Treatments were three mustard cover crop termination times and control with no cover crop. Mustard cover crop termination times were 56, 42, and 28 d before chile pepper seeding. Treatments were initially described as days before chile pepper seeding because days of initiation for beet leafhopper flights were unknown when trials were conducted. When year-specific temporal patterns of beet leafhopper flights were understood, treatments were described as both days before chile pepper seeding and days before beet leafhopper flights. At 56, 42, and 28 d before chile pepper seeding, mustard cover crop plants were at the rosette, stem elongation, and flowering stages, respectively. The least amount of time between cover crop termination and chile pepper seeding was 28 d because this amount of time was needed for the field to dry after the irrigation applied during cover crop termination.
Treatments were arranged in a randomized complete block design with four replications. Experimental units were 48 m long and 4 m wide; they are herein referred to as plots. During growing seasons for the cover crop, plots contained 22 rows of cover crop. Neighboring rows of cover crop were spaced 18 cm apart. During the growing season for chile pepper, experimental units contained four rows of chile pepper. Chile rows were centrally placed on raised beds that had a width of 0.8 m. Neighboring rows of chile pepper were spaced 1 m apart.
COVER CROP MANAGEMENT AND CHILE PEPPER SEEDING. Before sowing the cover crop, fields were prepared with a laser-guided land levelling system (Laser Alignment Inc., Grand Rapids, MI, USA). Dates for field preparation and subsequent management procedures are presented in Table 1. 'Caliente Rojo' brown mustard (Caliente Brand™; Stokes Seeds Inc., Holland, MI, USA) was seeded at 7 lb/acre with a mechanical grain drill (Model 450; John Deere, Moline, IL, USA). Within 48 h of seeding, fields were flood-irrigated. Flood irrigations occurred as needed to prevent crop mortality through fall and winter. Each irrigation was approximately 3 inches deep and saturated the soil.
The cover crop treatments were ended using a flail shredder (Model ORC12; RhinoAg Inc., Gibson City, IL, USA). Immediately after shredding, residues were incorporated into soil to the 15-cm depth with two passes of an offset tandem disk (Model 620; John Deere, Moline, IL, USA). Immediately after disking, raised beds were made using a lister (Dave Koenig Enterprises Inc., Mesilla Park, NM, USA). Within 2 h of listing, the furrows between raised beds were flood-irrigated. Each irrigation was approximately 3 inches deep and saturated raised beds.
For plots containing the control with no cover crop, the creation of raised beds and subsequent irrigation coincided with the third termination dates for mustard cover crops (31 Mar 2020 and 1 Apr 2021). On 28 Apr 2020 and 29 Apr 2021, raised beds were lightly disked and shaped using a bed shaper. 'Big Jim' chile pepper was seeded at 6 lb/acre to a depth of 1 inch using a mechanical seeder (MaxEmerge V R Plus; John Deere, Moline, IL, USA). Chile pepper rows were positioned in central areas of raised beds, with each bed containing one row of chile pepper. Furrow irrigation occurred immediately after seeding and as needed thereafter.
BEET LEAFHOPPER FLIGHTS. To measure changes over time in beet leafhopper abundance at the experimental farm, four yellow sticky traps (20 cm × 25 cm; Hummert International, Earth City, MO, USA) were placed 24 inches from the ground surface in four locations across the Leyendecker Plant Science Research Center. One of the four traps was 100 m north of the study site. Two traps were 350 m and 450 m northwest of the study site, and one trap was 450 m south of the study site. The total number of beet leafhoppers on each card was determined every 2 weeks from January to late March, and every week from late March to early July. On the days of collection, new traps were positioned to replace the collected traps.
SINIGRIN AMENDMENTS TO SOIL. To determine the amounts of sinigrin added to soil by cover crops ended on different dates, measurements of cover crop biomass at termination were combined with date-specific measurements of sinigrin concentrations in cover crop biomass. Measurements of cover crop biomass at termination generally followed the procedures of Nagila et al. (2022) (
link). Specifically, just before cover crop termination, aboveground biomass of mustard cover crops was clipped at the soil surface and collected from four 0.25-m 2 quadrats (0.5 m width × 0.5 m length) within each plot. The quadrats were evenly spaced along the central long axes of plots. Following collection, biomass samples were oven-dried at 60 C until they reached a constant weight and then weighed. The weights of the four samples were averaged before performing calculations to determine amounts of sinigrin added to soil and statistical analyses.
To quantify sinigrin in the aboveground and belowground mustard plant biomass, four entire mustard plants were collected from areas near the biomass harvest locations. These plants were placed on dry ice in cold storage containers and transported to the laboratory, where they were stored at À18 C. Cold-stored plants were crushed to a powder using a mortar and pestle in the presence of liquid nitrogen for 1 min (Doheny-Adams et al. 2017). Sinigrin was extracted from powdered samples following the cold methanol extraction method described by Doheny-Adams et al. (2017) . Sinigrin was quantified using high-performance liquid chromatography (HPLC) analyses following the methods of Wood et al. (2020) (
link). HPLC was conducted with an Agilent 1100 series HPLC (Agilent Technologies, Santa Clara, CA, USA) equipped with a Zorbax column (C18; 4.6 × 100 mm; 3.5 mm). Solvents in the HPLC analyses were 0.02 M tetrabutylammonium bromide and a 70:30 mixture of tetrabutylammonium bromide and acetonitrile. Data were recorded using software (Agilent Software Chemstation V.B.04.01). HPLC data were converted to sinigrin concentrations in plant samples using standard curves produced with five concentrations (1.25, 2.5, 5, 10, and 20 mM) of sinigrin standards (Sigma-Aldrich, St. Louis, MO, USA). Sinigrin concentrations for plants from the same plot were averaged before performing calculations to determine the amounts of sinigrin added to soil and statistical analyses.
Results of HPLC analyses were sinigrin concentrations based on fresh weight. To calculate the amount of sinigrin added to soil during mustard cover crop termination, dry weights of mustard cover crop samples were first multiplied by 10.7, which accounted for the moisture content of freshly harvested mustard biomass (89.3% water). These calculated fresh weights were then multiplied by HPLC-derived sinigrin concentrations.
WEED Weed density data were summed to determine cumulative weed density from 0 to 28 d after seeding and 29 to 56 d after seeding. Cover crop treatment effects on weed densities were determined with generalized linear models with negative binomial distributions developed using the R library mass. Trials were analyzed separately because preliminary analyses indicated that termination time effects on weed densities differed between the 2019-20 trial and 2020-21 trial. For each trial, predictor variables in generalized linear models were replicate and cover crop treatment including the no cover control. Parameter estimates from generalized linear models were used to assess possible differences among cover crop treatments. Specifically, parameter estimates with overlapping 95% confidence intervals indicated similarity among cover crop treatments. Parameter estimates with 95% confidence intervals that did not overlap indicated cover crop treatments with different weed densities.
To determine cover crop treatment effects on hoeing times and chile pepper stands, ANOVA tests were performed separately for trials. Predictor variables in ANOVA models were replicate and cover crop treatment. Visual inspections of residuals plotted against fitted values indicated that the log-transformation of response variables was necessary for ANOVA assumptions of the constant variance of errors. Thus, hand-hoeing time and chile stand data were log-transformed before the analyses. The annual total of beet leafhoppers was less in 2021 than in 2020. Year-to-year variability in beet leafhopper abundance in spring can be attributable to differences in host plant availability (Thomas and Martin 1971) (
link) that are partially consequences of annual differences in precipitation during the previous fall (Lehnhoff and Creamer 2020) (
link). In fact, levels of precipitation during the previous fall can be used to predict the relative abundance of beet leafhoppers in spring (Lehnhoff and Creamer 2020) (
link). Year-to-year variability in dates of initial beet leafhopper flights was associated with yearly differences in levels of precipitation at the study site during spring (Fig. 2). Spring 2021, which was relatively dry, featured an earlier beet leafhopper flight compared with the beet leafhopper flight that occurred in Spring 2020, which was relatively wet. This putative association between flight timing and precipitation level is consistent with the underlying causes of beet leafhopper flights. In New Mexico and other regions of the western United States, flights occur because adult beet leafhoppers depart from agricultural weeds and desert plants as this vegetation desiccates in early spring (Cook 1967; Davis 2010) . During this study, desiccation of agricultural weeds and desert plants likely occurred earlier during the spring with little precipitation rather than during the spring with more precipitation.
Bajagain A., Lehnhoff E.A., Creamer R., Steiner R.L., & Schutte B.J. (2024). Timing Termination of a Biofumigant Cover Crop for Weed Suppression in Chile Pepper. HortTechnology, 34(2), 142-152.
