Climate Change and Agriculture

Martin Parry and Cynthia Rosenzweig

Intergovernmental Panel on Climate Change(U.K. Meteorological Office and Goddard Institute for

Space Studies)

OUTLINE

• What we currently know about climate change and it implications for world food supply

• Ten lessons to help us respond

• Conclusions: challenges for the CGIAR

Agriculture: a driver of GHG and a key area of impacts

• Agriculture contributes 20% of GHG emissions globally, today.

• By 2020 this will be 15% or less; and by 2050 10% or less.

• Agriculture as a source of greenhouse gases IS NOT the main issue.

• Protecting agriculture from climate change IS the main issue

CLIMATE 1000 -2100ADIPCC 2001

Change in annual temperatures for the 2050s

The change in annual temperatures for the 2050s compared with the present day, when the climate model is driven with an increase in greenhouse gas concentrations equivalent to about a 1% increase per year in CO2. The picture shows the average of four model runs with different starting conditions.

Hadley Centre for Climate Prediction and Research

Observed change in annual precipitation for the 2050s

The change in annual precipitation for the 2050s compared with the present day, when the climate model is driven with an increase in greenhouse gas concentrations equivalent to about a 1% increase per year in CO2. The picture shows the average of four model runs with different starting conditions.

Hadley Centre for Climate Prediction and Research

Annual runoff : A measure of water supply (2080s)

Percentage change in 30-year average annual runoff by the 2080s.

University of Southampton

Projected increase in intensity of Indian monsoon, c. 2050 (HadRM2)

Change in Annual number of rainy daysChange in Annual number of rainy days Change in rainfall per rainy dayChange in rainfall per rainy day

Other projected changes of climate (by 2100)

• Sea levels will rise 0.1 to 0.8 metres• More tropical cyclones• More frequent very hot days• And… we can see some of these

changes already, occurring now.

Crop yield change: 2010, 2050, 2080

Percentage change in average crop yields for the climate change scenario (Hadley model). Effects of CO2 are taken into account. Crops modelled are: wheat, maize and rice.

2020s 2050s 2080s

% c

hang

e in

pric

es fr

om 1

990

base

line

10

8

6

4

2

0

–22020s 2080s 2020s 2080s

40004200

38003600340032003000280026002400

2050s 2050s

Cer

eal p

rodu

ctio

n (m

mt)

Referencescenario

Climate changescenario

2020s 2050s 2080sAddi

tiona

l milli

ons

of p

eopl

e at

risk

of h

unge

r

50454035302520151050

Additional people at risk of hunger(0 = reference case, i.e. about 300 million in 2000, FAO ).

Projected global cereal production for reference case and the climate change scenario.

Percentage change in global cereal prices under the climate change scenario (0 = reference case).

Response 1:Mitigation.Emissions and concentrations of CO2 from unmitigated and

stabilising emission scenarios

Unmitigated emissions 750 ppm stabilisation 550 ppm stabilisation

10

5

0

15

20

Ant

hrop

ogen

ic C

O2

emis

sion

s (G

tC/y

r)

2000 2050 2100 2150 2200 2250 2300 2350C

O c

once

ntra

tion

(ppm

)2

1000

950

900

850

800

750

700

650

600

550

500

450

400

3502000 2050 2100 2150 2200 2250 2300 2350

Hadley Centre for Climate Prediction and Research

Changes in crop yieldfrom the present day to the 2080s

Unmitigated emissions

Stabilisation of CO2 at 750 ppm Stabilisation of CO2 at 550 ppm

Potential change in cereal yields (%)

10 – 5

0 – -2.5

-5 – -10-2.5 – -5

-10 – -20

2.5 – 05 – 2.5

No data

Response 2: Sustainable DevelopmentThe effect of development pathways

example: IPCC’s SRES scenarios

A1 = World of Global Markets: high growth, high tech, low pop, A2 = World of Regional Enterprise: high pop. high growth (but inequitable) [current pathway]B1 = World of Global Sustainability: low pop,

moderate growth, global envtl agreements.B2 = World of Local Stewardship: low pop, moderate growth, local envt management.

0

100

200

300

400

500

600

700

800

900

1000

1990 2000 2010 2020 2030 2040 2050 2060 2070 2080

Year

Add

ition

al m

illio

ns o

f peo

ple

SRES A1 SRES A2 SRES B1 SRES B2

Future reference case estimates of the numbers of people at riskof hunger, for the four SRES marker scenarios (no climate change).

Response 3: ADAPT

• Successful adaptation needs clear picture of future agro-climate, at regional level

• So, what lessons have we learned about this, so far:

Climate Change and Agriculture‘Lessons’ Learned

Cynthia RosenzweigNASA Goddard Institute for Space Studies

Columbia University

www.itdg.org

Annual Temperature Trends 1973-2002

Impacts of Climate Change on Multiple Cropping Production Potential of Rain-fed Cereals

Climate change impacts are visualized using a normalized difference index. It is calculated by dividing the difference in cereal production capacity between future and current production potential by their sum.

Max-Planck Institute/ECHAM4 2080s

Agricultural Impacts are Heterogeneous

Magnitude and Distribution

of Impacts are Uncertain

Climate change impacts are visualized using a normalized difference index. It is calculated by dividing the difference in cereal production capacity between future and current production potential by their sum.

80

90

100

110

120

0 1 2 3 4 5 6 7 8 9 10

WORLD

Effects are Negative in Long Term Percent Change in Food Production Potential

PRODUCTION potential assuming LOW range end of CO2 concentrationPRODUCTION potential assuming HIGH range end of CO2 concentrationAREA EXTENT assuming LOW range end of CO2 concentrationAREA EXTENT potential assuming HIGH range end of CO2 concentration

0-10 = Severity of climate change (~time)

Developed and Developing Countries Diverge in Vulnerability

regions

be n

Corn

(2)(2)

(57)(49)(22)(25)(5)

no dataless than -30%-30 to - 10 %-10 to 0%0 to 10%10 to 30%more than 30%

Percent of change in yield

Both Impacts and Adaptive Capacity

IPCC, 2001

Multitudes of global and regional studies on rural groups,

agricultural systems

Possible decadalsurprises

0

20

40

60

80

100

1990 2000 2010 2020 2040 2040 2050Year

Per

cent

age

ReliabilityDemand met

400

300

200

100

50O N D

MonthsJ F M A M J J A S

Run

off(

cfs) Current

GFDLMPIHCChange in

seasonality

Water Resources are Key

Strzepek et al., 1999

Environmental StressDemand:Supply Ratio

Pests May Surprise!

199819931985

1971

Overwintering range of potato leafhopper under two doubled C02 climate change scenarios. (Stinner et al., 1989)

1 generation

2 generation

3 generation

4 generation

Range of expansion of soybean sudden death syndrome (Fusarium solani f.sp. glycines) in North America. (X.B. Yang).

Approximate distribution of European corn borer annual generations in the U.S. and Canada. (Mason, 1996)

GISS GFDL Present

Seasonal-to-interannual*

Decadal-to-century**Feb1998

Feb 2000

ENSO Uruguay

Baethgen, 2000

Risk of Current Climate Variability

is a Pathway to Climate Change

NOAA AVHRR

*Focus on Extreme events

Regional StakeholdersShort-term Decisions

Adaptation

**Focus on Mean changes

National PolicymakersLong-term Decisions

Adaptation & Mitigation

Ing. Juan Notaro, Uruguayan Minister of Agriculture in 1999/2000

"... The results of your work during the recent drought were useful for making both operational and political decisions.”

Need new models to integrate the two time-scales . . .

2100

4

3.5

3

2.5

2

1.5

1

0.5

19000

Future temperatureincrease

1950 2000 2050

Year

° Cab

ove

p re-

ind u

s tr ia

lglo

balm

ean

tem

per a

tur e

Business asusualStabilizationscenario

Farmers* need to Adapt** and Mitigate at the Same Time

NASA/GISS*Agricultural stakeholders **To both climate variability and change.

Responses

Adaptation

Two examples for the CCGS 2030s Scenario

10 Fargo ND Glasgow MT Boise ID

0

No adaptationAdaptation

%Yi

eld

c han

ge

Strategy: Early planting

Results: Successful heat stressavoidance

Spring wheat

10 Dodge City KS North Platte NE Goodland KS

0

No adaptationAdaptation

%Yi

eld

chan

ge

Strategy: Change of cultivarResults: Unable to reverse damage

due to low precipitation

Winter wheat

CC = Canadian Climate Centre GCM U.S. National Assessment; Tubiello et al., 2002

Adaptation is Important, But Not Always Possible, Complete, or Cheap

Climate Change and Adaptation Affects C-Sequestration

Corn Field, North Platte, NE

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

Chan

ge in

Soi

l C, T

C h

a-1 y

r-1

Current ClimateClimate Change, CCCMClimate Change, HAD

Irrigation

N Fertilization

Barriers and bridges projects to adoption of C-sequestration techniques by farmersTake changing climate and climate variability into account

Simulation Results (40 years)

How does C-sequestrationhelp farmers adapt to

climate extremes?

1973-2002 Annual Temperature TrendsInterpolated

Climate Change is Already Happening Phenology

Management practices, pests and diseases

Yields

Livestock

<-1.2C to >1.2C

Conclusions : 1

• Climate change is likely globally to : reduce potential agricultural output in the longer term , and increase risk of hunger.•Adverse effects, regionally and near-term, are especially marked in tropics and sub-tropics (eg especially in Africa)

Conclusions : 2• Most serious effects, sub-nationally, will probably

be at the social and economic margins (where adaptive capacity is low).

• Stabilisation at 750 ppm does not avoid most serious effects. Stabilisation at 550 ppm does, but cost will be great (= c.20 times Kyoto reductions).

• Therefore, successful mitigation will need to be part of a ‘Sustainable Development’ pathway.

Conclusions : 3• A combination of adaptation and mitigation is

necessary.• Invest in adaptation, to increase resilience to climate

change: a) technology (eg crop breeding for new climates; rural electrification ), b) management (egfarming systems that use water more efficiently), c) institutions (eg market and tariff structure).

• Many adaptations can be ‘win-win’ (eg drought-proofing for present weather can increase resilience to effects of a long-term drying trend).

Conclusions : 4• Need to foster adaptation in most vulnerable

regions: the poorest, small island, low-lying coats, andn semi-arid tropics/sub-tropics

• Should we concentrate more on non-optimising objectives? eg reduce risk, minimise yield reductions in drought years, develop resilient (rather than maximising) crop varieties and crop mixes.

• Regardless…Adaptation is needed now.

Examples of adaptation• New crop varieties: eg drought resistant (by

traditional plant breeding, or genetic modification);

• Irrigation: needs rural electrification (especially in Africa= a Millenium Development Goal)

• Shift of cropping zones: eg northwards in N. Hemisphere (but limited by soils)

• Reform global food system: [today, enough food is produced for all, but still 500m are hungry]

Changes in Drought Severity.

HadCM3 A2 scenario (Burke and Brown, 2006)

Additional number of people at risk of hunger in Africa under the climate change scenario (0 = Projected reference case).

2020s 2050s 2080s

454035302520151050

Africa

Addi

tiona

l milli

ons

of p

eopl

e at

risk

of h

unge

r

Projections for cereal production in Africa under the reference case and the climate change scenario

340360

320300280260240220200180160140

2020s2050s

2080s 2020s2050s

2080s

Cer

eal p

rodu

ctio

n (m

mt) Africa

scenario scenarioReference Climate change

Changes in runoff from the present day to the 2080s

Unmitigated emissions

Stabilisation of CO2 at 750 ppm Stabilisation of CO2 at 550 ppm

University of Southampton

–75 –50 –25 –5 to 5 25 50 75Change in annual runoff (%)

Millions at Risk in the 2080s

0

50

100

150

200

250

300

350

0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75 3 3.25Temperature Increase

Add

ition

al m

illio

ns o

f peo

ple

at r

isk

of h

unge

r, m

alar

ia

and

coas

tal f

lood

ing

0

500

1000

1500

2000

2500

3000

3500

Add

ition

al m

illio

ns o

f peo

ple

at r

isk

of in

crea

sed

wat

er s

hort

age Risk of water shortage

Risk of malaria

Risk of hunger

Risk of coastal flooding

Legend

Sta

b. 5

50

Sta

b. 6

50

Unm

itiga

ted

emis

sion

s - I

S92

a

Sta

b. 4

50

Sta

b. 7

50

Sta

b. 1

000

0

1000

2000

3000

4000

5000

6000

1990 2000 2010 2020 2030 2040 2050 2060 2070 2080

Year

Mill

ion

Met

ric T

onne

sSRES A1SRES A2SRES B1SRES B2

Future reference case estimates of cereal production under the four SRES marker scenarios (no climate change).

-20

0

20

40

60

80

100

120

140

160

180

1990 2000 2010 2020 2030 2040 2050 2060 2070 2080

Year

Cha

nge

in c

erea

l pric

es (%

) SRES A1SRES A2SRES B1SRES B2

Future reference case global cereal prices, relative to 1990 prices, for the four SRES marker scenarios (no climate change).

A2 in 2050s B2• Pop 11.3 billion• GDP 82 tr $• primary energy 970

GJ/yr• carbon 16 GtC/yr

• Pop 9.3 billion• GDP 110 tr $• primary energy 870

GJ/yr• carbon 11 GtC/yr

Aggregated developing-developed country differences (%) in average crop yield changes from baseline for the HadCM2 and HadCM3 scenarios

Potential change (%) in national cereal yields for the 2050s (compared with 1990) under the HadCM3 A1FI with CO2 effects

Parry et al. 2004

Development Path Matters