Abstract:

Four Midwestern oat varieties (A. C. Assiniboia, Bay, Riser, and Belle) used as cover crops were evaluated for their ability to suppress weed growth on an organic farm near Osceola, Wisconsin. Timing of mowing operations, mower type, and mowing height used on the cover crops were also evaluated. Timing of mowing operations had little effect on cover crop performance relating to weed suppression. Mower type choice made no difference on cover crop performance for weed suppression. Mowing height choice was also shown to not cause significant differences in weed growth. No differences between the oat varieties were seen for weed suppression on a consistent basis. However, the four oat varieties almost always suppressed weed growth more than plots without oats.

Introduction

Organic farmers in the Midwest use oat cover crops as a component of their weed management program (Sustainable Agriculture Network, 1998). Little information is available about which oat varieties are more useful as cover crops (Schonbeck et al. 1991).

Foxtail species are damaging to crops and comprise a major portion of the weed populations in agronomic systems in the upper Midwest (Forcella et al. 1992). Control of foxtail species is difficult and elimination is virtually impossible. All Midwest farmers are impacted by this group of weeds, but the organic farmers have fewer options for control. Costs of foxtail species include yield losses and the costs of controlling them (Swinton et al. 1994). Oat variety choice could impact weed-related damages and losses of revenue.

The oat project at the University of Minnesota has tested a number of oat lines that have demonstrated good competitiveness against foxtail weeds at an agricultural experiment station in Morris, Minnesota. This experiment evaluated four varieties of oats for their ability to suppress weed growth under different management practices to determine which if any varieties are best suited as a cover crop for organic farmers.

Methods and Materials

An on-farm study was conducted during the summer of 2001 to evaluate four oat varieties for their ability to suppress weeds. This study was conducted at Common Harvest Organic Farm, a CSA farm near Osceola, Wisconsin. On this farm, cover crops are composed oats, peas, and vetch and are mowed periodically to prevent weeds from seeding and to force the oats to produce more tillers.

Trials were held on three different fields to study the effects of different management practices on the oat cover crops as well as the weeds. In the first field, the oats were mowed three to five days earlier than normal, at the normal time, and three to five days later than normal. The cover crop is normally mowed when the growing point of the main stem of the oats is between ground level and one inch above ground. In the second field the type of mower used was evaluated. And in the third field the height of mowing was evaluated. In all three cases, oats were used as the solo cover crop species.

The oat varieties used in this study were A. C. Assiniboia, Bay, Riser, and Belle. A. C. Assiniboia is a grain variety from Canada that has wide leaves and an upright growth habit. It has shown good foxtail suppression in variety trials at the Morris experiment station. Bay is a Wisconsin variety often used as forage. It is characterized as a short stocky plant with wide leaves and an upright growth habit. Riser is a fast growing variety from South Dakota with thin stems and an upright growth habit. Belle, like A. C. Assiniboia, has wide leaves and an upright growth habit. It is a grain variety from Wisconsin. In the first two fields plots with no oats planted were used as a control.

Split plot and strip plot experimental designs were used in this study. In the first two fields the split plot design was used. Main plots were assigned to mowing treatments while oat varieties including the control treatment were assigned to the subplots. Three replications were used. A strip plot design was used in the third field with mowing treatments in main plots and the four oat varieties in subplots that were not randomized. There were two replications used in this field.

Planting, mowing and data collection were done in each field as needed. Oats were planted in the first two fields on April 27. In the third field the oats were planted on May 18. Seeds were planted in a freshly tilled seedbed with a six-foot wide seed drill with six-inch rows. Planting rates were two bushels per acre or sixty-five pounds per acre. Plot sizes were ten feet long by six feet in the first field and ten by ten in the second field. One variety was planted at a time and the drill was cleaned out by hand between varieties. Oat and weed stands were measured on May 25th in the first two fields and on June 12th in the third field. Stand measurements for oats were done once in each plot by counting plants using a one square foot frame and converting to plants per acre. Broadleaf and grassy weeds were measured separately in this study to help determine foxtail suppression. It was observed in all three fields that the grassy weed population was composed of over 90% green foxtail. The dominant broadleaf weeds in all three fields were common lambsquarters, Pennsylvania smartweed, and redroot pigweed. In the first two fields, from June 2nd to June 8th, oat and weed growth was measured before the first mowing operation. Oat height and tillering as well as weed height was measured per plot. The first mowing operations in the first field were done with a flail chopper and occurred between June 9th and June 16th. In the second field a flail chopper and rotary mower were used for main plot treatments, the first mowing was done on June 12th. In the third field, the first mowing date was June 26th. It was done with a flail chopper at three different heights. The three mowing heights were three inches, six inches, and no mowing. Samples were collected in the first and second field just prior to the first mowing. All of the above ground plant materials from one square foot per plot were removed and separated between oats, grassy weeds, and broadleaf weeds. These plant materials were dried at 95°F in a wind oven for three days and measured on a digital scale. Growth measurements for plant height for oats and weeds were done just before the second mowing in the first and second fields. The second mowing was done between June 23rd and June 30th in the first field, on June 26th in the second field. In the third field the second mowing was done July 7th. The final above ground biomass samples were collected from July 7th to July 20th in the first field, and on July 21st for the second and third fields.

Results

Data collected from the first field indicates main plot treatments had little effect while sub-plot treatments had more detectable differences . Main plot treatments were early, normal, and late mowing dates. Duncan's multiple range test was used to evaluate differences between treatment means in this field. Very few differences were seen in weed and crop growth and development between main plot treatments. Between the first and second mowing operations, there were detectable differences between main plot treatments for weed height. The plots mowed first had shorter weeds. This was expected because plots mowed first were mowed shorter than the others. The data also shows that there was more broadleaf pressure in the third main plot treatment. This was due to unevenness of weed populations within the field. There was not a subsequent difference in broadleaf growth demonstrated between main plots. Sub-plot treatments were oat variety. They were the four oat varieties and no oats. A. C. Assiniboia and Belle showed more tillering before mowing. At the same time Bay was the shortest variety. Before the first mowing, all varieties suppressed grassy weeds equally but only Belle was different than plots with no oats. Likewise, Belle and Riser were better than no oats at suppressing broadleaf weeds but were not better than the other varieties. After the second mowing there were no differences between varieties for grassy weed suppression but all were better than no oats. For broadleaf weed suppression, only A. C. Assiniboia was grouped with plots containing no oats.

In the second field main plot treatments had little effect and sub-plot treatments were more differences. Main plot treatments were type of mower. Before the first mowing, grassy weeds in the rotary mowed plots were taller and had more biomass compared to the flail chopped plots. This cannot be attributed to mowing type but rather some difference in microenvironments such as nutrient levels. Oat variety treatments caused more detectable differences. Unevenness in both grassy and broadleaf weed populations complicate any conclusions about weed suppression that can be drawn from this data about variety effects. As was seen in the first field, A. C. Assiniboia had more tillering than the other varieties before the first mowing. Also, before the first mowing there were no differences between varieties for grassy weed suppression but Bay was less effective than the other varieties at suppressing broadleaf weeds (Table 4). After the second mowing, there were no differences between varieties for weed growth suppression. However, Riser was shown to be no better than no oats at suppressing grassy weeds while A. C. Assiniboia was no better than no oats at suppressing broadleaf weed growth.

In the third field there were no differences between main plot treatments for weed suppression. There were more grasses in the plots mowed at three inches high than in the non-mowed plots. Conversely, there were fewer broadleaf weeds present in the plots mowed at three inches compared to the non-mowed plots. Regardless, there were no differences between main plot treatments for suppression of weed growth.

Sub-plot treatment effects on weed suppression were also complicated by unevenness in weed populations. Plots planted with A. C. Assiniboia had more grassy weeds than Bay and Belle. Likewise, Plots with Bay had more broadleaf weeds than the other varieties. Belle had the least grassy weed biomass while Riser had the most. Also, Belle and A. C. Assinibioa had less broadleaf biomass and Bay had the most.

The difference between oat varieties is well demonstrated in this field. Oat growth is ranked with Riser being the tallest, followed by Belle, then A. C. Assiniboia, and Bay was the shortest. A. C. Assinibioa and Riser had the more tillering early on. All four varieties were equal when grass and broadleaf weed heights were compared on June 23rd.

Discussion

The results of this study show that there were no patterns of weed growth suppression as a result of the choice of both cover crop management and oat variety. However, in most cases, plots with oats planted had significantly less weed growth compared to plots without any cover crop planted. Although generally there were no differences between oat varieties for weed suppression, there were measurable differences between the oat varieties themselves. Oat growth rates and tillering were markedly different. These differences could cause differences in weed suppression and may be measurable under a different experimental design or different environmental circumstances. This study was not able to quantify variety by site or variety by year interactions.

Literature Cited

Forcella, F. Wilson, R. G., Renner, K. A., Dekker, J., Harvey, R., Alm, D., Buhler, D., and Cardina, J. 1992. Weed Seedbanks of the U. S. Corn Belt: Magnitude, Variation, Emergence, and Application. Weed Science 40:63-644.

Schonbeck, M., Browne, J., Deziel, G., and De Gregorio, R. 1991. Comparison of Weed biomass and Flora in Four Cover Crops and a Subsequent Lettuce Crop on Three New England Organic Farms. Biological Agriculture and Horticulture. 8:123-143.

Sustainable Agriculture Network. 1998. Managing Cover Crops Profitably. Sustainable Agriculture Publications. Burlington, Vermont.

Swinton, S., Buhler, D., Forcella, F., Gunsolus, J., and King, R. 1994. Estimation of Crop Yield Loss Due to Interference by Multiple Weed Species. Weed Science 42:103-109.