A. HopkinsInstitute of Grassland and Environmental Research, North Wyke, Okehampton, Devon, EX20 2SB, United Kingdom
Introduction
Materials and methods
Results
Conclusions
References
Total nutrient supplies (fertilizers, excretal returns, biologically fixed nitrogen etc.) to grass/white clover swards commonly exceed the amounts removed from the system in the form of milk, meat and livestock by-products. Loss occurs through a variety of pathways, though leaching is undoubtedly a major factor affecting the supply of nitrogen and potassium. Ironically, swards containing white clover may release large amounts of N through winter leaching (Watson et al., 1992) yet fail to produce sufficient herbage to meet the demands for early season grazing or first-cut silage unless they receive fertilizer N in the spring (Morrison et al., 1983). Sward productivity and responses to available N may be affected by supplies of potassium and phosphorus, yet little of these elements are 'removed' from the system in the form of animal production (approximately 20 and 10 kg/ha, respectively, in a moderately intensive grassland system, averaged over a whole farm). The need to conserve nutrients within the soil-plant-animal system becomes of increasing importance in situations where measures are introduced to protect ground water supplies and water courses, and where increased regard is being taken of sustainable livestock production systems, e.g. on organic farms. Past research on grass/white clover swards has focused on the role of the legume as an N-fixing, high feed-value companion to the grass sward. A logical extension is to investigate whether these benefits can be enhanced by including additional companion species.
This paper reports some early results from an experimental programme which includes (i) small-plot trials in which winter-growing annuals, including rye (Secale cereale) are drilled into existing grass/clover swards, and (ii) investigations with deep-rooted herb species which may have the potential to recover nutrients from below the rooting depth of the grass and white clover components.
(i) In the winter-spring of 1991-92 an experiment was conducted on three sites, each having a sward of at least 3 years old and predominantly perennial ryegrass/white clover. Sites represented contrasting soil types. Treatments consisted of (1) control, (2) control sward with 50 kg N/ha in March 1992, (3) Rye (cv. Lovaszpatonai) sown at 180 kg/ha and (4) Italian ryegrass (cv. Tribune) at 15 kg/ha. Treatments 3 and 4 were sown into the existing grass/clover sward in early/mid-October using a Hunter strip-seeder. The treatments were repeated in autumn 1992 on site (C), and on an additional site (D) which included treatments 1-3 and also Westerwolds ryegrass alone and with rye. At site D a split-plot arrangement enabled the imposition of different harvest dates with first harvests commencing on 1 April 1993 or 14 April 1993. Assessments include: dry matter yield and herbage digestibility, N recovery in herbage, proportions of clover, rye, etc., clover stolon density and growing points and changes in soil N status.
(ii) Investigations of grass/white clover/herb associations include a mown plot trial sown in summer 1991 to white clover (cv. Menna) with either perennial ryegrass or meadow fescue, and companion herbs which include chicory (Cichorium intybus, cv. Puna), dandelion (Taraxacum officinale), yarrow (Achillea millefolium), a herb mixture and control. Assessments include dry matter yield and herbage digestibility, proportions of species in herbage harvested and content of N and major minerals. (Analysis of results from this investigation are not yet complete).
Table 1 summarizes herbage yield and N recovery from harvests 1 and 2 at these sites. At site A (poorly drained silty clay loam) establishment and growth of both rye and Italian ryegrass was relatively poor; neither treatments had advantages over the control and on the rye treatment only 9% of the herbage DM at harvest 1 consisted of rye. At site B (well drained stony brown earth over sandstone) the rye established better than Italian ryegrass and herbage yield and N recovery from this treatment at harvest 1 were higher than on the control (though this was partly offset at the second harvest). Rye contributed 49% of the herbage at the first harvest on this site.
Table 2 summarizes results from Site D in 1993 (a well drained coarse loamy brown alluvial soil). All strip-seeded treatments established well. The advantages of early growth and N recovery on the rye treatment were most pronounced at the early harvest date.
Clover stolon density and growing points per m2 were recorded from cores sampled after the third harvest (August 1992). On sites A and B introduction of rye was associated with a reduction in white clover developments. Similar assessments of the effects on clover will be made in 1993 (Table 3).
Table 1. Herbage DM yield (kg ha-1) and N recovery in herbage (kg N ha-1) from sites A and B, first and second harvests 1992
|
|
Control |
+ N50 |
Rye |
Italian ryegrass |
s.e.d |
|
|
Site A |
|
|
|
|
|
|
|
DM yield |
(1) |
3551 |
3403 |
3263 |
3294 |
366 |
|
" " |
(2) |
2096 |
2660 |
2026 |
2273 |
301 |
|
N recovery |
(1) |
88 |
110 |
80 |
86 |
11.8 |
|
" |
(2) |
46 |
59 |
42 |
49 |
6.5 |
|
Site B |
|
|
|
|
|
|
|
DM yield |
(1) |
1946 |
2443 |
3206 |
1501 |
408 |
|
" " |
(2) |
4074 |
4584 |
2957 |
4113 |
300 |
|
N recovery |
(1) |
54 |
77 |
71 |
46 |
13.5 |
|
" |
(2) |
69 |
77 |
49 |
59 |
4.5 |
Harvest dates: Site A (1) 22/4/92 (2) 29/5/92
Site B (1) 14/4/92 (2) 27/5/92
Table 2. Herbage DM and DOM yield (kg ha-1) and N recovery in herbage (kg N ha-1) from Site D, first harvest 1993
|
|
Control |
+ N50 |
Rye |
Westerwolds |
Rye + Westerwolds |
s.e.d |
|
|
DM yield |
(1) |
2129 |
2287 |
2564 |
2158 |
2055 |
** (164.9) |
|
" |
(2) |
3070 |
3078 |
3594 |
2915 |
3002 |
* (255.3) |
|
DOM yield |
(1) |
1494 |
1575 |
1806 |
1492 |
1465 |
* (124.0) |
|
" |
(2) |
2008 |
2011 |
2353 |
1941 |
2003 |
* (177.3) |
|
N recovery |
(1) |
64.5 |
85.2 |
71.7 |
57.8 |
55.8 |
*** (4.47) |
|
" |
(2) |
82.6 |
92.0 |
81.1 |
73.5 |
74.6 |
NS |
(1) harvested 1/4/93
(2) harvested 14/4/93
Table 3. White clover growing points (number m-2) from grass - clover swards sampled August 1992 after third harvest
|
|
Control |
Control + N50 |
Rye |
Italian Ryegrass |
|
Site A |
830 |
760 |
270 |
900 |
|
Site B |
2020 |
2040 |
1590 |
1130 |
Rye has the potential to advance herbage production and N recovery in spring and its introduction into grass-white clover swards may be appropriate in some situations, notably on light textured, free draining soils. However, the possibility of an adverse effect on subsequent clover performance needs further evaluation. Further research is needed to develop guidelines in terms of sowing date and subsequent management.
MORRISON J., DENEHY H. and CHAPMAN P.F. (1983) Possibilities for the strategic use of fertilizer N on white clover - grass swards. In: Corrall AJ. (ed.) Efficient Grassland Farming - Occasional Symposium No. 14. British Grassland Society, pp. 227-231.
WATSON C.J., JORDAN C., TAGGART P.J, LAIDLAW A.S., GARRETT M.K. and STEEN, R.J.W. (1992) The leaky N cycle on grazed grassland. Aspects of Applied Biology. 30. 215-222.
