Introduction
Adoption of cover crops (CCs) has steadily increased over the years in response to challenges facing farmers (Zhou et al. Reference Zhou, Guan, Wang, Jiang, Huang, Peng, Chen, Wang, Hipple, Schaefer, Qin, Stroebel, Coppess, Khanna and Cai2022), some of which include herbicide-resistant weeds and government-mandated herbicide restrictions (Peterson et al. Reference Peterson, Collavo, Ovejero, Shivrain and Walsh2018; Zilberman et al. Reference Zilberman, Schmitz, Casterline, Lichtenberg and Siebert1991). The use of CCs has grown due mainly to the soil health and nutrient-related benefits such as increased inorganic nitrogen and soil erosion prevention, but also for the ability of CCs to suppress weeds (Dabney et al. Reference Dabney, Delgado and Reeves2001; Myers and Watts Reference Myers and Watts2015; Norsworthy et al. Reference Norsworthy, Ward, Shaw, Llewellyn, Nichols, Webster, Bradley, Frisvold, Powles, Burgos, Witt and Barrett2012). Best practices to maximize CC benefits is largely dependent on the goal of the CC. For the purposes of weed suppression, the accumulation of heavy biomass has proven to be one of the most important factors (MacLaren et al. Reference MacLaren, Swanepoel, Bennett, Wright and Dehnen-Schmutz2019).
Biomass accumulation of a CC is largely dependent on timely planting and the time at which it is terminated (Ruis et al. Reference Ruis, Blanco-Canqui, Creech, Koehler-Cole, Elmore and Francis2019). To accumulate biomass, some growers have been experimenting with planting a “green,” or living CC, which means delaying CC termination until right before or just after the cash crop is planted. More commonly, the CC is terminated prior to cash crop planting, which becomes a “brown,” or dead CC. These two practices are both currently used by growers, but there is no clear consensus on whether one practice provides more benefits than the other. Maintaining a green canopy for longer may delay germination and emergence of small-seeded weeds because they are typically sensitive to light quality germination cues (Baskin and Baskin Reference Baskin and Baskin1977; Jha et al. Reference Jha, Norsworthy, Riley and Bridges2010). Conversely, there are concerns with soil moisture reduction via transpiration of the longer established CC leaving dryer conditions at the time of cash crop planting (Meyer et al. Reference Meyer, Bergez, Constantin, Belleville and Justes2020).
Similarly, management practices such as rolling the CC prior to cash crop planting have also been questioned for their effectiveness and need. Although some growers roll a CC after termination, it is unclear whether rolling more effectively suppresses weeds compared to a CC that is left standing and chemically terminated. A rolled CC may provide suppression through smothering, physical interference of weed emergence, and light interception. Although a standing CC does not physically suppress weeds, it can provide suppression by interfering with light availability to weed seeds; particularly small-seeded weeds such as Palmer amaranth (Amaranthus palmeri S. Watson) (Gallagher and Cardina, Reference Gallagher and Cardina1998; Jha et al. Reference Jha, Norsworthy, Riley and Bridges2010). The standing CC could potentially reduce light penetration for a longer period due to tall residue left in the field that intercepts light from the seeds and seedlings in the lower canopy.
Although these are complex systems with many factors, the importance in solidifying CC termination strategies lies in the economic implications that this information provides. Knowing best management practices at the time of termination could result in reduced labor associated with pesticide application and increased efficiency. Therefore, to answer these questions for CC growers in the Mid-Atlantic, an experiment was established to answer two main questions across a range of CC biomass: 1) Does rolling a CC or leaving it standing affect weed suppression and soybean yield?, and 2) Does planting into a green CC provide any weed suppression or soybean yield benefits that planting into a brown CC would not?
Materials and Methods
Field Experiments
Field experiments were conducted between 2021 and 2023 at the Southern Piedmont Agricultural Extension Station (SPAREC) near Blackstone, VA (37.087024°N, 77.956284°W), and at Kentland Farm near Blacksburg, VA (37.200091°N, 80.563504°W). Average temperatures for the experimental locations are −1 to 32 C, and −4 to 29 C, for the SPAREC and Kentland locations, respectively. Average rainfall is 114 cm and 101 cm, at the two locations, respectively.
Treatments were arranged in a four by two by two factorial design, for a total of 16 treatments, with four target levels of CC biomass (none, low, medium, or high), rolled vs. standing, and planted green vs. brown. Plots measured 3 × 9 m, with four replications per treatment arranged in a randomized complete block design. In fall 2021 and 2022, cereal rye CC (variety ‘Elbon’; Green 79 Cover Seeds, Bladen, NE) was planted to initiate the experiment at both locations. Cereal rye seed was planted at 84 kg ha−1 to a depth of approximately 2.5 cm using a 1.5-m-wide drill with 18-cm row spacing. Cereal rye planting dates were staggered to achieve a low, medium, and high CC biomass gradient. The first planting for the high biomass level took place around the last week of October of each year. The second planting followed 3 wk later, and the final planting for the low biomass took place 3 wk after the second planting. The CC was allowed to overwinter until 2 wk prior to soybean planting, at which point the brown termination treatments were sprayed with glufosinate ammonium at 656 g ai ha−1 (Interline®; UPL NA Inc., Mumbai, Maharashtra, India) + ammonium sulfate at 3.3 kg ha−1 (Growmark, Bloomington, IL) and glyphosate at 1.3 kg ae ha−1 (Roundup Powermax 3®; Bayer CropScience, Research Triangle Park, NC). All chemical applications were applied using a CO2 -pressurized backpack sprayer with a four-nozzle boom with 46-cm spacing. The boom was fitted with TeeJet XR 11002 flat-fan nozzles (Spraying Systems Co., Wheaton, IL) and calibrated to deliver 147 L ha−1 at 276 kPa. Soybean planting took place the first week of May in all experimental years. Green termination treatments were sprayed the day of soybean planting using the aforementioned methodology. Rolled treatments were established the day of soybean planting using a 1.5-m roller, while standing treatments were just passed over by the planter. XtendFlex® (Bayer CropScience) soybean seed was planted at 222,300 seeds ha−1 in both years of the study using a two-row planter (Kinze Evolution Series, Williamsburg, IA) on 76-cm spacing. Due to the size of the planter, some of the standing cereal rye was inevitably rolled down by the planting equipment and tractor wheels, but this also occurs with commercial-scale equipment. To allow for harvest, glufosinate at 656 g ai ha−1 + ammonium sulfate at 3.3 kg ha−1 was applied at 4 and 6 wk after planting to control the Palmer amaranth that was still present.
Data Collection
Total aboveground CC biomass was collected at the time of respective CC termination for every plot using a randomly placed 0.5-m−2 quadrat. At the time of CC biomass collection all aboveground CC biomass was collected and clipped at the soil surface using hand shears. Individual plot bags were collected to isolate the treatment samples. Sample bags were then dried at 52 C for 4 d to collect dry weight. Palmer amaranth emergence was recorded by counting emerged plants within a single, randomly placed, 0.25-m−2 quadrat for each plot at 4 and 6 wk after planting prior to the herbicide applications.
Light penetration was measured at 2, 4, and 6 wk after planting using an LI-191R Light Quantum Sensor (Li-Cor Biosciences, Lincoln, NE). Light penetration was measured on the photosynthetic active radiation spectrum as photosynthetic photon flux density (PPFD), expressed as μmol s−1 m−2, over a 1-m length. To observe light reaching the soil surface (emerging weeds) and above the soil (established weeds), PPFD was collected at 0 cm (at the soil surface but below CC residues), 15 cm, and 30 cm above the soil surface. To avoid rolled CC residue from tractor tire tracks, PPFD readings were collected in the middle rows. At the conclusion of the trial, soybean yield was calculated from the two middle rows of each plot by collecting grain using a plot combine (Classic Plus Combine; Wintersteiger, Ried im Innkreis, Austria). Yield for each plot was calculated by using total weight and adjusting for moisture.
Statistical Analysis
Palmer amaranth emergence and soybean yield data were analyzed using JMP Pro 16 software (v. 16; SAS Institute Inc., Cary, NC) to evaluate the main effects of termination timing, by rolled or standing method, and CC biomass at termination. Means were separated using Fisher’s protected least significant difference (LSD) test (α = 0.05). Replication was considered a random effect, whereas site year, location, and treatments were considered fixed effects. Soil-level PPFD data were compared across treatments, while PPFD at 15 and 30 cm above ground was compared only for the standing treatments because rolled CC residue was below these heights, and no reductions in PPFD were observed. Palmer amaranth emergence, soybean yield, and PPFD were fitted to quadratic, exponential decay, and hyperbolic decay models across CC biomass, as was performed in similar studies (Ma et al. Reference Ma, Wu, Jiang, Ma and Ma2015; Pacala and Weiner, Reference Pacala and Weiner1991) using SigmaPlot 15 software (Systat Software, San Jose, CA) depending on the best fit for each model.
Greenhouse Study
A greenhouse study was conducted in three experimental runs from 2021 through 2023, to further understand the effects of CC termination timing and cereal rye biomass on Palmer amaranth emergence. These studies were established the second week of May of each experimental year, and were carried out for 4 wk after initiation. Greenhouse day/night temperatures were 21 C to 32C for the duration of the trial, and no supplemental lighting was provided.
Trays measuring 54.5 × 27.8 × 6 cm were filled with Moisture Control Potting Mix (Scotts Miracle-Gro, Marysville, Ohio), and each tray contained 100 Palmer amaranth seeds that were mixed into the top 0.2 cm of the soil profile of the tray. Each tray and seeds were covered with cereal rye biomass according to a designated treatment (0, 2,242, 4,483, 6,725, 8,967, and 11,209 kg ha−1) for both green and brown CC residue types to resemble the trial established in the field. Trays were arranged in a completely randomized design with four replications per treatment. CC biomass was collected on the day of trial initiation from the field where both brown dead CC and green living CC were cut and transported to the greenhouse. In 2021 and 2022, weights of the CC were wet weights. However, in 2023 we accounted for the difference in water weight. Therefore, after testing for moisture (BHT6071 Silage, Crop, Hay Moisture Tester; Best Harvest, Largo, FL) of both brown and green CCs, the difference in water weight was taken into consideration, and CC residue amount was based on dry biomass.
Light penetration readings were measured using the Light Quantum (Li-Cor Biosciences) sensor equipment by placing the same levels of green and brown CC biomass on top of Plexiglass upheld by a wooden box. Boxes measuring 59.7 × 12.7 cm were created to allow the sensor to slide below the Plexiglass top. The boxes were enclosed to exclude light entering the box except through the Plexiglass top. Each level of CC biomass represented in the trays was also represented in the Plexiglass-covered wooden boxes to determine changes in PPFD as a result of CC biomass and termination timing. Boxes were arranged in a completely randomized design with two replications per treatment.
Greenhouse Data Collection
Newly emerged Palmer amaranth seedlings were counted and removed weekly for 4 wk after trial initiation. PPFD measurements were also recorded on a weekly basis for the duration of the trial.
Greenhouse Statistical Analysis
Emergence and light penetration data were analyzed using JMP Pro 16 software (v.16; SAS Institute Inc., Cary, NC). ANOVA was used to evaluate differences in Palmer amaranth emergence and light penetration as a result of biomass quantity, as well as the effect of biomass type (green or brown), and appropriate means were separated using Fisher’s protected LSD (α = 0.05). Year was considered a random effect, whereas treatment, which included CC biomass and termination timing, were considered fixed effects. Fisher’s protected LSD using an α = 0.05 level of significance was used to determine significant differences and means separation for main effects. Emergence and PPFD data were subjected to nonlinear regression across CC biomass levels with SigmaPlot 15 software using the aforementioned methodology.
Results and Discussion
Cover Crop Biomass
There was a site-year by CC termination timing interaction for biomass (P = 0.041) (Table 1). Generally, the green, terminated CC treatments had greater biomass accumulation across site-years with the exception of Blacksburg in 2023, when there was no difference between brown and green treatments. The trend of greater biomass in the green treatments was most likely a result of the green termination timing accumulating more growing degree days than the brown timing (Mischler et al. Reference Mischler, Curran, Duiker and Hyde2010).
Total cover crop biomass by site and termination timing from experiments conducted in Virginia between 2019 and 2021.a,b
a The no-cover crop comparison treatment was excluded from this analysis.
b Letters indicate significant differences (P < 0.05) in cover crop biomass within each site.
c Brown planting refers to cover crop terminated 2 wk prior to planting while green planting refers to cover crop terminated at the time of planting.
Palmer Amaranth Emergence 4 Wk after Planting
Emergence data indicated a two-way interaction between site-year and weeks after planting (P < 0.001), therefore, data were subsequently analyzed by site-year. At 4 wk after planting (WAP) at the Blackstone location in 2022, a two-way interaction between CC biomass and rolled or standing termination method was significant (P = 0.002). The rolled treatments decreased Palmer amaranth emergence as CC biomass increased, but CC biomass did not reduce Palmer amaranth emergence in the standing plots (Figure 1; Table 2).
Palmer amaranth emergence for significant sites and interactions at 4 wk after planting (WAP) across cover crop (CC) biomass for experiments conducted in Virginia between 2020 and 2023. The top graph shows the relationship between Palmer amaranth emergence for the rolled and standing treatments across the CC biomass gradient. The middle graph shows the three-way interaction relationship for Palmer amaranth emergence as a result of termination timing (brown: CC terminated 2 wk prior to planting; green: CC terminated at the time of planting), rolled and standing, and the CC biomass gradient. The bottom graph shows total Palmer amaranth emergence across the main effect of CC biomass. Dotted lines of same color indicate 95% confidence intervals (see Table 2 for equation parameters).
Summary of equation parameters in figures for trials conducted in Virginia between 2020 and 2023.a
a Abbreviation: PPFD, photosynthetic photon flux density; WAP, weeks after planting.
There was a three-way interaction between termination timing, rolled or standing termination method, and CC biomass (P = 0.003) at the Blacksburg 2022 site-year, but only differences below 1,500 kg ha−1 in Palmer amaranth emergence were evident. At greater CC biomass amounts, no differences in Palmer amaranth emergence were observed (Figure 1; Table 2). At the Blacksburg site in 2023, the main effects of CC biomass and termination timing were significant (P = 0.003 and P = 0.001, respectively). Regression analysis indicated that Palmer amaranth emergence decreased as CC biomass increased (Figure 1; Table 2). The main effect of termination timing showed the brown termination treatment having the greatest Palmer amaranth emergence (µ = 201 plants m−2) compared with the green termination treatment (µ = 137 plants m−2). The accumulation of at least 2,000 kg ha−1 CC biomass is necessary to receive observed benefits of weed suppression (Figure 1; Table 2). It is also important to note that when a CC is rolled, the weeds are suppressed as long as they are under the CC biomass, but at low CC biomass levels the soil is not completely covered. However, up to a biomass of 1,500 kg ha−1 the standing plots showed less Palmer amaranth emergence, but at greater CC biomass levels, the rolled plots showed greater suppression. This interaction indicates at greater biomass levels, rolled CC can provide greater suppression, but at lower biomass levels, the difference in Palmer amaranth emergence is insufficient to warrant CC rolling. Previous studies by Mischler et al. (Reference Mischler, Curran, Duiker and Hyde2010) showed weed suppressive effects of a rolled CC as being comparable to that of the application of an early postemergence herbicide. Although the results reported by Mischler et al. (Reference Mischler, Curran, Duiker and Hyde2010) are possible, as their study reports, the results are not always consistent across sites and are dependent on CC biomass.
Palmer Amaranth Emergence 6 Wk after Planting
At 6 WAP there was a three-way interaction between CC termination timing, rolled or standing CC, and CC biomass (P = 0.045) at the Blackstone 2022 site-year. More Palmer amaranth emerged when CC biomass was below 500 kg ha−1, particularly under the rolled brown treatment, but at greater CC biomass accumulations, there was no difference in emergence compared with the other treatments (Figure 2; Table 2). The same three-way interaction was observed at the Blacksburg location in 2022 (P = 0.018), when low CC biomass levels showed greater emergence for the green rolled and brown standing treatments, but at greater CC biomass levels there was no separation between treatments (Figure 2; Table 2). A two-way interaction between termination timing and whether the CC was rolled or standing was significant in both Blacksburg 2023 (P = 0.049) and Blackstone 2023 (P = 0.033) site-years (Table 3). At the Blackstone location in 2023, the brown rolled treatments exhibited greater Palmer amaranth emergence (µ = 26 plants m−2), while the green standing (µ = 12 plants m−2), green rolled (µ = 11 plants m−2), and brown standing (µ = 11 plants m−2) had the least Palmer amaranth emergence (Table 3); however, the brown rolled and green standing treatments did not differ at this site. Similarly, at the Blacksburg site in 2023, the brown rolled treatments produced the greatest Palmer amaranth emergence (µ = 37 plants m−2), whereas the green rolled treatments produced the least (µ = 7 plants m−2). The brown standing and green standing treatments were not different compared with any of the other treatments (µ = 20 plants m−2 and µ = 15 plants m−2, respectively) at this site. Overall, the effects of Palmer amaranth suppression by the CC treatments did not consistently separate by treatment at 6 WAP.
Regressions of significant three-way interactions at 6 wk after planting for Palmer amaranth emergence as a result of cover crop (CC) termination timing (brown: CC terminated 2 wk prior to planting; green: CC terminated at the time of planting), as well as rolled or standing, regressed across CC biomass for experiments conducted in Virginia between 2020 and 2023. Dotted lines of the same color indicate 95% confidence intervals (see Table 2 for equation parameters).
Palmer amaranth emergence at 6 weeks after planting for significant interactions of rolled versus standing CC when planted green versus brown in Virginia in Blackstone 2023 and Blacksburg.a,b
a Brown planting refers to cover crop terminated 2 wk prior to planting, whereas green planting refers to cover crop terminated at the time of planting.
b Letters indicate significant differences (P < 0.05) in Palmer amaranth plant counts within each site.
When compared with the 4 WAP data, the trends at 6 WAP were not consistent across the interactions. It is important to note that glufosinate was applied at 4 WAP, and thus these 6 WAP data are from a separate emergence cohort. Since the CC residue was all dead at this time, it is logical that there were no differences in termination timing (i.e., green versus brown). Additionally, the decomposition of the CC over time, particularly the rolled brown treatments, may have reduced treatment effects. With less biomass and more time as nonliving residue, it is likely that the brown rolled treatments allowed more Palmer amaranth emergence due to less biomass being present. This information agrees with that reported by Pittman et al. (Reference Pittman, Barney and Flessner2020) who noted that greater CC biomass leads to more weed suppression. Consistently, in this study and others (MacLaren et al. Reference MacLaren, Swanepoel, Bennett, Wright and Dehnen-Schmutz2019; Mirsky et al. Reference Mirsky, Curran, Mortenseny, Ryany and Shumway2011), the importance of sufficient CC biomass accumulation is highlighted, and the studies agree that it is perhaps the most important factor in suppressing weeds.
Field PPFD
At the soil level, PPFD data indicated a three-way interaction between termination timing, rolled or standing termination, and CC biomass (P = 0.006) (Figure 3; Table 2). There was also a two-way interaction between WAP and CC biomass (P < 0.001). The three-way interaction did not exhibit differences among treatments, but an overall trend of reduced PPFD with increasing CC biomass was observed (Figure 3; Table 2). The two-way interaction at the soil level showed greater light penetration at 2 and 4 WAP, but a decrease in light penetration at 6 WAP. The decrease in PPFD at 6 WAP could potentially be displaying the light interception via the soybean canopy as the crop developed.
Photosynthetic photon flux density (PPFD) at soil level for trials conducted in Virginia between 2020 and 2023. A) PPFD at the soil surface level as a result of three-way interactions between termination timing (brown: cover crop [CC] terminated 2 wk prior to planting; green: CC terminated at the time of planting), rolled or standing, and the CC biomass gradient. B) PPFD as a result of two-way interactions between weeks after planting and CC biomass. Dotted lines of the same color indicate 95% confidence intervals (see Table 2 for equation parameters).
When the various heights of standing plots were compared, PPFD data indicated a two-way interaction between termination timing and CC biomass (P < 0.001), as well as an interaction of WAP and CC biomass (P < 0.001) (Figure 4; Table 2). The two-way interaction of termination timing and CC biomass did not show light penetration differences until the biomass was greater than 7,000 kg ha−1. When biomass was less than 7,000 kg ha−1, there was no difference in PPFD between brown and green treatments. The two-way interaction of WAP and CC biomass indicated similar trends as the soil-level PPFD data: greater light penetration at 2 and 4 WAP, and decreased light penetration at 6 WAP. The main effect of height was also significant (P < 0.001). Height data showed greatest PPFD at the 30-cm height (µ = 1,059 μmol s−1 m−2), followed by the 15-cm height (µ = 968 μmol s−1 m−2), with the lowest PPFD at the soil level (µ = 794 μmol s−1 m−2), indicating that a reduction in light occurs through CC canopy that remains standing. This light reduction could potentially explain the reduction in weed emergence observed by Rector (Reference Rector2019) in standing CC compared with rolled CC at the soil surface. With light penetration decreasing as it intercepts the CC canopy, reduction in light, as well as potential changes in red to far-red ratios from green versus brown CC residues, may be changing both the quantity and quality of light reaching seeds in at the soil surface. The interception of light through the canopy is not present in a rolled crop; therefore, that light would encounter the rolled barrier only at the soil surface.
Photosynthetic photon flux density (PPFD) for all standing plots, including soil level, 15 cm, and 35 cm above the soil level for experiments conducted in Virginia between 2020 and 2023. A) Two-way interaction between termination timing (brown: cover crop [CC] terminated 2 wk prior to planting; green: CC terminated at the time of planting) and CC biomass. B) Two-way interaction between weeks after planting and CC biomass. Dotted lines of the same color indicate 95% confidence intervals (see Table 2 for equation parameters).
Overall, this trend is consistent with that reported in another study that documented light reduction under the presence of a CC (Wayman et al. Reference Wayman, Cogger, Benedict, Collins, Burke and Bary2015). The reduction in light under greater CC biomass accumulation could explain some of the reduction in weed emergence at greater CC biomass that was also observed in our experiment. However, further research would be beneficial in understanding the differences in quality of light as a result of CC species selection, and changes in red to far-red light spectrum ratios affecting troublesome weed species. Additionally, more information is needed to identify differences in the rate of decomposition between the green-terminated and brown-terminated CCs and their effect on small-seeded weeds such as Palmer amaranth.
Yield
Due to herbivory damage caused by deer and groundhogs, the Blacksburg site in 2022 was unable to be harvested to obtain yield data. The remaining site-years did not show significant main effects or interactions within each site-year; yield data indicated site-year as being significant (P < 0.001). These results indicate that yield was unaffected by any of the CC treatments of termination timing, rolled or standing, or by CC biomass. However, yield effects have been reported to be variable under CC presence (Shoup et al. Reference Shoup, Ciampitti, Kimball and Sassenrath2017; Singer and Kohler Reference Singer and Kohler2005). Average yield was 1,687 kg ha−1 from Blackstone in 2022, 807 kg ha−1 from Blackstone in 2023, and 1,805 kg ha−1 from Blacksburg in 2023. Although research has shown slight increases in yield under CC treatments (Cordeiro et al. Reference Cordeiro, Batista, Lopes and Echer2021; Seifert et al. Reference Seifert, Azzari and Lobell2018), there are many contingencies, including soil type, previous crop, and annual environmental conditions that must be taken into consideration. Conversely, multiple studies (e.g., Nascente and Crusciol Reference Nascente and Crusciol2012) have demonstrated that CC did not improve crop yields, similar to the results observed in our study. It is clear that there are no consistent results on how soybean yields are affected under CC rotations. A study reported by Fernando and Shrestha (Reference Fernando and Shrestha2023) showed that weed competition can be potentially suppressed through the use of a CC. There was no complete success between weed reductions and CC biomass, but greater weed suppression is often associated with greater CC biomass (Fernando and Shrestha Reference Fernando and Shrestha2023). Other studies, however, (e.g., Deines et al. Reference Deines, Guan, Lopez, Zhou, White, Wang and Lobell2023) have documented the potential reduction in soybean yield as a result of CC use. Deines et al. (Reference Deines, Guan, Lopez, Zhou, White, Wang and Lobell2023) mentioned that this reduction in soybean yield is largely dependent on seasonal weather variations, effect of CC on nitrogen dynamics, and soil conditions. Additionally, the issue of herbivory that CCs often attract is a known risk (DeYoung et al. Reference DeYoung, Fulbright, Hewitt, Wester and Draeger2019) that also played an important role in the present experiment. It has been documented that access to high-quality foods for deer will lead to more selective grazing by these herbivores (DeYoung et al. Reference DeYoung, Fulbright, Hewitt, Wester and Draeger2019). Therefore, the availability of a CC in the fall and crops such as soybeans in the summer, may invite yet another on-farm pest to manage. The variable results in soybean yield effects as a result of CC use are therefore to be expected in a large-scale field experiment, and the results of our study are therefore not unlike those found in other forms of CC literature.
Greenhouse Emergence
There was a three-way interaction between year, termination timing, and CC biomass on Palmer amaranth emergence (P < 0.001). Due to the Palmer amaranth emergence data being cumulative, the results at 4 wk after initiation (WAI) were chosen to represent the total emergence from Weeks 1 to 4 given that the significant treatment effects were the same. At 4 WAI in 2021 both CC biomass (P = 0.037) and termination timing (P = 0.020) were significant. The green termination timing showed greater cumulative emergence (µ = 290 plants m−2) compared with the brown termination timing (µ = 218 plants m−2). The CC biomass main effect showed that the 2,000 kg ha−1 treatment had the greatest cumulative emergence (µ = 323 plants m−2), whereas the 10,000 kg ha−1 treatment had the least (µ = 158 plants m−2), which is reflected by the negative relationship in Palmer amaranth emergence as CC biomass increased, as shown in the quadratic and hyperbolic decay models (Figure 5; Table 2).
Greenhouse cumulative Palmer amaranth emergence for each experimental run from initiation to 4 wk after initiation (left axis) with average photosynthetic photon flux density (PPFD) (right axis) regressed across total cover crop (CC) biomass for experiments conducted between 2020 and 2023. The 2021 graph shows the main effect of CC biomass on total Palmer amaranth emergence, while the 2022 and 2023 graphs show the two-way interaction between termination timing (brown: CC terminated 2 wk prior to planting; green: CC terminated at the time of planting) and CC biomass. Dotted lines of the same color indicate 95% confidence intervals (see Table 2 for equation parameters).
At 4 WAI the interaction of CC biomass and termination timing on Palmer amaranth emergence was significant (P = 0.001) in 2022. The greatest emergence was observed with the 2,000 kg ha−1 brown treatment (µ = 462 plants m−2), after which point the general emergence trend began to decrease with increasing biomass (Figure 5; Table 2). Up to 8,000 kg ha−1, the green and brown treatments did not differ, but at CC biomass levels greater than 8,000 kg ha−1 Palmer amaranth emergence was greater with the green treatments.
In 2023 at 4 WAI, the same interaction between CC biomass and termination timing was significant (P = 0.003). The same pattern observed in 2021 and 2022 was also present at 4 WAI, when the greater biomass treatments showed the greatest suppression, as the 8,000 kg ha−1 and 10,000 kg ha−1 green plots had the least Palmer amaranth emergence (µ = 85 and 33 plants m−2, respectively) (Figure 5; Table 2). Generally, less emergence occurred after the green treatments, but this did not emerge from the brown treatments until after 8,000 kg ha−1 was accumulated.
Overall, there was a clear pattern in the greater CC biomass treatments reducing Palmer amaranth emergence while more emergence occurred from the lesser CC biomass treatments. This information is in accordance with previous work such as the findings from Mirsky et al. (Reference Mirsky, Ackroyd, Cordeau, Curran, Hashemi, Reberg-Horton, Ryan and Spargo2017) that found increases in CC biomass will reduce weedy populations. Additionally, with the exception of 2021, both main effects of CC biomass and termination timing in the greenhouse are consistent with the field results that indicate that greater CC biomass increases Palmer amaranth suppression, and although it does not occur consistently, the green-terminated treatments also reduced Palmer amaranth emergence compared with the brown-terminated treatments. It is possible that this inconsistency between experimental runs in the greenhouse is due to the biomass amounts not accounting for moisture in 2021 and 2022 compared with 2023. Although observed emergence trends were similar, it is important to continue research with biomass consistently accounting for moisture weight to accurately compare differences between treatments.
Greenhouse Light Penetration
Light penetration data showed a year by biomass by termination timing interaction (P = 0.001). The average PPFD data for 2021 indicated a clear relationship between PPFD and CC biomass (Figure 5; Table 2). As the CC biomass increased there was a decrease in PPFD penetrating through the CC biomass. The same patterns were observed in 2022 and 2023 (Figure 3; Table 2). The main effect of termination timing was only significant in 2021 when the brown treatments received greater light intensity than the green treatments (µ = 261 and 249 µmol s−1 m−1, respectively). However, the effect of CC biomass was consistently the significant main effect that affected PPFD reaching the soil top layer. One important detail concerning the PPFD data in both the field and greenhouse studies is that these data do not speak to the quality of light penetrating the CC biomass. It is known that actively growing plants, or a CC in this case, would reduce the red to far-red light ratio (Batlla and Benech-Arnold Reference Batlla and Benech-Arnold2014). This change in quality of light can also affect weed seed germination. Additionally, Teasdale and Mohler (Reference Teasdale and Mohler1993) determined that dead plant residues do not affect these red to far-red ratios; however, living plant material could lead to different emergence cues for weedy species. Further research documenting the changes in red to far-red ratios under various CC termination timings is therefore necessary to have a more wholistic understanding of the changes in biological cues that CC management is having on weed seed emergence.
Practical Implications
Results suggest that CC management did not influence soybean yield. Although CCs can be another tool for weed suppression, sufficient biomass must be accumulated in order to reap Palmer amaranth suppression benefits. These results indicate that earlier plantings, as well as later termination timings, would allow farmers to accumulate biomass for results to be beneficial. This research shows the importance of CC biomass accumulation, as well as the variability with green and brown termination, in terms of weed suppression. Although planting a green CC may produce more biomass, this may not always provide more weed suppression. Therefore, from year to year, if biomass accumulation can be accomplished by other means, there would not be a strong incentive to delay CC termination.
Acknowledgments
We thank all the undergraduate assistants who helped with this project. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement. No conflicts of interest have been declared.
Funding
Funding was provided in part by the Virginia Soybean Board and by the U.S. Department of Agriculture–National Institute of Food and Agriculture through Hatch project 1026160.
Competing interests
The authors declare they have no competing interests.
Open access
