Previous PageTable Of ContentsNext Page

BUILDING UP SOIL NITROGEN WITH GRAIN LEGUMES

Jeffrey Evans

Agricultural Research Institute, PMB. Wagga Wagga, NSW 2650

Introduction

Nitrogen in the soil can be considered to be in two pools: an organic pool such as humus and plant debris, and a mineral pool containing nitrate and ammonium. Plants extract N from the mineral pool. This pool is rejuvenated by breakdown of organic N (mineralisation) and may be topped up further by addition of N fertilisers. The organic N pool is topped up by addition of plant debris containing fixed N2 (often the residues of legume pastures) or by organic manures.

Presently there is concern about low protein levels in the grain of cereal crops harvested in southern NSW, and low soil N fertility has been advanced as the reason. Reduced grain yield is also likely to occur as a consequence of lower N fertility.

This paper discusses effects of temperate crop legumes on the mineral N and organic N pools. The area sown to temperate crop legumes has increased dramatically (ca. 800% in the last 8 years), and lupin and field pea are the two most popular crop legumes. Therefore, the nitrogen utilisation of these two crops is compared.

Effects On The Mineral N Pool

Crop legumes increase soil mineral N in cereal rotations. Thus, in a long-term rotation involving crop legumes and wheat in alternate years at Tarlee in South Australia, the seven year average mineral N level to 60 cm under continuous wheat was 29 kg N/ha compared with 78 and 44 kg N/ha under lupin/wheat and pea/wheat respectively (J.E. Schultz, pers.comm.). Similarly, under continuous wheat at Esperance in WA the five year mean mineral N level was 14 ppm and 21 ppm under wheat/lupin (I.C. Rowland, pers. comm.). This is also evident from many trials defining the rate of N fertiliser required on continuous wheat to equate wheat yields after crop legumes: the level of N required averages about 40 kg N/ha.

The reasons for higher mineral N levels associated with crop legumes are twofold. Firstly, consider a period of two years. In year one portion of the soil organic N will mineralise to nitrate and ammonium. Research has shown that a crop of wheat sown in year one will use more of the mineral N available than a legume crop. Therefore in year two a wheat crop sown after wheat will have less carryover mineral N than after a legume. The saving relative to wheat sown in year one is shown for lupin and pea in Table 1 (column SN). The saved amount is variable between sites and seasons, but is commonly positive, and on average is larger after lupin (27 kg N/ha) than after pea (15 kg N/ha). Thus, in two crop sequences WWWW and WWLW (W wheat, L = legume) the fourth wheat crop in the rotation involving a legume receives not only mineral N from organic N mineralisation in year four, but a bonus of carryover mineral N from year three. Secondly, a fourth wheat crop might benefit from mineralisation of fixed N in the residue of previous legumes if the level of fixation was sufficient. Approximately 10-35% of the N in incorporated crop legume tops is used by a succeeding wheat crop by the time of anthesis (Amato and Ladd, 1983). Thus, as a typical lupin crop in the medium rainfall area of the wheat belt may leave 100 kg N/ha in its residue after grain harvest, about 20 kg N/ha could be used by wheat in the following year.

Table 1. Dependence of lupin and field pea crops on N2 fixation(F); contribution of the crops to increasing soil organic nitrogen (ONB); and the uptake of soil mineral N by lupin and field pea compared to wheat at the same site (SN).

Trial

   

ONB (kg N/ha)

SN (kgN/ha)**

 

Lupin

Pea

Lupin

Pea

Lupin

Pea

1

85

nd

138

nd

70

nd

2

64

nd

64

nd

49

nd

3

49

47

34

-11

38

73

4

51

34

38

-2

34

50

5

41

nd

31

nd

45

nd

6

92

81

56

30

22

19

7

97

95

60

39

11

11

8

90

60

126

-2

44

35

9

83

76

98

69

104

105

10

64

35

52

9

58

42

11

97

92

135

96

18

12

12

75

71

38

51

1

2

13*

nd

54

nd

nd

nd

nd

14*

30

65

-11

6

20

5

15*

46

50

22

1

50

28

16*

29

20

-41

-32

18

-24

17*

60

87

12

25

0

20

18

72

60

6

-5

1

-13

19

64

58

4

-13

-13

-32

20

59

73

14

17

-37

-52

21

81

78

71

59

7

2

22

65

nd

18

 

32

nd

23

42

44

-10

-4

22

-5

24

nd

44

nd

8

nd

7

MEAN

65.3

61.2

43.4

17.9

27.0

15.0

At pH (5

69.3

62.3

52.4

21.2

29.3

17.8

* Sites of pH > 6.8

** total wheat N minus soil derived N in the legume crops

Farmers can use the boost to mineral N with cropping grain legumes to increase the yield of succeeding wheat. Sometimes the protein content of the succeeding wheat has also been increased. This will depend, not only on soil N, but on the yield increase achieved and seasonal conditions, as discussed elsewhere in these proceedings. Together with the effect of legumes on reducing cereal disease the boost to N has resulted In yield increases in wheat after crop legumes ranging 11 to 165% compared with yields of wheat after wheat (Evans and Herridge, 1986). However, it is important to understand that these responses do not necessarily confirm an increase in soil organic N as they may be largely due to the saving in mineral N from the legume year. In the absence of N fertiliser addition soil mineral N levels (hence N fertility) will decrease if soil organic N is not maintained.

Effects On The Organic Nitrogen Pool

To increase the amount of N in the soil organic N pool, the amount of N fixed by Rhizobium must exceed the amount of N which will be removed when grain is harvested at crop maturity. This addition we will call the organic N benefit (ONB).

1. Magnitude of ONB

Experiments over the last five years have shown that ONB is a variable quantity (Table 1), with the average ONB for crops of lupin exceeding that for crops of field pea. It should be noted that the values shown assume that all plant residue after grain harvest is retained on site. The burning of residue or its removal by wind could markedly reduce the potential benefit. A negative value of ONB is possible and it means that the crop has exploited soil N to meet its N requirement for seed production.

ONB And The Stability Of The Soil N Pool In Cereal-Legume_Rotations

To avoid declines in soil N when rotating crops, any N benefit derived from a legume needs to equate the losses of N occurring in the rotation, for example in harvesting cereal grain, denitrification and leaching. Although fixed N may offset other crop N losses, a decline in soil N may still eventuate if mineral N is produced to such an extent that the crops in rotation cannot effectively use the amount available. Excess will then be leached beyond plant recovery. Thus, due to the various N gains and losses that characterise a site and rotation system, total soil N tends to an equilibrium level.

1. Long-term rotation study

Change to soil N has been measured in wheat and crop legume rotations. This approach integrates all gains and losses. However, the approach has limitations as it requires many resource-demanding field trials (to account for soil and climatic differences), and such trials need to be maintained perhaps 10-20 years to establish reliable trends in total soil N. Unfortunately few long-term trials have been established to assess soil N change in various grain legume-cereal rotations. I would like to mention two such studies which are providing valuable information. I refer to the ‘Esperance Rotation Trial’ managed by I.C. Rowland, Department of Agriculture, WA; and the ‘Tarlee Rotation Trial’ managed by J.E. Schultz, Department of Agriculture, SA. The two rotations contrast soil types: a gravelly sand of initially very low N fertility located at Esperance; and a more N fertile, hard-setting red-brown earth at Tarlee. The locations are of similar average annual rainfall (494 mm Esperance and 475 mm Tarlee). At Esperance, over a period of 9 years the total soil N in the top 10 cm under continuous cereal declined by 0.004%. By comparison, a rotation of wheat and lupin in alternate years has increased total soil N by 0.015% to reach a level of approximately 0.1%. At Tarlee, after 9 years of continuous wheat total soil N declined by 0.007%,whereas with wheat in rotation (1:1) with lupin, field pea or faba bean, soil total N appears to have been maintained at its 1978 level of about 0.09% (author’s interpretation). On the basis of this limited data set, alternate years of crop legume and cereal may allow equilibration of total N to approximately, and at least, 0.09-0.10% in the surface 10 cm.

2. Prediction of soil N change in rotation

The endeavour of this approach is to use climatic and soil characteristics and decisions of management to predict changes to soil N balance. Yet in its infancy, the procedure has potential to rapidly assess effects on soil N balance due to shifts in locality or composition and frequency of crops in rotation. Towards this goal, analysis of my trials has indicated that the quantity of fixed N in crops of lupin can be predicted from the volume and pattern of rainfall. For example, in Figure 1, historical rainfall records have been used to determine the expected variability and long-term average for N2 fixation in crops of lupin at Wagga Wagga.

Figure 1. The likely seasonal variability and long-term average fixed N for crops of lupin at Wagga Wagga estimated from rainfall variability between 1939-1984.

The predicted average is 130 kg N/ha. At a lower rainfall location, Condobolin in central-western NSW, the same procedure gave a mean for fixed N of 76 kg N/ha. Reducing these values by the expected average amounts

N in the harvested grain of lupin and its succeeding wheat crop estimates the expected crop effects on soil N balance in long-term rotations (assuming retention of legume and cereal stubbles). At Wagga Wagga these averages were obtained from the long-term rotation trial of wheat and lupin managed by A.C. Taylor (Department of Agriculture, NSW), and were 52 kg N/ha (lupin grain) and 70 kg N/ha (wheat grain). Thus the total exported grain N at Wagga Wagga averages 70+52 = 112 kg N/ha in a lupin-wheat (1:1) rotation. Against an input of 130 kg N/ha from N2 fixation the predicted change to soil N averages 130-112 = 18 kg N/ha each lupin-wheat cycle. Although the prediction suggests a slight benefit t.. soil N, under normal farm practice some legume residue is likely to be removed by wind or grazing; wheat stubble is often burnt, resulting in N loss; and lupin may be sown later than the earliest significant rainfall, reducing fixed N (see later). Therefore I would suggest that in a lupin-wheat (1:1) rotation at Wagga Wagga, with retention of most of the legume stubbles, the fixed N is capable of maintaining the soil N exploited by wheat. The likely total N equilibrium level for crop legume:cereal rotation at Wagga Wagga has yet to be measured or predicted. Clearly, the wider rotation LWW is likely to result in slowly declining soil fertility. Furthermore, and in acid soil, as field pea is less beneficial to increasing the soil organic N pool than lupin (ONB, Table 1), it is feasible that a field pea-wheat (1:1) rotation under these conditions will also not maintain soil N.

Factors Affecting The Magnitude Of ONB

From Figure 2 it can be seen that the magnitude of ONB is closely related to the amount of N fixed by legume crops grown at different sites.

Figure 2. Relation between the amount of N fixed by crops of lupin (a) and field pea (.) and the derived benefit of the crops to increasing soil organic nitrogen (ONB).

The amount of N fixed in turn is dependent on the total amount of dry matter (DM) produced by a crop together with the percentage of crop N derived from rhizobial N fixation (F) (the remaining crop N being derived from soil

N). It follows that increasing DM production and the level of reliance of legumes on N fixation for their source of N, will together increase ONB (Figure 3). influences on DM production and % N fixation (F) are therefore important concerns in regard to the level of N benefit, especially if they are manageable. These will be considered.

Figure 3. Relation between the derived benefit of lupin (o) and field pea (.) crops to soil N increase (ONB) and the reliance of the crops on fixed N2 (P).

(i) Influences on Dry Matter Production

The amount of DM produced by crop legumes is substantially influenced by the volume of rainfall (water availability). Lupin shows more sensitivity to this, possibly due to its longer period of growth and potential for greater OM production than field pea, except in neutral to alkaline soils. However, management decisions may influence DM production. Thus, the growth of legume crops may suffer the competition of weeds, the invasion of insects and diseases, the deficiency of essential nutrients, the effects of soil acidity and cold temperatures.

(a) Cold Temperatures: Lupin is not tolerant of cool temperatures. The longer sowing is delayed into winter the more likely crop growth will be reduced, and the later sown crops also appear to be less reliant on rhizobial N for growth. The result is a reduction of ONB. This can be seen in two recent trials at Wagga Wagga and Condobolin (Table 2).

Table 2. Effect of delayed sowing on the dependence cf lupin crops on N fixation (F) and its result on the contribution of lupin to increasing soil organic nitrogen (ONB)

Site

Sowing Date

F (%)

ONB (kg N/ha)

Wagga

6/5

6/5

65

49

64

34

Condobolin

5/5

65

18

 

16/5

42

-10

(b) Soil Acidity: Management should also avoid cropping lupins in neutral or alkaline soil. Thus, on the grey-black soils around Horsham (pH = 7.6, CaCl2) lupin produced only 36% the amount of DM as pea, and on sandy loam (pH = 6.8) at Walpeup the relative production was 74% of pea. The productivity (DM) of field pea appears to be similar to lupin at pH 4.8 but relatively less in strongly acid soil of pH 4.5 and lower.

(ii) Influences on % N2 Fixation (F)

(a) Dry Matter: In lupin, but not in field pea, larger crops tended to be more reliant on N2 fixation; and therefore tend to higher N returns (Figure 3).

(b) Soil Nitrate: Except in neutral to alkaline soil the level of reliance of field pea crops on fixed N2 is often exceeded by lupin crops (Table 1). One of the reasons appears to be greater sensitivity of field pea N fixation to soil mineral N. Soil nitrate and ammonium inhibit nodule formation and the N -fixing activity of nodules. In Figure 4 it can be seen that increasing amounts of plant available soil N reduces the value of F, thus reducing ONB (Figure 3); and for similar levels of available N lupin commonly achieves higher reliance on fixed N. Thus, low levels of mineral N in the legume year will be beneficial to both legume crops, and especially for field pea crops, in regard to increasing ONB. The use of a grain legume after 2-3 non-legume crops should result in increased reliance of legumes on fixed N A further strategy might be to incorporate cereal stubbles to immobilise mineral N, though the effectiveness of this has yet to be tested. Additionally, the sensitivity of N2 fixing symbiosis to soil nitrate may be reduced genetically. This remains to be achieved.

Figure 4. Reduction in reliance of field pea on N fixation due to Increasing availability of soil mineral (Condobolin, 1987).

Lupin - o Field Pea .

(c) Soil Acidity: The reliance of lupin on rhizobial N (F) can also be substantially less than field pea in neutral to alkaline soil, unless soil mineral N is high enough to suppress field pea N fixation. Thus, at the Horsham site described above, the reliance of lupin crop on fixed N2 was 30% compared with 65% for a field pea crop. On a sand ridge (pH = 8.0) at Walpeup the values were 60% (lupin) and 87% (field pea).

Conclusions

The best bet for increasing soil N is likely to be legume pastures of good quality and exceeding in years a ratio of 1:1 with cereal years (other than in very low fertility soil when legume crops are also likely to increase soil N). Later in the cereal phase legume crops will often enable increase in soil mineral N with a benefit to a subsequent cereal crop. If the legume crop is lupin and the soil mildly acid the boost to cereal yield is less likely to involve a decrease in soil organic N. When a target level of soil N has been achieved, intense cropping of lupin and wheat (1:1) could maintain soil N if the target is not excessively high and lupin trash is retained. With field pea, this achievement will require rigid control over the availability of soil mineral N.

Acknowledgements

Much of the data mentioned in this paper was made possible through financial assistance from the Wheat Research Council and the assistance of D. Coventry, B. Walker, J. Mahoney, D. Walscott (Research Agronomists with the Department of Agriculture and Rural Affairs, Victoria), N. Fettell,

E. Armstrong (Research Agronomists with the Department of Agriculture, NSW),

G. O’Connor (Technical Officer), A. Seymour (Field Assistant), and

G. Turner (Experimental Officer, CSIRO Canberra, ACT).

References

1. Amato, M. and Ladd, J.N. (1983). Report to Wheat Research Council.

2. Evans, J. and Herridge, D.F. (1986). In 'Nitrogen Cycling in Temperate Agricultural Systems' (Soil Science Soc., Riverina), 14.

Previous PageTop Of PageNext Page