Global Change and Wheat Production: Assessing the Future

1
Global Change and Wheat ProductionAssessing the
Future

• PD Jamieson (NZ)
• JR Porter (Denmark)
• MA Semenov (UK)

2
(No Transcript)
3
How is world wheat production likely to change in
the next century?
4
The drivers of change

• Atmospheric CO2
• Climate
• Geopolitics
• International trade
• Technology

5
The drivers of change

• Atmospheric CO2
• Climate
• Geopolitics
• International trade
• Technology

6
The drivers of change

• Atmospheric CO2
• Going up (definitely)
• Climate
• Warmer (probably)
• Changes in rainfall patterns
• Technology
• Information
• Agrichemicals
• Machinery
• Cultivars

7
Sensitivity to global change

• Changes in suitable land area
• Improvements associated with technology
• Direct responses to changes in CO2, temperature
  and water supply

8
Sensitivity to global change

• Changes in suitable land area
• Improvements associated with technology
• Direct responses to changes in CO2, temperature
  and water supply

9
On what basis can we make assessments

• Experimental work to determine and responses to
  temperature, water and CO2
• Simulation modelling that incorporates known
  responses
• The GCTE Networks
• Wheat
• SOM
• Rice
• Tropical cereals

10
On what basis can we make assessments

• Experimental work to determine and responses to
  temperature, water and CO2
• Simulation modelling that incorporates known
  responses
• The GCTE Networks
• Wheat
• SOM
• Rice
• Tropical cereals

11
GCTE International Wheat Network

• Formed Saskatoon Canada, 1992
• Meetings at
• Lunterin, Netherlands 1993
• Reading, UK 1995
• Casa Grande, Arizona, USA 1997
• Potsdam, Germany 1998
• KBS, Michigan, USA 2000
• To come
• Clermont Ferrand, France 2004

12
A loose alliance of researchers

• Experimenters
• USDA Arizona, FACE experiments
• Rothamsted Research CE experiments
• Others from NZ, Australia, Canada.
• Modellers from
• UK, USA, Canada, Germany, New Zealand, Denmark,
  Australia
• Publications from the collaborations
• Many in the international literature
• Plus many conference presentations

13
At the simplest

• production growth rate
• x growth duration
• x harvest index
• CO2 increase ? growth rate increase
• Temperature increase ? growth duration decrease
• Temperature increase ? harvest index change (?)
• Combined CO2 T increase ? Increased WUE
• 
  ? more suitable land (?)

14
Therefore

• The system is complex
• We need to use a physiological model to analyse
  impacts and assess the balance of competing
  effects

15
Some of the available models

• AFRCWHEAT2
• CERES-Wheat
• CropSim
• Demeter
• FASSET
• SWHEAT and the Dutch models
• Sirius

16
Some of the available models

• AFRCWHEAT2
• CERES-Wheat
• CropSim
• Demeter
• FASSET
• SWHEAT and the Dutch models
• Sirius

17
Some of the available models

• AFRCWHEAT2
• CERES-Wheat
• CropSim
• Demeter
• FASSET
• SWHEAT and the Dutch models
• Sirius

18
Some of the available models

• AFRCWHEAT2
• CERES-Wheat
• CropSim
• Demeter
• FASSET
• SWHEAT and the Dutch models
• Sirius

19
Some of the available models

• Sirius

20
Sirius

• Phenology based on leaf appearance and numbers
• Canopy growth a function of thermal time,
  modified by stress
• Biomass accumulation from light interception and
  light use efficiency
• Grain production based on simple partitioning
  rules
• Widely tested

21
Model Validation
22
What are the predicted impacts of global
atmospheric change?
23
Canterbury NZ - maturity
24
Canterbury NZ - Yields
25
On a straight race between yield reducing
duration changes (temperature) and yield
enhancing growth rate changes (CO2)

• CO2 wins

26
Lest you think this applies only to high
production systems

• Australia most likely climate change and
  double CO2
• Yield increases of 9-37 assuming current
  practice
• Yield increases of 13-46 with adapted management
  
• Howden, Reyenga and Meinke (1999) using APSIM
  I-Wheat
• Average yields lt 2 t/ha
• The variations are with location
• Risks may increase extreme events
• Temperature, droughts.

27
Technology effects

• Farmers and researchers live in largely the same
  environment, and this moulds their expectations
• These get revised upwards by both groups as they
  learn more from each other
• In NZ, 25 years ago wheat yields of 5 t/ha were
  being sought by both researchers and growers
• 10 years ago it was 10 t/ha
• Now it is 15 t/ha
• This needs attention to detail in management

28
Projected production increases by 2030 (million
tonnes pa)

• Developed world 308 to 440 (43)
• Developing world 272 to 418 (54)
• Mostly from increased yield/ha
• Marathée Gomez-MacPherson (2001 The World
  Wheat Book)
• By 2100?
• At least that much again is possible
• None of the above considers atmospheric changes

29
The Land Balance

• Combined CO2 temperature increase
• increased WUE
• more suitable land (?) but low production
• Rural depopulation and urban sprawl
• Reduction in available land area
• Magnitude?
• Losses are likely to bigger than gains

30
The Trade Balance

• Increasing demand from the developing world will
  increase its wheat deficit, and the amount of
  wheat traded
• Net trade (millions of tonnes) by 2030
• Developed world
• Exports increase from 64 to 155 (142)
• Developing world
• Imports increase from 61 to 152 (150)
• The biggest demand increases will be in China and
  North Africa/Near East
• Marathée Gomez-MacPherson (2001 The World
  Wheat Book)

31
Where to from here..?

• Whole systems soils, rotations..
• Food quality balancing up protein and
  carbohydrate production
• Genetics
• Keeping pace with global change
• Conventional plant breeding
• Biotechnology
• Scaling from field to landscape

32
Whole systemsR-W sequences in the IGP
Fallow Legume GreenManure Jute ChickPea Pea
OilSeed
33
Actual v potential yieldsThe yield gap
34
Simulation of Phenology(Ortiz-Monasterio et al.
1994)
35
Simulation of grain number
36
Simulation of yield
37
Causes of yield gap?

• The reasons that achieved yields are well below
  potential are not obvious from the experimental
  data
• What manifest as sowing date effects are likely
  to be soil and management influences
• Continuous cropping degrades soil
• We need better links between crop models and SOM
  models

38
Causes of yield gap?

• Severe biological and/or technological
  limitations to productivity, and . potential for
  substantial yield increase provided the
  environmental and management constraints can be
  identified and rectified, are evident
• Timsina and Connor, (2001).

39
The protein gap

• Many of the genetic advances that have raised
  wheat yields have resulted in reduced protein
  levels
• We believe this can be addressed by appropriate
  attention to N metabolism in wheat cultivars
• Higher production, whether from CO2 fertilisation
  or just better management, requires more N
  fertiliser

40
Where does the N go?
Jamieson and Semenov 2000. Field Crops Res.68,
21-29. Martre, Porter, Jamieson and Triboi 2003.
Plant Physiology, (In press)
1 of non-green biomass
15 kg/ha per unit green area
0.5 of non-grain
2 of Grain
41
Issues of scale

• Single plant scale is of most interest to the
  plant physiologist
• Plot scale is of most interest to the crop
  physiologist
• Field and farm scale is of most interest to the
  farmer
• Industries, commerce and governments must deal
  with regional scale

42
Regional scale

• Regional production depends on
• How much land is sown
• How much crop survives
• Storage and transportation
• Yield
• How best to represent/predict?

43
Spatial and temporal scales
44
Downscaling or upscaling?
45
Scaling up Simplifying a crop model

• Simplify a model by analysing model structure,
  model processes and its interactions

46
Scaling up Simplifying a crop model

• Simplify a model by analysing model structure,
  model processes and its interactions

Brooks, Semenov and Jamieson, 2001. Eur. J.
Agron. 14, 43-60.
47
Spatial downscaling HadRM climate regional model
HadCM3 global model
48
Temporal downscalingLARS-WG stochastic weather
generator

• Generates precipitation, min and max temperature
  and radiation
• Modelling is based on wet/dry series
• Flexible semi-empirical distributions are used
• Temperature and radiation are cross-correlated

www.iacr.bbsrc.ac.uk\mas-models\larswg.html
49
Climate change impact assessment high temporal
and spatial resolutions
GCM
Observations
low resolution, 300 km
high resolution, 1 km
50
Effect of changes in climatic variability on
simulated grain yield (Nature, 1999)
51
Conclusions

• Global atmospheric changes will have mostly
  positive effects on wheat production
• Other influences are likely to be just as big and
  just as positive
• It aint all doom! But..
• Risks may increase
• We need to continue to improve the methodology

52
Thanks to.

• GCTE Conference organisers
• International Wheat Network collaborators