Temporal efficacy of commercially available pheromone lures for monitoring Helicoverpa zea (Lepidoptera: Noctuidae)

Abstract

The performance of commercially available corn earworm, Helicoverpa zea Boddie (Lepidoptera: Noctuidae), female sex pheromone lures that attract male moths from four different commercial vendors was evaluated to determine the duration of catch in the field. Lures were aged outdoors beneath Scentry Heliothis traps for a period of 0, 2, 4, 6, and 8 weeks before standardized field testing in New York and North Carolina. Scentry Heliothis traps baited with aged lures were monitored for male H. zea for 14 d across two independent replications in each state. Gas chromatography-mass spectrometry study (GC-MS) was used to quantify the release rates and residual pheromone content in the dispenser of 11-Z-hexadecenal (Z11-16Ald), the major pheromone compound, for a subset of field-aged lures. We found an unequal decline in H. zea attraction efficiency among commercial lure types (significant lure type × age interaction). Lure type and age were also related to trap capture (model main effects), which are two important factors when selecting lures and determining replacement intervals in the field. Release of pheromone over 24 h was significantly different among lure types. Notably, discrepancies between residual pheromone and actual emission rates in septa lure highlights the need for better commercial dispensers with longer longevity and higher pheromone controlled release. Results highlight clear differences in lure performance that could directly affect the probability of H. zea detection in the field. Impacts of these temporal emission differences may increase the frequency of lure replacement in field applications.

Introduction

Monitoring approaches using synthetic analogs of sex pheromones targeting different taxonomic groups have been a component of Integrated Pest Management (IPM) programs for decades (Baker et al. 2009, Rizvi et al. 2021). For lepidopteran pests, pheromone-based trapping systems are used as a decision support tool, often deployed in coordinated trapping networks that inform management decisions in both perennial and annual cropping systems (Trumble 1997, Witzgall et al. 2010, Hodgson et al. 2023). Given the widespread use of these tools in agriculture, demand for field-deployable pheromones is significant with multiple commercial lure vendors providing different products. Although the specific chemistry of the sex pheromone attractant may be identical, proprietary carrier materials and pheromone loading amounts can vary among lures (Table 1). Differences in lure composition may affect trapping efficiency which complicates efforts to move toward universal recommendations for management thresholds based on insect trap capture.

Table 1.

Commercially available lures tested

Lure type	Vendor	Pheromone blend	Reported pheromone load (mg lure−1)	Dispenser type
Alpha Scents Corn Earworm	Alpha Scents, Inc., Canby, Oregon	97% Z11-16Ald, 3% Z9-16Ald	3	Rubber septa
Hercon Zealure Luretape	Hercon Environmental, Emigsville, Pennsylvania	87% Z11-16Ald, 3% Z9-16Ald, 2% Z7-16Ald, 8% 16Ald	2.5	Laminated polymer
Scentry Corn Earworm	Scentry Biologicals Inc., Billings, Montana	97% Z11-16Ald, 3% Z9-16Ald	5	Rubber septa
Trécé Pherocon Corn Earworm	Trécé Inc., Adair, Oklahoma	97% Z11-16Ald, 3% Z9-16Ald	3	Rubber septa

Lures targeting male corn earworm, Helicoverpa zea Boddie (Lepidoptera: Noctuidae), are one example of lure composition differences that may translate into efficacy variation among lure vendors. In a recent study, Mahas et al. (2025) evaluated the performance of four commercially available H. zea lures deployed on three common trap types (ie Hartstack, Scentry Heliothis, and Universal unicolor forest green bucket trap). Results showed that average percent moth capture per night was significantly different among four commercially available H. zea lures. Using a standard Hartstack trap, the authors found a significant difference in trap efficiency among lure types across locations in New York, Maryland, Delaware, Virginia, and North Carolina. In a similar study, Kwadha et al. (2025) documented greater H. zea trap captures increased when traps were modified with colored cloth. A series of thorough experiments show that visual contrast is an important factor that could improve trap capture (Kwadha et al. 2025). Together, these two studies highlight that the efficacy of different commercially available lures and traps vary, which is an important consideration for stakeholders using pest activity to determine potential crop injury risk. These results are especially important for crop scouts that deploy H. zea pheromones to inform insecticide interventions in sweet corn, Zea mays L. convar. saccharata Koern. In the 1990s, dynamic trap-based H. zea thresholds were established specifically for sweet corn to reduce the prevalence of ear injury (Dively 1996). The thresholds were developed using a single pheromone lure type (ie Hercon Zealure Luretape, Hercon Environmental, Emigsville, Pennsylvania) in combination with standard Hartstack traps developed in the late 1970s (Hartstack et al. 1979). These thresholds have been used successfully in sweet corn for decades, however, stakeholders often consider lures interchangeable which could result in misleading threshold interpretation based on unequal lure efficiencies (Mahas et al. 2025).

Documenting the performance of commercial pheromone lures over time is critical to recommendations for trap capture interpretation and replacement frequency. Studies into the effect of lure age have been the basis for IPM programs in annual crops (Adams et al. 1989, Showler et al. 2005, Armstrong et al. 2006), perennial crops (Anshelevich et al. 1994, Knutson et al. 1998, Kovanci et al. 2006, Hossain et al. 2008, Stelinski et al. 2009, Sullivan et al. 2023), forestry (Evenden et al. 1995), and stored product systems (Mullen et al. 1991, Guarino et al. 2020). These lure optimization studies provide useful information about the duration of effective trapping for different pheromone blends, lure components, and trap combinations. For H. zea, studies documented the efficacy of 11-Z-hexadecenal (Z11-16Ald) in the 1970s when it was embedded into a septa dispenser (Sekul et al. 1975, Klun et al. 1979, 1980) or a laminar polymer matrix such as Hercon disperser (Leonhardt and Moreno 1982), which became the basis of many commercially available pheromone monitoring tools available today.

Despite the identification of an effective pheromone component that is attractive to male moths, considerable knowledge gaps exist regarding the differences in individual product efficacy and the duration of lure effectiveness. In this study, we evaluated aged commercially available H. zea pheromone lures outdoors for 0 to 8 wk and tested their moth capture potential following the aging period. To quantify differences in pheromone release following aging, we also conducted a gas chromatography-Mississippi study (GC-MS) to assess pheromone emissions over 24 h and quantified the remaining pheromone in the lure dispenser. We hypothesized that unaged commercially available lures would have equivalent trapping efficiency. Further, we hypothesized that the rate of capture would vary by commercial vendor and lure age (age by lure interaction).

Materials and Methods

Field Evaluation

Lures from different suppliers were aged in two separate replicates at each study location (Table 1). Exposure dates and average daily environmental conditions for each experimental replicate are reported in Table 2. An additional six replicates per treatment were aged simultaneously for chemical analysis. All lures were secured to Scentry Heliothis traps (Scentry Biologicals, Inc., Billings Montana) using binder clips to enable adequate air flow and exposure to the environment during the aging period. Three lure types are rubber septa (Table 1); these lures were clipped so the primary opening was positioned in a downward orientation to reduce complications with moisture interacting with interior of septa. The other lure type is laminated polymer and was clipped to the trap in the same position as the rubber septa. The lures were positioned at the same level as the opening of the trap. Lures were aged for 2, 4, 6, or 8 wk in the field beginning in spring 2023 (Table 2). Lures were stored in a −20 °C freezer until they were used for either chemical analysis or trapping experiments.

Table 2.

Lure aging intervals, trapping periods, and environmental conditions during the lure aging period in New York and North Carolina

Location	Lure aging replicate	Aging start	Aging end	Trapping start	Trapping end	Average daily temperature (°C ± SD)	min. (°C)	max (°C)
NY	A	19 May 2023	14 July 2023	28 August 2023	4 September 2023	18.2 (4.1)	1.7	31.7
	B	2 June 2023	28 July 2023	11 September 2023	25 September 2023	19.8 (3.2)	7.2	31.7
NC	A	16 June 2023	11 August 2023	11 August 2023	25 August 2023	26.9 (1.9)	16.7	36.7
	B	30 June 2023	25 August 2023	25 August 2023	8 September 2023	27.7 (1.5)	18.3	37.8

Temperature data during the aging interval at each location was obtained from the Global Historical Climatology Network (Menne et al. 2012).

Twenty Scentry Heliothis traps were positioned in a standardized linear transect spaced 100 m between traps along the edge of field corn or sweet corn fields in North Carolina and New York, respectively. Lure by age treatments (4 × 5 factorial randomized design) were assigned randomly to traps for evaluation of H. zea capture efficiency. Traps were monitored for H. zea capture over 14 d for two experimental exposures in each state. Treatments were rerandomized on day seven to limit the potential for local landscape effects (eg proximity to alternate habitats) affecting the probability of trap capture. Studies were timed to coincide with typical flight windows of H. zea in each state (Table 2). Trap capture was recorded daily in New York and every other day in North Carolina.

Preliminary evaluation of trap data revealed significant location-wise differences in trap counts over time, which complicated reasonable treatment comparisons. To standardize these H. zea counts among states and experimental runs, the average proportion total of moths captured per night was calculated for each treatment by dividing the average number of moths captured per night for each treatment by the sum of all averages for the 20 treatments (total moth catch). This enabled straightforward comparison of treatment effects (lure type by age combinations) while accounting for differences in moth abundance between locations.

Pheromone Emission

To estimate pheromone release differences among aged lures, we conducted two specific chemical analyses. First, the release rate over 24 h was determined under controlled laboratory conditions (hereafter “emission analysis”). An estimate of the pheromone remaining in the lure was also conducted (hereafter “residual analysis”). For pheromone emission analysis, lure replicates were placed in a sealed jar connected to an air pump (500 ml/min) at the inlet and an adsorbent (activated carbon, Supelco Orbo 32) at the outlet. Following 24 h at 22.2 °C (72°F), the adsorbent was eluted with 1 ml of hexane in a vial for subsequent GC-MiS analysis. To estimate the residual pheromone in the lure, the lure was suspended in 20 ml of hexane at 22.2 °C for 9 d, after which a 1 ml aliquot of the extract was transferred into a GC-Mississippi vial for further GC-MS analysis.

All samples were analyzed using an Agilent 7890B GC coupled with a 5977C mass spectrometer (MS) (Santa Clara, California). Liquid samples were analyzed using the following instrument parameters: 1 µl of liquid sample was injected at a 10:1 split onto a 60 m × 0.25 mm × 0.25 µm HP-5MS column (Agilent Technologies, Santa Clara, California) with a notched glass inlet liner (Gerstel) at a constant He flow of 1.2 ml/min. The inlet was held at 40 °C for 0.5 min and then ramped at 12 °C/s to a final temperature of 280 °C, where it was held for the remainder of the run. The GC oven program began at 45 °C, then was ramped according to the following protocol: 10 °C/min to 100 °C; 3 °C/min to 130 °C; 10 °C/min to 180 °C; and finally 30 °C/min to 280 °C, where it was held for 5 min. The MS was operated in scan mode with a m/z range of 29–550 and a solvent delay of 5.5 min. Agilent MassHunter Quantitative Analysis software (Version 12.1, Agilent, Santa Clara, California) was used to analyze all data based on a calibration curve of 11-Z-hexadecenal (Z11-16Ald, Sigma-Aldrich reference 249084) which is the major component of the pheromone lure blend. For the calibration, sample analysis involved injecting 3 replicates of Z11-16Ald standard solution in hexane at concentrations of 0.5 μg/ml, 1 μg/ml, 5 μg/ml, and 10 μg/ml for the emission analysis and at concentrations of 1 μg/ml, 5 μg/ml, 25 μg/ml, and 100 μg/ml for the residual analysis. The ion counts of these known concentration standards collectively served as the basis for constructing calibration curves for the absolute quantification of Z11-16Ald.

Statistical Analysis

To determine H. zea trap efficiency in the field experiment, we tested the main effects of lure type (categorical variable), lure age (continuous variable), and their interaction. Here, we analyzed the effect of lure type on trap capture while accounting for lure age as a continuous covariate with an ANCOVA modeling framework. The response variable was average proportion of total moths caught per night. Because proportional data was doubly bounded between 0 and 1, we analyzed the data with beta regression (Ferrari and Cribari-Neto 2004) using the betareg package in R (Cribari-Neto and Zeileis 2010). Residual distribution and associated diagnostic plots were visually examined using the DHARMa package (Hartig 2024). Uniformity of residuals was tested using a Breusch and Pagan test in the lmtest package in R (Zeileis and Hothorn 2002). The overall model fit statistic was calculated using a Cox and Snell pseudo R² using the rcompanion package (Mangiafico 2025). Tests for statistical significance for lure type, age and their interaction were calculated using the joint_tests function in the emmeans package (Lenth 2025). Because we observed differences among lure types, we also calculated categorical multiple means separations for lure type using Sidak tests in emmeans. Regression model fits were calculated using the predict function in base R. Differences in average proportion of total moths caught per night between aging extremes (0 or 8 wk aged lures) were calculated from predicted fits for each lure type and used to calculate fold change in moth capture (fold change = [0 age lure—8 wk age lure]/8 wk age lure).

Emission and residual pheromone data were analyzed with two different generalized linear models that tested for main effects of lure type, a continuous variable for age, and their interaction. A gaussian distribution with an identity link function provided the best fit with uniformity of residual distribution following fitting. Model diagnostics followed the same general approach described above. Categorical multiple means separations for lure type using Sidak tests in emmeans.

Results

Field Studies

Across two experimental runs nested within two locations, we observed a significant interaction between lure type and age on the average proportion of total moths caught per night (χ² = 13.74; P = 0.003). This result suggests that the effect of lure type on average proportion of total moths caught per night depends on lure age (Fig. 1). Specifically, that the probability of trap capture is not the same for all lures across 0 to 8 wk of lure aging with Hercon lures capturing the most moths over the study duration (Fig. 1). We also observed significant main effects of lure type (χ² = 67.38; P < 0.001) and lure age (χ² = 53.48; P < 0.001). Important differences exist between lure types, which shows that lure selection in general can affect the probability of moth capture (Fig. 2A). Moreover, the general effect of aging time shows that extended periods of monitoring could lead to declining trap capture as a function of pheromone depletion or degradation driven by environmental conditions (Fig. 2B).

Fig. 1.

Effect of lure type and aging duration on average proportion of total H. zea trap capture per night. Regression fits illustrate a significant lure type by age interaction. Please note, observed data has been jittered to improve visual comparison.

Fig. 2.

Effect of lure type on average proportion of total H. zea trap capture per night. Mean separation using Tukey HSD (α = 0.05) A). Temporal change in average proportion of total trap capture per night across 8 wk of artificial lure aging B). Please note, observed data has been jittered to improve visual comparison.

GC-MS Emissions and Residual Assays

The released and remaining quantities of the Z11-Ald corn earworm pheromone varied significantly among lure types over an 8-wk period (lure × time interaction: F = 31.29; df = 3, 112; P < 0.001). Main effects of time (F = 264.48; df = 1, 112; P < 0.001) and lure type (F = 53.26; df = 1, 112; P < 0.001) were also significant, which align with insect trap capture results presented above. Results indicated that the Hercon lure emitted approximately 1.5 µg per 24 h for a lure with an unaged lure (Fig. 3A). Emissions declined to 0.78 and 0.038 µg per 24 h in weeks 4 and 8, respectively. The lures from Scentry had a higher pheromone load of 5 mg, twice the amount as the lures from Hercon (Table 1), and an unaged lure emitted around 0.86 µg per 24 h. Emissions declined to 0.52 and 0.18 µg per 24 h in weeks 4 and 8, respectively. This lure had the shallowest slope, which indicated the most consistent emission profile over the study period. The Alpha Scents and Trece lures, both containing 3 mg of pheromone (Table 1), an unaged lure emitted approximately 0.5 µg per 24 h, with emissions reduced to 0.3 µg after 4 wk of aging. Interestingly, Hercon lure, with the lowest pheromone loading amount (2.5 mg), exhibited the highest initial emission rate during the first 4 wk, but declined at a rapid rate between 6 and 8 wk (Fig. 3A).

Fig. 3.

Effect of lure type and age on emissions of Z11-16Ald measured from headspace using GC-MS over 24 h A). Remaining Z11-16Ald measured in a 1.5 ml aliquot of hexane used to extract residual pheromone over a 7 d period following initial headspace measurements B). Please note, observed data has been jittered to improve visual comparison.

We observed a significant time by lure type interaction for the amount of Z11-Ald corn earworm pheromone remaining in lures (F = 3.22; df = 3, 112; P = 0.026). Main effects of time (F = 233.92; df = 1, 112; P < 0.001) and lure type (F = 80.69; df = 1, 112; P < 0.001) were also significant. The amount of residual pheromone after the emission experiment revealed similar trends, with relatively consistent amounts of pheromone remaining in three lures and near depletion by week eight in the Hercon lure (Fig. 3B) whereas other three lures still contained a large quantity of pheromone despite the low emission rate at week eight. It is important to note that three of the lure types (Alpha Scents, Scentry, and Trécé) were traditional rubber septa, which had similar trends in lure amount reduction over time (Table 1). In contrast, the laminated polymer used by Hercon had a very different retention profile over time, exhibiting more pronounced decay pattern than the three the septa lures.

Discussion

Here, we use field-based observation methods to clarify how relative attraction to H. zea varies among different commercially available lures. Similar approaches, referenced above, have been applied for other taxa to compare efficacy of pheromones and associated lure types. Our results show that the emission rates, not the concentration inside the lure, are associated with lure performance (Fig. 1, Table 1). Similarity of model fits between moth capture data and 24 h emission measured by GC-MS suggest that pheromone emission rate at a biologically relevant overnight temperature is a reasonable indicator of lure efficacy in field conditions. These findings underscore that pheromone lure selection is important for stakeholders aiming to maximize H. zea detection probability without incurring additional costs of frequent lure replacement. This result is important because current sweet corn spray thresholds recommend using Hercon lures to determine specific spray intervals while the crop is in the susceptible reproductive stage. Our results provide trap capture evidence that Hercon lures may have the longest biological activity across the 8-wk period. Specifically, relative change in average proportion of total moths caught per night between week 0 and 8 were considerable based on fitted values. Alpha Scents and Trécé lures declined 20- and 41-fold, respectively (Fig. 1). In contrast, both Scentry and Hercon lures only declined by approximately 2-fold over the same period. The large decline in trap efficiency suggests that some lures may need to be changed more frequently to ensure equivalent probability of moth capture. As a result, trap network operators may need to carefully consider the necessary frequency of lure changes, the likelihood that lures are operating at a level sufficient for accurate monitoring, and the cost of lures.

Differences in lure performance in this study may be attributed to pheromone blend composition and the rate of release. Helicoverpa zea sex pheromone consists of four compounds (Z11-16Ald, Z9-16Ald, Z7-16Ald, and 16Ald) (Klun et al. 1979). Among the lure tested, three of the four commercial lures include a blend of only two primary components (Z11-16Ald and Z9-16Ald), whereas the Hercon lure contained all four components (Table 1). Prior research has documented that many lepidopteran pests demonstrated stronger behavioral response to a complete pheromone blend than to partial blends containing only primary components (Linn and Roelofs 1989 and references therein). For example, Linn et al (1986) observed that the addition of minor pheromone components significantly enhanced Trichoplusia ni (Hübner) (Lepidoptera: Noctuidae) and Argyrotaenia velutinana (Walker) (Lepidoptera: Totricidae) and Grapholita molesta (Busck) (Lepidoptera: Totricidae) male sensitivity and recognition of pheromone and pheromone source location, even at lower doses. The pheromone release by H. zea females varies over time, with average release rate (±SE) of primary component (Z11-16Ald) ranging from 0.89 ± 0.25 ng min⁻¹ (3.20 ± 0.9 µg per 24 h) during high release period (first 2 h of scotophase) to 0.17 ± 0.03 ng min⁻¹ (0.61 ± 0.1 µg per 24 h) during lower emission period (between 5 and 21 h after the onset of scotophase) (Pope et al. 1984). The mean release rate of Hercon and Scentry lure, 0.78 and 0.62 µg per 24 h respectively, was observed either at or above the H. zea females low emission rate during the first 4 wk. In contrast, the rate of release of Alpha and Trécé lures was already closer to the H. zea females low emission rate when they had not been aged. This suggests that matching both blend complexity and realistic emission rates is essential for sustained field attraction, though more detailed behavioral validation remains needed.

The observed discrepancy among lures indicates that pheromone emissions vary not only independently from initial loading, but also in their longevity profiles. This finding has practical implications for lure replacement schedules, particularly in climates with large temperature fluctuations. Pheromone evaporation is higher during the day when temperatures peak, leading to more significant daytime emissions, whereas nighttime emissions are comparatively lower (McDonough et al. 1989). This variation underscores the influence of temperature on pheromone release, a factor that may be underappreciated in predicting lure longevity and efficacy for monitoring the activity of nocturnal insect pests. Future research should quantify the impact of temperature-driven emission patterns across a broader range of environmental conditions interact with pheromone chemistry (ie chain length, degree of unsaturation, functional groups, volatility, and boiling point) to refine lure application strategies.

Controlled-release lures generally rely on diffusion through a barrier (dispenser material); if the concentration gradient remains constant, release follows zero-order kinetics and is steady over time, but most formulations instead follow first-order kinetics where release rates decline as residual pheromone decreases (McDonough 1991). Our data support that aging alters release profiles differently depending on dispenser shape (Fig. 3A). Pheromone mobility is influenced by its volatility and interactions with the matrix, but dispenser design can help improving first-order kinetics with a faster pattern of release. We observed that pheromone near the surface of a flat dispenser evaporates steadily, while pheromone deeper in a septum may remain trapped and unavailable for diffusion. The laminar Hercon design increases surface area, promoting more stable and controlled release even at lower loadings (Leonhardt and Moreno 1982). Additional factors, such as matrix porosity and polymer type, could also modulate release rates over time (McDonough 1991, Hossain et al. 2008, Stelinski et al. 2009, Sullivan et al. 2023, Hellmann et al. 2024). Overall, this work emphasizes the importance of considering environmental and temporal (lure age) factors as well as dispenser design when evaluating H. zea lure performance in field conditions.

Despite our new understanding about pheromone release rates and trap catch data over time, we are unable to estimate the point at which the threshold of attraction required by H. zea intersects with pheromone emission levels as lures age. Our results suggest that the optimal replacement timing likely differs among vendors (Fig. 1). However, the specific timing of the change is difficult to assess given the variation among sample locations that necessitated temporal averaging for the analysis. Measuring insect responses to specific pheromones at different emission rates involve gas chromatography coupled with electroantennography (GC-EAD) recordings (Almaas et al. 1991, Myrick and Baker 2012) or challenging behavioral assays in wind tunnels (Miller and Roelofs 1978, Vickers et al. 1991). While these studies improve understanding at an individual level, field population studies are a gold standard to assess lure technology and performance in pest monitoring programs. New portable electroantennograms (pEAG) technology could further understanding of pheromone mediated behavior in the field (Pawson et al. 2020), but specific approaches to use this tool are not widely available at this time. Future field studies could explore when lure types lose attractiveness, but this approach is complicated by dynamic adult H. zea activity patterns related to generational turnover over time (Jackson et al. 2008, Head et al. 2010, Lawton et al. 2022). Because adult populations are not temporally static throughout the growing season, additional levels of control are needed to decouple population dynamics from lure efficacy. To address activity differences, studies could integrate a black light trap to estimate population activity independent of response to pheromone emission. Although more complex, this approach would help establish a baseline for comparing male moth captures across multiple lure types in time and would provide reasonably unbiased threshold for effective lure longevity. Analyses could include both environmental conditions (e.g., temperature fluctuations) and black light captures as additional covariates to address pitfalls in the present study design. Outcomes would better estimate when pheromone emission depletion begin to misrepresent true moth activity.

The information generated by the current study provides important context for lure selection to maximize capture probability. As a result, selection of the most efficacious lure type could increase the odds of early detection and, in turn, enable stakeholders to make informed management decisions based on established economic thresholds. In practice, understanding lure type differences clearly matters for monitoring efforts that provide H. zea activity information for multiple crops. Evaluation of both trap design (Kwadha et al. 2025, Mahas et al. 2025) and lure optimization are important to provide stakeholders with reasonable estimates of trap sensitivity that affect recommendations for H. zea monitoring and management.

Author Contributions

Anders Huseth (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Funding acquisition, Writing—original draft, Writing—review & editing, Supervision), Christophe Duplais (Conceptualization, Data curation, Investigation, Methodology, Funding acquisition, Writing—review & editing), Sujan Panta (Investigation, Formal Analysis, Writing—review & editing), Kanika Jakhmola (Investigation, Writing—review & editing), Lucas Seybert (Investigation, Formal analysis, Writing—review & editing), and Brian Nault (Conceptualization, Data curation, Investigation, Methodology, Funding acquisition, Writing—review & editing)

Funding

This research is supported by the Specialty Crop Research Initiative, project award no. 2023-51181-4115, from the U.S. Department of Agriculture’s National Institute of Food and Agriculture. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and should not be construed to represent any official USDA or U.S. Government determination or policy.

Conflicts of Interest

The authors have no relevant financial or non-financial interests to disclose.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.