Several changes in agronomic and preventative control practices have brought about the need to reevaluate and improve insect pest management in Maryland soybean production. The proposed research effort will focus on three major questions driven by conversations with farmers, crop consultants, and extension agents. First, what is the current soybean insect pest risk, in terms of species present, abundance, and seasonal timing? Second, how do different full-season soybean planting dates change the pest damage risk and do sensitive crop stages overlap with significant pest activity? The activity of damaging pests depends primarily on seasonal timing, weather conditions, and pest life cycles, so any shift in planting date can alter the likelihood of pest damage. Third, what is the control efficacy and economic benefit of adding a pyrethroid insecticide to postemergence herbicide applications and are there any negative consequence from this tank mix approach? Whether insect pest pressure is present or not, the use of an early season insecticide as a preventative measure has been shown to cause more serious secondary pest outbreaks by knocking out natural enemies. The upcoming field season (2022) will be year one of a three-year project, to capture the range of annual variability in pest risk and thoroughly address these questions.

In Maryland, full season soybean planting dates are being pushed earlier, even into April, increasing the likelihood of cool and wet environmental conditions during the seedling stage, which will likely increase the risk of slug damage and soil-borne pathogens. Maryland extension agents and crop consultants report that stink bugs represent the insect pest complex of highest concern at this time, and that stink bugs are a major reason why insecticides are added to the postemergence herbicide application. Indeed, there is clear evidence that stink bug populations have increased in Maryland soybean crops, due to increasing average temperatures year-round, with warmer winters resulting in more stink bugs successfully overwintering. However, stink bugs prefer to feed on fruiting structures and thus populations reach highest densities during the reproductive growth stages when pods are forming (stages R3-R4) and seeds are filling (R5-R6). Furthermore, stink bugs are well-known to readily disperse across the farmscape to utilize available host plants, so it is unlikely that an early insecticide treatment mixed with the herbicide will have any significant suppressive effects on late season populations. Even if stink bugs are present when the postemergence herbicide is applied, the question is whether they are actually causing economic injury to the young soybean plants. Moreover, stink bug populations are subject to significant field edge effects, and their invasion into soybean fields is usually limited to less than fifty feet from the field edge (Venugopal et al., 2014). With these factors in mind, we aim to determine the population densities of stink bugs and other pest species that are present when postemergence herbicide tank mixes may be applied, as well other insect pest risks that could develop later in the season as a result of this treatment approach. For example, spraying an early season insecticide can eliminate the beneficial organisms that aid in aphid and spider mite control, resulting in later outbreaks of these pests.

An in depth understanding of the likelihood and timing of pest pressure increases the efficacy of management tactics and maximizes profits. In combination, our objectives will provide important information necessary for producers to optimize insect pest management in full season soybean. This project will also support extension and education efforts on soybean insect pest management.

To better understand full season soybean pest risks and the timing of management interventions, we will 1) assess the effects of two planting dates on the timing, abundance, and economic impacts of slug, insect, and early season pathogen populations across two soybean planting dates; 2) determine the control efficacy and economic benefits of adding a pyrethroid insecticide to the postemergence herbicide application; and 3) ascertain whether there are any increased pest risks later in the growing season that are triggered by the insecticide-herbicide tank mix approach.

Field experiments will be conducted at two University of Maryland Research and Education Centers in Beltsville and Queenstown, MD. We will use an untreated, high-yielding, late-three maturity group, full season soybean variety. Each experiment will be conducted at two separate locations per research center (four separate experiments per year), and we will coordinate with farm directors to ensure consistent plot design, row spacing, and plant populations (seeds/acre). Standard no-till Maryland full season soybean production practices will be used. Because drilled or narrow soybeans are relatively common and pest pressure varies with increased spacing, we are currently planning to use a drilled system in this study. However, we will consult the Maryland Soybean Board to ensure our production practices are as similar as possible to current standards. Depending on research land availability, plots will be as large as possible to improve yield measurements and better capture the pattern of pest densities and plant injury. Each experiment will be laid out in a two-factor split plot design. The main factor will be planting date (two dates) and the subplot will be insecticide versus no insecticide mixed with the postemergence herbicide application. The first planting will be planted as early as possible in (target mid to late April), and the second planting date will be approximately one month later. Main plots will measure at least 60’ x 60’, with subplots at least 30’ x 60’. The layout of each experiment will be positioned along non-cropped bordering areas such that we can test for edge effects in at least two directions. Each planting date by insecticide treatment combination will be replicated four times in each experiment.

Slug shelter traps will be installed at planting and checked weekly from planting until the V3 stage, and slug injury on plants will also be assessed. Stand counts, plant height, and damage from soil pests will be measured a few weeks after emergence. Each plot will be sampled at least biweekly from VE to R5 to measure the timing, abundance, and economic impacts of insect pest populations, using sweep netting and visual inspections of insect numbers. Recorded data will be compared to soybean thresholds. Plant injury including damage incidence (% of plants) as well as severity (amount of damage) will also be quantified. In addition to other measurements, average soybean growth stage will be recorded on each sampling date. To evaluate the efficacy of pyrethroid insecticide applications timed with postemergence herbicides, we will monitor for potential pest targets before and after sprays, additionally quantifying pest outbreaks that may result from reduced natural enemy activity. Due to their rapid population growth, aphids and spider mites most commonly outbreak because of insecticide applications.

Intensive sampling will also take place during the pod-filling stages, because soybean is vulnerable to losses in seed number and quality, particularly from stink bugs, at these stages. Therefore, we will visually inspect pods for damage incidence and severity. At harvest maturity, plant subsamples will be examined in each plot to record the number of immature pods and mature pods as well as seed number. Each plot will be machine-harvested to measure soybean grain weight and a random sample of at least 300 seeds will be taken to determine moisture, test weight, as well as be scored for seed damage and quality. We will use the damage and quality categories developed by Venugopal and Dively unpublished: 1) stink bug damaged seed, distinguished by a puncture scar and often surrounded by a discolored cloudy area; 2) moldy seed, characterized by having milky white or grayish crusty growth on surface, sometimes with cracks and fissures; 3) shriveled seed that appeared wrinkled and often undersized; 4) purple seed recognized as purple or pink areas on the seed coat due to the fungus Cercospora kikuchii; 5) green seed showing discolored green tissue in cross section, rather than the normal yellow; and 6) normal, undamaged seed.

Update:
We proposed to measure the timing, abundance, and economic impact of slugs, insects, and early-season pathogens across two planting dates and two farms and to assess the control efficacy and economic benefits of adding a pyrethroid insecticide to the post-emergence herbicide application. The plot design divides the soybean into treatment plots and control plots. The first plantings went in late April/early May (4/29 at WyeREC and 5/2 at CMREC) and the second plantings were done the first week of June (6/1 at WyeREC and 6/7 at CMREC). The early planted soybeans achieved canopy closure between 7/11-7/18 at CMREC and between 7/13-7/20 at WyeREC, and most plants are in the early stages of reproduction. We are tracking the important plant characteristics of height, width (until canopy closure), and growth stage in addition to measuring pest activity.

Since emergence, we have tracked slug numbers in shelter traps on a weekly basis as well as the captures of important moth species (fall armyworm, western bean cutworm, and corn earworm). Slug numbers were relatively high in the early season and significant losses in stand count were observed in the early planted soybean. Sticky cards and visual assessments have been used to identify and monitor the abundance of pests and natural enemies. The incidence of pest damage has also been evaluated with visual stand counts that determine the number of plants exhibiting different types of damage. Severity has been evaluated by measuring the amount of leaf area removed on a subset of damaged leaflets. Earlier in the season we did this nondestructively using transparent guides to estimate square centimeters of missing tissue, now we are collecting leaflets, digitizing them, and using LeafByte to analyze the area and % of consumed tissue. Quantitative defoliation measurements can be correlated back to the economic thresholds for defoliating pests. We are also sweep-netting, which is one of the most common sampling methods used in soybean economic thresholds. We plan to compare our data to these established thresholds to see if management would be recommended.

To determine the potential benefit of tank mixing insecticide with the post emergence herbicide, and how planting date may affect the return on investment, we are comparing herbicide alone to herbicide plus warrior. The highest labeled rate of Warrior II (1.92 fl oz/acre) was applied to the insecticide+herbicide treated plots. This is the rate recommended for grasshoppers, stink bugs and spider mites, which were identified by MSB in 2021 as key pests of Maryland soybean. We used 2,4-D and glyphosate herbicides. Treatments occurred at the end of June (6/28 at CMREC and 6/30 at WyeREC) in the early plots and the late planted treatment plots were sprayed 7/21 at CMREC and 7/22 at WyeREC. In order to detect pest control benefits, we focused our most intensive sampling efforts (sweep netting, defoliation assessments and visual counts of insects and damaged plants) on the dates immediately before spraying and about 2 weeks after spraying. We will then compare the insecticide sprayed plots to the herbicide only plots to determine if there was an effect.

Soybean is most vulnerable to yield losses from insect feeding during the reproductive stages and especially during pod-filling. The majority of the soybean in the early planted plots is in R1-R2. Some of the late planted soybeans are beginning to enter R1-R2 but most are still in the vegetative stages. The chosen soybean variety is indeterminate meaning that it continues to grow after flower production. For the remainder of the growing season we will continue to monitor growth stage as well as plant height and canopy closure, as well as insect timing, abundance, and impact. Some early season pathogens have been detected (Septoria and Cercospora) in the early planted plots, so we will continue to monitor for disease symptoms which often don’t appear until later growth stages. At harvest, we plan to destructively sample 3 feet of row in at least 4 sites per plot and record the number of immature and mature pods as well as any signs of disease or insect feeding. We will also take a random sample of at least 300 seeds per plot to measure moisture and weight and score for damage and quality. Once field work is complete, we will focus on analyzing the data and producing a more complete report to submit to the MSB board when applying for year 2 of funding this fall.

Update:
To better understand full season soybean pest risks and the timing of management interventions, our goals were to 1) assess the effects of two planting dates on the timing, abundance, and economic impacts of slugs, insects, and pathogens across two soybean planting dates; 2) determine the control efficacy and economic benefits of adding a pyrethroid insecticide to the postemergence herbicide application; and 3) ascertain whether there are any increased pest risks later in the growing season that are triggered by the insecticide-herbicide tank mix approach. Because pest pressure varies from site to site and year to year, multiple years of data in multiple locations will be required for robust results.

The combination of two planting date treatments [as early as possible (late April/early May) and one month later (June)] as well as two pesticide treatments [insecticide (Warrior II, lambda-cyhalothrin, 1.92 fl oz/acre) versus no insecticide mixed with the postemergence herbicide application] were replicated at two University of Maryland Research and Education Centers. Two field pers site with three replicates per field provided a total of six replicate subplots (45 ft by 64-79ft) planted at 15” row spacing using P38A54E full season soybeans planted to a population of 120,000 seeds per acre. To evaluate seasonal pest pressure and damage, we used monitoring traps to measure slug and podworm pressure. In addition, visual sampling, sticky cards, and sweep nets samples collected pest and beneficial abundance data. Plant injury including the damage incidence (% of plants) as well as its severity (amount of damage) was quantified for leaves and pods. We recorded seed quality and damage, plant lodging, as well as Dectes stem borer infestation at the end of the season. Plant growth was taken to better understand how pest pressure overlapped with soybean development. Finally, yield was compared.

We found few differences between planting date and postemergence pesticide treatments (insecticide+herbicide vs. herbicide alone). There were no differences due to our treatments in plant stand at the end of the season, Dectes stem borer infestation, test weight, or yield. For example, averaged across the entire study (24 plots each), earlier planted plots yielded 70.3 ± 3.1 bu/ac whereas later planted plots averaged 69.1 ± 3.6 bu/ac. Comparing the pesticide treatments (24 plots each), the insecticide plus herbicide plots averaged 69.4 ± 3.3 bu/ac and the herbicide only plots averaged 70.0 ± 3.4 bu/ac. However, fields and sites did vary (Table 1). Earlier plots had more lodged stems at the end of the season with 1.83± 0.38 lodged plants per 36 ft of row compared to later plots (n =24) that had 0.83± 0.30 lodged plants per 36 ft of row. Earlier plots also experienced significantly higher slug pressure (4.27 ± 0.40 slugs per trap per week) compared to later planted plots (2.83 ± 0.37 slugs per trap per week). The intensity of leaf defoliation was periodically measured from just before canopy closure through R2/R3 by selecting 15 representative damaged leaflets per plot. This conservatively estimates amount of damage because we looked for damaged leaflets. There was little overall difference in the earlier plots, with an average of 4.3 ± 0.5% in the insecticide treated compared to 4.4 ± 0.3% leaf area removed; numerically it did appear that insecticide reduced damage for a few weeks post treatment but this effect did not persist and overall damage was low. In the later plots, insecticide had a bit more effect with an average of 2.7 ± 0.2% in the insecticide treated compared to 3.5 ± 0.4% leaf area removed. Defoliation treatment thresholds are around 15-20% for the reproductive stages and higher for the vegetative stages. Our highest percent damage for this portion of the season was 11.6%, so there was not an economic level of defoliation and our treatments did not impact stand or yield.

A proposal for a second year of this study is currently in review.

View uploaded report

Update:
Introduction
To better understand full season soybean pest risks and the timing of management interventions in Maryland full season soybean, we 1) assessed the effects of two planting dates on the timing, abundance, and economic impacts of slugs, insects, and pathogens across two soybean planting dates; 2) determined the control efficacy and economic benefits of adding a pyrethroid insecticide to the postemergence herbicide application; and 3) ascertained whether the insecticide application resulted in secondary pest outbreaks. Because pest pressure varies from site to site and year to year, multiple years of data in multiple locations will be required for robust results.

Methods
The combination of two planting date treatments [as early as possible (late April/early May) and one month later (June)] as well as two pesticide treatments [insecticide (Warrior II, lambda-cyhalothrin, 1.92 fl oz/acre) versus no insecticide mixed with the postemergence herbicide application] were replicated at two University of Maryland Research and Education Centers, for a total of four treatment combinations. Two fields per site with three replicate plots per field provided a total of six replicate plots (45 ft by 64-79ft) per treatment combination, with P38A54E full season soybeans planted at 15” row spacing for a population of 120,000 seeds per acre.

To evaluate seasonal pest pressure and damage, we used monitoring traps to measure slug (weekly until 6 weeks after planting the later planting) and podworm (weekly for the entire season) pressure. In addition, visual sampling, sticky cards, and sweep nets samples collected pest and beneficial abundance data. We conducted four visual assessments between early June and early August, in which stand count, the number of plants damaged, pathogen incidence, and insects present were quantified. Samples were taken for lab identification when pathogen symptoms were observed. Plant growth was measured weekly for the majority of the season to better understand how pest pressure overlapped with soybean development. Sticky cards were deployed weekly from early June to mid-August (11 weeks) and sweep netting was done five times starting mid-June and ending mid-September.

Plant injury including the frequency of damage and severity (leaf: % area removal, pod: % damage) was quantified and described for leaves and pods. The intensity of leaf defoliation was measured every three weeks from just before canopy closure through R2/R3. In three different locations per plot, five different plants were selected as having representative types of damage. The uppermost/newest trifoliates were examined, and the leaflet with the median amount of damage was used to quantify leaf area removed, for a total of 15 leaflets per plot. The area defoliated was estimated for the first month after planting using transparencies marked with a 1cm x 1cm grid. For samples taken after the first month, defoliation was quantified destructively by collecting the leaflets, taking photos of them in the lab, and using the Leafbyte iPhone app to calculate the total area of the leaf as well as the percentage of leaf removed. Pods were destructively sampled in late August, about 2 months after the herbicide+insecticide spray for the early planting treatment and about 1 month after the spray for the later planting treatment to evaluate pod damage incidence as well as severity.

We additionally recorded plant lodging, Dectes stem borer infestation, and seed quality and damage at the end of the season. At one site, the entire plot was harvested and yield was measured using a truck scale. At the other, the center 30 ft of the plot was harvested and yield was measured using a weigh wagon. To evaluate seed quality, test weight, and percent moisture a subsample of seeds from each plot were collected from each plot after yield was weighed. Test weight and percent moisture were taken the day of harvest using a DICKEY-john® GAC 2100. The remaining seeds were stored for later damage assessments. Three hundred of these seeds were then examined individually for overall damaged, split, and undamaged values, the “Enlist shadow” visible on some seeds was categorized as undamaged. Damage categories including “stink bug”, “chalky white”, “purple stain”, and “brown stain” are being assigned to the unsplit damaged seeds.

Results and Discussion
We found few differences between the planting date and postemergence pesticide treatments (herbicide + insecticide vs. herbicide alone). Treatments did not impact final plant stand, Dectes stem borer infestation, test weight, or yield. For example, when averaged across the entire study (24 plots per treatment combination), earlier planted plots yielded 70.3 ± 3.1 bushels per acre while later planted plots averaged 69.1 ± 3.6 bushels per acre. Comparing the pesticide treatments, the herbicide plus insecticide plots averaged 69.4 ± 3.3 bu/ac and the herbicide only plots averaged 70.0 ± 3.4 bu/ac. However, fields and sites did vary, with one field experiencing (WyeREC Field 2) relatively lower yield than the others due to a combination of environmental and soil conditions. Earlier (May) plots had more lodged stems at the end of the season with 1.83± 0.38 lodged plants per 36 ft of row compared to later (June) plots that had 0.83 ± 0.30 lodged plants per 36 ft of row. Earlier plots also experienced significantly higher slug pressure (4.27 ± 0.40 slugs per trap per week) compared to later planted plots (2.83 ± 0.37 slugs per trap per week).

Plant damage severity (amount of leaf area consumed, Figure 1) and incidence (% of plants damaged, Figure 2) were lower in the insecticide treated plots for the first sampling date after treatment for both planting dates. However, our defoliation and damage incidence measurements conservatively estimate the amount of damage because we looked for damaged leaflets to quantify severity. For incidence, any plant that was damaged more than 0.4 in2 (2.6 cm2) in the seedling stage and more than 1 in2 (5 cm2) in later vegetative stages was considered damaged. Defoliation treatment thresholds are around 15-20% for the reproductive stages and higher for the vegetative stages. Even with our conservative methodology, our highest percent damage for this portion of the season was 11.6%. Therefore, we were suppressing subeconomic levels of foliar pest damage. When the June plots were first sampled, they were in the early vegetative stages (V1/V2) while the May plots were in the late vegetative stage (V5+). The earlier planting experienced heavier damage incidence than the later planting (Figure 2), with slightly higher damage severity (Figure 1). The higher incidence of damage in the earlier planted treatment persisted through flowering and the beginning of pod-fill, with the first week of August corresponding with R2-R3 in later plots and R4-R5 in early plots.

Visual samples of arthropod abundance revealed no differences in the mean number of pests, though the earlier planting date had slightly higher abundances than the later planting date (Figure 3). Beneficial abundance was relatively similar across treatments, and there may have been some reduction of beneficials after the insecticide treatment in the later planted plots (Figure 4). Sticky cards (912 cards) and sweep net samples (648 samples) are frozen and still being identified.

When pods were sampled for damage, stink bug and pathogen damage occurred most frequently. Overall pod damage incidence was under five percent across all treatments and pesticide treatment did not impact pod damage incidence. Earlier plantings did experience higher damage (3.68% ± 0.41%) compared to the later plantings (1.04% ± 0.19%). However, pod damage never exceeded the five percent economic threshold which is used for scouting for bean leaf beetle pod feeding, and is the lowest published economic threshold for soybean pods. Therefore, there was not an economic level of infestation that would justify management for pod-feeders in any of the treatments.

Preliminary visualizations indicate slight differences in the mean number of damaged seeds, related to both planting date and pesticide treatment, with a higher percent damaged in the early planting (67.3% ± 4.3%) than the later planting (58.8% ± 3.3%) and a lower percent damaged in the herbicide + insecticide treatment (61.0% ± 3.4%) than the herbicide treatment (65.1% ± 4.4%), with planting date having a greater effect than spray treatment. Seed quality analysis is ongoing, but brown stain appears to be the most common damage type, followed by stink bug damage. Although the amount of seed damage seems concerning, and earlier plantings exhibited poorer seed quality than later plantings, damage severity was very low and these seeds would not have warranted a quality inspection by seed buyers.

Conclusion
In summary, soybean planted in late April/early May seems to experience slightly higher pest pressure compared to June plantings; especially slug pressure, pods damaged, lodged stems, and damaged seeds. We did not detect a yield benefit from using insecticides at the postemergence herbicide timing, with small reductions in defoliation severity and incidence. As for potential non-target effects, there was no evidence of late season pest outbreaks, and we observed a small reduction in beneficials in our visual samples. Sweep net and sticky card analyses are ongoing, and we plan to repeat this study in the 2023 season to capture year to year variation in pest pressure.

View uploaded report

After one year of data collection in two fields at two sites in Maryland full season soybeans, we found little value in adding a pyrethroid insecticide to the postemergence herbicide application. Pest pressure was not economic and yield was not improved. This effect was consistent across two planting dates, one in early May and the other in early June. Earlier planted plots did experience slightly higher pest pressure; however, planting date also did not impact yield in this study. Another year of data will help determine whether this application regularly mismatches with Maryland insect pest pressure.

An in depth understanding of the likelihood and timing of pest pressure increases the efficacy of management tactics and maximizes profits. In combination, our objectives will provide important information necessary for producers to optimize insect pest management in full season soybean. This project will also support extension and education efforts on soybean insect pest management.