Previous PageTable Of ContentsNext Page

Nitrogen fixation inputs from lucerne-dominated pastures in the central-east of NSW

G.J. Butler, T.Christian, G.D. Schwenke and D.F. Herridge

NSW Agriculture, Tamworth Centre for Crop Improvement, RMB 944, Tamworth, NSW.

ABSTRACT

In research to improve the nitrogen (N) management of phased lucerne-crop systems in the central-east of NSW, participating farmers and NSW Agriculture advisers and researchers identified the need to quantify the inputs of fixed N by the lucerne-dominated pastures and relate those estimates to soil N accumulation. In the first year of the pasture phase (1999), estimates of total N fixed (shoot + root) ranged between 113-499 kg N/ha at 6 different field sites. The % legume N derived from N2 fixation ranged from 20-91% for lucerne and 48-91% for the annual legumes present in a number of the pastures. Annual shoot dry matter production varied from 5.5 to 12.3 t/ha. Rainfall, lucerne density, % grass composition, soil N fertility and sub-surface soil constraints were all likely factors affecting total N fixed.

KEY WORDS

Nitrogen, N2 fixation, lucerne.

INTRODUCTION

Farmers from the mixed animal production-cropping region of the central-east of NSW were keen to quantify amounts of N fixed during the pasture phase of their systems to enable them to match pasture length to crop N requirements. Total N fixed by legumes is a function of dry matter (DM) produced and the proportion of N in the DM derived from atmospheric N2 (%Ndfa). Research in southern Australia indicates about 20-25 kg of shoot N is fixed per tonne shoot DM produced (1); however, other work at Condobolin in the central-west of NSW suggests the value could be as low as 9 kg N/t DM (2). In a low productivity environment, the uncertainty of fixed N inputs has implications for the overall N management of the system, which in turn can have a major impact on rotation gross margins.

MATERIALS AND METHODS

Six wheat paddocks, undersown with pasture, were selected in December 1998 within a 100km radius of Gilgandra in the central-eastern cropping belt of NSW. Site selection was mainly determined by soil type. Three of the sites were grey vertosols (Roc, Col and Tug), one was a red vertosol (Won), and two were red chromosols (Wes and Tex). Soil pH in the top 1 m ranged from 5.5 to 8.0 across all sites. At 1.0-1.5 m depth, pH remained neutral-alkaline (range 6.2-8.1), except at Col where pH was 4.2 at 1.5 m. Nitrate N (0-1.5 m) was negligible at 5 of the 6 sites; at Col, 40 kg N/ha was present below 1.0 m. Lucerne was the dominant legume species at all sites. Minor legume species were annual clovers and/or medics. Won, and to a lesser extent Tex, also contained phalaris and ryegrass (not measured) in the pasture. Four 1 m2 exclusion cages were placed in each paddock for estimating shoot DM. About 4-6 weeks after major rain events, all pasture foliage within each cage was harvested. The leguminous material (plus phalaris at Won) was dried at 70c for 48 h, ground and analysed for %N and 15N. The %Ndfa and the amount of N2 fixed (kg N/ha) was then calculated using the natural 15N abundance method, as described in Unkovich et al (3). To estimate total N fixed, it was assumed 50% of total plant N was present in the root system (1).

RESULTS AND DISCUSSION

The %Ndfa for lucerne and annual legumes, pasture DM and total N fixed during 1999 are shown in Figs 1and 2. Dry matter cuts were not undertaken at Won and Col in February due to insufficient growth. After initial variations, the %Ndfa remained relatively constant for lucerne at each site (Fig 1a). Values ranged between about 25% for the Col site to about 85% for Won and Tex. In early autumn there was a peak in %Ndfa at 5 of the 6 sites, most likely because of low available nitrate. The Col site had access to deep nitrate and gave the lowest %Ndfa values. Monitoring during 2000 will determine whether %Ndfa remains relatively constant for each site or if there are seasonal fluctuations in plant reliance upon N fixation for growth. The %Ndfa for annual legumes was similar at all sites in October (85%) (Fig 1b), coinciding with peak periods of growth. Pasture DM production varied substantially between sites (5.5-12.3 t/ha), with most of it derived from lucerne (70-100% at all sites except Won at 33%) (Fig 2). There was similar variation in total N fixed, ranging from 113 kg N/ha at Col to 499 kg N/ha at Wes. At 5 of 6 sites, total N fixed was consistent with DM production.

Figure 1. Changes in % Ndfa in (a) lucerne and (b) annual legumes at the 6 sites in 1999.

Figure 2. Dry matter production and total N fixed (shoot + root) for lucerne-based pastures at the 6 sites in 1999.

CONCLUSIONS

The factors most likely determining amounts of N fixed by the 6 lucerne-dominated pastures are outlined in Table 1. Each site was given + to +++ for each factor, which were then summed (final column). For instance, rainfall ranged across the sites from 550 mm (+) to 875 mm (+++), while lucerne density varied between 10 (+) and 25 (+++) plants/m2. The different factors were given equal weighting, which probably resulted in an underestimation of the impact of a high grass component at Won and sub-surface constraints at Col. Notwithstanding the limitations of the rating approach, the final ranking of the sites (final column, Table 1) is similar to the ranking of sites for total N fixed (Fig 2).

Table 1. Factors determining N fixed at each site, mediated through yield and %Ndfa (more +'s is better).

Site

Yield

%Ndfa

Total

 

Rainfall

Density

Low grass/weed density

Lack of sub-surface constraints

Grass/weed
competition for soil nitrate

Low soil nitrate fertility

 

Won

+

+

+

+++

+++

++

11

Wes

+++

+++

+++

++

+

+++

15

Roc

+++

++

+++

+++

+

+

13

Col

+

+++

+++

+

+

+

10

Tug

+++

+

++

+++

++

+

12

Tex

+++

++

++

++

++

+++

14

REFERENCES

1. Peoples, M.B., Bowman, A.M., Gault, R.R., Herridge, D.F., McCallum, M.H., McCormick, K.M., Norton, R.M., Rochester, I.J., Scammell, G.J. and Schwenke, G.D. 2000. Plant and Soil 2000 (In Press).

2. Roesner, E.A., 2000. Proc. 15th Ann Conf. Grassland Soc. NSW, 2000, pp123-4

3. Unkovich, M.J., Pate, J.S. and Sanford, P. 1997. Aust J. Agric. Res. 48, 267-93.

Previous PageTop Of PageNext Page