Defining a role for legumes in future food security
& in the adaptation & mitigation of climate change
Mark Peoples
CSIRO – Canberra, Australia
PLANT INDUSTRY
Why legumes?
Food security & adaptation to future climates
• Leguminosae 20,000 species (cf 10,000 Gramineae & 3,500 Brassicaceae)
• Many hundreds of species used for human &/or animal food globally
• Only 20 species grown commercially over large areas
• Underexploited legumes an untapped pool of genetic diversity for adaptation
Climate change & variability
Combat incursions of new diseases & pests
Provide options to grow more food in hostile or deficient soil environments
+P
-P
+P
-P
+P
-P
Photo: courtesy of Janet Sprent, Dundee, UK
Photo: courtesy of the late Peter Hocking, CSIRO
Source: Herridge et al (2008) Pl. & Soil 311:1-18; Peoples et al (2009) Symbiosis 48:1-17
Why legumes?
√
Food security & adaptation to future climates
•
•
•
•
•
Leguminosae 20,000 species (cf 10,000 Gramineae & 3,500 Brassicaceae)
Many hundreds of species used for human &/or animal food globally
Only 20 species grown commercially over large areas
Underexploited legumes an untapped pool of genetic diversity for adaptation
N content of foliage & seed protein not as affected by e[CO2] as non-legumes
Mitigating climate change
•
√ •
√ •
•
Biological N2 fixation
underpins benefits
Lower fossil energy use than N-fertilized systems
Lower green-house gas emissions than N-fertilized systems
Contribute to C sequestration in soil
Opportunities to replace petroleum products - as a source of feedstock for
biofuels & biorefineries
Biological N2 fixation
underpins benefits
Source: Herridge et al (2008) Pl. & Soil 311:1-18; Peoples et al (2009) Symbiosis 48:1-17,
Rogers et al (2009) Pl. Physiol. 151:1009-1016; Jensen et al (2012) Agron. Sustain. Dev. 32: 329-364.
Stable isotopes of Nitrogen
Element
Isotope
% Atmospheric
abundance
Nitrogen
14N
99.6337
15N
0.3663
Range
Variable
in plants
-2 to +26‰* Plant species
* δ15N (‰) = 1000 x (sample abundance – 0.3663) / (0.3663)
Quantification of N2 fixation requires a measurable difference in 15N content
between atmospheric N2 & available soil N
% legume N fixed = 100 x (15N non-legume – 15N legume) / (15N non-legume)
Where 15N non-legume is assumed = 15N plant-available soil N used by legume
Source: Peoples et al (2009) Am. Soc. Agron. Monograph 52: 125-170
Examples of amounts of N2 fixed measured in farmers’ fields
Goondiwindi
QUEENSLAND
Moree
Walgett
Pulses = 0-211 kg N/ha
Narrabri
Gunnedah
Trangie
NEW SOUTH WALES
Pulses = 63-135 kg N/ha
Pastures = 21-167 kg N/ha
Pulses = 12-183 kg N/ha
VICTORIA
Horsham
Pastures = 13-82 kg N/ha
Pulses = 26-111 kg N/ha
Condobolin
Cootamundra
Sydney
Junee
Wagga Wagga
Canberra
Pastures = 99-238 kg N/ha
Pulses = 59-244 kg N/ha
Rutherglen
Pastures = 2-160 kg N/ha
Pulses = 30-174 kg N/ha
Melbourne
0
50 100
Source: Peoples et al (2001) Plant & Soil 228: 29-41 & unpublished data.
200 Km
Two factors determine the amounts of N2 fixed
%Legume N derived from
atmospheric N2
Amount N fixed = Legume N x %Ndfa
N accumulated
during growth
Legume N = Legume DM x %N
Levels of %Ndfa detected in farmers’ pastures
Legume
number of
pastures
Mean
%Ndfa
Proportion with %Ndfa
<65%
>66%
Subclover
310
76
0.23
0.77
Annual medic
41
73
0.29
0.71
Lucerne (Alfalfa)
116
69
0.33
0.67
White clover
75
64
0.50
0.50
Factors contributing to low %Ndfa values:
• Low numbers or poor effectiveness of rhizobia
• Nutrient deficiencies
• High soil N status
Source: Sanford et al (2010) Aust.J.Agric.Res. 45:165-181; Peoples et al (1998) Aust.J.Agric.Res. 49:459-474;
Riffkin et al (1999) Aust.J.Agric.Res. 50:273-281; Bowman et al (2004) Aust.J.Expl.Agric. 42:439-451
When is there a need to inoculate?
Numbers of rhizobia per gram soil
10000
Critical level
needed for
nodulation
1000
100
Rate of decline
greater in acid soils
10
1
0
2
4
6
8
10
Years since legume last grown
Source: Collated from published Australian data for a range of different legume species & soil types
12
Effect of soil pH on the persistence of soil rhizobia
9 months after removal of a pasture
Rate of lime applied at pasture establishment
nil
1.5
3
6 t/ha
Soil pHCa (0-10cm) in final year of pasture
4.66
4.93
5.23
5.47
Rhizobial numbers per gram soil
9 months after pasture
Rhizobia for medics
Following lucerne
Rhizobia for clovers
Following subclover
12
18
274
2,753
3
185
416
1,412
Source: Roesner et al (2005) Aust. J. Expl. Agric. 45: 247-256
Is there enough rhizobia in the soil?
Field pea rhizobia in soils of southern Australia
Around 30% of 115 fields did not have sufficient rhizobia for peas!
South Australia
(n=48)
New South
Wales
Western Australia
(n=29)
Adelaide
5
Perth
Victoria
(n=37)
Commercial
inoculant strain
100 km
SU303
70 km
V117
V127
V132
%Symbiotic
100%
50%
22%
89%
performance compared to commercial inoculant strain
Source: Drew et al (2012) Crop & Pasture Sci. (in press)
Melbourne
Effect of inoculation on shoot dry matter (DM), grain yield
& N2 fixation : No legume for >10 years (acid soil)
Species
Inoculation
Faba bean
LSD (P<0.05)
Shoot
Grain
(t DM/ha)
(t/ha)
-
5.97
1.75
23
4
+
11.61
2.70
202
17
2.40
0.44
56
2.6
An addition tonne of grain/ha worth A$300/ha
+ 180 kg additional N fixed/ha for only ~
A$7/ha investment in inoculant
Source: Denton & Peoples (unpublished data)
Amount N fixed
(kg N/ha) (kg N/t DM)
Effect of inoculation on shoot dry matter (DM), grain yield
& N2 fixation : No legume for >10 years (acid soil)
Species
Inoculation
Faba bean
Shoot
Grain
(t DM/ha)
(t/ha)
-
5.97
1.75
23
4
+
11.61
2.70
202
17
2.40
0.44
56
2.6
-
7.76
3.50
37
5
+
9.18
3.70
169
18
NS
NS
56
3.3
LSD (P<0.05)
Lupin
LSD (P<0.05)
Amount N fixed
(kg N/ha) (kg N/t DM)
No yield effects, but
130 kg additional N fixed/ha
Source: Denton & Peoples (unpublished data)
Impact of legume productivity
Source: Peoples & Baldock (2001) Aust. J Expl Agric 41: 327-346
Relationship between amounts of shoot N fixed
& shoot dry matter (DM) production
300
subclover
y = 0.56 + 20.2x
R2 = 0.82
Medic
N fixed (kg/ha)
200
150
100
24 kg N/t DM
y = -0.19+24.3*x
R2=0.89
250
200
150
100
50
50
0
0
0
2
4
6
8
10
12
14
0
2
4
Shoot DM (t/ha)
N fixed (kg/ha)
N fixed (kg.ha)
Source: Unkovich et al (2010) Plant & Soil 329: 75-89
250
300
20 kg N/t DM
6
8
10
12
Shoot DM (t/ha)
300
Lucerne, white clover
250
y = 0.15+18.7*x
R2=0.82
200
18 kg N/t DM (lucerne)
150
100
50
0
0
2
4
6
8
Shoot DM (t/ha)
10
12
14
14
Estimates based on legume shoot DM & N
will underestimate the total amounts of N2 fixed
% Total plant N in nodulated roots
Subclover
42%
Annual medics
21-27%
Yellow serradella 30-46%
Vetch
32%
Lucerne
50%
White clover
41% (+ stolons)
Source: summarised in Unkovich et al (2010) Plant & Soil 329: 75-89
Sources of variation in legume biomass (1)
Example of ameliorating P-deficient acid soils
on subclover content & N2 fixation in permanent pastures
Location
Braidwood, NSW
Treatment
Nil
+Lime
+P
+Lime+P
Total herbage
dry matter
(t DM/ha)
2.73
2.72
4.35
6.50
Source: Peoples et al (1995) Soil Biol. Biochem. 27: 663-671
Pasture
content
(%legume)
10
18
44
63
Amount
shoot N fixed
(kg N/ha)
4
6 (+50%)
32 (+700%)
72 (+1100%)
Sources of variation in legume biomass (2)
Effect of legumes species
• Different tolerance to hostile soils
• Better agronomic properties
• Differences in duration of growth – annual vs perennial legumes
Sources of variation in legume biomass – annual vs perennial
Non-legume DM
Sub
Lucerne
Sub
Lucerne
Lucerne
Lucerne
Sub
Sub
Sub
Lucerne
Sub
Lucerne
Subclover
Legume DM
Subclover
Shoot N fixed
(kgN/ha per yr)
Herbage DM production
(t/ha per yr)
Comparison of annual (subclover) & perennial (lucerne) legumes
Year Rainfall (mm)
 1992
 1993
 1994
 1995
L-t Ave
Source: Peoples et al (1998) Aust. J. Agric. Res. 49: 459-474
GSR
359
374
123
359
303
Annual
746
604
406
663
531
Adaptation to future climates - Elevated [CO2]
Further studies needed to quantify impacts on N2 fixation
• Genetic variation in response to e[CO2]
<+10% to >+40% by 5 different field pea varieties cf ambient over 2 years
• Effects of soil type
+20% to +86% for chickpea
+44% to +51% for field pea
+114% to +250% for annual medic
• Effects of pasture composition
Stimulation negated by :
• Low supply of available P
• Elevated temperatures
• What other factors?
+29% for subclover
Source: Lilley et al (2001) New Phytol. 150: 385-395; Lam et al (2012) Crop & Pasture Sci. 63: 53-62;
Fitzgerald et al (2012) Proc. Aust. Agron. Conf.
Quantifying the partitioning & fate of legume N
Faba bean harvest
Above-ground residues
At grain harvest of
following wheat crop
2 kgN/ha
Seed N removed
131 kgN/ha
Taken up by wheat
44 kgN/ha
22 kgN/ha
Soil organic N
20 kgN/ha unaccounted
51 kgN/ha
Below-ground N
(23% total plant N)
Source: Khan (2000) PhD thesis University of Melbourne
4 kgN/ha
43 kgN/ha
Taken up by wheat
Soil organic N
4 kgN/ha unaccounted
Adaptation to future climates – Elevated CO2
Further studies required to quantify impact on soil N-dynamics
• Below-ground partitioning of legume N
e[CO2] often increases nodulation (but not always).
Effects of N rhizodeposition?
• Subsequent availability of legume N
Gross N mineralization rates unaffected by e[CO2] , but N immobilization
can be 30% higher.
In controlled conditions wheat obtained 11% of its N supply from
previous field pea under e[CO2] cf 20% under ambient conditions.
Are these few results representative of other legume systems?
Source: van Groenigen et al (2006) Ecol. Studies 187: 373-391;
Rogers et al (2009) Pl. Physiol. 151: 1009-1016; Armstrong et al (unpublished data)
Total N2O emissions per growing season or year
Category
Number of site-years
of data
(171 in total)
Measured emissions
(kg N2O-N/ha)
Range
Mean
Legume-based pasture
N-fertilized grass pasture
25
19
0.10-4.57
0.30-18.16
1.38
4.49
Crop legumes
N-fertilized crops
46
48
0.03-7.09
0.09-12.67
1.02
2.71
Soil: no legume or added N
33
0.03-4.80
1.20
Source: Jensen et al (2012) Agron. Sustain. Dev. 32: 329-364.
A role for legumes in mitigating climate change?
Further studies required to study N2O losses
• Identify the sources of N2O emissions during legume growth
• Quantify the subsequent losses of N from legume residues
N2O vs losses of other forms of N?
• Determine the likely impact of future climates
Effect of increased microbial immobilization of N under e[CO2]?
Effect of higher temperatures &/or more variable rainfall?
Isotopes of Carbon
Element
Isotope
% Atmospheric
abundance
Carbon
12C
13C
98.98
1.11
Range
Variable
in plants
-20 to -34‰ C3 photosynthesis
-9 to -17‰ C4 photosynthesis
14C
Adaptation to future climates – Variable rainfall
Use of 13C to identify improved water use efficiency
 Discrimination ( ∆13C ) is LESS in species or
varieties which have GREATER transpiration
efficiency.
 Greater transpiration efficiency (lower
may result from:
(1) Lower stomatal conductance,
(2) Greater photosynthetic capacity,
(3) Some combination of these.
∆13C)
Transpiration efficiency (g/kg)
 C3 plant species discriminate against 13CO2
during photosynthesis
6.5
6.0
5.5
5.0
4.5
4.0
21
22
23
Discrimination (103 x ∆)
Source: Farquhar & Richards (1984) Aust. J. Pl. Physiol. 11: 539-552
Identifying genetic differences in plant water-use efficiency
13C
discrimination
As a selection tool
WUE by semi-leafless field pea superior to conventional genotypes.
Drought adaptation in peanut & cowpea.
• Comparative measures of WUE
Between seasons & between species.
However, some legume studies have failed to find correlations with water use
Further studies required
• Explain apparent inconsistencies
Intra- & inter-specific comparisons.
Define influence of environment, management & sampling protocols.
Source: Armstrong et al (1994) Aust. J. Pl. Physiol. 21:517-532; Wright et al (1994) Crop Sci. 34: 92-97;
Hall et al (1994) Field Crops Res. 36: 125-131; Turner et al (2007) J. Integ. Pl. Biol. 49: 1478-1483
A role for legumes in mitigating climate change?
Following the fate of legume organic C – contributions to soil C
• Exploiting differences in 13C content of C3 & C4 species
Long-term lucerne (alfalfa)-maize rotations - >50% C in lucerne
contributed to soil organic C cf <15% of C from maize residues
• 14C methodologies
Quantify the extent of losses of C from legume residues & flow of C
through different soil organic matter pools
Further studies required
• Quantify the effect of different strategies on rate of change of soil C stock
Legumes or species combinations x management
Influence of environment
Source: Gregorich et al (2001) Can. J. Soil Sci. 18:21-31;
Bhupinderpal-Singh et al (2005) Eur. J. Soil Sci. 56: 329-341
Response to future climates?
The fate of organic C in e[CO2] environments
• Meta-analysis of soil C affects under e[CO2] : not legume specific
Total soil C increase by 4% on average, but dependent upon N supply
nil difference from ambient @ inputs <30 kgN/ha per year
+4% @ inputs 30-150 kgN/ha per year
+8% @ inputs >150 kgN/ha per year
Increase in soil C largely in labile C pools
Further studies required
• Determine how legumes influence soil C accretion under e[CO2]
• Identify strategies to increase sequestration of organic C into more stable pools
Source: van Groenigen et al (2001) Ecol. Studies 187:373-391
Conclusions
Factors regulating inputs of fixed N
• Legume reliance upon N2 fixation for growth
Rhizobia numbers & effectiveness in soil
Decline over time in absence of legume
Benefits of rhizobial inoculation
• Close relationship between N2 fixation & legume biomass
Pasture legume content
Nutritional disorders
Species differences
Longer duration of growth by perennial pasture species
Conclusions
Adaptation to future climates
• N2 fixation & below-ground partitioning of N
Quantify affects of e[CO2]
Explore genetic variability in response
Determine how response is modified by GxExM interactions
• Subsequent availability legume N
Quantify importance of legume below-ground N for following crops
Define how soil N-dynamics in legume-based systems are affected by e[CO2]
Conclusions
Adaptation to future climates
• Water use efficiency (WUE)
Can 13C discrimination be used as a quantitative tool for legumes?
Explore genetic variability in WUE.
Explain apparent inconsistencies between studies.
Identify conditions & sampling protocols where 13C is most reliable.
Conclusions
Mitigation of climate change
• N2O losses from legume-based systems
Confirm origin of N2O release
Consider how N2O emissions might be influenced by future climates
• Quantifying the fate of legume organic C
Determine how rate of change in soil C is modified by GxExM interactions.
Define how soil C-dynamics are affected by e[CO2].
Evaluate whether it is possible to sequester more C into stable pools.