The Evolving Role of IA and Emerging Critiques

John P. Weyant Stanford University

Presentation at the Eighth Annual Meeting

of the Integrated Assessment Modeling Consortium

November 17, 2015 Seminaris SeeHotel Potsdam, Germany

Outline

• Types and uses of IAMs • Critiques of the IAM modeling community • Critiques of the IAMs • Critiques of the use of the IAMs • A couple of huge challenges along the way

• “Integrated” climate change impacts analyses • Equity and income distribution

• The alternatives to the models OR new directions in modeling?

Types and Uses of IAMs • Detailed Process IAMs

• Use of DP IAMs in Mitigation Analyses • Use of DP IAMs for Climate Impacts Analysis • Use of DP IAMs to Study Interactions between Mitigation and

Impacts/Adaptation

• Applications of Simple Aggregate Benefit-Cost Analysis IAMs • Computing “Optimal” Climate Polices with BC IAMs • Using Results from BC IAMs to Assess “Non-Optimal” Climate Policies • Projecting the Social Cost of Carbon Using BC IAMs

• Regional IAMS, Partial IAMs and Linkages to Sustainable Development

Source: Model developed in Nordhaus, A Question of Balance, Yale University Press, 2007, with some runs omitted. 0

1

2

3

4

5

6

Tem

pera

ture

cha

nge

(C)

Optimal Baseline < 2 deg C < 2x CO2

0

50

100

150

200

250

300

350

400

450

500

2005 2015 2025 2035 2045 2055 2065 2075 2085 2095 2105

Car

bon

pric

e (2

005

US$

per

ton

C)

Optimal < 2 degrees C < 2x CO2

Example: DICE B/C RESULTS

Global Emissions and Radiative Forcing by Policy

Kriegler, Weyant et al. Note: LowEI assumed, not priced

G8 Policy Scenario not Consistent with 2°C.

Fragmented Policy well above 550 ppm CO2-e.

Energy efficiency improvements alone are not enough to stabilize

climate change.

5

EMF27–The role of technology in mitigation

6

Kriegler, Weyant et al, Climatic Change, forthcoming Krey, Luderer et al, Climatic Change, forthcoming

BECCS is very valuable mitigation strategy, particular for 450 ppm

2000 2020 2040 2060 2080 2100

GH

G e

mis

sion

s (G

tCO

2 eq

uiv.

)

0

10

20

30

40

50

60

70

80 AIM-Enduse DNE21 GCAM IMAGE MERGE-ETL MESSAGE REMIND WITCH Baseline High 2030 Low 2030 Optimal

Lack of short-term reductions (2030)

World GHG emissions Focus on

2030

(extension of “pledge

effort”) Implications for costs, technology needs, and

attainability of long-term stabilization goals?

AMPERE WP2 Study (Riahi et al.)

See presentation by Nils Johnson in session on informing near term international policy discussion

DP IAM Example: MIT IGSM

The Land Use Implications of Stabilizing at 450 ppm When Terrestrial Carbon is Valued

9

0%

20%

40%

60%

80%

100%

1990 2005 2020 2035 2050 2065 2080 2095

UrbanLandRockIceDesertOtherArableLandTundraShrubLandUnmanagedForestForestGrassLandUnmanagedPasturePasturePurGrownBioRiceSugarCropOtherGrainOilCropMiscCropFodderCropFiberCropCornWheat

Other Unmanaged Land

Unmanaged Forests

Managed Forests

Pasture

Crops

Unmanaged Pasture

Desert

Bioenergy Crops

0%

20%

40%

60%

80%

100%

1990 2005 2020 2035 2050 2065 2080 2095

Other Unmanaged Land

Unmanaged Forests

Managed Forests

PastureCrops

Unmanaged Pasture

Desert

Bioenergy Crops

0%

20%

40%

60%

80%

100%

1990 2005 2020 2035 2050 2065 2080 2095

Other Unmanaged Land

Unmanaged Forests

Managed Forests

Pasture Crops

Unmanaged Pasture

Desert

Bioenergy Crops

Reference Scenario

450

ppm

Sta

biliz

atio

n S

cena

rio

Whe

n A

LL

Car

bon

is V

alue

d

450

ppm

Sta

biliz

atio

n S

cena

rio

Whe

n Te

rres

tria

l Car

bon

is

NO

T V

alue

d

DP IAM Example: GCAM

Critiques of the IAM Community • The models are not documented

• There has always been a fair amount of this, but probably too fragmented • ADVANCE On-line documentation program as big step forward

• The models are not open source • IP and tech support issues • Can a bad modeler run a good model well?

• The models are not “validated” • IAMs are very very different than science models in this regard • Some useful benchmarks still highly desirable • Can a bad modeler run a good model well?

• The community does not adequately consider scenario uncertainty • Community is getting better • Community needs more of a strategy for this • At some point this DOES probably involve subjective probabilities

• The models don’t get the right answers • Hmmmm…… • Community needs to be more savvy about motives

Critiques of – Challenges for - the IAMs

• What to count and how to count it • Heroic aggregation assumptions • Tipping points, fat tails, and potential catastrophes • Treatment of geographical equity: regional, national, international • Treatment of inter-temporal discounting and intergenerational equity • Projections of baseline drivers and policy implementation details • Dealing with uncertainty and risk

Huge Challenge #1: Integrated Climate Change Impacts Assessments

• Multi-sector impacts may be significant (system boundaries) • Energy, land, water, food, climate, poverty, health, SLR, etc. • Could lead to significant competition, re-allocations, transfers of inputs

• Substitution of outputs could also be significant • General equilibrium effects (consumption, production, supply chains) • Transfers, inter-state commerce, international trade and aid, etc. • Can often ameliorate net impacts • But can also provide external shocks from outside regions

• Mitigation and impacts/adaptation interactions can be large • Land and water for biofuels squeeze agricultural/food markets • Climate change leads to energy supply and demand impacts

• Climate change feedbacks • Global earth system and back down • Regional

• Policy synergies • Land, agriculture, forest, energy, air quality, climate • Examples include climate change and air quality targeted policies.

Critiques of the use of the IAMs

• Not fully communicating what’s considered and what’s not • Inadequate treatment of equity/income distribution considerations • Incomplete coverage of full uncertainty space • Lack of a strategy for setting priorities • An aversion to subjective probability assessments

On Maps and Navigators Maps and navigators are good, but they may not be enough

Map/Modeling guides would be useful as well Psalm 25 Of David. 1 In you, LORD my God, I put my trust. 2 I trust in you; do not let me be put to shame, nor let my enemies triumph over me. 3 No one who hopes in you will ever be put to shame, but shame will come on those who are treacherous without cause. 4 Show me your ways, LORD, teach me your paths. 5 Guide me in your truth and teach me, for you are God my Savior, and my hope is in you all day long. 6 Remember, LORD, your great mercy and love, for they are from of old. 7 Do not remember the sins of my youth and my rebellious ways; according to your love remember me, for you, LORD, are good.

Huge Challenge #2: A Revealed Preference Study of Climate Change Policy Analyses

Class ofWorld Citizen

Typical OECD (AEA Member?)Analysis

ROW Analysis

2 Billion PeopleWithout Markets

What 2 BillionPeople?

High Priority:Reduce TheirVulnerability

2 Billion In or NearPoverty with FragileMarkets

They Don’t Countfor Much!

High Priority:Reduce TheirVulnerability

2 Billion PotentialDecaf LattéDrinkers

Half Are Stuck InTransition, But theRest We Can Help

They Can TakeCare ofThemselves

Just a Few Frustrating Modeler Perspectives Still Occasionally Heard Around The Community

• Its not in my model, so it is not important • It is in my model, but it is not important in my model, so it is not

important in the real world • The real world has produced different outcomes than my model has

projected, so the real world must be seriously incorrect

Summary: New Models and Methods? • Models will become more data driven • Models need to become more transparent (traceable accounts idea) • Equity/income distribution needs to be considered more directly • Integration will need to be treated more seriously • Uncertainty needs to be treated more systematically • Model results need to be explained and qualified more carefully • The idea of probabilities will have to be exploited one way or the other

The End Questions?

Two things fill the mind with ever-increasing wonder and awe, the more often and the more intensely the mind of thought is drawn to them: the starry heavens above me and the moral law within me.

Immanual Kant

Cost/Benefit Modeling Approach: Balancing the Costs of Controlling Carbon Emissions Against the Costs of the Climate impacts They Cause

Value/Cost of Emissions Reductions

Carbon Emissions

Marginal Cost of Climate Impacts

Marginal Cost of Emissions Control

21

• ca. 950 mitigation scenarios AMPERE: https://secure.iiasa.ac.at/web-apps/ene/AMPEREDB

LIMITS: https://secure.iiasa.ac.at/web-apps/ene/LIMITSDB

EMF27/ RoSE to be published by Summer / Fall 2014

• Major contribution to IPCC AR5 report of Working Group 3 AR5 Scenario Database: https://secure.iiasa.ac.at/web-apps/ene/AR5DB

Climate Change Economics 4(4)/5(1)

(LIMITS)

Climatic Change 123(3-4) (EMF27)

Tech. Forecasting and Soc. Change

(AMPERE)

New studies of the energy-land transformation

Climatic Change (RoSE)

22

Numerous new insights on the implications of, inter alia, • short term climate policies (until 2030), • availability of low carbon technologies and • different assumptions about future emissions drivers on the costs and achievability of long term climate targets

Agreement in 2015 and 2oC

(LIMITS)

Role of Technology Availability (EMF27)

Global policy landscape & timing

(AMPERE)

` Studies of the energy-land transformation

Role of emissions drivers (RoSE)