z-tests were conducted using the ‘compParm’ function (Ritz et al. 2015) to compare LC₅₀ values to further analyze susceptibility of different field populations at each time point as well as potential sex differences. Insecticide concentration, population, and sex were used as predictor variables, while the proportion of dead males and females in each bioassay vial was used as the response variable. Proportions were weighted by the total sample size for each sex. Resistance ratios were calculated by dividing the LC₅₀ and LC₉₀ of the resistant field population by the LC₅₀ and LC₉₀ of the susceptible Wolfskill population.

To test the overall effects of year and seasonal time point on fly susceptibility, three datasets were created: 1) only data from the organic fields, 2) only data from the conventional fields, and 3) combined data. Each dataset was separately analyzed using a GLM with a binomial distribution in which insecticide concentration, population, year, and seasonal time point were used as predictor variables, and the proportion of dead flies, regardless of sex, was used as the response variable. Then, the estimated marginal means (EMMs) were obtained from each model using the function ‘emmeans’ followed by the function ‘pairs’ for pairwise comparisons of the EMMs across different combinations of year and seasonal time points. EMMs are recommended for examining the main effects when there are unequal sample sizes to account for the unbalanced design (Lenth et al. 2020). After statistical analysis, resulting EMMs were back-transformed into their original scale for interpretability and plotting.

Results

Discriminating Dose Bioassays

Flies collected in the untreated Wolfskill orchard displayed 90–100% female mortality and 93–100% male mortality against the discriminating dose of spinosad throughout the study period. In contrast, mortality from the Monterey Bay field populations ranged from 68 to 90% in 2018, from 36 to 88% in 2019, and from 31 to 98% in 2020 when exposed to the discriminating dose (Fig. 1). In general, fly susceptibility decreased from year to year, and within each year, with the greatest decrease occurring between early and late collection time points.

In 2018, D. suzukii collected from Conv 1 (n = 260) and Org 1 (n = 170) showed no significant change in susceptibility throughout the season (P > 0.05). Flies collected from Conv 2 (n = 199) during the mid-season time point exhibited lower susceptibility against the discriminating dose than early season flies (z = 2.611, P = 0.0245), while Org 2 flies (n = 281) were more susceptible at mid-season compared to the early collection (z = −3.474, P = 0.0015). At both locations, however, late season D. suzukii showed intermediate susceptibility and did not differ from either of the earlier collections (Fig. 1).

In 2019, Conv 1 (n = 239) early season flies were more susceptible than those collected during the mid-season (z = 2.967, P = 0.0085) and late time points (z = 3.720, P = 0.0006), however, the difference between middle and late collections was not significant (P > 0.05). Late collections from Conv 2 (n = 163) and Org 1 (n = 181) both had lower susceptibilities compared to the mid-season collections (Conv 2: z = 4.082, P < 0.0001; Org 1: z = -5.9444, P < 0.0001). In the Org 2 field (n = 181), no significant difference was observed in susceptibility of flies between the three seasonal time points (P > 0.05) (Fig. 1).

In 2020, the seasonal collections from Conv 1 (n = 230) did not differ in susceptibilities to the discriminating dose of spinosad (P > 0.05). Early and mid-season collections from the Conv 2 field (n = 360) had more susceptible D. suzukii adults than the late collection (early vs late: z = 4.979, P < 0.0001; middle vs late: z = −4.779, P < 0.0001). Middle and late collections from the Org 1 field (n = 280) were less susceptible than early collections (early vs middle: z = 5.195, P < 0.0001; early vs late: z = 4.927, P < 0.0001), and significant difference in susceptibility was observed between early and late collections from the Org 2 field (n = 250, z = 3.465, P = 0.0015) (Fig. 1).

Dose–response Bioassays

Dose–response assays indicated that Wolfskill flies were more susceptible to spinosad than all the field populations at each time point within each year (Table 1 and Supp Table 1 [online only]). LC₅₀ values of adults from the conventional sites, Conv 1 and Conv 2, ranged from 163.0 to 228.7 mg l⁻¹ in 2018, from 59.6 to 665.6 mg l⁻¹ in 2019, and from 154.0 to 2700.8 mg l⁻¹ in 2020. For the organically managed fields, Org 1 and Org 2, LC₅₀s of adults ranged from 105.0 to 300.0 mg l⁻¹ in 2018, from 91.5 to 1291.5 mg l⁻¹ in 2019, and from 99.7 to 2547.1 mg l⁻¹ in 2020 (Table 1). The differences between the LC₅₀ values of field populations and Wolfskill adult flies were significant, except for the early season collections from the Conv 1 (t = −1.834, P = 0.0666) and Org 2 (t = −0.390, P = 0.6968) fields in 2019 and mid-season collection from the Conv 2 field in 2020 (t = −0.2767, P = 0.7820) (Table 1 and Supp Table S1 [online only]). Significant differences were also observed among the LC₅₀ values of the field populations, and these differences became more prominent in 2020 (Table 1 and Supp Table S1 [online only]). No significant differences were observed in mortality between males and females (P > 0.05), except for the mid-season Conv 2 flies in 2018, with females being more susceptible than males (t = −1.995, P = 0.0460), and the late-season Conv 1 flies in 2019, with males being more susceptible than females (t = 2.294, P = 0.0218). When male and female data were combined, the resistance ratios based on the LC₅₀ values ranged from 3.8 to 10.7-fold in 2018, from 0.6 to 13.2-fold in 2019, and from 0.6 to 16.9-fold in 2020. Resistance ratios at 90% mortality reached 23.9-fold in 2018, 18.9-fold in 2019, and 124-fold in 2020 (Table 2).

Overall Effects of Year and Seasonal Time Point on Adult Susceptibility

The GLM analyses showed that the proportion of dead adults differed significantly between seasonal time points of early and late for both organic (z = -9.834, P < 0.0001) and conventional fields (z = −7.640, P < 0.0001), and when data from both field types were combined (z = −11.877, P < 0.0001). The difference between early and middle time points was only significant for organic and combined data (conventional: z = −0.330, P = 0.741; organic: z = −6.486, P < 0.0001; combined: z = −4.587, P < 0.0001). The model also indicated significant differences in susceptibilities between 2018 and 2019 for organic and combined data (conventional: z = 0.333, P = 0.739; organic: z = −3.908, P < 0.0001; combined: z = −2.465, P = 0.0137), and between 2018 and 2020 for conventional, organic, and combined data (conventional: z = −11.320, P < 0.0001; organic: z = −11.889, P < 0.0001; combined: z = −16.238, P < 0.0001).

The estimated marginal means (EMMs) and pairwise P value comparisons of these EMMs showed that flies collected late season in 2020 were the most tolerant flies to spinosad (Fig. 2; Table 3).

Comparison of early season fly susceptibilities among three years indicated that there was no change in susceptibility between 2018 and 2019, except when only data from the organic fields were considered (conventional: z = −0.333, P = 1; organic: z = 3.908, P = 0.0030; combined: z = 2.465, P = 0.2490). However, significant reductions in early season susceptibility were observed between 2018 and 2020 (conventional: z = 11.320, P < 0.0001; organic: z = 11.889, P < 0.0001; combined: z = 16.238, P < 0.0001), and between 2019 and 2020 (conventional: z = 10.903, P < 0.0001; organic: z = 7.287, P < 0.0001; combined: z = 12.659, P < 0.0001). The same results and test statistics were obtained when mid-season or late season fly susceptibilities were compared among three years, showing significant reductions in susceptibility between 2018 and 2020, and between 2019 and 2020 (Fig. 2; Table 3).

Within each year, fly susceptibility also declined significantly from early to late (conventional: z = 7.640, P < 0.0001; organic: z = 9.834, P < 0.0001; combined: z = 6.264, P < 0.0001), and from middle to late season (conventional: z = −8.575, P < 0.0001; organic: z = −4.686, P < 0.0001; combined: z = −9.252, P < 0.0001). Early to mid-season susceptibilities within each year were also significant, except when only data from the conventional fields were examined (conventional: z = 0.330, P = 1; organic: z = 6.486, P < 0.0001; combined: z = 4.587, P = 0.0002) (Fig. 2; Table 3).

Discussion

Diagnostic bioassays using the spinosad LC₉₉ × 2 concentration provided rapid documentation that D. suzukii flies collected in the Monterey Bay region, California, at different times over the course of the 2018–2020 growing seasons exhibit spinosad resistance, further validating the utility of this simple and low-cost method to screen D. suzukii field populations for insecticide resistance.

Our study showed that spinosad susceptibility in the D. suzukii field populations generally declined during the fruit production season. Insecticide applications for D. suzukii management are typically initiated with fly detection and fruit ripening and continue until the harvest is completed (Asplen et al. 2015). In caneberries and strawberries grown on California's central coast, insecticides are applied to the fields for pest management from April through November. Although D. suzukii is not a target of many of these insecticide applications, indirect selection pressure could also be imposed on the fly populations. The pesticide use data obtained from the California Pesticide Information Portal (CALPIP, 2021) of the California Department of Pesticide Regulation (CDPR) for spinosyn insecticides (spinosad and spinetoram [Delegate WG, Corteva Agriscience, Indianapolis IN]) confirm the extensive use of these insecticides in caneberry production areas in Santa Cruz County (Supp Table S2 [online only]). Furthermore, Van Steenwyk and Bolda (2014) showed that spinosyn insecticide use in California caneberries in 2012 was 3.3 times greater than in 2007, a year before the D. suzukii invasion. In strawberries, however, spinosyn insecticide (spinetoram, Radiant SC formulation) use did not increase after the invasion of D. suzukii since it was already widely used for control of the western flower thrips, Frankliniella occidentalis (Pergande) (Thysanoptera: Thripidae), and several lepidopteran pests including the light brown apple moth, Epiphyas postvittana (Walker) (Lepidoptera: Tortricidae) (Zalom 2010, Van Steenwyk and Bolda 2014). The increased resistance of D. suzukii flies to spinosad during a season may thus be the result of repeated insecticide applications within the same season across berry crops grown in the region. In addition, lower levels of resistance to spinosad early in the season might be due in part to greatly reduced insecticide use during the winter. Similar seasonal susceptibility trends have been observed in other pests. For example, resistance to insecticides in pear psylla, Cacopsylla pyricola Foerster (Hemiptera: Psyllidae), was low at the beginning of the growing period in spring and increased in fall (Buès et al. 1999). Follett et al. (1985) also reported differences in susceptibility of adult pear psylla to insecticides early in the summer compared with late summer. Another study demonstrated that resistance in field populations of the Hawaiian flower thrips, Thrips hawaiiensis (Morgan) (Thysanoptera: Thripidae), on banana crops developed quickly within a season, and was attributed mainly to both the intense use of insecticides and short generation time of thrips (Fu et al. 2019). Wu and Jiang (2002) also showed that insecticide resistance in the diamondback moth, Plutella xylostella (L.) (Lepidoptera: Plutellidae), was high from September through the following April and the lowest in August. The short generation time of D. suzukii and its high reproductive rate are also expected to contribute to the rapid development of resistance during the season due to greater opportunity for mutations as has been shown in other Drosophila species, such as Drosophila melanogaster (Meigen) (Diptera: Drosophilidae) (Daborn et al. 2007, Sun et al. 2019). Moreover, higher susceptibilities we observed early in the season could be explained by the migration of susceptible flies from untreated noncrop D. suzukii host plants to berry fields when fruit is ripening.

Our results indicated a trend for increased resistance to spinosad from year to year as indicated by increasing resistance ratios. Since flies collected early in 2020 were more resistant to spinosad than those collected early in 2018 and 2019 in both organic and conventional fields, we believe that continuous and extensive spinosyn use in the area has likely contributed to the overall increased resistance to spinosad in D. suzukii. Our assumption is that susceptible flies immigrating to caneberries from other treated D. suzukii hosts have developed some levels of resistance to spinosyns over consecutive years and the number of resistant flies has generally increased in the region through mixing of these newly resistant flies with resistant individuals from previous years or accumulation of resistant alleles overtime. Disi and Sial (2021) also indicated that susceptibility of D. suzukii populations significantly decreased after 10 generations of laboratory selection, suggesting that resistance alleles exist in the field populations from which those colonies were established. The LC₅₀ values increased from 50.7 and 26.0 ppm in the F₃ selected generations to 167.3 and 105.8 ppm in the F₁₀ selected generations for females and males, respectively (Disi and Sial 2021). A similar trend was shown by Van Timmeren et al. (2019) when spinosad susceptibility of D. suzukii populations from Michigan were tested from 2016 to 2018. The LC₅₀ value for females was 16.3 mg l⁻¹ in 2016 when mortality was assessed after 6 h of exposure to spinosad. LC₅₀s based on 8 h of exposure, as we did in the present study, increased from 13.1 mg l⁻¹ in 2017 to 40.1 mg l⁻¹ in 2018 (Van Timmeren et al. 2019). LC₉₀ values of females ranged from 10.8 mg l⁻¹ in 2016 to 476.2 mg l⁻¹ in 2018. Lethal concentrations found in Michigan populations were much lower than those obtained in the present study, further emphasizing the significance of spinosad resistance in California populations and the urgency for insecticide resistance management (IRM). It is important to note that the LC₅₀ values obtained in our study in 2020 far exceed than the maximum label concentration for spinosad (Entrust 24 SC) at the typical application volume used on California berries which is 112.6 mg l⁻¹, perhaps explaining why growers have been reporting control failures.

Clearly, implementation of resistance management practices to preserve or restore spinosyn susceptibility in D. suzukii populations would be warranted. The concern is particularly urgent for the organic sites which had greater resistance ratios than the conventional sites in both 2019 and 2020. However, removing selection pressure in the organic system might prove more difficult than in the conventional system due to lack of effective registered organic insecticides, so altering aspects of the production system by cultural methods or behavioral and biological controls to reduce D. suzukii populations and accepting low levels of damage might be necessary as alternative practices. Research on such strategies is progressing (Tait et al. 2021), and will hopefully become feasible in the near future.

Understanding the seasonal and annual susceptibility dynamics of D. suzukii to spinosad is essential to develop effective strategies for using this insecticide in order to avoid control failures due to D. suzukii resistance in the field. In addition, determining resistance levels to spinosad can guide the need for using insecticides from other chemical classes as well as alternative nonchemical control measures, such as cultural, behavioral, and biological control strategies for the integrated management of D. suzukii.

Acknowledgments

We thank Mark Bolda and Hillary Thomas for their assistance in arranging access to our field locations. We also thank Chase Matterson for his assistance in collections, colony maintenance, and bioassay setups. This research was supported by grants 2015-51181-24252 and 2020-51181-32140 from the USDA-NIFA Specialty Crops Research Initiative, grant 2018-51300-28434 from the USDA-NIFA Organic Research and Extension Initiative, and grant 18-0001-053-SC from the CDFA Specialty Crops Block Grant Program.

References Cited