The long-term goal of this project is to identify environmental factors associated with increased onion maggot pressure and control failures in NY commercial onion. This information will be used to identify fields at risk for substantial onion maggot damage for improved management. To achieve this goal, I propose the following objectives:
1) Evaluate associations between environmental factors, including precipitation, soil temperature, soil organic matter, surrounding landscape, and temporal factors such as planting date, on the occurrence and relative pressure of onion maggot across onion producing regions of New York.
2) Assess uptake and dissipation of novel, reduced-risk seed treatments used in the control of onion maggot.
The purpose of this project is to evaluate how environmental factors impact the abundance of onion maggot, Delia antiqua (Diptera: Anthomyiidae) and to assess the performance of insecticide seed treatments used for its control. The goal of this research is to improve prediction and management of this pest and to reduce grower reliance on chlorpyrifos, an organophosphate insecticide applied as a drench treatment at planting in an attempt to improve onion maggot control provided by insecticide seed treatments.
New York accounts for over 90% of production in the Northeast and ranks 6th in national onion production. With an annual value of 40-50 million dollars, onion is one of the most important vegetable crops grown in the state (NASS 2017). Closely related to other root feeding maggot species, including seedcorn maggot (Delia platura) and cabbage maggot (Delia radicum), onion maggot feeds on the below-ground portion of onions (roots and bulb). As adults, flies lay their eggs on or at the base of onion plants and once hatched, the larvae move into the root zone, enter the basal plate, and begin feeding. Onion seedlings are most susceptible to damage by onion maggot, which can result in crop losses greater than 50% (Taylor et al. 2001; Nault et al. 2006a). For this reason, protecting the crop against this pest is essential.
Growers manage onion maggot using a combination of cultural controls (e.g. crop sanitation) and insecticides applied at planting, primarily as seed treatments. Currently, the most widely adopted seed treatment package is Syngenta’s FarMore®FI500, which includes two active ingredients against onion maggot, thiamethoxam (neonicotinoid) and spinosad (spinosyn). Thiamethoxam alone is not effective against onion maggot, suggesting that spinosad is the most important of the two (Nault et al. 2006b). Spinosad, a reduced-risk insecticide, is labeled as a foliar spray on many fruit and vegetable crops but is labeled as a seed treatment exclusively on onion. For many growers, FarMore®FI500 is highly effective, and growers achieve >90% control of onion maggot. However, a subset of NY growers consistently experiences onion maggot control failures. This is surprising given that those with damage and those without plant onions in the same soil type (muck soils), employ similar management strategies, cultivate the same varieties, plant at the same seed density, and seldom rotate out of onions. Consequently, growers facing control failures often resort to augmenting seed treatments with in-furrow drenches of chlorpyrifos (Lorsban®), sometimes at rates far greater than those recommended. Field trials have shown that chlorpyrifos is not very effective against onion maggot and that D. antiqua are resistant to chlorpyrifos in New York (Nault et al. 2006c). Because growers are limited in their management options, identifying at-risk fields is essential.
The information generated from this study can be used to inform more sustainable management recommendations for the control of this pest moving forward. Understanding what factors put growers at risk for damage by onion maggot can inform prophylactic measures such as crop rotation or planting with tranplants – both approaches are typically cost prohibitive but have been shown to dramatically reduce onion maggot damage.These findings will lay the groundwork for better predictive abilities and more precise control measures against this devastating pest.
Objective 1: To assess the effect of environmental factors on adult abundance and larval damage by onion maggot.
(Methods and Materials from a manuscript in preparation)
Site selection and description
Environmental factors, fly activity, and damage were monitored in commercial onion fields in six counties in central New York in 2018 (n=15) and 2019 (n=12). Fields which had a history of onion production were monitored in Orleans (n=4, 2018-19), Linwood (n=1, 2018-19), Yates (n=2, 2018-19), Steuben (n=2, 2018; n=0, 2019), Oswego(n=4, 2018; n=3, 2019), and Wayne (n=2, 2018-2019) counties. Except for field sites in Steuben county, three sites in Oswego county, and one site in Orleans county, all field sites were sampled in both 2018 and 2019. Field sites were placed a minimum of 1.5 km apart and were located within continuous pockets of Palms or Carlisle muck (SSURGO). Muck soils are high organic matter histosols, whose origin is organic material from woody and herbaceous materials, primarily from drained lakes or wetlands (NRCS). All fields were direct-seeded with dry bulb onions in April and May. In 2018, all seeds were treated with FarMore® FI500 (Syngenta), and 80 percent of growers (12/15) applied chlorpyrifos (Lorsban®, Corteva) at rates of 1-2 fl oz per acre. In 2019, 75% of participating growers (9/12) treated seeds with FarMore® FI500, while the remainder used Trigard® (ADAMA, USA). The majority of growers (10/12) applied chlorpyrifos at planting at the recommended rates (1-2 fl oz per acre) in 2019 as well. Despite these differences in management, small plot field studies suggest that Trigard® and FarMore® FI500 perform equivalently (Moretti & Nault, 2020), and the addition of chlorpyrifos does not improve onion maggot control.
Field sites were sampled from mid-May (15 May) to mid-July (17 July) in 2018 and 2019, corresponding to the first flight of adults from overwintered pupae and subsequent first generation of onion maggot larvae in the field.
Fly abundance. Fly activity was monitored along the edge of each field, bordered by a treeline or shrubby area, nearest to where onion maggot damage was later evaluated. To monitor activity, three 15 X 15 cm yellow sticky cards (Olson Products, Medina, OH, USA) were spaced 30 m apart along the field edge (Werling et al. 2005). Cards were fastened to wooden stakes at a height of ~25 cm (approximating onion height) using spring loaded clamps (Woodworkers’ Supply, Casper, WY, USA). Sticky cards were collected and replaced on a weekly basis, and flies were identified to species using Hucket (1992) and Savage et al. (2016).
Larval damage. Larval damage was evaluated weekly in each field by walking transects totaling 100 m. In 2018 two 50 m transects and in 2019 four 25 m transects were evaluated weekly. In both years, transects were placed a minimum of 25 m apart and extended from the field edge toward the center of the field. The number of damaged plants in the four rows of onions adjacent to each transect was counted weekly. Onion maggot damage was identified by plants with flaccid leaves; plants were removed and assessed for active feeding by larvae or in the absence of larvae, larval entry or exit wounds.
Because onion maggot and seedcorn maggot (Delia platura) can co-occur in onion fields, in 2019, we confirmed the species identification of larvae in heavily infested fields. Larvae from severely infested plants were collected in tubs and returned to the lab. Larvae were reared in larval rearing boxes fitted with a screened lid lined with 3 cm of moistened unwashed mason sand. Larvae were fed halved yellow bulb onions, and were maintained at 22oC until emergence. Emerged flies were frozen, pinned, and identified to species.
Precipitation, temperature, and plant size. Precipitation and average weekly soil temperature were monitored each week. Precipitation was assessed using a simple rain gauge consisting of an inverted 2 L bottle with the bottom removed and fitted with a funnel to direct rainfall into the bottle. Rain gauges were affixed ~1.5 m above the ground to a steel pole, placed along the edge of each field, unobstructed by trees or other structures. Soil temperature was monitored using continuous data loggers (i-button®, Thermochron, DS1921G, OnSolution Pty Ltd) which recorded temperature every 20 min. Data loggers were placed in 50 mL Falcon Tubes (Corning™) filled with rinsed and dried calcined clay (Turface® MVP, Profile Products LLC, Buffalo Grove, IL, USA). Tubes were sealed with Parafilm to exclude moisture and were buried at a depth of 5 cm, the depth at which most larvae can be found (Loosjes, 1967).
Plant growth was assessed each week by haphazardly selecting 10 plants from damage evaluation transects within each field. Plant size was estimated using neck diameter, measured at the widest point of the neck with digital calipers (Lancaster et al. 1996)
Soil collection and analysis
Soil was sampled at field sites the week of 9 July in 2018 and 17 July in 2019. Soil was only sampled once for each field site over the two-year sampling period; those sampled in 2019 were field sites that had been moved or added. A total of nine samples were taken in each field in a 100 m x 100 m grid. Samples were taken at 0, 50, and 100 m along the grid, generating a 3×3 sampling grid. Five soil cores (5 cm diameter, 6-10 cm depth) were taken within a 1 m radius at each sampling location, and cores were taken between onion rows to avoid roots. Cores were bulked, and soil was air-dried and stored in plastic bags until processing. For analysis, visible roots were removed from samples and soil was passed through a 2 mm sieve. Water holding capacity (WHC) and pH were quantified following standard procedures (Barrett et al. 2009), and organic matter was determined using loss on ignition at 500 oC for 2 hr (Storer, 1984).
Land cover around all sites was quantified using ArcGIS (ArcGIS version10.7.1, ESRI, 2018) with data from the United States Department of Agriculture Cropland Data Layer (CDL; USDA National Agriculture Statistics Service, 2018). Land cover was designated as agriculture (all crop systems excluding onion production), forest, and developed, and each land cover was quantified in a 1.5 km radius around each field site. This distance is the best estimate of how far adult flies can travel (Martinson et al. 1988). Because onion fields surveyed were all surrounded by simple landscapes composed primarily of only forest or agriculture, relationships between only the amount of surrounding forest were analyzed.
Analyses were performed in R (R Core Team, 2019). To test the relationships between fly abundance, climate factors, temporal factors, landscape, and soil organic matter and damage observed in fields across the first generation of onion maggot, we used generalized linear mixed effects models analyzed using the lme4 package and function glmer() (Bates et al. 2015). Because damage was lower across fields sites in 2019, years were analyzed separately. For each model, the cumulative number of plants damaged in each field over the observation period served as the dependent variable and was modeled with a Poisson distribution. Average fly abundance across the observation period; climate factors, including soil temperature averaged across the nine-week sampling period (“soil temp”) and cumulative rainfall (“precip”); temporal factors, including planting date (“plant date”) and neck diameter (“neck diam”) during the week of peak flight; proportion of forest at 1500 m (“forest”); and soil organic matter (“OM”) served as fixed effects for each model, respectively. Both proportion of forest in the landscape and soil organic matter were arcsine square root transformed and fly abundance (average flies per card per week) was log transformed prior to analysis. Dependent variables were mean centered and scaled, and in all models, field site was included as an observation level random effect to adjust for overdispersion. The relationships between soil OM, pH, and water holding capacity (WHC) were analyzed using linear regressions with Pearson correlation coefficients.
To assess which factors or combinations of factors are most important in determining larval damage in fields in each year, we used best subsets regressions. Prior to analysis, collinearity between continuous variables was assessed with variance inflation factors (VIF). Using the dredge() function (‘MuMIn’, Barton, 2013) all possible mixed effects models were constructed with fixed effects “soil temp”, “precip”, “plant date”, “neck diam”, “forest”, and “OM”. All models included a minimum of one fixed effect, were limited to one, two-way interaction, and only included factors or interactions that were statistically significant (a < 0.05). Model selection was based on the lowest delta Akaike’s information criterion value corrected for small sample sizes (AICc), and uncertainty of model selection was measured with Akaike’s weight. Marginal R-squared values for each model were determined using the ‘MuMIn’ package (Barton, 2009).
Objective 2: Assess uptake and dissipation of novel, reduced-risk seed treatments used in the control of onion maggot.
Spinosad uptake assay field study 2019
In order to determine if and to what extent spinosad, the primary component of FarMore® FI500, is taken up by onion plants during plant growth and how this uptake corresponds to onion maggot damage in the field, we conducted a uptake and damage evaluation study in 2019.
Onion maggot damage and fly activity
Onion maggot control was evaluated on a commercial onion field with a history of onion production and high pest pressure near Oswego, NY in 2019. On May 16, dry bulb onion seeds, cultivar “Highlander”, were planted into muck soil with a hand-pushed cone-seeder at rates of 30 seeds/m. Each plot consisted of two 9.1 m rows spaced 25.4 cm apart, and plots were separated from each other by a 0.91 m alley of bare soil. There were two treatments: spinosad-treated and untreated. Spinosad-treated seeds were commercially treated with Regard SC (Syntenta, 0.2 mg/seed), and both the spinosad-treated and untreated seeds were commercially treated with fungicides penflufen and thiram for seedling disease control. Treatments were replicated five times.The number of plants dead or damaged from onion maggot was recorded weekly at five timepoints: 39, 42, 46, 54, and 62 days after planting. Damage by onion maggot was identified by plants with flaccid leaves; plants were removed and assessed for active feeding by larvae, or in the absence of larvae, entry or exit wounds.
Fly activity was assessed at the field site using yellow sticky cards. To monitor activity, three 15 X 15 cm yellow sticky cards (Olson Products, Medina, OH, USA) were spaced 30 m apart along the field edge closest to the experimental plots (Werling et al. 2005). Cards were fastened to wooden stakes at a height of ~25 cm (approximating onion height) using spring loaded clamps (Woodworkers’ Supply, Casper, WY, USA). Sticky cards were collected and replaced on a weekly basis, and flies were identified to species using Hucket (1992) and Savage et al. (2016).
In order to evaluate the dissipation of spinsad in the soil and plant tissue, adjacent to the onion maggot damage plots, two, 27 m rows spaced 25.4 cm apart were planted with spinosad-treated seeds (those used in the damage evaluation described above). At five timepoints (25, 40, 47, 54, and 61 days after planting) a subset of plants was sampled from the two rows. At each timepoint, 10 plants were haphazardly collected using a small trowel. The entire plant and the soil clinging to the roots around the base of the plant were collected and transported back to the lab. Leaf stage of the sampled onions was recorded, and the soil was gently separated from the plants and roots. Aboveground tissue was separated from belowground material (seedling base and roots), and belowground tissue was rinsed with DI water to remove any soil. Tissue samples (aboveground and belowground) were massed, placed into 50 mL Falcon Tubes, and frozen at -20oC until analyzed. At the first timepoint, because the onions were at first leaf, samples were bulked. Tissue of five plants was combined into a single sample (for belowground and aboveground tissue) for a total of 2 replicates. At the following sampling timepoints, tissue of two plants was combined into sample for a total of 5 replicates. Soil, corresponding to the area surrounding the belowground tissue of each plant sample was homogenized by vigorously shaking the sample for one min. Soil surrounding two sampled plants was bulked together, for a total of 5 replicates at each timepoint. A 5 g aliquot of soil for each replicate was placed into a 50 mL Falcon Tube and frozen at -20oC until analyzed. Another 5 g aliquot of soil was dried, and percent moisture was determined. Spinsad content of the belowground and aboveground tissue and soil was quantified using a QueChERS extraction method.
RESULTS: (from manuscript in preparation)
Fly abundance and larval damage
In 2018 and 2019 average fly capture during weeks of adult fly activity ranged from 0.33-17.9 and 0-16.2 D. antiqua flies per card per week, respectively. Cumulative number of damaged plants observed in field sites over the nine-week sampling period ranged from 0-853 in 2018 and 1-384 in 2019. Adult flies were successfully reared from larvae collected in fields from four sites in Oswego county, one site in Wayne county, and three sites in Orleans county. At each field site, the majority of flies (> 85%) were identified as D. antiqua, suggesting that onion maggot is the dominant pest species in these systems followed by D. platura. In 2018, fly abundance and larval damage were strongly positively associated (F1, 12 = 86.191, p < 0.001, R2m = 0.89). Similarly, in 2019, in which larval damage was lower across field sites, fly abundance and larval damage were still associated, but the relationship was less strong (F1,9 = 4.4769, p = 0.034, R2m = 0.29)
Soil temperature and rainfall
In 2018, neither cumulative precipitation (F1,10 = 0.0704, p = 0.58) nor the interaction between precipitation and average soil temperature (F1,10 =0.6431, p = 0.42) had a significant relationship with larval damage, but there was a significant effect of average soil temperature (F1,10 =14.14, p <0.001). Fields with higher average soil temperatures over the nine-week observation period had lower damage (Figure). However, this relationship was not upheld the following year. In 2019, there was no significant relationship between larval damage observed in fields and average soil temp (F1,8 = 0.0022, p = 0.96), cumulative precipitation (F1,8 = 2.9759, p = 0.13), or their interaction (F1,8 = 3.2439, p = 0.072).
In both 2018 and 2019, planting date had a significant positive relationship with larval damage. Fields that were planted later had higher damage than those planted earlier in the spring (2018: F1,10 = 4.0523, p = 0.046; 2019: F1,8 = 6.626, p = 0.039). Peak fly activity occurred later in 2019 than in 2018. In 2018, peak fly activity occurred, on average, between weeks 4 and 5, while in 2019, peak fly activity occurred between weeks 5 and 6. However, in both years, peak damage occurred, on average, between weeks 6 and 7. In both years, neither plant size (neck diameter) at peak fly activity (2018: F1,10 = 0.0158, p = 0.90; 2019: F1,8 =0.0138, p = 0.91) nor the interaction between plant size and planting date (2018: F1,10 = 3.5667, p = 0.06; 2019: F1,8 =0.4135, p = 0.52) had a significant relationship with cumulative larval damage, although the interaction was marginally significant in 2018.
Forest and agriculture (not including onion cropping systems) constituted the majority of the landscape surrounding field sites. The amount of forest surrounding each field site in a 1500 m radius ranged from 6 – 62%, and agriculture ranged from 2 – 64%. Proportion of agriculture and forest were strongly negatively correlated (Pearson correlation: r = -0.60, p = 0.02); that is, fields surrounded by more forest had less agriculture, and conversely, fields surrounded by more agriculture had less forest. Because of this close relationship, only the relationship between surrounding forest and onion maggot damage was analyzed. The landscape in a 1500 m radius surrounding each field site had a significant association with larval damage. Field sites surrounded by a greater proportion of forest had more larval damage in both 2018 (F1,12 = 15.866, p<0.001, R2m = 0.54) and 2019 (F1,10 = 7.4216, p = 0.007, R2m = 0.39).
Muck soils are both acidic and have high organic matter. Organic matter in soils from observed fields ranged from 40-85%, and all soils were acidic, with pH ranging from 4.8-6.3. As expected, organic matter and soil acidity were negatively correlated (Pearson’s Correlation, r = -0.624, p = 0.006) and OM and WHC were strongly positively correlated (r = 0.834, p < 0.001). In 2018, OM was had a significant positive relationship with larval damage (F1,12 = 8.6984, p = 0.003, R2m = 0.39), but in 2019, there was no effect (F1,10 = 0.2995, p = 0.58).
Best predictors of larval damage
In 2018, larval damage was best predicted using the model that included landscape (“forest”), planting date (“plant date”), their interaction (“forest * plant date”), and average soil temperature (“soil temp”). Together, these variables explained a substantial amount of variation in larval damage observed across field sites (R2m = 0.90). Looking across the top three best models in 2018, landscape, planting date, soil temperature, and organic matter appeared to be important factors affecting larval damage. In 2019, landscape and planting date made up the best two models for estimating larval damage. However, neither factor explained more than 40 percent of the variation in larval damage or were substantially better than the null model (~1) based on AIC weight.
Briefly, these results suggest that contrary to the literature, which suggest that precipitation and temperature are key predictors of onion maggot damage severity, this study found that surrounding landscape and planting date are associated with increased onion maggot damage and that precipitation and temperature were less important. In both 2018 and 2019, fields that were planted later in the season and those whose surrounding landscape was dominated by forests had more damage than fields that were planted earlier in the season and had a surrounding landscape dominated by agriculture.
The spinosad uptake study demonstrates that spinosad can be recovered from both the aboveground belowground tissue as well as the surrounding soil (Figure 1). At the first sampling point (25 days after planting), when onions at first leaf stage, the greatest amount of spinosad was found in the soil surrounding the plants, followed by belowground tissue and aboveground tissue. However, despite the high amount found in the soil, spinosad dissipated in the soil rapidly at the second sampling point (40 days after planting).
The field study examining damage from onion maggot to spinosad-treated versus untreated plants overlayed with the spinosad content found in tissue and soil indicates that the seed treatment provides a brief window of protection after planting (Figure 1). While the spinosad treated seeds had on average 20% less damage at each sampling point compared with the untreated seeds, damage to the spinosad treated seeds increased at approximately the same rates as the untreated seeds. This may indicated that the protection afforded by spinosad earlier wears off.
Education & Outreach Activities and Participation Summary
On 22 August 2018, I presented at the Oswego Onion Growers Twilight Meeting, which took place in Oswego County and hosted approximately 60 growers from Oswego and other adjacent counties, industry representatives, and extension personnel. This meeting occurs annually and serves as a way to discuss the growing season and major findings from research conducted during the summer. Using a summary handout, I briefly explained the goal of the project and some of our preliminary findings regarding precipitation and larval damage and fielded questions from growers regarding the project.
I wrote an article for VegEdge, the Cornell Cooperative Extension vegetable crops newsletter for growers and extension personnel. In the article, I provide information on seed treatment recommendations for onion maggot management based on field trials conducted by Brian Nault.
I presented this research at the the annual meeting of the Entomological Society of America in St. Louis, MO in November in the student 10 minute paper competition.
Upcoming, I will be presenting this research at the Cornell Entomology annual graduate student symposium (17 January 2020) as well as growers at the Empire State Producers Expo in Syracuse, NY (15 January 2020). At the Expo, I will give a 25 minute talk sharing the results and management implications of the research with onion growers from across the state.
In the summer and fall of 2018, the project outcomes were mainly in the form of grower contact. Working on this project, we met with growers sometimes on a weekly basis while monitoring their fields. These served as valuable opportunities to discuss the project with growers and to hear their concerns and management approaches. We were able to use these discussions to remind growers of the importance of rotating their fields and using different seed treatments.
This year (2019), our main project outcomes are from the environmental factors study. These results, which will be shared with growers on 15 January at the Empire State Producers Expo, will hopefully open conversations about effective management of this pest, and we look forward to the responses from growers, to whom this research has not been presented in its entirety
This summer concluded the bulk of the study identifying which environmental factors are associated with onion maggot damage. The results of the study, which found that planting date and surrounding landscape, rather than temperature and precipitation, were associated with onion maggot damage were surprising. While the literature points to a few reasons why this may be the case, the mechanisms behind these relationships is not fully elucidated. Nevertheless, these findings highlight areas of study on which research as not focused, which may be helpful for management in the future.
Similarly the findings from our preliminary spinosad uptake field study and damage evaluation, revealed some key findings about how spinosad seed treatments work to protect plants from onion maggot damage. Specifically, we confirmed that spinosad from seed treatments is taken up by the plant and is found in both the aboveground and belowground tissue. Spinosad is also found in the soil surrounded plants. In order to determine how onion maggot is exposed to the seed treatment we are currently working on both topical and ingestion bioassays with spinosad. The results of those bioassays in the form of LC50s for both modes of exposure compared with the amounts of spinosad recovered from the plant tissue and soil should help us determine where the maggot is most-likely being exposed to the insecticide and how susceptible they are to the exposure.