Direct Seeding in Forage Based Crop Rotations

Heather Loeppky, Adrian Johnston and Manjula Bandara

Agriculture and Agri-Food Canada, Melfort, SK

Introduction

Perennial forages play a vital role in diversifying the Saskatchewan farm economy. They contribute to the improvement of the soil's physical properties (Blackwell et al., 1990), soil organic matter (Campbell et al. 1990) and reduce soil erosion (Stinner and House, 1989). Legumes, such as alfalfa, also improve soil fertility by adding nitrogen. In addition, the year-round crop stand of these perennial crops may prevent the development of soil salinity (Forage Crops, 1991). Snow trapping, and ultimately favourable moisture conditions, are another benefit of growing forages. One unique attribute of forages is their ability to suppress weeds (Harvey and McNevin, 1990). The frequent cutting and heavy growth of forages reduces both the vigour and seed production of weeds. Even alfalfa seed crops have been found to decrease population of annual weeds. Moreover, one of the most important benefits of forage for farmers is higher yield of grain crops following forages in the rotation (Baddarudin and Meyer, 1990). In north-eastern Saskatchewan, forage seed production, dehydrated alfalfa, alfalfa pellets and sun-cure hay provide an opportunity for producers to tap into international markets while improving the fertility and physical condition of their soils. This is particularly important on Gray-wooded soils, which are inherently low in organic matter, and have surface and sub-surface soil structure that poses unique challenges for crop production.

Traditionally, intensive tillage has been used in both the establishment and termination phases of the forage based rotation. Forage crops have been seeded into soil that has been tilled a number of times to control weeds and to prepare a fine, firm seedbed. These conventional seedbed preparation techniques dry the seedbed and increase the risk of erosion, organic matter loss, and therefore, loss of soil productivity. Over tilling can also lead to crusting which result in poor seedling emergence and ultimately poor crop stands. Establishment has been recognized as the major constraint in forage production (Waddington and Bittman, 1983) and is critical to the successful seed and hay production in subsequent years. Forage seedlings are especially vulnerable to soil moisture deficits because the small seeds are sown near the soil surface (Sheaffer, 1989). Zero tillage has been demonstrated as a feasible alternative to conventional tillage (Allen and Entz, 1994). Numerous studies have shown zero tillage to give similar plant populations to conventional tillage (Wolf et al., 1985). Direct seeding of forage crops into stubble is beneficial in terms of moisture conservation (Allen and Entz, 1994), and alleviating problems associated with conventional tillage such as soil erosion. The stubble can also be used for snow trapping to improve spring moisture conditions.

Weed control is essential for successful forage establishment because the seedlings are weak and non-competitive. Weeds compete with crops for light, water, and nutrients and can reduce the vigor and density of the forage seedlings in the establishment and subsequent years. Weeds also reduce the quality and value of both forage and seed crops. Alfalfa seed yield can also be reduced when weeds compete for the attention of pollinating insects. The weeds can also interfere with harvesting and seed cleaning operations. Without tillage, weed control techniques are limited to herbicides (which are limited in some forage crops) and factors that improve the crop's competitive ability. Although early spring seeding is often preferred, the crop can also be established in late fall, just before freeze up. Late fall seeding helps to ensure germination early in the spring so the seedlings can take advantage of spring moisture. Another strategy is to delay seeding until most of the weeds have germinated then control these weeds with tillage or herbicides just before seeding.

Terminating an alfalfa stand has also traditionally been quite tillage intensive, starting after the first cut in July followed by several tillage operations through the summer, fall and early spring before an annual crop is seeded. Recently herbicides have been identified that control alfalfa effectively (Button 1991), but this has led to questions regarding weed control (particularly dandelions) and crop selection following alfalfa. Two studies were initiated in NE Saskatchewan to address the issues of reducing or eliminating tillage in both establishment and termination of forages in crop rotations.

A. Forage Establishment

Objectives:

Northeastern Saskatchewan is one of the major forage producing areas of Canada. Despite this fact, no studies have reported on direct seeding and forage establishment in this region. To address the issue, this project was conducted to evaluate the impact of reducing or eliminating tillage in seedbed preparation and seeding date on: a) crop establishment, b) forage and seed production, and c) weed control.

Three seeding dates, late fall, early spring (May) and late spring (June) were included to determine whether seeding date influences crop production under these tillage systems.

Materials and Methods

Field experiments were established at two locations: a sandy site near Gronlid and a clay site near Ridgedale in northeastern Saskatchewan, beginning in the fall of 1993. Both sites were in the gray soil zone characterized by low organic matter. Treatments consisted of three seeding dates (October 1993, May 1994, June 1994) and different tillage operations ranging from zero to four tillage operations applied in spring and/or fall (see Table 1). Treatments were arranged in a randomized block design with four replicates at each site. Tillage (conventional tillage CT) consisted of discing and/or harrowing prior to seeding with the exception of zero-till (ZT) where there was no pre-seeding tillage. If only one tillage operation was performed, the plots were cultivated with a tandem disc followed by a diamond harrow and were packed prior to seeding. Plots requiring multiple tillage treatments were first prepared using a cultivator mounted with a tine harrow, followed by 1 to 3 passes with a diamond harrow and soil packing before seeding.

Fleet meadow bromegrass and Heinrichs alfalfa were planted into plots of 4.3 x 7.7m at a rate of 8 and 2kg ha^-1 respectively, at a depth of 1cm with a ConservPak air seeder with knife openers spaced at 23cm. The alfalfa was inoculated before seeding. At seeding, the fertilizer mixture of 11-51-0 was banded at a rate of 30kg P ha-1 at 5 cm below and 2.5 cm to the side of the seed. Plots were subsequently broadcast fertilized according to soil test recommendations. The same herbicides were applied to all plots. Herbicide selection was based on the weed species present. Although there were differences in the number of weeds present in certain treatments these differences did not warrant changes in the use of herbicides. Herbicides and insecticides were applied across all treatments according to label instructions. Plant stands were determined by counting all plants (crop and weeds) in each of four 0.25 m² quadrates each year just before herbicides were applied and again 6 weeks after application. Forage plots were harvested twice each year at the late bud to 1/10 bloom stage. Leafcutting bees were placed in alfalfa stands to promote seed production. Seed was harvested from adjacent plots at maturity in 1995 and 1996.

Results and Discussion

In the establishment year, alfalfa plant densities were higher under conventional tillage than under zero-till at Ridgedale, but similar at Gronlid (data not shown). Conventional tillage also resulted in higher bromegrass plant densities at both sites. In contrast with plant density, alfalfa and bromegrass forage yields in the establishment year were higher under zero-till than conventional tillage at both locations, though the difference in bromegrass forage yield at Gronlid was not significant (Tables 1 &2). This indicated that while plant establishment under conventional tillage may be superior over zero-till, the benefits associated with zero-till such as higher soil moisture (Allen and Entz, 1994), may enhance forage production and overcome the initial reduction in plant population. In the case of alfalfa, reduced plant density and enhanced vegetative growth on zero-till on one hand, and higher plant population but lower vegetative growth on conventional till on the other, resulted in similar alfalfa seed yields in both the second and the third production years at both sites (Table 3). The location x tillage operation interaction for bromegrass seed yield was significant in the third year, 1996. At Ridgedale, bromegrass seed yield in 1996 was higher on conventional tillage but at Gronlid bromegrass seed yield in 1996 was higher on zero-till. Thus the effect of tillage system depends on site-specific characteristics, which are influenced by environmental factors.

Between 1995 and 1996, alfalfa seed yield increased under both tillage systems at both locations. The effect on bromegrass seed yield was different in that at Gronlid bromegrass seed yield decreased regardless of the tillage system, while at Ridgedale, bromegrass seed yield increased under conventional tillage but decreased under zero-till.

The overall effect of these increases and decreases on bromegrass between the two production years was that, zero-till resulted in reduced seed yield at Gronlid, but increased seed yield at Ridgedale. Tillage system had no effect on alfalfa seed yields at either location.

Spring seeding resulted in higher alfalfa and bromegrass plant densities regardless of tillage system at both locations. This effect was also observed in subsequent years. However, this superiority was not reflected in forage yields. Fall-seeded alfalfa resulted in higher forage yield under conventional tillage at both sites, but spring-seeded out-yielded fall seeding under zero-tillage at Gronlid, though the difference was not significant. Variations in the effect of date of seeding were also observed within spring seeded treatments. At both sites, May-seeded alfalfa resulted in higher forage yields than June-seeded alfalfa regardless of tillage system in the establishment year, exhibiting fluctuations in subsequent years at Gronlid under zero-tillage. However, June seeding resulted in higher forage yield under conventional tillage at Gronlid, although the only significant difference was exhibited in the total alfalfa yield for 1996. At Ridgedale, June seeding was consistently superior over May seeding in both conventional and zero-till systems. Date of seeding in spring had no effect on alfalfa seed yield at both locations regardless of the tillage system. Fluctuations in the effect of date of seeding within spring seeding were also observed in bromegrass seed yield.

Summary

It can be concluded from this study that while conventional tillage enhanced plant population in the establishment year, this may not necessarily result in increased forage and seed yield of alfalfa and bromegrass. This indicates that reduced tillage, which is more beneficial in the long run, can replace conventional means of sward establishment. However, depending on site-specific characteristics, success may vary. Under zero-till, fall seeding may enhance establishment year forage yield but not necessarily so in subsequent years as this seems to depend on prevailing soil and climatic conditions.

Table 1. The effects of tillage and seeding date on alfalfa forage yield.

ns, *, and ** denote non-significant at p=0.05, significant at p=0.05 and 0.01, respectively.

Table 2. The effects of tillage and seeding date on bromegrass forage yield.

ns, *, and ** denote non-significant at p=0.05, significant at p=0.05 and 0.01, respectively.

Table 3. The effects of tillage and seeding date on alfalfa and meadow bromegrass seed yield.

ns, *, and ** denote non-significant at p=0.05, significant at p=0.05 and 0.01, respectively.

B. Alfalfa Termination

Objectives:

To evaluate the effect of time and method of alfalfa termination on alfalfa re-growth and weed control, and performance of succeeding crop established in the year following termination of alfalfa.

Materials and Methods

Field studies were conducted on two sites in northeast Saskatchewan (Tisdale- termination in 1993, cropped in 1994; Gronlid - termination in 1996, cropped in 1997). Alfalfa fields were selected based on the infestation of dandelion, which is typical of alfalfa stands at the time of breaking. Other perennial weed problems such as quackgrass were avoided. Termination treatments include different chemical treatments alone, tillage alone, or herbicide followed by fall tillage, performed at three different times (see below).

Treatments

a) Method of termination: Approximate Cost ($/acre)

1. Tillage alone * $15-18

2. Roundup (glyphosate) (2.5 L h-1) + 2,4-D-amine (1.6 L h-1) $12-15

3. Roundup (glyphosate) (2.5 L h-1) + 2,4-D-amine (1.6 L h-1) + fall till $15-18

4. Lontrel (clopyralid ) (0.37 L h-1) + 2,4-D-ester (0.99 L h-1 ) $12

5. Lontrel (clopyralid) (0.37 L h-1) + 2,4-D-ester (0.99 L h-1) + fall till $15

* tillage termination involved plowing, discing, cultivating in fall and/or spring and harrowing before seeding at Tisdale. At the Gronlid site plowing was replaced by discing.

b) Time of termination

1. After first cut

2. After second cut

3. After seed harvest (treatment 1 applied in fall, treatments 2, 3,4 and 5 applied the following spring)

Each treatment plot was 6m x 15m in size. The effect of different termination methods on subsequent crop performance was evaluated by establishing four different field crops, wheat cv. Katepwa, barley cv. Harrington, canola cv. Parkland and pea cv. Princess, across the treatment plots, in the following spring. Treatments were arranged in a randomized block design with 4 replicates. Data collections include weed density (including volunteer alfalfa), stand density and seed yields of wheat, barley, canola and peas.

Results and Discussion

Although there were some similarities in response to these treatments at the two sites, there were some very significant differences. At Tisdale in 1995 reducing or eliminating tillage had few detrimental effects on subsequent crop performance. At Gronlid in 1997, however, eliminating tillage from the termination process reduced yield of all crops significantly.

1. Study at Tisdale in 1993/94

The effects of timing and method of termination treatment on plant density and crop yield were species-specific. In wheat, there were significant differences in plant density, however, there was no obvious pattern to the differences and grain yield was not affected. The lowest seed yield (1816 kg/ha compared to the mean of 2580 kg/ha) was harvested from plots treated in the spring with Lontrel + 2,4-D. In barley, this treatment resulted in significant yield reductions (Table 4). Competition from dandelions can sometimes interfere with crop establishment and influence yield, however, dandelion control was particularly good in plots treated with Lontrel + 2,4-D in the spring compared to other treatments (Table 6). Pea yields were also significantly lower after spring application of Lontrel + 2,4-D regardless of whether the plots were tilled after application or not (Table 5).

Table 4. The effects of timing and method of termination on barley yield (kg ha^-1). Tisdale, 1995.

Table 5. The effects of timing and method of termination on pea yield (kg ha^-1). Tisdale, 1995.

Table 6. The effects of timing and method of termination on dandelion density (#m^-2). Tisdale, 1995.

2. Study at Gronlid in 1996/97

At Gronlid, all annual crops were affected by method of breaking . In general, termination without tillage reduced plant density and yield (Table 7-10). In wheat, plots treated with Roundup or Lontrel +2,4-D without tillage, yielded only 59% -66%, respectively, as much grain as plots treated with the same herbicide with tillage. Barley responded in a similar fashion, with termination with herbicides alone producing approximately 60% of the grain yield of herbicide treatments combined with tillage. The herbicides alone treatments also reduced canola yield by 70% when treated with Roundup + 2,4-D, and 50% when treated with Lontrel + 2,4-D, compared to the same herbicides with tillage. Combinations of herbicides with fall tillage, however, produced yields that were as high or higher than tillage alone (Table 9). Peas followed a similar pattern to canola with both method and time of termination affecting plant density (data not shown) and yield (Table 10).

Termination with Roundup alone resulted in particularly high numbers of dandelion, especially when termination occurred after first cut (Table 11). This treatment provided effective control of alfalfa, which provided the remaining dandelions with a relatively competition free environment during the termination season. As a result, the dandelion population increased over the period between termination and seeding and this increase may explain why pea yields are particularly low following termination with Roundup + 2,4-D after first cut.

The Lontrel +2,4-D termination treatment resulted in a decline in yield the magnitude of which increased as the time between termination and seeding date decreased. This was expected because Lontrel residues are known to damage peas. Tillage after herbicide treatment improved the safety of this treatment, but significant yield losses were still encountered.

Summary

Timing and method of termination had a significant influence on both alfalfa and dandelion re-growth, plant density and seed yield of succeeding wheat, barley, canola and peas. The effects of termination treatments were site-specific. Herbicide treatments without tillage were equally effective as tilled treatments in controlling alfalfa and dandelion re-growth and also maintaining the plant population and seed yields of subsequent crops grown at the Tisdale site. At Gronlid, however, the herbicides without tillage produced poor control, particularly of dandelion re-growth, poor plant density resulting in reduced seed yields, compared to tilled treatments.

Table 7. The effects of timing and method of termination on wheat yield (kg ha^-1). Gronlid 1997.

Table 8. The effects of timing and method of termination on barley yield (kg ha^-1). Gronlid 1997.

Table 9. The effects of timing and method of termination on canola yield (kg ha^-1). Gronlid 1997.

Table 10. The effects of timing and method of termination on pea yield (kg ha^-1). Gronlid 1997.

Table 11. Effect of time and method of termination on alfalfa and dandelion control in subsequent cropping at Gronlid in 1997.

Conclusions

• reducing tillage generally had a positive rather than a negative effect on alfalfa and bromegrass forage and seed yield however both significant increases and decreases did occur when tillage was eliminated during seed bed preparation.
• reducing tillage in terminating alfalfa did not have a negative impact on subsequent crop production except when Lontrel + 2,4-D was applied in the spring before seeding peas. Eliminating tillage altogether, however, did have a negative impact on performance of wheat, barley, canola and peas in one of two site years regardless of whether Roundup + 2,4-D or Lontrel + 2,4-D was used to the alfalfa.

References

Allen, C.L., and M.H. Entz. 1994. Zero-tillage establishment of alfalfa and meadow bromegrass as influenced by previous annual grain crop. Can. J. Plant Sci. 74:521-529.

Badaruddin, M., and D.W. Meyer. 1990. Forage legume effects on soil nitrogen and grain yield, and nitrogen nutrition of wheat. Crop Sci. 30: 819-825.

Blackwell, P.S. , T.W. Green, and W.K. Mason. 1990. Responses of biopore channels from roots to compression by vertical stresses. Soil Sci. Soc. Am. J. 54:1088-1091.

Button, Roy. 1991. Chemical Control of Alfalfa. Final Report ADF Project # ADF-D-SC-0966. pp 13.

Campbell, C.A., R.P. Zentner, H.H. Janzen, and K.E. Bowren. 1990. Crop rotation studies on the Canadian prairies. Agriculture Canada Pub. 1841/E.Ottawa. On. pp 133.

Entz , M.H., P.D. Ominski, R. Mohr, A. Schoofs, D. Forster, W.J. Bullied, S. Shirtliffe, and K.C. Bamford. 1997. Forages improve efficiency of prairie cropping systems. The 9th Annual meeting, conference and trade show of the Saskatchewan Soil Conservation Association Proceedings. p. 101-109.

Harvey, R.G., and G.R. McNevin. 1990. Combining cultural practices and herbicides to control wild-proso millet (Panicum miliaceum L.). Weed Technol. 4:433-439.

Sheaffer, C.C. 1989. Legume establishment and harvest management in the U.S.A.. p. 277-289.

Stinner, B.R., and G.J. House. 1989. The search for sustainable agroecosystems. J. Soil Water Conserv. 44:111-116.

Waddington, J. and S. Bittman 1983. Bromegrass and alfalfa establishment with a wheat companion crop in northeastern Saskatchewan. Can. J. Plant Sci. 63:659-668.

Wolf, D.D., Hagwood, E.S. Jr. and M. Leter. 1985. No-till alfalfa establishment as influenced by previous cover crop. Can. J. Plant Sci. 65:609-613.