© 2007 Plant Management Network.
On-Farm Trial Assessing Efficacy of Three Insecticide Classes for Management of Stink Bug and Fruit Damage on Processing Tomatoes
Eileen M. Cullen, 1630 Linden Drive, Department of Entomology, University of Wisconsin, Madison 53706; and Frank G. Zalom, One Shields Avenue, Department of Entomology, University of California, Davis 95616
Cullen, E. M., and Zalom, F. G. 2007. On-farm trial assessing efficacy of three insecticide classes for management of stink bug and fruit damage on processing tomatoes. Online. Plant Health Progress doi:10.1094/PHP-2007-0323-01-RS.
Five foliar insecticides representing a neonicotinoid (thiamethoxam), a pyrethroid (lambda-cyhalothrin), a neonicotinoid plus pyrethroid (thiamethoxam and lambda-cyhalothrin), and two organophosphates (dimethoate and methamidophos, respectively) were evaluated for relative efficacy in managing consperse stink bug (Euschistus conspersus Uhler) on California processing tomatoes. E. conspersus density and percentage fruit damage were measured at harvest in an on-farm experiment at two locations in 2002, and small plot experiments in 2002 and 2003. Results showed that thiamethoxam plus lambda-cyhalothrin, lambda-cyhalothrin, and methamidophos can provide equivalent control of E. conspersus on processing tomatoes. Relative efficacy of the same treatments was inconsistent when applied before completion of small nymph development as estimated by the E. conspersus phenology model. Results can be utilized by growers with canning contracts that limit total organophosphate active ingredient per season, or in cases where stink bug treatment thresholds are reached within the processor preharvest interval for organophosphates. Insecticide efficacy evaluation is discussed within the context of continued research to provide IPM-compatible insecticide options to growers facing low processor tolerance for stink bug damaged fruit, and organophosphate insecticide restrictions in tomato canning contracts.
California produces 95% of the processing tomatoes (Lycopersicon esculentum Miller) grown in the United States (10,11). Consperse stink bug (Euschistus conspersus Uhler) nymphs and adults (Fig. 1) feed on fruit, resulting in feeding scars surrounded by spongy tissue beneath the skin (Fig. 2). Damage acceptable on tomatoes intended for paste or juice production can lead to down-grading or rejection for whole pack use (15).
Stink bug damage tolerance levels have not been defined by the processing tomato industry, and an economic injury level has not been published for this pest (18,20). However, Zalom et al. (18) established the relationship between canopy shake samples in tomatoes over the growing season and fruit damage at harvest. A mean of one-third to one-half stink bug per tray shake will result in about 5% damaged fruit, the maximum level acceptable to many growers (18). One-third to one-half bug per tray (nymphs and adults) is recommended as the stink bug treatment threshold for whole pack or dice canning fields in California (20). Approximately half (54%) of California’s processing tomato crop is treated with the organophosphate insecticides dimethoate or methamidophos (15). Within this estimate, recurrent stink bug problems account for an average of 10% of California’s processing tomato acreage treated annually with methamidophos (14). While processing tomato companies tolerate variable, yet low, levels of stink bug damage in whole pack and dice canning contracts, processors may voluntarily impose restrictions on organophosphate applications that are more strict than presented on the US federal label (8,13,19).
Methamidophos, for example, is included on the Food Quality Protection Act (FQPA) Group 1 list of pesticides for registration reassessment (13,19). Methamidophos is currently registered by the U.S. Environmental Protection Agency (EPA) and Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) under a special local need 24(c) supplemental label in California for use on tomatoes with a 7-day preharvest interval (2). To limit pesticide residues in processed products, processors may require growers to obtain prior approval for methamidophos use, limit total amount applied per growing season, extend the preharvest interval to at least 14 days, or limit treatment to field margins only (F. G. Zalom, personal communication). In contrast to organophosphates, the pyrethroid active ingredient lambda-cyhalothrin is registered by EPA under a FIFRA section 3 label for use on tomatoes for stink bug with a 5-day preharvest interval, and no additional processor restrictions (2) (F. G. Zalom, personal communication).
Continued field evaluation is required to compare efficacy of different insecticide classes that can be incorporated into practical integrated pest management (IPM) sampling and treatment decision protocols in situations where processor contracts restrict, or do not allow, application of an organophosphate insecticide. Although foliar applied neonicotinoid insecticides provide excellent potato aphid (Macrosiphum euphorbiae Thomas) control on processing tomatoes, these reduced-risk materials alone provide limited stink bug control (15). Plant systemic neonicotinoid efficacy in combination with insect contact activity materials has not been documented for stink bug control under commercial production conditions. The objective of this study was to evaluate relative field efficacy of foliar applied neonicotinoid, pyrethroid, and organophosphate insecticides for stink bug control on processing tomatoes.
In 2002, an on-farm experiment was conducted at two locations within 5 km of each other in Yolo County, California. Study sites were identified by road names adjacent to each field, Ensley (24.3 ha) and Becker (42.5 ha). Tomatoes were direct seeded (variety ‘Heinz 9663’) on 1.5-m row centers 10 and 25 April, harvested 12 and 26 August, respectively. Euschistus conspersus were monitored weekly at each field site using Euschistus spp. pheromone traps (IPM Tech., Inc., Portland, OR) and direct canopy shake samples (11) from beginning bloom (mid-June) to ripe fruit stage (mid- to late August) (5) (Fig. 3). Based on previous studies (3,5,6), accumulated E. conspersus degree-days were recorded (12) from observed occurrence of female-biased, maximum pheromone trap capture, 13 June (Ensley) and 18 June (Becker), respectively (Fig. 4). Accumulation of 310 degree-days (DD) was used according to the phenology model published for this species to forecast small nymph (first to third instar) development within the field (3,12). In addition to insecticide class, treatment efficacy was evaluated by application timing relative to 310 DD required for E. conspersus egg incubation to completion of third instar (3) as small nymph stages are most susceptible to control by any available insecticide (9,20).
The on-farm experiment consisted of three foliar insecticide treatments representing a neonicotinoid tank-mixed with a pyrethroid (thiamethoxam and lambda-cyhalothrin); a pyrethroid (lambda-cyhalothrin); and the cooperating grower’s organophosphate standard (methamidophos) for stink bug control (Table 1). Treatments were arranged in a randomized complete block, including an untreated control, and replicated four times at each location. Each plot measured 8 rows (12.2 m) by 61 m length (0.07 ha). Treatments were applied by the cooperating grower’s commercial applicator with ground rig spray equipment at a 234-liter/ha volume, 8010 flat fan nozzles (TeeJet West, Meridian, ID) and sprayer pressure of 70 to 80 psi. Treatments were applied when E. conspersus densities reached one-third to one-half bug per tray in all plots, Ensley and Becker combined, F = 0.65, df = 7, 24, P = 0.7124 (a = 0.05; PROC GLM v8.2, SAS Institute Inc., Cary, NC). A single application was made at Ensley (5 July) and Becker (17 July), 226 and 327 E. conspersus degree-days from the respective female-biased, maximum pheromone trap capture dates (13 and 18 June) observed in each field (6,17) (Fig. 4).
Table 1. Insecticides evaluated on-farm for efficacy to reduce E. conspersus density and fruit damage at harvest on processing tomatoes, Yolo Co., CA, 2002.
Small Plot Experiments
In 2002 and 2003, complementary small plot experiments were conducted at the University of California Vegetable Crops Research Farm in Davis, CA. Tomatoes (variety ‘Heinz 8892’) were direct seeded (2002) or transplanted (2003) on 1.5 m row centers. Individual plots measured 4 rows (6.1 m) by 3 m length (0.0019 ha). Euschistus conspersus populations were established by artificial infestation from a laboratory colony (3) started each spring with adults collected in Yolo Co., CA. Plots were infested when fruit development reached the blush stage. In 2002, each plot was infested with three E. conspersus egg masses on 13 July, and five second to third instar nymphs on 15 July. In 2003, one E. conspersus egg mass was placed along each 3-m length of tomato row within the experimental planting between 22 July and 25 July, for a total of 4 egg masses per plot.
Experiments consisted of four foliar insecticide treatments representing three different insecticide classes and one tank mix of two classes (Table 2). Treatments were arranged in a randomized complete block including an untreated control, and blocks were replicated four times in 2002 and three times in 2003. Treatments were applied on 22 July 2002 and 1 August 2003, respectively, using an Echo Duster/Mister backpack sprayer at a volume equivalent to 468 liter/ha.
Table 2. Insecticides evaluated in small plot trials for efficacy to reduce E. conspersus density and fruit damage at harvest on processing tomatoes, Davis, CA, 2002-2003.
Treatment Efficacy Evaluation and Statistical Analysis
To evaluate efficacy in the on-farm experiment, live E. conspersus nymphs and adults were counted from 10 canopy shake samples taken from the two center rows of each plot 7 and 14 days after treatment. Additionally, one week prior to harvest at each location, 100 red fruit were randomly selected from the two center rows of each plot and individually rated for stink bug damage. Fruit were scored as undamaged (0 stink bug feeding scars) or damaged (≥ 1 stink bug feeding scars).
In the small plot experiments, live E. conspersus nymphs and adults were counted from 5 canopy shake samples taken from the two center rows of each plot on 16 August 2002 and 11 August 2003. In addition, 100 red fruit samples were rated from each plot as previously described on 19 August 2002 and 10 September 2003.
On-farm experimental data (2002) for mean E. conspersus/tray and proportion of stink bug damaged fruit were pooled, respectively, from both locations and analyzed using concatenated analysis of variance with insecticide and accumulated E. conspersus degree-days as classification variables (a = 0.05; PROC GLM v8.2, SAS Institute Inc., Cary, NC). Small plot experimental data (2002, 2003) for mean E. conspersus/tray and proportion of damaged fruit were analyzed separately for each year using analysis of variance with insecticide as the classification variable (a = 0.05; PROC GLM v8.2, SAS Institute Inc., Cary, NC). When necessary to stabilize variances, mean bugs/tray (log[x+0.1]) and proportional fruit damage means (arcsine [square root (y)]) were transformed before analysis of variance. Means were separated by protected least significant difference test (a = 0.05; PROC MEANS v8.2, SAS Institute Inc., Cary, NC), but only as appropriate after a significant F-test.
Effects of Insecticides on E. conspersus density and Fruit Damage Under Commercial Production Conditions, On-Farm Experiment, Yolo Co., CA, 2002
Euschistus conspersus densities exceeded one-third to one-half bug/tray (11) in untreated control plots at Becker (327 DD) and Ensley (226 DD), 7 and 14 days after treatment, respectively. Mean percentage fruit damage at harvest exceeded 50% in the untreated controls at both locations (Table 3). Seven days after treatment, there was a significant treatment effect on E. conspersus density by location (F = 3.12; df = 7, 24; P = 0.0173). At both locations, untreated control plots had the highest mean density of E. conspersus, but were not significantly different than thiamethoxam plus lambda-cyhalothrin, or lambda-cyhalothrin alone applied at 226 DD. Mean densities were lowest, and did not exceed the one-third bug/tray treatment threshold, for thiamethoxam plus lambda-cyhalothrin and lambda-cyhalothrin alone at 327 DD; and for methamidophos applied at 226 and 327 DD. At 14 days after treatment, there was a continued significant treatment effect on E. conspersus density by location (F = 3.64; df = 7, 24; P = 0.0081). Treatment means separated according to the same significance pattern described for 7 days after treatment, with the exception that methamidophos applied at 226 DD did exceed the one-half bug/tray treatment threshold (Table 3). Treatments also had a significant effect on percentage of fruit damage at harvest (F = 8.33; df = 7, 24; P < 0.0001). All treatments had a lower percentage of fruit damage than the untreated control for either location, except lambda-cyhalothrin applied at 226 DD. Thiamethoxam plus lambda-cyhalothrin and lambda-cyhalothrin alone, each applied at 327 DD, resulted in fruit damage reductions equivalent to methamidophos at 226 or 327 DD (Table 3).
Table 3. Density of E. conspersus, 7 and 14 days after treatment (DAT), and percentage fruit damage at harvest on processing tomatoes in on-farm demonstration experiment, Yolo Co., CA, 2002.
Means within a column followed by the same letter are not significantly different (a = 0.05; least significant difference test, LSD, PROC MEANS v8.2, SAS Institute Inc., Cary, NC).
x Along with insecticide class, on-farm treatments were evaluated by application timing relative to respective field estimates of 310 degree-days accumulated for E. conspersus egg incubation to completion of third instar.
Effects of Insecticides on E. conspersus Density and Fruit Damage in Small Plot Experiments, UC Davis Vegetable Crops Research Farm, Davis, CA, 2002 and 2003
At the UC Davis Vegetable Crops Research Farm insecticide treatments did not have a significant effect on E. conspersus density 24 days after treatment, 16 August 2002 (F = 2.25; df = 4, 19; P = 0.1119); or 10 days after treatment, 11 August 2003 (F = 1.27; df = 4, 14; P = 0.3444). In both years, mean E. conspersus densities exceeded the one-half bug per tray threshold for all treatments (Table 4). There was a significant treatment effect on percentage fruit damage at harvest in 2002 (F = 4.06; df = 4,19; P = 0.0200) and 2003 (F = 6.86; df = 4,14; P = 0.0063). As expected, the untreated controls had the highest percentage fruit damage, but only thiamethoxam plus lambda-cyhalothrin, and lambda-cyhalothrin had significantly lower fruit damage than the untreated control. This result was consistent over both trial years (Table 4).
Table 4. Density of E. conspersus, 24 (2002) and 10 (2003) days after treatment, and percentage fruit damage at harvest on processing tomatoes in small plot experiments, UC Davis Vegetable Crops Research Farm, Davis, CA, 2002-2003.
Means within a column followed by the same letter are not significantly different (a = 0.05; least significant difference test, LSD, PROC MEANS v8.2, SAS Institute Inc., Cary, NC).
Our on-farm experiment showed that all three treatments can provide the same efficacy against E. conspersus on processing tomatoes (Table 3). Interestingly, the most efficacious treatments included a neonicotinoid plus pyrethroid tank mix, a pyrethroid, and an organophosphate applied at the Becker location 327 DD after observed maximum E. conspersus pheromone trap catch. Relative efficacy of the same neonicotinoid plus pyrethroid, and pyrethroid treatments was inconsistent at the Ensley location where insecticides were applied to plots 226 DD after observed maximum E. conspersus pheromone trap catch. At the 226 DD application timing only the organophosphate methamidophos resulted in a significant reduction in fruit damage relative to the untreated control.
Although significant treatment effects were maintained for E. conspersus density in the on-farm demonstration experiment with 0.07-ha plot size, direct canopy shake sampling in our small plot (0.0019 ha) trials may not be a reliable indicator of insecticide efficacy due to adult movement between plots (F. G. Zalom, personal communication). Based on the current study, and our previous experience, percent fruit damage at harvest is a more accurate measure of insecticide efficacy in small plot trials. Using mean percent fruit damage, the most efficacious insecticides against E. conspersus over the two small plot experiment years included the same neonicotinoid plus pyrethroid tank mix and pyrethroid treatment identified in the on-farm demonstration experiment. Thiamethoxam plus lambda-cyhalothrin and lambda-cyhalothrin alone resulted in significantly better fruit protection than the untreated controls, and equivalent (2002) or better (2003) fruit protection relative to the organophosphate dimethoate.
While fruit damage values in the on-farm experiment (Table 3), and small plot trials (Table 4), exceeded the 5% damage level acceptable to many growers (18), these data represent treatment efficacy obtained in heavily infested fields areas, and not a whole field average. In 2000, Cullen et al. (4) documented a highly aggregated field distribution pattern for E. conspersus in processing tomatoes, particularly in field sides or corners adjacent to stink bug overwintering habitat. Aldrich et al. (1) and Zalom et al. (18) indicated that E. conspersus aggregation pheromones further reduce the tendency for within-field dispersal and resulting fruit damage. Presumably, statistically significant fruit damage reductions observed in heavily infested commercial field areas and small plots in the current study will contribute to lowering overall stink bug fruit damage incidence for the harvested field.
Conclusions and Recommendations
This study reports new information on a neonicotinoid plus pyrethroid combination, and a pyrethroid, as effective foliar applied insecticide treatments against E. conspersus on processing tomatoes compared to an organophosphate standard under commercial production conditions. However, our results found no statistically significant advantage to the neonicotinoid plus pyrethroid combination versus a pyrethroid alone. In our small plot experiments, no apparent differences in E. conspersus control, as measured by percent fruit damage at harvest, were observed between a neonicotinoid alone and the untreated control. Although the reduced risk neonicotinoid active ingredient thiamethoxam (Actara, Syngenta, Greensboro, NC) was registered for foliar use on tomatoes for stink bug control during the period in which the current study was completed, the manufacturer has since voluntarily cancelled foliar use on tomatoes (16). Currently, thiamethoxam (Actara) is registered for foliar use on peppers against stink bug at the same rate used in the current study. On tomatoes, thiamethoxam (Platinum, Syngenta, Greensboro, NC) is now registered as a soil applied formulation for aphids, flea beetles, Colorado potato beetles (Leptinotarsa decemlineata Say), and whiteflies, but not stink bug.
This study represents an important initial step by documenting an effective chemical class substitution (7) for organophosphates in tomatoes for E. conspersus control. In the near term, this information can be utilized by tomato growers and consultants with whole pack or dice canning contracts that limit organophosphate applications, or in cases where stink bug treatment thresholds are reached within the processor pre-harvest interval for organophosphates. Pyrethroids are broad spectrum materials that also control other insect pests of processing tomatoes such as potato aphid and tomato fruitworm (Helicoverpa zea Boddie) (11). However, pyrethroids are not considered reduced risk as applications will also kill natural enemies such as insect predators and parasitoids, which can play an important role in regulating certain tomato insect pest species (11).
More research is needed in this area to provide IPM-compatible insecticide options for stink bug control on processing tomatoes to growers who simultaneously face low processor tolerance for stink bug damaged fruit, and organophosphate insecticide restrictions in whole pack or dice tomato canning contracts. Our preliminary field efficacy evaluation results support previous laboratory and field studies (3,5,6) suggesting improved E. conspersus control provided by insecticide application timed to the third instar nymphal stage, approximately 310 DD after an estimated pheromone trap catch biofix date. A “biofix” is simply a biological marker that initiates the beginning of degree day accumulation used to forecast temperature-dependent insect developmental stages (17). In this and previous studies, the biofix is presumably the date of early season maximum pheromone trap capture significantly comprised of reproductively active females (5,6). Refined field efficacy evaluation efforts can further validate the E. conspersus phenology model under commercial production conditions by repeated trials with earlier and more frequent pheromone trap catch records (e.g., beginning in late May, collected every 2 to 3 days thereafter) to identify true “peak” catch as a field scouting and treatment decision biofix in each field. Conceivably, improved phenology-based treatment timing toward the more susceptible small nymph stage will provide an appropriate method for continued field evaluation of a wider range of insecticide active ingredients, including reduced-risk materials, for stink bug control in processing tomatoes.
Acknowledgments and Disclaimer
This study was supported by the USDA CSREES Pest Management Alternatives Program and the California Tomato Research Institute. The authors thank tomato grower D. Richter for allowing us to conduct the on-farm experiment at two locations in 2002, and Toby Glik for his field sampling assistance in 2002 and 2003.
Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement.
1. Aldrich, J. R., Hoffman, M. P., Kochansky, J. P., Lusby, W. R., Eger, J. E., and Payne, J. A. 1991. Identification and attractiveness of a major pheromone component for nearctic Euschistus spp. stink bugs (Heteroptera: Pentatomidae). Environ. Entomol. 20:477-483.
3. Cullen, E. M., and Zalom, F. G. 2000. Phenology-based field monitoring for consperse stink bug (Hemiptera:Pentatomidae) in processing tomatoes. Environ. Entomol. 29:560-567.
4. Cullen, E. M., Zalom, F. G., Flint, M. L., and Zilbert, E. E. 2000. Quantifying trade-offs between pest sampling time and precision in commercial IPM sampling programs. Agric. Sys. 66:99-113.
5. Cullen, E. M., and Zalom, F. G. 2005. Relationship between Euschistus conspersus (Hem., Pentatomidae) pheromone trap catch and canopy samples in processing tomatoes. J. Appl. Entomol. 129:505-514.
6. Cullen, E. M., and Zalom, F. G. 2006. Euschistus conspersus female reproductive morphology and attraction to methyl 2E, 4Z – decadienoate pheromone baited traps in processing tomatoes. Entomol. Exp. Appl. 119:163-173.
7. Epstein, L., and Bassein, S. 2003. Patterns of pesticide use in California and the implications for strategies for reduction of pesticides. Annu. Rev. Phytopathol. 41:351-375.
8. Flint, M. L., Cullen, E., Zilbert, E., Zalom, F. G., Miyao, G., and Coviello, R. 1998. Pest management record-keeping duties shift with computerization. Calif. Agric. 52:27-31.
9. Hoffman, M. P., Wilson, L. T., and Zalom, F. G. 1987. Control of stink bugs in tomatoes. Calif. Agric. 41:4-6.
10. Rickard, B. J., and Sumner, D. A. 2006. EU support reductions would benefit California tomato growers and processors. Calif. Agric. 60:207-210.
11. University of California. 1998. Integrated Pest Management for Tomatoes, 4th Ed. Publ. No. 3274, Univ. of Calif. Div. of Agric. and Natur. Resourc., Oakland, CA.
13. US EPA. 1999. Implementing the food quality protection act: progress report. US Environ. Prot. Agency Off. of Prevention, Pesticides, and Toxic Substances. EPA Publ. 735-R-99001, Washington, DC.
15. Van Steenwyk, R. A., and Zalom, F. G. 2002. Tomatoes (fresh and processed). In: The Economic Importance of Organophosphates in California Agriculture. Online. M. Metcalfe, B. McWilliams, B. Hueth, R. VanSteenwyk, D. Sunding, D. Zilberman, eds. Calif. Dept. of Food and Agric., Sacramento, CA.
17. Wilson, L. T., and Barnett, W. W. 1983. Degree-days: An aid in crop and pest management. Calif. Agric. 37:4-7.
18. Zalom, F. G., Smilanick, J. M., and Ehler, L. E. 1997. Fruit damage by stink bugs (Hemiptera:Pentatomidae) in bush-type tomatoes. J. Econ. Entomol. 90:1300-1306.
19. Zalom, F. G. 2003. Pests, endangered pesticides and processing tomatoes. Acta Hort. 613:223-233.
20. Zalom, F. G., Trumble, J. T., Fouche, C. F., Summers, C. G., and Toscano, N. C. 2005. UC pest management guidelines: Tomato, stink bugs. Online. UC ANR Publ. 3470. Agric. and Natur. Resourc., Davis, CA.