Exclusion of the Brown Marmorated Stink Bug from Organically Grown Peppers Using Barrier Screens (2024)

Kentucky site.

All greenhouse plantings and field trials were carried out at the University of Kentucky Horticultural Research Farm in Lexington (lat. 37°58′23.06″N, long. 84°32′8.18″W) in the organically certified section. During Apr. 2013 and 2014, bell pepper seeds were grown in 72-cell trays in 0.1N–0.02P–0.04K potting mix (Organic Choice Potting Mix; Miracle-Gro, Marysville, OH), following standard practices in the organic greenhouses on the farm. Before the construction of plastic-covered, raised growing beds in the field, 4N–0.9P–3.3K organic fertilizer (Fertrell Co., Bainbridge, PA) was incorporated into the 0.2-acre field site to a depth of 6 inches at a rate of 80 lb/acre N using a rotary tiller. Four pairs of 120-ft-long plastic-covered beds with underlying drip tape were laid using a plastic mulch layer. Pairs of rows were separated by 48 inches while rows within a pair were separated by 18 inches. The bell pepper plants were transplanted on the plastic-covered beds using a water-wheel setter on 14 May 2013 and on 20 May 2014. The bases of the peppers were covered with soil to prop the plants off the black plastic.

Cage frames were composed of two, overlapping pieces of 15-ft-long, 1/5-inch diameter reinforcing rebar, each end driven into the plastic-covered row to a depth of 8 inches, creating a domed shape. Electrical wire was wrapped around the top of the cage at the intersection of the rebar pieces for stability and was also wrapped around the circumference of the frame to keep the screens from encroaching on the peppers. Screen materials were cut into 16 × 14-ft pieces and draped over the cage frames. The screens were held to the ground by covering excess netting with 14 paving stones (8 × 6 by 3 inches) spaced ≈18 inches apart around the exterior of the cage. Each cage measured 6.5 × 6.5 ft, was ≈4.5 ft in height, and covered 24 pepper plants. Control plots did not contain a cage or frame but were delineated by research flags at their corners. Screens were constructed in the field on 23 May 2013 and 27 May 2014 before the initiation of flowering. The delay in the screening of plots likely did not influence the chance of trapping stink bugs inside the cages due to the low probability of their presence before flowering (Lee et al., 2013a).

To monitor beneficial insects, one yellow 15 × 30-cm sticky card (IPM Corn Rootworm Card; Great Lakes IPM, Vestaburg, MI) was secured with binder clips on an 18-inch wooden stake and was collected and replaced in each plot weekly. The stakes with yellow sticky cards were set in the middle of one plastic-covered bed within each plot, with the card positioned parallel to the rows of peppers. Collected yellow sticky cards were covered in thin plastic film, labeled with the appropriate collection date and plot, and stored in a cooler until they could be analyzed. All beneficial insects on the yellow sticky cards were identified to the family level. Visual searches within the plots were also conducted weekly to detect the presence of stink bugs. During these searches, the four middle plants in each plot were thoroughly examined and the species and life stage of any stink bugs found in the plots were recorded. Though sweep net sampling is more reliable for measuring the density of brown marmorated stink bug before they drop to the ground (Quarles, 2014), the limited space inside the plots and the risk of damaging the fruit lead to visual counts being employed instead for this study.

The plots were harvested three times during each year to assess the quality and quantity of peppers from each plot. The peppers from each plot were first counted and weighed together. Then, marketable peppers were divided into size categories according to the U.S. Department of Agriculture size categories of “Fancy,” “No. 1,” and “No. 2” (USDA, 2005). Culled, unmarketable peppers were counted and assessed for injury. Any incidence of sunscald and chewing insect injury was documented. The number of stink bug feeding sites on each pepper was recorded to determine damage severity. Fruit was harvested on 15–17 July, 30 July, and 12 Aug. 2013; and 8–10 July, 22–23 July, and 12–13 Aug. 2014. To reduce the movement of insects into the plots during harvest, researchers kept the cages closed except when entering and exiting the plots.

Tennessee site.

Field trials were conducted at the Organic Crops Unit of the East Tennessee AgResearch and Education Center in Knoxville (lat. 35°52′48.00″N, long. 83°55′47.95″W). In 2013, bell pepper seeds were sown in a greenhouse in 128-cell plug trays (Griffin Greenhouse Supply, Knoxville, TN) using 0.02N–0.03P–0.06K potting mix (Premium Lite Organic Grow Mix; McEnroe Organic Farm, Millerton, NY) and then potted up to 50-cell plug trays. In 2014, seeds were sown in 72-cell plug trays. Before the construction of plastic-covered, raised growing beds in the field, the field site was top-dressed with 8N–2.2P–4.2K organic poultry litter fertilizer (Nature Safe Natural and Organic Fertilizers, Cold Spring, KY) at a rate of 90 lb/acre N. The soil was rototilled to a depth of 6 inches using a rotary tiller (Bush Hog, Selma, AL) and a compact tractor (B Series; Kubota Tractor Corp., Torrance, CA). Four pairs of plastic-covered rows (1-mil thick), with underlying drip tape (10-mil thick, 12-inch emitters, T-Tape), were laid to a length of 150 ft. Pairs of rows were separated by 36 inches and rows were separated by 24 inches within a pair. Pepper plants were transplanted by hand on 27 May 2013 and 7 May in 2014.

Cage frames were constructed using 1/2-inch diameter metal electrical conduit and 1/2-inch polyvinylchloride elbows, with an insertion of 1/2-inch rubber hosing to secure the metal conduit. The cages were 4 ft high, 5 ft long, and 6 ft wide. Unscreened plots were delineated by the same cage frames as the screened plots. The cages were constructed on 8 July 2013 and 24 June 2014 and each cage covered 13–15 plants.

To monitor for pests and beneficial insects inside the covered and open plots, yellow sticky cards on wooden stakes were placed in the center of each plot and replaced weekly for 6 weeks. Yellow sticky cards were labeled, placed in clear plastic bags, and frozen until they could be analyzed. Insects on sticky cards were identified to family and all brown marmorated stink bug within the plots were counted on each harvest date.

The plots were harvested on 30 July, 13 Aug., 27 Aug., and 10 Sept. 2013; and on 17 July, 31 July, 14 Aug., 28 Aug., 10 Sept., and 19 Sept. 2014. The harvests were analyzed in the same way as the Kentucky sites, though peppers that contained stink bug and other types of injury were assessed and weighed separately. Damage recorded on peppers included, but was not limited to, zippering, cracking, splitting, soft rots and bacterial spots (E. carotovora), and fungal diseases [e.g., early blight (Alternaria solani) and gray mold (Botrytis cinerea)].

Data analysis.

Data collected during pepper harvests (including the number, weight, and sizes of peppers and the instances of various types of damage) along with data collected from yellow sticky cards (numbers of sampled beneficial and pest insects) were compared among treatments in SAS (version 9.3; SAS Institute, Cary, NC). A mixed model analysis of variance was performed on the data with year and treatment classified as fixed effects and block and block by year designated as random effects (Saxton and Auge, 2014). A Fisher’s least significant difference means separation was used to determine significant differences between treatment effects at α = 0.05 and test the null hypothesis that treatment effects would not be significantly different among treatment plots.

Kentucky site.

Though a black, fine mesh was tested in 2013 and a white, fine mesh was tested in 2014, their individual effects on the resultant pepper crops and their exclusion of natural enemies were mostly consistent both years. Therefore, the fine-mesh treatments were combined for analysis in Kentucky so they could be more easily compared with data from the Tennessee site, which tested white, fine mesh both years.

Beneficial insects identified on the yellow sticky cards included lady beetles (Coccinellidae), hover flies (Syrphidae), green lacewings (Chrysopidae), damsel bugs (Nabidae), damselflies (Order Odonata, Suborder Zygoptera), minute-pirate bugs (Anthocoridae), long-legged flies (Dolichopodidae), big-eyed bugs (Geocoris), ground beetles (Carabidae), and spined soldier bug (Podisus maculiventris). Significantly more beneficial insects were identified in the unscreened plots compared with the screened plots and significantly fewer beneficial insects were identified in the fine-mesh plots than in the other plots (Table 1). Significantly higher percentages of peppers with stink bug injury were found in the 1/8-inch mesh and unscreened plots (Table 2). The highest number of stink bug feeding sites on peppers was observed in the unscreened and 1/8-inch mesh plots and the lowest was observed in the 1/6-inch and fine-mesh plots. The highest incidence of insect chewing injury was observed in the unscreened and 1/6-inch mesh plots. Significantly more sunscald was observed in the unscreened plots when compared with the screened plots.

No stink bugs were observed in the Kentucky plots during weekly searches in 2013. Most of the stink bugs in 2014 were identified inside the 1/8- and 1/6-inch mesh plots, with three stink bugs identified in both treatment plots. A second instar, third instar, and adult green stink bug were identified in the 1/8-inch mesh plots and a third instar green stink bug and two adult brown stink bugs were identified in the 1/6-inch mesh plots. One adult brown marmorated stink bug was identified in the unscreened plots and no stink bugs were identified in the fine-mesh plots.

The total pepper yield in numbers and weight in Kentucky was significantly higher in the unscreened plots compared with the screened plots (Table 4). Though the number of marketable peppers was not significantly different among treatments, the percentage of marketable peppers was significantly higher in the screened plots compared with the unscreened plots. No significant differences were observed among treatments for the numbers of fancy or number 2 grade peppers (Table 5). There were significantly more number 1 grade peppers harvested from the unscreened plots when compared with the screened plots.

Tennessee site.

Insects identified on the yellow sticky cards included brown marmorated stink bug, fleahoppers (Halticus), flea beetles (Chrysomelidae), leafhoppers (Cicadellidae), lygus bugs (Lygus sp.), lady beetles, green lacewings, pollinators, big-eyed bugs, damsel bugs, carabids, and parasitoids. Parasitoids on the yellow sticky cards were not identified to family, though parasitoids from the genera Ooencyrtus, Telenomus, and Trissolcus were identified in the same location during a parallel study surveying egg parasitoids of the brown marmorated stink bug (Ogburn et al., unpublished data). Significantly fewer pest and beneficial insects were identified in the fine-mesh plots when compared with the other treatment plots (Table 1). Significantly more beneficial insects were identified in the 1/8-inch mesh and unscreened plots though beneficial insects identified in the 1/6- and 1/8-inch mesh plots were not significantly different from each other.

Most of the peppers with stink bug damage in Tennessee were found in the unscreened, 1/6-, and 1/8-inch mesh plots (Table 2). The percentages of peppers with stink bug damage among these three treatment plots were not significantly different and the percentages identified in the 1/6-inch and fine-mesh plots were not significantly different. The same trend was observed for stink bug injury severity with the exception of the 1/8-inch mesh plots that were not significantly different from the 1/6-inch and fine-mesh plots. There were no significant differences in the number of peppers that were culled for defects other than stink bug injury among the treatment plots.

Significantly, more brown marmorated stink bug adults were identified in the unscreened plots in Tennessee and their numbers were not significantly different among screened plots (Table 3). There were no significant differences for the total numbers or weight of peppers harvested from the treatment plots in Tennessee (Table 4). The highest numbers of marketable peppers were harvested from the 1/6-inch and fine-mesh plots. By percentage, more marketable peppers were harvested from the fine-mesh plots. Significantly, fewer fancy grade peppers were harvested from the unscreened plots when compared with the screened plots (Table 5). Significantly, more number 1 grade peppers were harvested from the fine-mesh plots and fewer number 1 grade peppers were harvested from the unscreened plots. No significant differences were observed among treatments for the number 2 grade peppers.