Fall Applied Weed Management Systems for Tall Fescue Seed Production in Northeast Missouri

Introduction:
Tall fescue is one of the most important cool-season forages in the United States occupying over 35 million acres with seed production on 370,000 to 490,000 acres each year (Young, 1997; Wheaton, 1999). Missouri is the leading fescue seed producer in the Midwest and harvests a majority of tall fescue seed from existing pasture or forage land as uncertified ‘Kentucky 31'. Annual tall fescue seed production in Missouri has ranged from 170,000 to 320,000 acres with average yields from 170 to 220 lbs/acre (Young, 1997). Missouri ranked second in the number of total cattle in the United States in 2002 (Anonymous 2002). Integrated livestock and seed producers may utilize baled stubble aftermath as a forage source and stockpile forage for winter grazing from specialized seed production fields for a "triple-crop" production system. Since the 1960's, specialized fescue seed production systems in the Pacific Northwest have averaged from 580 to 1300 lbs/acre (Young 1997). Methods to increase seed and forage yields may add value to current forage and livestock production systems in Missouri.

Research has evaluated the effects of N (Fairey and Lefkovitch, 1998; Young III et al., 1998b; Young III et al., 1999), defoliation (Watson and Watson 1982), cultivar selection (Young III et al., 1998a; Young III et al., 1998b; Young III et al., 1999), row spacing (Fairey and Lefkovitch 1999; Young III et al., 1998b), defoliation (Watson and Watson, 1982), residue management methods (Mueller-Warrant et al., 1995; Young III et al., 1998a; Young III et al., 1999), and weed management (Johnson and Carrow, 1995; Peters et al., 1989) in tall fescue seed production systems. Some weed management systems may reduce seed head density, forage production and quality (Moyer and Kelley, 1995) which may reduce seed yields. However, no research has evaluated weed management systems in specialized fescue seed production systems in the Midwest. Downy brome and cheat (Bromus secalinus L.) are common winter annual weeds in fescue seed production fields in Missouri and are difficult to separate from tall fescue seed. No research or recommendations were available on the impact of downy brome on fescue seed yields in specialized seed production fields in the Midwest (Wheaton, 1999). The objective of this study was to evaluate preemergence, postemergence, and preemergence followed by postemergence weed management systems on control of downy brome and tall fescue injury, forage, and seed yield in specialized seed production fields in Northeast Missouri.

Materials and Methods:
Field research was conducted in 2002, 2004, 2005, and 2006 at the University of Missouri Greenley Research Center near Novelty, MO. The soil was a Kilwinning silt loam (Fine, montmorillonitic, mesic, Vertic Ochraqualfs). The soil had textural fractions of 32% sand, 36% silt, and 32% clay with a 6.0 pHs and 2.7% organic matter. This research was arranged as a randomized complete block design in plots 10 by 40 ft with four replications in the non-grazed experiment and three replications in the grazed experiment. ‘Kentucky 31' was drill¹ seeded in separate experiments at 8 lbs/acre in 15 inch wide rows on 11 September 2000 (non-grazed stockpiled forage) and 29 August 2002 (grazed stockpiled forage). The grazed stockpiled forage site established in 2002 was intensively grazed by mature cows (Bos taurus L.) prior to the postemergence herbicide application. The seedbed was diskharrowed three times and cultipacked twice in 2000 and no-till seeded in 2002. Plots were broadcast seeded with downy brome at 30 lbs/acre in the spring of each establishment year. Experiments were initiated in the fall after establishment and following forage aftermath removal. Maintenance fertilizer was split-applied at 50 lb N/acre, 40 to 80 lb P/a, and 125 to 180 lb K/acre in late summer (August) and 40 to 50 lb N/acre in early spring (March) based on soil test recommendations, forage aftermath removal, and expected seed yield. Prohexadione calcium at 280 g ai/ha plus nonionic surfactant² at 0.25% v/v plus diammonium suflate at 1.1 kg/ha was broadcast applied to all plots in 2002. Trinexapac-ethel at 210 g ai/ha was broadcast applied to all plots in 2004, 2005, and 2006. Plant growth regulators were applied the first week of May when the flag leaf was initiated to reduce lodging and allow direct harvesting (Young III et al., 1999). Forage aftermath was mowed, raked, baled, and removed from mid-July to early- August each year.

Application information and dates, weed height and density, and fescue height at the time of application are recorded in Table 1. Herbicide treatments and application timings are listed in Table 2. All treatments were applied with a CO2 propelled hand-boom. Fescue injury was evaluated 7 days after the postemergence application while visual weed control was evaluated in the following spring (mid-May). Visual evaluations were based upon a scale of zero (no effect) to 100 (complete weed or crop death) percent. Prior to seed harvest, 12 by 30 inch quadrats were randomly placed perpendicular to the row in the center of every plot, and forage biomass was harvested and fresh weight determined. Plots were direct harvested with an Allis Chalmers K2³ plot combine. Yield was adjusted to 12% moisture and purity adjustment was determined based on cleanout analysis by a local seed buyer prior to analysis.

All data were subjected to analysis of variance using PROC ANOVA (SAS Inst., 1999) and subjected to an F Max test for homogeneity (Kuehl 1994). Data were combined over years and locations since variances were homogenous. Visual injury data were transformed to the arc sine prior ANOVA. This transformation did not affect conclusions; therefore, original means were reported. Means were separated using Fisher's Protected LSD at p<0.05.

Results and Discussion:
All treatments except metribuzin alone or tank mixed with oxyfluorfen injured non-grazed fescue less than 10% (Table 1) while oxyfluofen and metribuzin applied alone or tank mixed injured grazed fescue 3 to 11% (Table 2). Diuron alone or tank mixed with oxyflurofen and cultivation only controlled downy brome greater than 79% in non-grazed tall fescue (Table 1). Oxyflurofen plus metribuzin applied postemergence and pendimethalin preemergence followed by metribuzin, oxyfluorfen plus metribuzin, diuron, or oxyfluorfen plus diuron controlled downy brome greater than 85% in grazed tall fescue. All weed management systems evaluated had forage aftermath and seed yields similar to the non-treated control.

Similar weed management systems have been utilized for control of volunteer tall fescue (Mueller-Warrant et al., 1995). Limited volunteer tall fescue has become established in treated or non-treated plots; however, grazing timing and intensity may reduce the incidence of volunteer tall fescue (personal observation). Watson and Watson (1982) evaluated defoliation treatments; therefore, late grazing may reduce seed yields and recommended no defoliation in the spring after tillers elongate. Additional research is needed to evaluate the impact of grazing timing on seed production and the interaction with other management decisions.

Acknowledgements:
The authors would like to thank Matthew Jones, Randall Smoot, Dana Harder, Chris Bliefert, Adam Jones, Heather Collier, Sandra Devlin, Sheena Mitchell, Erik McGuire, and Steve Webb for their technical assistance with this research.

References:
Anonymous. 2002. Missouri Farm Facts. Missouri Department of Agriculture, Jefferson City, MO; United States Department of Agriculture, Washington, D.C.

Fairey, N.A. and L.P. Lefkovitch. 1998. Effects of method, rate and time of application of nitrogen fertilizer on seed production of tall fescue. Can. J. Plant Sci. 78:453-458.

Fairey, N.A. and L.P. Lefkovitch. 1999. Crop density and seed production of tall fescue (Festuca arundinacea Schreber). 1. Yield and plant development. Can. J. Plant Sci. 79:535-541.

Johnson, B.J. and R.N. Carrow. 1995. Reduced preemergence herbicide rates for large crabgrass (Digitaria sanguinalis) control in six tall fascue (Festuca arundinacea) cultivars. Weed Technol. 9:716-723.

Kuehl, R. O. 1994. Statistical Principles of Research Design and Analysis. Duxbury Press, Belmont, CA. pp. 686.

Moyer, J.L. and K.W. Kelley. 1995. Broadleaf Herbicide Effects on Tall Fescue (Festuca arundinacea) Seedhead Density, Forage Yield, and Quality. Weed Technol. 9:270-276.

Mueller-Warrant, G.W., W.C. Young III, and M.E. Mellbye. 1995. Residue Removal Method and Herbicides for Tall Fescue Seed Production: I Weed Control. Agron. J. 87:551-558.

Peters, T.J., R.S. Moomaw, and A.R. Martin. 1989. Herbicides for Postemergence Control of Annual Grass Weeds in Seedling Forage Grasses. Weed Sci. 37:375-379.

Watson, Jr., C.E. and V.H. Watson. 1982. Nitrogen and Date of Defoliation Effects on Seed Yield and Seed Quality of Tall Fescue. Agron. J. 74:891-893.

Wheaton, H.N. 1999. Seed production of tall fescue and other cool season grasses. University of Missouri Ext. Pub. G4670.

Young III, W.C. 1997. Festuca arundinacea Schreb. (Tall Fescue) in the USA. Pages 287-296 in Fairley, D.T. and J.G. Hampton eds. Forage Seed Production Volume 1: Temperate Species. CAB International, New York, NY.

Young III, W.C., G.A. Gingrich, and T.B. Silberstein. 1998a. Post-harvest residue management of creeping red and chewings fescue seed crops. Agron. J. 90:69-73.

Young III, W.C., H.W. Youngbert, and T.B. Silberstein. 1998b. Management Studies on Seed Production of Turf- Type Tall Fescue: I. Seed Yield. Agron. J. 90:474-477.

Young III, W.C., M.E. Mellbye, and T.B. Silberstein. 1999. Residue management of perennial rye grass and tall fescue seed crops. Agron. J. 91:671-675.

¹Great Plains Manufacturing Inc., P.O. Box 218, Assaria, KS 67416
²Nonionic surfactant was Activator-90, a mixture of alkyl polyoxyethylene ether and free fatty acids, Loveland Industries Inc., P.O. Box 1289, Greeley, CO 80632.
³AGCO Corp., 4205 River Green Parkway, Duluth, GA 30096.

Table 1. Herbicide treatment information for preemergence (PRE) and postemergence (POST) application timings, downy brome density and height, and tall fescue height and leaf number at the time of application.

Table 2. Tall fescue response, weed control, seed and forage yield with weed management systems for seed production at the nongrazed site established in fall, 2000 at Novelty. Data were averaged over 2002, 2004, and 2005.^a

^aAbbreviations: BROTE, downy brome; DAT, days after treatment.

Table 3. Tall fescue response, weed control, seed and forage yield with weed management systems for seed production at the grazed site established in fall, 2002 at Novelty. Data were averaged over 2004 and 2005. ^a