Climate Vulnerability of Farm Households: New Methods and Evidence

John M. AntleProfessor of Applied Economics

Oregon State University

1

Presented at Resources for the Future January 28 2015

2

Acknowledgements

AgMIP & collaborators (agmip.org)

USDA Ag Research Service

UKAID (DFID)

USAID

REACCH-PNA & collaborators

3

Themes

A new “community of science” approach to agricultural systems research (AgMIP)

New approaches & methods to regional and global climate change assessment

Results from two new regional assessment studies

Looking ahead … US assessment, AR6, and beyond

4

Agricultural Model Inter-comparison and Improvement Project (AgMIP)• A new global community of science: climate,

water, soils, crops & livestock, economics,

pests & diseases…led by:

• Cynthia Rosenzweig, GISS and Columbia U

• Jim Jones, U Florida

• Jerry Hatfield, Ag Research Service, USDA

• John Antle, Oregon State U

• More than 600 participating scientists

• Crop, livestock and economic model

inter-comparisons

• Protocol-based regional and global

integrated assessments

• Next Generation Ag System Models

• Sustainable Ag SystemsAgMIP.org

5

Ag, Food and Climate Change

The Goal: sustainable food & nutritional security under future bio-physical and socio-economic conditions

National, local and household relevance (global?)

Beyond commodity production, to the entire food system

Assessment not yet feasible: major data and methodological challenges remain

Vulnerability: who is at risk of loss, and who can gain? Urban consumers: primarily price effects?

Rural ag households: production and price changes affect income, availability, stability

Mitigation and adaptation: what can we do, sustainably?

6

REACCH - Regional Approaches to Climate Change in Pacific Northwest Agriculture

5-year project funded by USDA-NIFAUniversity of IdahoOregon State UniversityWashington State UniversityUSDA-ARS+ 100 scientists & students

Large-scale wheat-fallow and annual cropped systems typical of “industrial commodity agriculture”

7

AgMIP Regional Assessment Teams5-year project, DFID funded8 regional teams, 18 countries, ≈ 200 scientistsData, models, scenarios designed & implemented by multi-disciplinary teams & stakeholders

Small-scale, mixed crop and crop-livestock systems; principal crops vary by region (maize, millet/peanut, rice, wheat) typical of “semi-subsistence agriculture”

8

For the full AgMIP story:

9

AgMIP regional integrated assessment method: beyond average impact to vulnerability

A. Global & national prices, productivity and representative ag

pathways and scenarios (RAPS)

B. Complex farm household systemsC. Heterogeneous regions

System 1: > 0(losers)

System 2: < 0 (gainers)

(ω)

(losses)

D. Technology adoption and distribution of

economic, environmental and

social impacts

E. Linkages from sub-national regions to national and global

10

AgMIP regional integrated assessment method: beyond average impact to vulnerability

A. Global & national prices, productivity and representative ag

pathways and scenarios (RAPS)

B. Complex farm household systemsC. Heterogeneous regions

System 1: > 0(losers)

System 2: < 0 (gainers)

(ω)

(losses)

D. Technology adoption and distribution of

economic, environmental and

social impacts

E. Linkages from sub-national regions to national and global

A. Global & national prices, productivity and representative ag

pathways and scenarios (RAPS)

B. Complex farm household systemsC. Heterogeneous regions

(ω)

(losses)

TOA-MD model simulates gains and losses tradeoffs.oregonstate.edu

0Average impact

Loss

11

TOA-MD Model: vulnerability before and after adaptation

Distribution of losses before adaptation

Distribution of losses after adaptation

12

Relative yield distributions: linking bio-physical and economic models to represent heterogeneity and vulnerability

Source: Author and collaborators, REACCH-PNA Project

05

10

15

De

nsity

.9 1 1.1 1.2 1.3 1.4

GCM 1

02

46

810

De

nsity

.6 .8 1 1.2 1.4

GCM 2

1.5 - 1.6

1.4 - 1.51.3 - 1.4

1.2 - 1.3

1.1 - 1.21 - 1.1

.9 - 1

.8 - .9

.7 - .8

GCM 1

1.5 - 1.6

1.4 - 1.51.3 - 1.4

1.2 - 1.3

1.1 - 1.21 - 1.1

.9 - 1

.8 - .9

.7 - .8

GCM 2

1.5 - 1.6

1.4 - 1.51.3 - 1.4

1.2 - 1.3

1.1 - 1.21 - 1.1

.9 - 1

.8 - .9

.7 - .8

GCM 9

1.5 - 1.6

1.4 - 1.51.3 - 1.4

1.2 - 1.3

1.1 - 1.21 - 1.1

.9 - 1

.8 - .9

.7 - .8

GCM 10

02

46

8

De

nsity

.6 .8 1 1.2 1.4

GCM 9

02

46

8

De

nsity

.6 .8 1 1.2 1.4

GCM 10

(Using Conventional Tillage)

Relative Yields of Spring Pea Projected in 2050 at RCP 8.5

13

Representative Ag Pathways: assessing CC impacts under plausible future bio-physical and socio-economic conditions

Many regional assessments simulate impacts of future climate under current socio-economic conditions

AgMIP RIA methods create plausible future scenarios Global “Shared Socio-Economic Pathways”

Agriculture-specific bio-physical and socio-economic pathways (RAPs)

Created by regional teams with stakeholder input

14

Extent of Vulnerability without adaptation: AgMIP Sub-Saharan Africa & South Asia Teams

0.0

20.0

40.0

60.0

80.0

100.0

120.0

-40.0 -30.0 -20.0 -10.0 0.0 10.0 20.0 30.0 40.0 50.0

Exte

nt o

f Vul

nera

bilit

y (%

hou

seho

lds

vuln

erab

le to

loss

)

Net Economic Impact (% of farm income)

Q1 Q2

Each point a regional scenario:Blue = CC + current socio-econ conditionsRed = CC + future socio-econ conditions

(higher prices, incomes)

Compare to IPCC projected range of average yield changes (-40 % to 0)

15

Extent of Vulnerability without adaptation: REACCH-PNA winter wheat – fallow

0

10

20

30

40

50

60

70

80

-40 -30 -20 -10 0 10 20 30 40 50

Exte

nt o

f Vul

nera

bilit

y (%

hou

seho

lds

vuln

erab

le to

loss

)

Net Economic Impact (% of farm income)

Q1 Q2 - HH Q2 - LL Q1 - Small Q2 - Small - High Q2 - Small - Low

Maroon = large farms: CC + current conditions & pricesBlue = large farms: CC + future cond + high pricesBlack = large farms: CC + future conditions + low pricesGreen = small farms: CC + current conditions & pricesRed = small farms: CC + future conditions + high pricesTan = small farms: CC + future conditions + low prices

Future scenarios:• Dysfunctional world• Business as usual• Sustainable world

16

Extent of Vulnerability without adaptation: AgMIP Regions + REACCH

0.0

20.0

40.0

60.0

80.0

100.0

120.0

-40.0 -30.0 -20.0 -10.0 0.0 10.0 20.0 30.0 40.0 50.0

Exte

nt o

f V

ulne

rabi

lity

(% h

ous

ehol

ds v

ulne

rabl

e to

loss

)

Net Economic Impact (% of farm income)

Q1 Q2 Q2 PNW

Blue = AgMIP CC + current socio-econ conditionsRed = AgMIP CC + future socio-econ conditionsGreen = REACCH all scenarios

17

Magnitude of Vulnerability without adaptation: AgMIP Sub-Saharan Africa & South Asia Teams

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

-40.0 -30.0 -20.0 -10.0 0.0 10.0 20.0 30.0 40.0

Leve

l of V

ulne

rabi

lity

(% lo

ss in

farm

inco

me)

Net Impact (% of farm income)

Q1 Q2

Blue = CC + current socio-econ conditionsRed = CC + future socio-econ conditions

18

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

-40.0 -30.0 -20.0 -10.0 0.0 10.0 20.0 30.0 40.0

Leve

l of

Vu

lne

rab

ilit

y (%

loss

in f

arm

inco

me

)

Net Impact (% of farm income)

Q1 Q2

0.0

10.0

20.0

30.0

40.0

50.0

60.0

-40.0 -30.0 -20.0 -10.0 0.0 10.0 20.0 30.0 40.0 50.0

Leve

l of V

ulne

rabi

lity

(% lo

ss in

farm

inco

me)

Net Impact (% of farm income)

Q1 Q2 Q2 PNW

Magnitude of Vulnerability without adaptation: AgMIP Teams + REACCH

Blue = AgMIP CC + current socio-econ conditionsRed = AgMIP CC + future socio-econ conditionsGreen = REACCH all scenarios

19

Taking Action: Designing & Testing Meaningful Mitigation and Adaptation Strategies

• Large-scale models lack sufficient detail for this purpose!

• Systems approach essential to design & test sustainable mitigation and adaptation options• Agronomic adaptation: variety choice, timing of operations, etc

• Economic adaptation: intensive margin (within system crop choice, land allocation) and extensive margin (between system)

• Sustainability: genetic, soil, water resources, health & nutrition, …

• Future society: industry structure, infrastructure, policy, institutions

• AgMIP Regional Teams, REACCH• strategies developed with local stakeholders

• use plausible future scenarios…”Representative Ag Pathways”

20

Looking ahead…the AgMIP agenda

• Translating impacts on income into food & nutritional security, financial, environmental outcomes

• Expand coordinated regional-global integrated assessments• Protocol-based for transparency, comparability

• New generation of modular, open-source, inter-operable models & data

• Evaluation of climate, model, scenario uncertainty

• Relevance: stakeholder-designed adaptation mitigation strategies

• Working with many partners towards next US assessment, AR6 and beyond!

21

Impact of CC on yields?

Median yield changes (%) for RCP8.5 (2070–2099 in comparison to 1980–2010 baseline) with CO2 effects over five GCMs x

seven GGCMs for rainfed maize (35 ensemble members). Hatching indicates areas where more than 70% of the ensemble

members agree on the directionality of the yield change. Gray areas indicate historical areas with little to no yield capacity

(Rosenzweig et al., PNAS 2013).

Source: IPCC AR-5, WGII, Ch 7.

22

Are we at a turning point for agricultural prices?

Source: IPCC AR-5, WGII, Ch 7

• AgMIP global econ model inter-comparison (Nelson et al. PNAS 2014): • Without climate change, trend to 2050 highly uncertain (+/- 50%)• Effect of climate change likely positive, but also highly uncertain (0 to + 50%)• Results strongly determined by assumed productivity trends