Introduction
Weeds have been a major production challenge for sugarbeet since the crop was first widely grown in Europe in the late 1700s (Schweizer and May Reference Schweizer, May, Cooke and Scott1993). Weed management in sugarbeet is especially challenging because of its low growth habit, slow canopy development, and limited POST herbicide options (Bollman and Sprague Reference Bollman and Sprague2007). Inter-row cultivation and hand-weeding were the primary weed control methods prior to the development of herbicides, but these methods took on a lesser role in sugarbeet weed management as more herbicides were developed. Desmedipham and phenmedipham were the primary herbicides used to control Amaranthus and Chenopodium species in sugarbeet from the 1970s to 2000s (Dale et al. Reference Dale, Renner and Kravchenko2006; Dexter Reference Dexter1977, Reference Dexter1994), with inter-row cultivation to supplement their use. The chloroacetamide herbicides S-metolachlor and dimethenamid-P were registered for sugarbeet in the mid-2000s, which led to their brief use with desmedipham and phenmedipham before the commercialization of glyphosate-resistant (GR) sugarbeet (Bollman and Sprague Reference Bollman and Sprague2007).
The commercialization of GR sugarbeet cultivars in 2008 resulted in a sweeping change in sugarbeet weed management. Glyphosate became the primary tool used for weed control, as it was cheaper, safer, and more effective than desmedipham, phenmedipham, and inter-row cultivation. Guza et al. (Reference Guza, Ransom and Mallory-Smith2002) reported that greater than 95% redroot pigweed (Amaranthus retroflexus L.) and common lambsquarters (Chenopodium album L.) control could be achieved with two POST applications of glyphosate in sugarbeet. Dexter and Luecke (Reference Dexter and Luecke2000) reported that glyphosate improved sugarbeet tolerance and weed control compared to the conventional micro-rate program, which contributed to significantly greater root yield. Other research suggested that inter-row cultivation for controlling weeds was unnecessary and possibly detrimental to yield (Dexter et al. Reference Dexter, Luecke and Smith2000). Survey data from 2007 indicated that 99% of North Dakota and Minnesota sugarbeet hectares were inter-row cultivated (Carlson et al. Reference Carlson, Luecke, Khan and Dexter2008), but only 11% of hectares were cultivated in 2011 following the release of GR cultivars (Stachler et al. Reference Stachler, Carlson, Khan and Boetel2011). Unfortunately, GR weeds, including waterhemp, had already migrated into the upper Midwest by the time GR sugarbeet cultivars were released and have become progressively more problematic in recent years. Weed control trials in 2016 reported that glyphosate-only treatments in sugarbeet controlled only 30% to 40% of waterhemp by late August (Peters et al. Reference Peters, Lueck and Groen2017). In addition to the diminishing effectiveness of glyphosate, the label registrations for desmedipham and phenmedipham were not renewed in the mid-2010s (EPA 2014), leaving sugarbeet producers with even fewer POST control options.
Use of the chloroacetamide herbicides S-metolachlor, dimethenamid-P, and acetochlor applied early POST has increased from 15% in 2014 to 91% in 2018, according to surveyed producers in regions where producers identify waterhemp as their primary production challenge (Carlson et al. Reference Carlson, Peters, Khan and Boetel2015; Peters et al. Reference Peters, Khan and Boetel2020). Chloroacetamide herbicides are activated into soil solution by rainfall and provide residual control of emerging small-seeded broadleaf weeds, including Amaranthus species. The current recommendation to Minnesota and North Dakota producers for waterhemp control in sugarbeet is to apply S-metolachlor and/or ethofumesate PRE followed by layered applications of a chloroacetamide herbicide early POST (Peters et al. Reference Peters, Lueck and Groen2017). Layering residual chloroacetamide herbicides throughout the season will prevent weed emergence until the sugarbeet crop canopy provides shade to suppress further weed growth. Chloroacetamide herbicides require 10 to 20 mm of precipitation for activation into soil solution and do not control already emerged weeds (Anonymous 2014, 2017). Herbicide-resistant weed escapes are a concern when limited rainfall results in poor herbicide activation or when excessive rainfall makes timely herbicide applications challenging. Many sugarbeet producers have used inter-row cultivation to remove glyphosate-resistant weeds that escaped the residual chloroacetamide herbicide layer.
Inter-row cultivation mid-season has benefits and drawbacks. The greatest benefit is nonselective removal of weeds between crop rows that herbicides did not or cannot control. Other benefits include drying and loosening of the soil and incorporation of fertilizer and soil-active herbicides. In some weed species, disturbance of the soils by tillage can increase the germination and emergence of weed seeds in the seed bank; in other species, tillage reduces emergence (Egley and Williams Reference Egley and Williams1990). Over a 5-yr average, however, Egley and Williams (Reference Egley and Williams1990) concluded that tillage or tillage depth did not significantly affect weed emergence for numerous species including redroot pigweed. Weed dormancy and emergence is a complex interaction of soil moisture, temperature, and light exposure (Alm et al. Reference Alm, Stoller and Wax1993; Baskin and Baskin Reference Baskin and Baskin1990; Kemp Reference Kemp2000). For many weed species, especially small-seeded broadleaves, tillage has limited effects on weed emergence. Oryokot et al. (Reference Oryokot, Murphy and Swanton1997) reported that tillage had minimal effect on redroot pigweed emergence. Soil samples taken to a 2.5-cm depth on plots with and without tillage indicated that tillage did not change soil temperature or soil moisture. This incidence is due to a relationship between the top 2.5 cm of soil and the atmosphere that creates an equilibrium of temperature and moisture. Amaranthus species are physiologically limited to germination within the top 2.5 cm of soil as a result of the seed’s small endosperm, explaining tillage’s limited effect on redroot pigweed emergence (Oryokot et al. Reference Oryokot, Murphy and Swanton1997). Exposure of weed seeds to light during tillage can also affect weed emergence. Buhler (Reference Buhler1997) reported that common lambsquarters emergence increased nearly 250% when tillage was performed in the light compared to the dark, demonstrating the influence of infrared light on weed dormancy and germination.
Numerous studies have evaluated the effect of inter-row cultivation on sugarbeet yield and quality. Results of these studies generally demonstrate that early-season cultivation has little effect on recoverable sucrose yield, but cultivation later in the season is detrimental to yield and quality (Dexter et al. Reference Dexter, Luecke and Smith2000). Dexter (Reference Dexter1983) reported that sugarbeet yield tended to increase with up to three cultivations but decreased after four cultivations. Giles et al. (Reference Giles, Cattanach and Cattanach1987) reported increasing cultivation number from one to four numerically reduced yield in one of two environments. Giles et al. (Reference Giles, Cattanach and Cattanach1990) reported that one to three cultivations had no effect on sugarbeet yield, but there was an increasingly negative effect on sugarbeet yield as cultivation number increased from four to seven in one of two environments.
Root yield loss from inter-row cultivation later in the season is probably due to two factors: physical damage to the sugarbeet plant tissue and increased infection of Rhizoctonia solani (Kühn), the causal agent of Rhizoctonia crown and root rot. Giles et al. (Reference Giles, Cattanach and Cattanach1990) excavated roots in mid-July and observed less root development in the area from surface to 7 cm deep of soil in treatments receiving a large number (four to seven) of cultivations. The physical act of driving a mechanical implement between crop rows can also crush beet leaves that extend across field rows. The trend for reduced yield could also be related to soil-borne diseases. Cultivation when the sugarbeet plants are near canopy closure may deposit soil on the crown of the sugarbeet roots, potentially moving pathogens nearer their host. Schneider et al. (Reference Schneider, Ruppel, Hecker and Hogaboam1982) reported that covering sugarbeet roots with soil (hilling) in mid-August caused a significant increase of root rot from R. solani. However, hilling did not cause greater disease pressure in all location-years, suggesting that environmental factors may also contribute to disease severity. Cultivation at reduced ground speeds is recommended to reduce the chance of R. solani infection due to soil hilling near the sugarbeet crown (Schneider et al. Reference Schneider, Ruppel, Hecker and Hogaboam1982; Windels and Lamey Reference Windels and Lamey1998).
Sugarbeet producers in the late 2010s frequently applied glyphosate and chloroacetamide herbicides in layers until crop canopy closure. Inter-row cultivators are used after herbicide application to remove herbicide-resistant weed escapes or to control weeds when inconsistent control occurs with herbicides. Producers have inquired whether inter-row cultivation is a viable tool to remove weeds that glyphosate did not control. They are also interested in knowing whether or not a delayed cultivation will expose weed seeds in untreated soil, resulting in additional weed seed germination and emergence. Many producers are also concerned that inter-row cultivation will reduce sugarbeet yield and quality because of the results of earlier research by Dexter and Giles (Dexter Reference Dexter1983; Dexter et al. Reference Dexter, Luecke and Smith2000; Giles et al. Reference Giles, Cattanach and Cattanach1987, Reference Giles, Cattanach and Cattanach1990). Most producers consider one to two cultivation passes mid-season a “rescue” strategy rather than a contributor to an integrated weed management strategy. Therefore, two experiments were developed to address these concerns: (1) “delayed-cultivation efficacy” and (2) “cultivation tolerance.” The objectives of the “delayed-cultivation efficacy” experiment were to evaluate (1) the effectiveness of cultivation at removing herbicide-resistant weeds and (2) the effect of delayed cultivation on weed emergence. The objective of the “cultivation tolerance” experiment was to evaluate the effect of inter-row cultivation timing and number on sugarbeet yield and quality.
Materials and Methods
Site Description
Delayed-Cultivation Efficacy
Field experiments were conducted on four environments in grower fields; near Renville, MN (44.78°N, 95.14°W) in 2017, near Nashua, MN (46.05°N, 96.33°W) in 2018, near Lake Lillian, MN (44.88°N 94.98°W) in 2019, and near Galchutt, ND (46.38°N 96.84°W) in 2019. Each site–year combination is considered an environment. All environments were chisel-plowed in the fall and prepared for spring sugarbeet planting with a field cultivator. The environments in this experiment have a history of recurrent glyphosate use and presence of glyphosate-resistant waterhemp. Detailed soil descriptions for each environment can be found in Table 1.
Table 1. Soil description across environments including series, texture, subgroup, organic matter (OM) and pH; 2017, 2018, and 2019.
Cultivation Tolerance
Field experiments were conducted on six environments in grower fields; near Glyndon, MN (46.86°N, 96.52°W) in 2018, Galchutt, ND (46.38°N 96.84°) in 2019, Hickson, ND (46.70°N, 96.80°W) in 2018 and 2019, and Amenia, ND (47.00°N, 97.10°W) in 2018 and 2019. Previous crops grown in fields were soybean [Glycine max (L.) Merr.], soybean, sugarbeet, and wheat (Triticum aestivum L.) at the Glyndon, Galchutt, Hickson, and Amenia sites, respectively. Each site–year combination is considered an environment. All environments were chisel-plowed in the fall and prepared for spring sugarbeet planting with a field cultivator. Detailed soil descriptions for each environment can be found in Table 1.
Experimental Procedures
Delayed-Cultivation Efficacy
The experiment was a 2 × 4 factorial split-block design with four to six replications, depending on environment. Each replication (block) was grid split, where the horizontal factor was cultivation at two levels and the vertical factor was herbicide at four levels. Plots were 3.3 m wide and 9.1 m long. Sugarbeet was planted to a density of approximately 152,000 (± 1,000) seeds ha–1 in six rows spaced 56 cm apart, and S-metolachlor (Dual Magnum; Syngenta Crop Protection, Greensboro, NC) at 534 g ai ha–1 was applied PRE within 48 h after planting in all environments.
Herbicide treatments were applied to 8- to 10-cm weeds with a bicycle wheel-type sprayer with a shielded boom to reduce particle drift at a volume of 159 L ha–1 (Table 2). The center four rows of each six-row plot were sprayed using pressurized CO2 at 241 kPa through 8002XR nozzles (XR TeeJet® Flat Fan Spray Tips; TeeJet® Technologies, Glendale Heights, IL). Half of the plots were cultivated approximately 2 wk after herbicide application using a modified Alloway 3130 cultivator (Alloway Standard Industries, Fargo, ND) with 38-cm sweep shovels spaced at 56 cm with a ground depth of 4 to 5 cm at 6.4 km h–1. Dates of planting, herbicide application, cultivation, and crop stage at herbicide application can be found in Table 2.
Table 2. Description of treatments applied in the delayed cultivation efficacy experiment; timing of cultivations and herbicide treatments applied to 8- to 10-cm weeds at Renville-2017, Nashua-2018, Lake Lillian-2019, and Galchutt-2019.
a All herbicide treatments included ethofumesate at 140 g ai ha−1 (Ethofumesate 4SC, Willowood LLC, Roseburg, OR), high-surfactant methylated oil concentrate at 1.75 L ha−1 (Destiny HC, Winfield Solutions LLC, St. Paul, MN), and ammonium sulfate liquid solution at 2.5% v/v (N-Pak AMS liquid, Winfield Solutions LLC, St. Paul, MN). Herbicide treatments were applied June 26 at Renville-2017, June 12 at Nashua-2018, June 26 at Lake Lillian-2019, and June 17 at Galchutt-2019.
b Monsanto Company, St. Louis, MO.
c Syngenta Crop Protection, Greensboro, NC.
d BASF Corp., Research Triangle Park, NC.
Cultivation Tolerance
The experiment was a randomized complete block with four replicates. Plots were 3.3 m wide and 9.1 m long. Treatments were applied on 14-d intervals throughout the growing season starting June 22 and ending August 17. Treatments were a combination of the cultivation date, number of cultivations, and an untreated control (Table 3). Cultivation date and frequency were reflective of current grower practices (Peters et al. Reference Peters, Khan and Boetel2018). Inter-row cultivation was performed using a modified Alloway 3130 cultivator (Alloway Standard Industries, Fargo, ND) with 38-cm sweep shovels spaced at 56 cm with a ground depth of 4 to 5 cm at 6.4 km h–1.
Table 3. Numbers and dates of cultivations in cultivation tolerance experiment, Amenia, Hickson, and Glyndon, 2018 and Amenia, Hickson, and Galchutt, 2019.
a Treatments were cultivated within 5 d (±) of date.
The sugarbeet cultivar ‘Crystal 355RR’ (American Crystal Sugar Company, Moorhead, MN) was planted approximately 3 cm deep to a density of approximately 152,000 (± 1,000) seeds ha–1 in six rows spaced 56 cm apart (Table 4). Sugarbeet seeds were treated with penthiopyrad (Kabina ST; Sumitomo Corp., New York, NY) at 14 g per 100,000 seeds, hymexazol (Tachigaren 45; Mitsui Chemicals Agro, Tokyo, Japan) at 45 g per 100,000 seeds, and clothianidin and beta-cyfluthrin (Poncho Beta; Bayer Crop Science, Research Triangle Park, NC) at 60 g and 8 g per 100,000 seeds, respectively. Nitrogen, phosphorus, and potassium fertilizer was applied based on spring soil tests and incorporated prior to planting. Weeds and disease were controlled so that crop injury from cultivation could be detected without interference from other yield-limiting factors. Weeds were controlled using glyphosate (Roundup PowerMAX; Monsanto Company, St. Louis, MO) at 1.26 kg ae ha–1. One to three glyphosate applications were made at each environment, and herbicide-resistant waterhemp plants were removed by hand-weeding. Root disease pressure from R. solani was controlled with two soil-applied applications of azoxystrobin (Quadris; Syngenta Crop Protection, Greensboro, NC) at Amenia and Hickson. Disease pressure from Cercospora beticola was controlled with season-long foliar applications of triphenyltin hydroxide (Super Tin 4L; United Phosphorus, Inc., King of Prussia, PA), thiophanate methyl (Topsin 4.5FL; United Phosphorus, Inc., King of Prussia, PA), prothioconazole (Proline; Bayer CropScience LP, Research Triangle Park, NC), and difenoconazole/propiconazole (Inspire XT; Syngenta Crop Protection, Greensboro, NC).
Table 4. Planting and harvest dates, previous crop, and sugarbeet population prior to first cultivation treatment in the cultivation tolerance experiment at six environments, 2018 and 2019.
Data Collection and Analysis
Delayed Cultivation Efficacy
Percent visual waterhemp control was evaluated 14 and 28 (± 3) d after the cultivation treatment (DAC). Evaluation was on a scale of 0% (no control) to 100% (complete control) relative to the untreated check rows between treatments. Waterhemp in the 2.2-m by 9.1-m treatment area of each 3.3-m by 9.1-m plot was counted 14 and 28 DAC at the Renville-2017 and Nashua-2018 environments because of low weed pressure, whereas a 0.25-m2 quadrat, placed twice between the two middle plot rows, three paces apart, was used at Lake Lillian-2019 and Galchutt-2019, where weed pressure was greater. Rainfall data were collected from nearby weather stations operated by the North Dakota Agricultural Weather Network (NDAWN), National Weather Service (NWS), and Southern Minnesota Beet Sugar Cooperative (SMBSC) and are presented in Table 5.
Table 5. Weekly and monthly rainfall in delayed cultivation efficacy experiment conducted in four environments compared with 30-yr averages, 2017, 2018, and 2019. a
a Nashua and Galchutt climate data collected by the North Dakota Agricultural Weather Network (NDAWN); Renville climate data collected from Olivia, MN airport (NWS); Lake Lillian climate data collected from on-site weather station operated by Southern Minnesota Beet Sugar Cooperative (SMBSC).
b Distance (km) and direction of weather station from trial site.
c Rainfall data in parentheses are within the accumulation period between planting and last evaluation.
d 30-yr average is National Weather Service (NWS) average at Wahpeton, ND.
Data were subjected to analysis using the GLIMMIX procedure in SAS (v. 9.4, SAS Institute, Cary, NC) to test for treatment effects and significant interactions. Data were analyzed as a split-block randomized complete block design with expected means squares as recommended by Carmer et al. (Reference Carmer, Nyquist and Walker1989). Significantly different treatment means were separated using the Tukey-Kramer procedure at the P ≤ 0.05 level. Waterhemp control data were arc sine square-root transformed {arcsin[(Y/100)1/2]}, and waterhemp density data were square-root transformed ([Y+0.5]1/2) to better fit assumptions of the model analysis, and untransformed means are presented. The cultivation and herbicide treatment factors were considered fixed effects, whereas replicate, environment, and interactions containing replicate and environment were considered random effects. Results from Levene’s test for homogeneity found that waterhemp control data could be combined across four environments (P values = 0.141 and 0.408, 14 and 28 DAC), but waterhemp density data required separate analysis for each environment. Only main effects are presented, as no significant cultivation-by-herbicide interactions were detected.
Cultivation Tolerance
Sugarbeet density was collected in the center two rows prior to the start of cultivation treatments and prior to harvest to determine percent stand mortality throughout the season (Equation 1). At harvest, sugarbeet was defoliated and harvested mechanically from the center two rows of each plot and weighed. A sample weighing approximately 10 kg was collected from each plot and analyzed for sucrose content and sugar loss to molasses by American Crystal Sugar Company (East Grand Forks, ND). Sugarbeet roots were visually analyzed for Rhizoctonia root and crown rot, but no visible infection was observed. Root yield (kg ha–1), purity (%), and recoverable sucrose (kg ha–1) were calculated using Equations 2, 3, and 4, respectively. Monthly rainfall data were collected for each environment by NDAWN and are presented on Table 6.
$${\rm{Stand}}\;{\rm{mortality}} = \left[ {1 - \left( {{{{\rm{Density}}\;{\rm{prior}}\;{\rm{to}}\,{\rm{harvest}}} \over {{\rm{Density}}\;{\rm{prior}}\;{\rm{to}}\;{\rm{cultivation}}}}} \right)} \right]\; \times \,100$$
$${\rm{Root}}\;{\rm{yield}}\;\left( {{\rm{kg}}/{\rm{ha}}} \right){\rm{ }} = {\rm{ }}{{{\rm{Harvested}}\;{\rm{plot}}\;{\rm{weight}}\;\left( {{\rm{kg}}} \right)} \over {{\rm{Hectare}}\;{\rm{area}}\;{\rm{of}}\;{\rm{harvested}}\;{\rm{plot}}}}$$
$${\rm{Purity}}\;\left( \% \right) = {\rm{ }}{{\% \;{\rm{Sucrose}}\;{\rm{content}}\; - \;\% \;{\rm{sugar}}\;{\rm{loss}}\;{\rm{to}}\;{\rm{molasses}}} \over {\% \;{\rm{Sucrose}}\;{\rm{content}}}} \times \;100$$
$$\matrix{
{{\rm{Recoverable}}\;{\rm{sucrose}}\;\left( {{\rm{kg}}/{\rm{ha}}} \right) = } \hfill \cr
{\left( {{{\left[ {\left( {\% \;{\rm{Purity}}\;/\;100} \right)\; \times \;\% \;{\rm{sucrose}}\;{\rm{content}}} \right]} \over {100}}} \right)} \hfill \cr
} {\rm{ }}\; \times \;{\rm{root}}\;{\rm{yield}}$$
Table 6. Monthly rainfall, cultivation tolerance experiment, 2018 and 2019. a
a Climate data collected by instrumentation managed by the North Dakota Agricultural Weather Network.
b Distance (km) and direction of weather station from trial site.
c Rainfall data in parentheses are within the recorded period between planting and harvest.
d 30-yr average is National Weather Service (NWS) average at Fargo, ND.
Data were subjected to analysis using the GLIMMIX procedure in SAS (v. 9.4, SAS Institute, Cary, NC) to test for treatment effects, and means were separated using the Tukey-Kramer procedure at P ≤ 0.05. Cultivation treatment was considered a fixed effect, whereas environment and replicate were considered random effects. Single degree-of-freedom contrasts were used to compare the effect of cultivation number on sugarbeet density and yield components. Levene’s test for homogeneity was conducted to determine which environments could be combined for each independent variable at the P ≤ 0.05 level.
Results and Discussion
Delayed Cultivation Efficacy
Waterhemp Control
Visual waterhemp control data were analyzed across environments, but data from each environment are also reported (Table 7). Cultivation significantly improved waterhemp control 11% and 12%, 14 and 28 DAC, respectively, across environments (Table 7). Herbicide treatment did not affect waterhemp control, nor was there any interaction between herbicide and cultivation treatments (Table 7). A cultivator removes about two-thirds of weeds in fields with 56-cm crop spacing (38-cm shovels cover 68% of area in 56-cm rows). One of the major drawbacks of inter-row cultivation is that it removes only the weeds in between rows (VanGessel et al. Reference VanGessel, Schweizer, Wilson, Wiles and Westra1998), necessitating the use of other means to remove the remaining weeds. Hand-weeding to remove weeds that escaped inter-row cultivation was a common strategy prior to the development of GR sugarbeet in 2008. Cultivation may have reduced herbicide efficacy on waterhemp control in this experiment, as previous research demonstrated improved waterhemp control from a chloroacetamide plus glyphosate treatment as compared to glyphosate alone (Peters et al. Reference Peters, Lueck and Groen2017).
Table 7. Waterhemp control in response to cultivation and herbicide treatment, 14 and 28 d after cultivation treatment (DAC). a
a Numbers within a main effect and environment column followed by different letters are statistically different at P ≤ 0.05 (Tukey’s test).
b Cultivation was approximately 2 wk after spray treatment.
c All herbicide treatments included ethofumesate (140 g ai ha−1), high-surfactant methylated oil concentrate (1.8 L ha−1), and liquid ammonium sulfate at 2.5% v/v.
d Evaluation on scale of 0% (no control) to 100% (complete control).
Waterhemp Density
Data were not combined across environments based on Levene’s test and were analyzed by environment. Density of waterhemp was relatively low (<2 plants m–2) at the Renville-2017 and Nashua-2018 environments, and plants were counted on a per-plot basis, whereas a 0.25-m2 quadrat was used at Lake Lillian-2019 and Galchutt-2019 to measure waterhemp density. Results from Renville-2017 indicated that cultivation reduced waterhemp density 63% and 53% 14 and 28 DAC, respectively (Table 8). Waterhemp density at Nashua-2018 was not affected by cultivation. Results from Lake Lillian-2019 indicated no effect of inter-row cultivation 14 DAC, but cultivation reduced waterhemp density 59% 28 DAC (Table 8). Inter-row cultivation increased waterhemp density 600% and 196% at Galchutt-2019 14 and 28 DAC, respectively (Table 8). Herbicide treatment did not affect waterhemp density at three of four environments (Table 8), but the effect of herbicide at Renville-2017 was not meaningful, as recorded plants were already emerged at time of herbicide application.
Table 8. Waterhemp density in response to cultivation and herbicide treatment, 14 and 28 d after cultivation treatment (DAC). a
a Numbers within a main effect and environment column followed by different letters are statistically different at P ≤ 0.05 (Tukey’s test).
b Cultivation was approximately 2 wk after spray treatment.
c All herbicide treatments included ethofumesate (140 g ai ha−1), high-surfactant methylated oil concentrate at 1.8 L ha−1, and liquid ammonium sulfate at 2.5% v/v.
Waterhemp density with and without cultivation is likely due to an interaction between the crop stage and canopy at the time of cultivation and the amount of rainfall received following the cultivation event. Renville-2017 and Lake Lillian-2019 were cultivated with a developed canopy of 8- to 12-leaf and 8- to 10-leaf (10-leaf predominant) sugarbeet (Table 2) that provided natural shade in between the sugarbeet rows and received approximately 9 mm and 11 mm of rainfall in the 14 d following cultivation at Renville-2017 and Lake Lillian-2019, respectively (Table 5). The interaction of a developed canopy and low precipitation following cultivation at Renville-2017 and Lake Lillian-2019 likely made environments nonconducive to further weed germination. Nashua-2018 was cultivated with an underdeveloped canopy of 6- to 8-leaf (8-leaf predominant) sugarbeet (Table 2) and received approximately 28 mm of cumulative rainfall to 14 d following cultivation (Table 5), but rainfall did not trigger further weed emergence. The waterhemp density at Nashua-2018, however, was relatively low compared to the other environments. Galchutt-2019 was cultivated with an underdeveloped canopy of 6- to 8-leaf (6-leaf predominant) sugarbeet (Table 2) and received 105 mm of precipitation in the 14 d following the cultivation event (Table 5). The interaction of an underdeveloped crop canopy that left much soil exposed at the time of cultivation and the 105 mm of precipitation that followed are likely responsible for the explosion of weed emergence following the cultivation.
Cultivation works by stirring soil and carrying plants to the soil surface, where they will ideally die of desiccation. In the case of Galchutt-2019, however, the cultivation on an underdeveloped canopy likely mixed soil, exposing seeds in the seed bank to infrared light that stimulates germination, and subsequent rainfall created an environment conducive to further weed emergence. Furthermore, observations throughout the years have shown that a timely precipitation event following cultivation can result in failed control by causing plants uprooted by cultivation to re-root (Mohler et al. Reference Mohler, Iqbal, Shen and DiTommaso2016).
Cultivation Tolerance
Sugarbeet Stand Mortality
Stand mortality percentage was calculated from the ratio of sugarbeet density before cultivation treatments and at harvest. Inter-row cultivation did not affect sugarbeet stand mortality at any environment in 2018 and 2019 (Table 9). The relatively high stand mortality at Hickson-2018 and Galchutt-2019 probably occurred because sugarbeet and soybean, respectively, were the crop grown on the field sites in the prior year. Planting sugarbeet into sugarbeet residue greatly increases chance of infection from R. solani, which can cause significant stand loss (Windels and Brantner Reference Windels and Brantner2008).
Table 9. Sugarbeet stand mortality in response to cultivation timing and number, 2018 and 2019. a
a Abbreviations: NC, Noncultivated treatment; C, cultivated treatments; ST, single-timing treatments; DT, double-timing treatments; TT, triple-timing treatment.
b Percent stand mortality is calculated using the ratio of harvest stand and pre-treatment stand.
Harvested sugarbeet roots were visually inspected for root and crown rot from R. solani, but no significant infection due to inter-row cultivation was observed at any environment. Damage from R. solani in the field primarily manifests in the form of stand mortality (M. Khan 2018, personal communication), which was observed on all treatments at Hickson-2018 and Galchutt-2019. Inter-row cultivation has historically been associated with root and crown rot, because cultivation may physically deposit soil onto a beet crown, moving soil-borne pathogens nearer their host. Schneider et al. (Reference Schneider, Ruppel, Hecker and Hogaboam1982) reported that covering sugarbeet roots with soil with a cultivator moving 13 km h–1 in mid-August resulted in greater root rot due to R. solani in two of three field environments. Windels and Lamey (Reference Windels and Lamey1998) reported that reducing cultivation ground speed reduces the likelihood of infection from R. solani. Some soil movement onto beet crowns was observed in this experiment, but the cultivation speed of 6.4 km h–1 used in this experiment may not have been fast enough to cause significantly different root rot infection compared to the untreated control.
Root Yield
Inter-row cultivation did not affect root yield at any environment or environment combination (P values = 0.271 to 0.863) (Table 10). Inter-row cultivation only disturbs soil between the sugarbeet rows and does not significantly affect root growth or yield. Giles et al. (Reference Giles, Cattanach and Cattanach1990) conducted root excavations on sugarbeet in late July and reported less root development and yield with treatments receiving five to seven weekly cultivations throughout the season in one of two environments. Giles et al. (Reference Giles, Cattanach and Cattanach1990) cultivated to a similar depth of 4 to 5 cm, but ground speed was 5 km h–1. Significant root yield reduction was not observed with up to three cultivations in this experiment, cultivating 4 to 5 cm deep and 6.4 km h–1. The yield loss Giles et al. (Reference Giles, Cattanach and Cattanach1990) reported in one of two environments was likely due their five to seven cultivations as compared to one to three implemented in our experiment.
Table 10. Sugarbeet root yield and recoverable sucrose in response to cultivation timing and number, 2018 and 2019. a
a Abbreviations: NC, Noncultivated treatment; C, cultivated treatments; ST, single-timing treatments; DT, double-timing treatments; TT, triple-timing treatment.
Sucrose Content
Inter-row cultivation timing and number significantly reduced sucrose content at one environment, Hickson-2019 (Table 11). At Hickson-2019, only the June 22 and July 20 single-cultivation treatments had sucrose content similar to the untreated control. Single degree-of-freedom contrasts at Hickson-2019 showed that cultivation reduced sucrose content by an average of 0.8% compared to sugarbeet that were not cultivated (P value = 0.007) (Table 11). A single mid-season cultivation of sugarbeet reduced sucrose by 0.7% (P value = 0.031), two mid-season cultivations reduced sucrose content by 1.0% (P value = 0.004), and three mid-season cultivations reduced sucrose content by 0.9% compared to the control that was not cultivated (P value = 0.010). These data suggest that mid-season cultivation, especially multiple cultivations, can reduce sucrose content of harvested sugarbeet in certain environments. The reason for Hickson-2019 showing these differences and other environments not showing differences is unknown.
Table 11. Sugarbeet sucrose content in response to cultivation timing and number, 2018 and 2019. a
a Abbreviations: NC, Noncultivated treatment; C, cultivated treatments; ST, single-timing treatments; DT, double-timing treatments; TT, triple-timing treatment.
We observed on multiple occasions that cultivation at later dates can damage leaf tissue by tearing or ripping leaf tissue from the plant (Figure 1). Sugarbeet plants compensate for the foliar damage by producing new leaves, utilizing sucrose stored in roots as energy source and thus potentially lowering sucrose content. Leaf tissue is the medium by which photosynthesis and sucrose production is conducted in the plant, and cultivation damaging leaf tissue would logically reduce percent sucrose. Three layers of sugarbeet leaf tissue are required for optimal sucrose production in sugarbeet (K. Fugate 2019, personal communication), and we observed cultivation causing most damage to the bottom layer. Foliar damage was also noted from the tractor wheels traveling between plot rows. The tractor wheels in this experiment traveled on the outside of the plot area to remove the effect of the wheels on these results. Most producers operate with cultivators the same size as their planter to reduce the amount of unnecessary tire tracks and canopy damage in a field. Further research should determine the correlation between severity and timing of foliar damage and sucrose reduction.
Figure 1. Defoliation of sugarbeet leaf tissues caused by an inter-row cultivation event on August 1, 2018 near Amenia, ND.
Recoverable Sucrose per Hectare
Inter-row cultivation did not affect recoverable sucrose per hectare (RSH) at any environment or combination of environments (Table 10). RSH is a calculation derived from root yield and sucrose content (Equation 3) and is considered the most important metric in sugarbeet production. No treatment differences were measured in any environment (P values = 0.193 to 0.847). This result was expected, because recoverable sucrose is most heavily weighted by root yield (Equation 3), and the effect of root yield was insignificant.
Practical Implications
These data demonstrate that inter-row cultivation is a valuable tool to remove weeds that herbicide did not/could not control but can cause further weed emergence under certain environmental and crop conditions. Cultivation improved visual waterhemp control by 11% to 12% when performed 2 wk following a POST application, but an increase in waterhemp emergence was observed in one environment with an underdeveloped canopy and excessive precipitation. Inter-row cultivation did not affect sugarbeet stand mortality, root yield, or recoverable sucrose across six environments in 2 yr, but cultivation significantly reduced sugarbeet sucrose content in one environment, which could be attributed to foliar destruction from the cultivator. Although no change in recoverable sucrose (considered the most important yield metric for producers) was observed from cultivation, we caution growers that any operation that damages leaf tissue comes with the risk of reducing sucrose content as the plant seeks to replace damaged leaves. Most producers in 2018 and 2019 used cultivation only to remove weeds that glyphosate did not control, so it is unlikely that any sugarbeet producer would cultivate a field more than three times in one season. Most cultivations in 2018 and 2019 were also done after the sugarbeet canopy closed in mid-July. The effect of inter-row cultivation on yield is likely a complex interaction of cultivation timing, soil type, environmental conditions, disease pressure, cultivation speed, and cultivation equipment.
Sugarbeet producers are concerned about yield loss from inter-row cultivation because of previous research reported by Dexter et al. (Reference Dexter, Luecke and Smith2000) and Giles et al. (Reference Giles, Cattanach and Cattanach1990). Although the cultivation methods and procedures used in our experiment were similar to what Dexter and Giles implemented in their experiments, our timing of cultivation differed. Dexter and Giles conducted their cultivations on weekly intervals with the same start date, whereas our cultivations were 2 wk apart with staggered starting dates and timings as late as August 16. Furthermore, certain aspects of sugarbeet production that could affect disease pressure differ from the 1980s and 1990s, such as diploid genetics, seed treatments, and soil-applied applications of azoxystrobin. Our results show that cultivation 4 to 5 cm deep at 6.4 km h–1 did not affect recoverable sucrose in 2018 and 2019, but further research is needed in future years with different ground speeds, cultivator configurations, fungicide applications, and environmental conditions to determine how and when cultivation could affect sugarbeet yield.
Acknowledgements
No conflicts of interest have been declared. Financial support for this research was provided by the Sugarbeet Research and Education Board of Minnesota and North Dakota, using check-off dollars from growers representing American Crystal Sugar Company, Minn-Dak Farmers Cooperative, and Southern Minnesota Beet Sugar Cooperative. Thanks to Alexa Lystad, Jeff Stith, Peter Hakk, and David Mettler for assistance in the field. Thanks to Drs. Alan Dexter and Caleb Dalley for their review of this manuscript.
