Vertical tillage effects on crop production and pest management in Pennsylvania

Abstract

Within the last four decades, widespread transition to no-till corn (Zea mays L.) and soybean [Glycine max (L.) Merr.] production in Pennsylvania has improved soil conservation and soil quality but can result in residue and pest management challenges. To effectively manage residue in no-till cropping systems, some growers have adopted vertical tillage, a residue management practice characterized by cutting and incorporating crop residue within the top 5–10 cm of soil. Despite few studies documenting effects on crop production and soil conservation, vertical tillage has become widespread. Replicated on-farm trials were conducted over a 2-year period in 2021–2022 to improve grower and consultant decision-making regarding the role of vertical tillage relative to continuous no-till on southeast Pennsylvania farms located within the environmentally sensitive Chesapeake Bay Watershed. We assessed the effects of vertical tillage on corn residue cover, winter annual weed abundance, slug damage, and soybean performance in 40 paired strip trials comparing spring vertical tillage to no-till using three different vertical tillage tools used by farmer cooperators. Vertical tillage equipment type is a driver of variation in changes to surface residue cover. Baseline surface residue cover was similar among no-till strips, but a greater proportion (32%) of strips had mean surface residue cover levels below a 60% conservation program compliance threshold when a Kuhn-Krause Excelerator was used. Relative to no-till, vertical tillage resulted in a 50% reduction in winter annual weed cover, a 24% reduction in slug damage, and no significant differences in soybean stand establishment or grain yield.

1 INTRODUCTION

Adoption of no-till practices has steadily increased in the Mid-Atlantic over the last 40 years, and approximately 70% of cropland in Pennsylvania was managed using no-till practices in 2017 (USDA NASS, 2017). In addition, the average corn (Zea mays L.) grain yield in the United States increased 40% over a 25-year period (1990–2015), and Pennsylvania corn grain yield increased from approximately 6.5 to 9.5 Mg ha⁻¹ (Adler et al., 2015) in the same time frame. One of the most evident management challenges of no-till crop production in high-yielding corn environments is the accumulation of corn stover left on the soil surface postharvest (Jeschke & Heggenstaller, 2012; Lorenz et al., 2010). While surface residue cover in no-till systems reduces soil erosion, promotes water infiltration, and reduces evaporation (Adler et al., 2015; Wilhelm et al., 2010), excess residue creates additional management challenges. In the last two decades, a subset of Mid-Atlantic growers has practiced a form of shallow non-inversion tillage, commonly known as “vertical tillage,” to manage corn stover and prepare seedbeds for planting (Chen et al., 2016; Smith & Warnemuende-Pappas, 2015). Although vertical tillage is now a widespread agronomic practice in the Mid-Atlantic, little is known regarding its effects on soil conservation, pest management, and crop production goals.

Vertical tillage implements are designed to be residue management tools for preparing an adequate seedbed by cutting and incorporating crop residue within the top 5–10 cm of soil to speed decomposition (Chen et al., 2016; Schomberg et al., 1994). To meet policy thresholds for soil conservation incentive programs in Pennsylvania (i.e., the Resource Enhancement and Protection (REAP) program guidelines for “low-disturbance residue management equipment”), disk blade angles of vertical tillage tools must not exceed 5° and have no concavity, working depth must not exceed 10 cm, and surface residue cover must not fall below 60% throughout the year (PA State Conservation Commission, 2019).

The use of vertical tillage as an integrated weed management (IWM) tool in no-till production is largely unexplored. No-till production results in the concentration of weed seeds in the upper (0–10 cm) soil profile (Yenish et al., 1992) and selects for small-seeded species that are more susceptible to mechanical disturbance at early growth stages (Gallandt, 2006; Nichols et al., 2015). The intensity and mixing efficiency of tillage, as well as the timing of tillage operations relative to weed emergence and growth stages, influence the effectiveness of shallow tillage as an IWM tool. For example, Bates et al. (2012) reported that control of small, early-emerging annual weeds with a preplant rotary harrow equipped with non-concave and non-angled coulters was comparable to burndown herbicides. Use of shallow tillage implements (<10-cm working depth) has produced variable levels of Erigeron canadensis L. control (Brown &Whitwell, 1988; Chahal & Jhala, 2019; Vanhie et al., 2021), which is a facultative winter annual with variable emergence patterns (Buhler & Owen, 1997).

Accumulation of crop residues at the surface may also create a favorable environment for no-till pests that complete part of their life cycle in soil. A recent meta-analysis did not find evidence that arthropod pests and slugs (Mollusca:Agriolimacidae, Arionidae) were more abundant in no-till systems compared to non-inversion and inversion-tillage systems (Rowen et al., 2020). However, over 80% of no-till growers in Pennsylvania reported slugs as one of the greatest pest management challenges in their cropping systems (Douglas & Tooker, 2012). Slugs favor cool, wet conditions in high-residue environments and use previous crop residues as refuge habitat (Douglas & Tooker, 2012). By incorporating crop residue into the soil and speeding decomposition in spring, vertical tillage may mitigate early-season slug abundance and damage to crop seedlings in no-till cropping systems.

Core Ideas

• Vertical tillage reduced no-till corn residue cover by 16% points relative to no-till across 40 strip trials.
• The most aggressive vertical tillage tool in the study reduced corn residue cover below state conservation program compliance thresholds (<60%) in 32% of strips.
• Vertical tillage resulted in a 50% reduction in winter annual weed cover relative to paired no-till strips.
• Vertical tillage resulted in a 24% reduction in the incidence of slug damage relative to paired no-till strips.
• Vertical tillage did not influence soybean crop establishment or grain yield relative to no-till production practices.

Trends from a small body of research suggest that vertical tillage may improve crop performance, including crop stand establishment and crop yield (Van Dee, 2005; Watters & Douridas, 2013). Many growers use vertical tillage tools to create a warming and drying effect on soil conditions to achieve timelier planting and more even crop emergence and establishment in high-residue environments. Assessing the impact of shallow non-inversion tillage on crop emergence and establishment, as well as on crop yield, is of keen interest to growers and agronomists.

Greater understanding of the impact of vertical tillage on soil conservation, pest management, and crop production may inform development of best management practices (BMPs) for effectively integrating vertical tillage into no-till cropping systems while adhering to soil conservation and surface residue cover thresholds. Toward this end, a multi-criteria assessment of vertical tillage was conducted with the use of replicated on-farm paired strip trials across 2 years (2021–2022) in grain corn residue prior to full-season soybean [Glycine max (L.) Merr.] in southeast Pennsylvania. Relative to no-till production, we report the effects of spring-implemented vertical tillage on the change in (1) corn residue surface cover; (2) winter annual weed abundance at the time of planting; (3) incidence and severity of leaf defoliation due to slug feeding; (4) soybean emergence and establishment; and (5) soybean grain yield. We hypothesized that relative to no-till, vertical tillage would significantly reduce surface residue cover, winter annual weed abundance, and the incidence and severity of slug damage, while improving soybean establishment, including stand density and growth stage, and soybean grain yield.

2 METHODS

2.1 Strip-trial locations

Crop production and pest management effects of vertical tillage were compared to no-till practices within a grain corn to full-season soybean crop rotation using an on-farm strip trial approach in southeast Pennsylvania (Lancaster and Chester Co.) in the 2021 and 2022 growing seasons. Most of the farms in this study were located on very deep, well-drained soils on uplands, formed in residuum or colluvium from limestone, micaceous limestone, calcareous schist, micaceous schist, siltstone, shale, or similar parent materials. These silt loam and loam soils are located in a moderate climate and frequent manure applications contribute to relatively fertile and historically high-yielding crop production environments on all cooperating farms.

Cooperating farms consisted of cash grain operations raising corn and soybean in rotation with other cash crops such as winter wheat (Triticum aestivum L.) and winter barley (Hordeum vulgare L.). On each cooperating farm, paired strips were established in fields rotating from grain corn, where corn residue was left unharvested and undisturbed over winter, to full-season soybean. Fields were left fallow in the corn to soybean transition except for two fields in 2021 and one in 2022, which had a late-planted cover crop terminated early in the spring of the following year.

2.2 Experimental design

Vertical- and no-till treatments were imposed in 20 field-length paired strips each year (Table S1) using a nested experimental design to control for multiple sources of variation (Figure 1). To account for sources of variation in treatment responses due to baseline soil conditions, a soil management legacy factor (n = 2) was nested within each year. Soil management legacy was identified as either (1) long-term no-till practiced for more than 10 years or (2) vertical tillage occurring generally on an annual basis for at least the previous 8 years. Cooperating farms were nested within soil management legacy to account for farm-level management sources of variation. Paired strips were nested within farm to account for crop management and soil-related sources of variation. The number of paired strips within farm differed each year due to field limitations, with paired strips replicated across single- to multiple-fields per farm or, in some cases, replicated within larger fields that contained variability in landscape position. In 2021, strip trials occurred on a total of nine farms and in 12 fields. In 2022, strip trials occurred on a total of 12 farms and in 17 fields.

FIGURE 1

Nested experimental design with fixed effects included tillage treatment (VT, vertical till; NT, no-till) and random effects included paired strips nested within farm, soil management legacy, and study year. Colocated transects located across landscape positions are nested within paired strips. Equipment type varied across farms by year and is included as an additional blocking factor with three brands of tools used by farmer cooperators (Kuhn-Krause Excelerator, Great Plains Turbo Till, and Salford Independent).

2.3 Vertical tillage treatment

On-farm cooperators used owned or rented implements at an average working depth of 5 cm to manage corn residue in the early spring (April) prior to establishing full-season soybean (Table S1). The vertical tillage tools were compliant with standards for low-disturbance residue management equipment as defined by the REAP program guidelines for fiscal year 2019 (PA State Conservation Commission, 2019). Three different vertical tillage tools were used on cooperating farms in 2021, including a Salford Independent (Salford Group, Inc.), a Great Plains Turbo-Till (Great Plains Manufacturing, Inc.), and a Kuhn-Krause Excelerator (Kuhn North America, Inc.). In 2022, cooperating farms used a Great Plains Turbo-Till or a Kuhn-Krause Excelerator. Vertical tillage equipment often consists of straight (i.e., non-concave) disk blades or “coulters,” either individually spring-mounted or collectively gang-mounted, followed by one or more rows of spiked wheels or coil tines, and then followed by a set of rolling baskets. The Salford and Great Plains Turbo-Till have individually mounted non-concave coulters that are fixed at a 0° angle. The Kuhn-Krause Excelerator models used in this study have slightly concave gang-mounted coulters that may be angled up to 5°. Variations in equipment design across brands create varying levels of aggressiveness when cutting residue and creating soil disturbance and mixing. When manufacturing of vertical tillage tools first began several decades ago, these tools traditionally limited the horizontal shearing and mixing of soil, instead creating soil disturbance on a vertical rather than a horizontal plane (Chen et al., 2016).

Vertical tillage treatments occurred in the spring approximately 14–21 days prior to planting in a randomly located field-length strip (Table S1), which created a paired vertical tillage and no-till strip. The width of each strip in the field was based on the size of available harvesting equipment, so one or two combine passes could be completed within tillage treatment strips. Strip length ranged from approximately 130 to 690 m. In most site years, vertical tillage was completed once a year in the spring. At one location in 2022, vertical tillage was conducted twice in the spring prior to planting full-season soybean. At three locations in 2022, corn residue was shredded in the previous fall with a flail mower.

Each cooperator implemented their own fertility, soybean variety selection, and crop protection programs for full-season soybean (Table S1). Preplant burndown and preemergence residual herbicides, along with any postemergence in-crop herbicides, were applied at the discretion of the cooperator. Most strips in 2021 and all strips in 2022 received a burndown herbicide application, a preemergence soil-applied residual herbicide and a postemergence herbicide (Table S1). All burndown herbicide applications occurred after vertical tillage but prior to planting. Due to low weed pressure in the early spring at two locations in 2021, no burndown herbicide was applied after vertical tillage.

2.4 Data collection

To account for within-field sources of variation, three data collection transects were established in unique field positions (e.g., summit, shoulder, backslope, foot slope, and toe slope) within each strip and colocated between paired strips (Figure 1). The location of transects was marked using georeferencing software (QGIS Geographic Information System, 2022) and a GPS receiver (Garmin GLO Portable GPS and GLONASS receiver, Garmin Ltd.) and waypoints were used for data collection throughout the trial.

Surface residue cover (%) was determined each spring approximately 21 days after vertical tillage and just prior to soybean planting (Table S1). Surface residue cover was quantified using the line-transect method described by the United States Department of Agriculture-Natural Resources Conservation Service standard surface residue cover assessment protocol (USDA-NRCS, 1984). One 15-m transect was established at each waypoint location (n = 3) within each strip, and the presence or absence of corn residue along the transect was recorded every 15 cm. Surface residue cover was expressed as a proportion of the total observations per transect (n = 100).

Weed abundance was measured at the same time, using the same transects that were used to measure surface residue cover. The belt-transect sampling method was used to quantify weed abundance at each waypoint location (n = 3) within each strip, so weed abundance could be correlated with surface residue cover. Within each belt transect, the presence or absence of weeds located within 15 cm on each side of the transect was recorded every 15 cm. Weed cover was expressed as a proportion of the total observations per transect (n = 100).

Slug damage was assessed in late spring at the V1–V3 soybean growth stage by establishing two transects at each waypoint location (n = 3) within each strip. Within each transect, the number of soybean plants per 3 m of row was counted and assigned a slug damage severity rating based on leaf defoliation (%) via slug feeding, where 0 = no apparent damage, 1 = 0%–25% defoliation, 2 = 26%–50% defoliation, 3 = 51%–75% defoliation, and 4 = 76%–100% defoliation (Douglas & Tooker, 2012). The total number of plants per 3 m of row and the number of plants damaged within a given leaf defoliation category were averaged across the six transects assessed within each strip and the incidence and severity of slug damage expressed as percentages.

Soybean emergence was assessed in late spring (June), approximately 30 days after planting, by measuring the emerged plant population and soybean growth stages, which ranged from the cotyledons expanded (VC) growth stage to the V3 growth stage at the time of the assessments. In fields with crop rows spaced 38 cm apart, two 5.3-m-long transects were established within each strip at each waypoint location (n = 3). In fields with crop rows spaced 76 cm apart, one 5.3-m-long transect was established. Emerged soybean plants were counted along transects, and soybean stand density was expressed on a plants per square meter basis.

Soybean stand development was also assessed during the population assessments using the same transect. Twenty plants were counted along the transect, their growth stages recorded, and an average growth stage was determined by averaging the growth stage measured at each waypoint location (n = 3) within each strip.

Crop yield was measured at harvest in the fall using a combine yield monitor or by weighing grain harvested from each strip using a weigh wagon or truck scale. The calibration status of the yield monitors was not always known but should not compromise the validity of results as differences in crop grain yield between strips were of greater importance than the absolute grain yield from a given strip. All harvest and weighing equipment were provided by on-farm cooperators, local custom harvest operators, or agricultural supply retailers. Crop moisture was determined using either a combine yield monitor if collecting yield data using the combine or by a grain moisture tester (Moisture Chek PLUS SW08120, Deere & Company) if collecting yield data using a weigh wagon or truck scale.

2.5 Statistical analysis

The effects of vertical tillage on soil conservation, pest management, and crop performance metrics were analyzed with linear mixed-effects (LME) models using the lme package (Bates et al., 2015) in R Statistical Software (v4.2.1; R Core Team, 2022). Prior to analysis, each response variable was expressed as the difference between vertical tillage and no-till treatments measured at paired waypoints within paired strips. The treatment difference (vertical tillage − no-till) was then modeled using a random intercept model with paired strip nested within farm, farm nested within soil management legacy, and soil management legacy nested within year. Equipment type was fit as an additional random intercept term to account for sources of variation attributed to the intensity of vertical tillage treatments (Figure 1).

With use of this random intercept model, fixed effects were limited to the population-level intercept, which is an estimate of the average difference between tillage treatments across grouping levels. A one-sided t-test (df = 39) was used to determine if population-level intercepts were significantly different from zero (p < 0.05), thereby providing a statistical test of vertical tillage treatments relative to the no-till control strips. Due to significant within- and among-strip variation in pest metrics, winter annual weed cover and slug damage incidence were subjected to linear regression using the described mixed effects model by including no-till weed cover or slug damage incidence in the no-till strip, respectively, as a continuous predictor variable. The effect of baseline pest conditions in no-till strips on the magnitude of change after vertical tillage was evaluated using log-likelihood ratio tests (Wald χ²) to compare the regression model to the null random-effects model using the anova function.

Two test statistics were calculated to describe sources of variation in the model. First, the conditional (R²c) coefficient of determination was calculated to describe the proportion of the total variance in the response variable attributable to estimated random effects using the MuMin package (Nakagawa & Schielzeth, 2013). Next, variance partition coefficients, or intraclass correlation coefficients (Nakagawa & Schielzeth, 2013), were calculated by extracting variance estimates for each grouping level from lme models using the VarCorr function and then expressing each estimate as a proportion of the total variance, or sum of variance parameters. Variance partition coefficients for each model describe what proportion of the total variance in measured metrics can be attributed to variation within- or between-grouping levels (i.e., year, soil management legacy, farm, equipment, paired strip, and within-strip variation). Finally, conditional means and 95% confidence intervals were extracted from lme models for random effects of interest. Conditional means describe the deviation of observations in a group level from the population-level effect and can be used to draw inferences about differences between levels within a grouping factor (Harrison et al., 2018).

3 RESULTS

3.1 Surface residue cover

Results from the on-farm strip trials support the hypothesis that vertical tillage reduces surface residue cover, winter annual weed abundance, and the incidence of slug damage.

The use of vertical tillage in the spring reduced surface residue cover in comparison to paired no-till strips (t-test = −1.9, df = 39, p = 0.03). Averaged across paired strips (n = 40), surface residue cover was 16% points lower in vertical tillage treatments. However, the difference in residue cover between vertical tillage and no-till treatments varied considerably among and within paired strip locations (Figure 2a). Differences in surface residue cover among tillage treatments ranged from a 35% point reduction in surface residue cover to no change in residue cover.

FIGURE 2

Tillage treatment effects on surface residue cover, including (a) mean (±1 SE) difference in surface residue cover (%) between paired vertical tillage (VT) and no-till (NT) strips which are color-coded by equipment type; (b) variance explained (% of total) by random effects in analysis of mean difference in surface residue cover; (c) conditional mean (±95% confidence interval) diffeconfrence from the population mean difference in surface residue cover between vertical tillage and no-till strips by equipment type; and (d) surface residue cover (%) observations at the transect level by tillage treatment and equipment type; conservation compliance thresholds for residue cover (<60%) in Pennsylvania are denoted by a darkened horizontal line. Great Plains TT, Great Plains Turbo Till; Kuhn-Krause Ex, Kuhn-Krause Excelerator.

Random effects included in the model (R²c) accounted for 77% of the total variation in the difference in surface residue cover among treatments. Assessing how each of the random components in the model influenced the change in surface residue cover with the use of variance partition coefficients revealed that equipment type accounted for 46% of the total variance (Figure 2b). Other grouping factors accounted for a smaller proportion of total variance, with year (12%), farm (9%), strip (6%), and soil management legacy (4%) explaining less than half of the variation in surface residue cover.

Analysis of conditional mean differences for equipment type shows the magnitude of vertical tillage effects on surface residue cover for each brand of vertical tillage tool (Figure 2c). The Kuhn-Krause Excelerator had a significantly greater reduction in residue cover compared to the population-level mean (−10%), with an overall reduction in residue cover of approximately 26% points compared to no-till. Furthermore, according to 95% confidence intervals of the conditional mean differences, the Kuhn-Krause Excelerator had a greater reduction in residue cover than the Salford Independent, whereas the Great Plains Turbo-Till had an intermediate reduction in residue cover that was not significantly different from the other equipment types.

Evaluation of surface residue at the transect level by equipment type demonstrates the range of baseline surface residue cover in no-till treatments and surface residue cover following vertical tillage treatments (Figure 2d). Baseline surface residue cover was similar among no-till strips, but among strips subjected to vertical tillage, a greater proportion (32%) of strips had mean surface residue cover levels below a 60% threshold when a Kuhn Krause Excelerator was employed.

3.2 Preplant winter annual weed abundance

The use of a single vertical tillage pass in the spring reduced (t-test = −2.2, df = 39, p < 0.02) winter annual weed cover compared to no-till treatments and the benefit of vertical tillage was greater at higher weed pressure (Figure 3a). Random effects included in the model accounted for 57% of the variation (R²c) in winter annual weed cover. Assessing how each of the random components of the model influenced the change in winter annual weed cover revealed that variation between farm (23%), year (19%), and strip (16%) accounted for approximately 60% of the total variance in winter annual weed cover, whereas soil management legacy and equipment type had a negligible effect.

FIGURE 3

Linear relationships (p < 0.001) between (a) the difference in winter annual weed cover (%) in vertical tillage and no-till paired strips respective to baseline winter annual weed cover (%) in the no-till strip and (b) the difference in the incidence of slug damage (%) in vertical tillage and no-till paired strips respective to baseline slug damage (%) in the no-till strip. Transect-level observations are denoted and color-coded by equipment type used in vertical tillage strips. Great Plains TT, Great Plains Turbo Till; Kuhn-Krause Ex, Kuhn-Krause Excelerator; NT, no-till; VT, vertical tillage.

Winter annual weed cover varied considerably in no-till treatments among paired strips, which can be attributed to the patchy distribution of winter annual weeds in no-till production fields. Consequently, we regressed the difference in winter annual weed cover between vertical tillage and no-till strips by winter annual weed cover in the no-till strip to evaluate the proportional change in weed cover as a result of vertical tillage. Vertical tillage resulted in about a 50% reduction in winter annual weed cover compared to the no-till strip when controlling for variation in weed abundance within and among strips (Wald χ² = 73.8; p < 0.001; Figure 3a).

3.3 Slug feeding damage

The use of a single vertical tillage pass in the spring reduced (t-test = −4.5, df = 39, p < 0.001) the incidence of slug damage compared to no-till treatments and the benefit of vertical tillage was greater at higher slug pressure (Figure 3b). Random effects included in the model accounted for only 6% of the variation (R²c) in the incidence of slug damage, whereas within-strip variation (residual) accounted for 94% of the total variance in slug damage incidence. The regression model indicated that vertical tillage resulted in about a 24% reduction in slug damage compared to the no-till strip when controlling for variation in damage within and among strips (Wald χ² = 13.8; p = 0.001; Figure 3b). In addition to slug damage incidence, the severity of slug damage was quantified based on an estimate of leaf defoliation. Mean slug damage severity was lower (p < 0.05) in vertical tillage treatments. In the no-till treatment, approximately 50% of plants were in the 0%–25% damage category and approximately 4% of plants were in the 26%–50% damage category. The vertical tillage treatments resulted in 6% and 2% point reductions in these categories, respectively, which though statistically different, is unlikely to be biologically important.

3.4 Soybean performance

Vertical tillage did not improve early-season soybean development (i.e., mean growth stage) of soybean (t-test = 0.67, df = 39, p = 0.74), nor stand establishment (t-test = 0.007, df = 39, p = 0.50) when compared to paired no-till strips. Additionally, no vertical tillage treatment effect was observed on soybean grain yield (t-test = 0.26, df = 32, p = 0.60) when compared to paired no-till strips (Figure 4). Random effects included in the model accounted for 25% of the variation (R²c) in soybean yield, with farm and soil management legacy accounting for more variation (23%) than equipment type (<1%). In 2021, mean soybean grain yield across all sites was 5677 kg ha⁻¹ in no-till treatments and 5759 kg ha⁻¹ in vertical tillage treatments. In 2022, mean soybean grain yield across all sites was 4814 kg ha⁻¹ in no-till treatments and 4757 kg ha⁻¹ in vertical tillage treatments. Soybean grain yields were substantially and unusually lower in 2022 across several locations due to a mid- to late-summer drought in localized areas (Table S2).

FIGURE 4

Soybean yield (kg ha⁻¹) in no-till (NT) and vertical tillage (VT) treatments. Box-plots include strip-level observations (n = 33). Horizontal lines within each boxplot indicate the median and whiskers extend to 1.5 × interquartile range. Treatments are not significantly different (p > 0.05) at the population level. NT, no-till; VT, vertical tillage.

4 DISCUSSION

Relative to no-till, vertical tillage reduced surface residue cover, winter annual weed cover, and slug damage but did not influence soybean performance. This study was completed within high-yielding grain corn fields without removing stover and represents one of the highest residue scenarios currently found in row–crop systems in the Mid-Atlantic region. Policy and management implications drawn from this study should be placed in the context of the production environment.

This study found that vertical tillage reduced surface residue cover by 16% points. Other vertical tillage studies with comparable baseline residue conditions found similar reductions in corn residue cover after one pass with a vertical tillage tool (Chen et al., 2016; Klingberg & Weisenbeck, 2011; Smith & Warnemuende-Pappas, 2015; Whitehair & Presley, 2010). However, this study showed that the type of vertical tillage tool and the intensity of use significantly impacted the magnitude of vertical tillage effects. Mean surface residue reduction was 26% points when the Kuhn-Krause Excelerator was employed, which resulted in residue cover below state conservation program compliance thresholds (<60%) in approximately one-third of strips using this tool. Both the design of these tools, and the way in which growers use them, dictates the amount of disturbance created by vertical tillage operations, and therefore residue levels remaining on the soil surface. The magnitude of residue incorporation and soil disturbance can be manipulated by operating the Kuhn-Krause Excelerator at varying depths and disk blade angles with the direction of travel. When operating the Great Plains Turbo Till and Salford tools, the depth of tillage can be adjusted, but disk blade angle cannot be adjusted as the disks on these tools are fixed at a 0° angle. This likely added to the observed variability in residue cover as on-farm cooperators were instructed to operate the vertical tillage tools using their standard practice.

Vertical tillage resulted in about a 50% reduction in winter annual weed cover relative to paired no-till strips. While such reductions are not likely to lead to reduction in preplant burndown herbicide inputs, vertical tillage could be considered a tool for multi-tactic weed management of winter annual weed species in no-till systems. Well-timed tillage can enhance herbicide-based weed control through additive effects (Buhler et al., 1992), reducing the herbicide resistance evolution rate (Liebman & Gallandt, 1997). However, use of fall-sown cover crops provides similar additive effects (>50%) for winter annual weed suppression (Essman et al., 2020; Vollmer et al., 2020; Wallace et al., 2019), is increasingly promoted as a proactive herbicide resistance management tactic (Bunchek et al., 2020), and can provide multiple ecosystem services (Schipanski et al., 2014) in no-till systems. Furthermore, our study highlights the spatial variability of winter annual weed recruitment patterns at field scales (Somerville et al., 2020), which suggests that precision weed management strategies should be favored when developing IWM programs rather than the use of shallow non-inversion tillage tools such as vertical tillage that require implementation at field scales (Buhler, 2002), potentially impacting soil and water quality.

Vertical tillage resulted in about a 24% reduction in the incidence of slug damage relative to paired no-till strips. Assessing the absolute level of slug damage was more challenging as slug feeding in soybean fields can occur on young seedlings close to the soil surface and partially buried by crop residue (Douglas & Tooker, 2012). Vertical tillage is used by a subset of growers in the Mid-Atlantic region specifically for slug control via residue management. Our results suggest that using vertical tillage tools across large acreages solely for slug management may not be efficient due to patchy distribution patterns within and among fields.

Vertical tillage had no effect on soybean population, crop growth stage, or grain crop yield. Other studies measuring impact of vertical tillage on crop yield report varying results. In one study, vertical tillage increased soybean yield (Watters & Douridas, 2013), while another study reported increases in corn yield after vertical tillage but not soybean yield (Van Dee, 2005). Several cooperating farms within this study have invested in planter technologies, such as automatic and pneumatic adjustment of row-unit down pressure and closing wheel systems of various types. These are designed to improve planter performance in heavy corn residue. Such investment may represent an alternative method for maintaining crop performance under higher crop residue environments (Drewry et al., 2020).

In response to our study, farmer cooperators emphasized the utility of vertical tillage to quickly prepare seedbeds before planting as they attempt to achieve timely crop establishment under challenging springtime weather conditions. Historical soybean research data from Pennsylvania and several Midwest states indicate yield potential reductions for soybean planted after May 10 (Roth, 2020). More recent soybean production research conducted in Pennsylvania indicates that early and timely crop establishment, in order to maximize solar radiation capture, is a contributing factor to grain yield potential (Faé et al., 2020). The ability of vertical tillage to hasten soil warming and drying, allowing producers to establish crops earlier, should be a topic of future research. In addition to seedbed preparation, producers also perceive an ability of vertical tillage to incorporate manure or fertilizer and alleviate surface compaction. Extending on-farm evaluation of vertical tillage to other production regions, and under alternative crop management (e.g., planting date and springtime weather) scenarios, is needed to fully characterize its effects on cash crop performance.

5 CONCLUSIONS

Results from this 2-year, on-farm strip trial study in corn residue in southeast Pennsylvania suggest that vertical tillage can reduce surface residue cover below state conservation compliance thresholds (<60%), depending on equipment type and aggressivity of use, and may have marginal utility as IWM or slug management tools. While vertical tillage may locally influence residue cover, weed control, and slug damage, the effect is likely not large enough to substantially alter chemical weed management or mitigate slug damage. In addition, vertical tillage did not improve soybean stand establishment or soybean grain yield. Given these results, growers should consider the limitations associated with performing vertical tillage for purposes other than residue management. Vertical tillage may not eliminate an early-season burndown herbicide application and may not eliminate slug damage. Vertical tillage reduces surface residue cover which creates soil conservation concerns. In high-yield crop production environments in southeast Pennsylvania, vertical tillage may not improve soybean stand establishment or grain yield.

Questions remain regarding the tradeoffs associated with re-introducing a form of minimum tillage into an otherwise no-till system for the purposes of seedbed preparation, manure or fertilizer incorporation, compaction alleviation, or improvement of soil health, and these questions merit further investigation as producers and researchers seek a crop production model that is profitable and conservation minded.

AUTHOR CONTRIBUTIONS

Andrew Lefever: Conceptualization; data curation; formal analysis; funding acquisition; investigation; project administration; visualization; writing—original draft; writing—review and editing. John Wallace: Conceptualization; formal analysis; methodology; project administration; resources; supervision; validation; visualization; writing—original draft; writing—review and editing. Paul Esker: Conceptualization; formal analysis; methodology; visualization; writing—review and editing. Charles White: Conceptualization; methodology; resources; visualization; writing—review and editing. Sjoerd Willem Duiker: Conceptualization; methodology; resources; visualization; writing—review and editing. John F. Tooker: Methodology; visualization; writing—review and editing.

ACKNOWLEDGMENTS

This thesis research was supported with funding from the Northeast Sustainable Agriculture Research and Education (SARE) graduate student research grant program (GNE21-263-35383). SARE is an outreach program of the National Institute of Food and Agriculture (NIFA) and the U.S. Department of Agriculture (USDA). The mention of specific brands of agricultural equipment does not constitute an endorsement of any brand or type of vertical tillage tool, nor do the tools discussed serve as an exclusive list of tools generally considered to be within a “vertical tillage” category of tools. The authors would like to thank Penn State Department of Plant Science research technicians, post-doctoral research scholars, graduate students, undergraduate students, Penn State Extension Agronomy Educators, and family members who helped with field data collection. Finally, the author would like to thank cooperators on 16 farms in Lancaster and Chester Counties in southeast Pennsylvania who allowed to establish strip trials on their farms, who gave not only the land to use but often donated their time, equipment, and other resources to further these research efforts.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.