Examples  of  Other  Par1cipa1ng  Models  

January  17,  2012  

The  Following  Models  will  be  Par1cipa1ng  in  LAMP  

Model   Ins*tute   Country  

ReMIND   Potsdam  Ins1tute  (PIK)   Germany  

IMAGE   Netherlands  Environmental  Assessment  Agency  (PBL)  

The  Netherlands  

WITCH   Fondazione  Eni  Enrico  MaTei  (FEEM)   Italy  

MESSAGE   Interna1onal  Ins1tute  for  Applied  Systems  Analysis  (IIASA)  

Austria  

MERGE   Model  for  Evalua1ng  Regional  and  Global  Effects  of  GHG  Policies  

USA  

Potsdam Institute for Climate Impact Research Research Domain Sustainable Solutions

The ReMIND-MAgPIE Integrated Assessment Model System

Potsdam Institute for Climate Impact Research

Research Domain III: Sustainable Solutions

January 2012

Elmar Kriegler Potsdam Institute for Climate Impact Research

Key Design Characteristics Model: ReMIND (Refined Model of Investment and Technological Development)

Model Type: Ramsey-type macro-economic growth model coupled to energy system model, climate box model, (agro-economic land use model MAgPIE)

Model team: Gunnar Luderer, Marian Leimbach, Nico Bauer, Tino Aboumahboub, Tabare Arroyo-Curras, Lavinia Baumstark, Christoph Bertram, Anastasis Giannousakis, David Klein, Ioanna Mouratiadou, Robert Pietzcker, Franziska Piontek, Jana Schwanitz, Elmar Kriegler

Time Step: 5 years

Time Frame: 2005 to 2100

Solution Type: Intertemporal Optimization

Equilibrium Type: Intertemporal equilibrium of capital, energy, goods markets (Pareto optimum)

Underlying Computing Framework: GAMS

Elmar Kriegler Potsdam Institute for Climate Impact Research 5

Welfare

Labour Capital

Energy system costs

Output

Consumption Investments

Final energy

Energy transformations and conversion technologies

Fuel costs

Investment costs

Operation and Maintenance

costs

Labour efficiency Emissions

Learning by doing

Ressource and potential

constraints

Macro Economy

Energy system

Exogenous Data

Energy efficiency

Trade

Trade

Trade

Climate module

ReMIND model overview

Leimbach,Bauer, Baumstark, Edenhofer, O. (2009)

Elmar Kriegler Potsdam Institute for Climate Impact Research

ReMIND regions

USA - USA EUR - EU27 JAP - Japan CHN - China IND - India RUS - Russia AFR - Sub-Saharan Africa (excl. Republic of South Africa) MEA - Middle East, North Africa, central Asian countries OAS - Other Asia (mostly South East Asia) LAM - Latin America ROW - Rest of the World (Canada, Australia, New Zealand, Republic of South Africa, Rest of Europe).

Regional detail

Elmar Kriegler Potsdam Institute for Climate Impact Research

Inputs and Outputs Key inputs •  Demographics: Population by region •  Economic: Labour & energy

productivity •  Resources: Depletable resources by

marginal extraction cost curves (after Rogner; coal, oil, gas, uranium); Renewable resources by grade (wind on & off-shore, solar, geothermal, hydro).

•  Emissions / forcing: CO, NOx, VOC, halocarbons, SF6, OC, BC

•  Technology: see sketch. CCS for coal, gas, biomass-fired power

plants, coal & biomass to liquid technologies, H2 production

Key outputs •  Economic: GDP, Consumption, Investment, Shadow prices, Factor shares, Trade flows •  Energy: see sketch •  Agriculture (via MAgPIE): Land use pattern, Production costs, Shadow prices Investment in productivity improvements •  Emissions: CO2 (by source), CH4, N2O, SO2 •  Climate: GHG Concentrations, Radiative Forcing, Temperature, (Ocean pH, Sea level rise)

Elmar Kriegler Potsdam Institute for Climate Impact Research

Cereals

Oilseeds

Pulses

Sugar beets

Crop yields Land & Water constraints

+200 mm-200 -100 0 +100

CCSR

ECHAM4

Climate

LPJ (50x50 km grid)

Biophysical inputs

Income vs. Food consumption

0

500

1000

1500

2000

2500

3000

3500

4000

0 5000 10000 15000 20000 25000 30000 35000 40000

GDP / Cap / Year

Kca

l / C

ap /

Day

kcal_cap (105 countries, 1990/2000) kcal_cap (fitted values)

kcal = 802 * gdp^(0.142327) [R^2 = 0.66]

0

2

4

6

8

10

12

14

16

1900 1920 1940 1960 1980 2000 2020 2040 2060 2080 2100

Billion

Low fertility, low mortality High fertility, high mortality Central fertility, central mortality (Lutz et al. 2001)

Demography

Income and diet

Food demand, production costs

Socioeconomic inputs

MAgPIE – a global land use optimisation model •  spatially explicit (0.5°), 10 economic regions •  30 production activities (13 crops, livestock, irrigation, bioenergy, land conversion) •  internal feed balances, international trade •  endogenous land expansion •  endogenous technological change

Bioenergy

Lotze-Campen , Popp et al. (2008), Agricultural Economics

Elmar Kriegler Potsdam Institute for Climate Impact Research

Land

use

dyn

amic

s

Income vs. Food consumption

0

500

1000

1500

2000

2500

3000

3500

4000

0 5000 10000 15000 20000 25000 30000 35000 40000

GDP / Cap / Year

Kca

l / C

ap /

Day

kcal_cap (105 countries, 1990/2000) kcal_cap (fitted values)

kcal = 802 * gdp^(0.142327) [R^2 = 0.66]

0

2

4

6

8

10

12

14

16

1900 1920 1940 1960 1980 2000 2020 2040 2060 2080 2100

Billion

Low fertility, low mortality High fertility, high mortality Central fertility, central mortality (Lutz et al. 2001)

Cereals

Oilseeds

Pulses

Sugar beets

Crop yields Land & Water constraints

+200 mm-200 -100 0 +100

CCSR

ECHAM4

Climate change (GCM) Demography

Income and diet

Food demand, production costs

LPJ (50x50 km grid)

Biophysical inputs

Socioeconomic inputs

Bioenergy

Lotze-Campen et al. (2008), Agricultural Economics

Land use pattern

2035

Latin American Modeling Project: The IMAGE Model

February 3, 2012

10

1. The IMAGE modeling Framework

February 3, 2012 Latin American Modeling Project: The IMAGE model

11

Aim: § Dynamic and long-term perspectives on the systemic consequences of global change §  Insights into the impacts of global change §  A quantitative basis for analysing the relative effectiveness of various policy options addressing global change Components: § Socio-Economic System § Earth System § Policy options

2. IMAGE Regional Breakdown

February 3, 2012 Latin American Modeling Project: The IMAGE model

12

Four Latin American regions of which two are individual countries:

1.  Brazil

2.  Rest of South America

3. Mexico

4.  Rest of Central America

3. Reductions in 2020 for different scenarios Non-­‐Annex  I  -­‐  Reductions  -­‐  (GtCO2eq)

-­‐2 0 2 4 6 8

Low  Pledges

High  Pledges

High  Pledges  incl.REDD  in  theCmarket

Two  Degrees

REDD  ReductionsOther  ReductionsSale

Brazil  -­‐Reductions  -­‐  (GtCO2)

-­‐1 0 1 2

REDD  ReductionsOther  Reductions

Sales

February 3, 2012 Latin American Modeling Project: The IMAGE model

13

• 2020 emissions reductions from the pledges are too low to reach 2oC pathways • Non-Annex-1 and Brazil are sellers in all cases looked at • In Brazil, more than 80% of reduction from REDD depending on the scenario • REDD actions are conditional to different mechanisms of financing

4. LAMP model comparison/research (brainstorm)

§  Mitigation options of Latin America countries? –  Land Use –  Energy

›  (BE)CCS ›  Renewables

–  Non-GHG §  Fossil fuel trade balance (exporting/importing) and energy security §  Land use, land use change and forestry dynamics in Latin American

countries? §  Can we provide insight on the competition for land due to REDD,

biofuel and food production in Latin America? §  Integration of current climate policy proposals in the energy and

land use systems of IAMs §  Conditions for REDD to play a major role as mitigation option

February 3, 2012 Latin American Modeling Project: The IMAGE model

14

LAMP

WITCH: model and applications

WITCH key features

•  Model name: WITCH

•  Number of Regions: 14

•  Time Step: 5 yrs

•  Model Type: Optimal growth inter-temporal optimization

•  Features: Endogenous Technical Change. Game theoretic set-up

•  Computing: GAMS. Parallel optimization of regions

WITCH regions

 •  2  new  regions  (India  and  Indonesia)  already  separated  out  •  Plans  to  split  South  America:  Brasil,  and  maybe  one  more  divison.  

WITCH analysis in the region

CARRIED  OUT  SO  FAR  •  Assessment  of  forestry  (both  deforestaJon  and  afforestaJon)  as  a  

miJgaJon  opJon  •  Specific  analysis  into  REDD  using  different  data  sources  on  REDD  

availability  and  costs  (IIASA,  Wood  Hole,  Ohio  State  univ.)  and  different  policy  regimes  (e.g  with  and  without  internaJonal  banking  of  permits)  

INTERESTED  IN  CARRYING  OUT  •  InnovaJon  and  technical  change  in  clean  energy  and  bio-­‐energy  •  Regional  South  America  carbon  trading  mechanisms  •  Further  analysis  on  REDD    

The MESSAGE Model

A Brief Overview and Some Recent Applications

Peter Kolp, Volker Krey, David McCollum, Yu

Nagai, Shonali Pachauri, Shilpa Rao, Keywan Riahi, Manfred Strubegger, Oscar van Vliet

International Institute for Applied Systems Analysis (IIASA)

LAMP Kick-off Meeting, 17-19 January 2012, Belo Horizonte, Brazil

IIASA Integrated Assessment Framework

20/21

Scenario Storyline • Economic development • Demographic change • Technological change • Policies

Population

Economy

DIMA Forest Management Model

AEZ-BLS Agricultural Modeling Framework

Downscaling Tools Spatially explicit and national scenarios

MESSAGE-MACRO Systems Engineering / Macro-Economic Modeling Framework (all GHGs and all sectors)

National, regional & spatially explicit socio-economic drivers

Spatially explicit socio-economic drivers

Consistency of land-cover changes (spatially explicit maps of agricultural, urban, and forest land)

Potential and costs of forest bioenergy and sinks

Carbon and biomass price

Feedbacks

Agricultural bioenergy potentials and costs Drivers for land-use related non-CO2 emissions

Feedbacks

Global and Regional Scenarios

CLIMATE and ACIDIFICATION IMPACT MODELS

NATIONAL POLICY MODELS (GAINS)

Emissions Emissions & Abatement Costs

The MESSAGE Model ●  11 World Regions ●  Time horizon: 1990-2100 in 5/10 year steps ●  Systems engineering model of the global energy system ●  Soft-linked with IIASA Agriculture and Forestry models ●  Includes ~400 individual energy conversion and end-use

technologies ●  Inter-temporal and recursive-dynamic (myopic) versions ● Cost optimization approach ● Demand response from aggregated macroeconomic

model

Regional Representation in MESSAGE

NAM PAO

WEU

EEU

FSU

MEA

AFR

LAM

SAS

PAS

CPA

1 NAM North America 2 LAM Latin America & The Caribbean 3 WEU Western Europe 4 EEU Central & Eastern Europe

5 FSU Former Soviet Union 6 MEA Middle East & North Africa 7 AFR Sub-Saharan Africa 8 CPA Centrally Planned Asia & China

9 SAS South Asia 10 PAS Other Pacific Asia 11 PAO Pacific OECD

OECD

REFS

ALM

ASIA Breaking out Brazil from Latin America Region planned

CCT

gas well Extraction Treatment

Conversion Technologies

Distribution Technologies

Final Energy

End-Use Technologies

Energy Services

coal mine

oil well

agro- forestry

gas coal bio- mass oil

power plant

refin- ery

grid/ truck grid truck

gas elect- ricity

kero- sene

air craft, light bulb, furnace, air conditioner, oven, automobile etc.

synfuel plant

grid/truck

synthetic fuel

sun- light

hydrogen plant

hydrogen

grid/ on site

Energy Conversion Sector

Primary Sources

MESSAGE: Reference Energy System

Supply Side Illustration: Low-Carbon Technologies

●  Biomass Biofuels (2nd generation) Power, CHP, Heat Hydrogen Production

●  Hydro power ●  Wind power (onshore) ●  Solar

PV CSP Solar Thermal (heat/water)

●  Geothermal Power, CHP, Heat

●  Carbon Capture and Storage (CCS) Coal- and Gas-fired Power Plants + CHP Fossil Synfuel Production Fossil H2 Production Cement Production

●  Nuclear Energy Light Water Reactors Fast Breeder High Temp. Reactor

+CCS

in selected scenarios

●  Natural Gas: 3 conv. + 5 unconv. categories

price response (e.g. correction of AEEI)

Iteration

Demand Side Illustration: Transportation Sector and Service Demand Response

MESSAGE

energy supply

structure

coal

natural gas IC

Gasoline/Diesel IC

fuel oil IC

ethanol IC

methanol IC

electricity

hydrogen FC

ethanol FC

methanol FC demand for

transportation

Shares and growth rates constrained

"backstop"

modal split, transitions and behavioral changes implict in

demand projection (scenario-specific)

Scenario Generator

relative efficiencies

MACRO

Prices

IC = Internal Combustion Engine FC = Fuel Cell

(2070)

• Forest land-cover is decreasing • However, deforestation is slowing down (increasing affluence - trend is more rapid in other scenarios)

RCP8.5 – Forest Cover

Source: Riahi et al., 2011, Climatic Change

0,0

0,6

1,2

1,8

2,4

3,0

3,6

4,2

Base  2005 No  new  policies  2030 50%  Fuel  Subsidy   2030

Microfinance  @15%   only  2030

Microfinance  @15%   +  50%  Fuel  Subsidy  

2030

Popu

la*o

n  in  Billions

Solid  Fuel  Users  Sub -­‐ Saharan  Africa Solid  Fuel  Users  Pacific  Asia Solid  Fuel  Users  South  Asia Liquid  Fuel  Users  Sub -­‐ Saharan  Africa Liquid  Fuel  Users  Pacific  Asia Liquid  Fuel  Users  South  Asia

Access  to

 Mod

ern  

Fuels

Solid

 Fue

l  Users

Alternative Policies to Overcome Energy Poverty

Source: Riahi et al., forthcoming, Global Energy Assessment, Methodology: Ekholm et al., 2010, Energy Policy

0.0%

0.2%

0.4%

0.6%

0.8%

1.0%

1.2%

Only  Energy  Security Only  Air  Pollution  and  Health Only  Climate  Change All  Three  Objectives

Total  G

lobal  Policy  C

osts  (201

0-­‐2030)

All objectives fulfilled at Stringent level

At least one objective fulfilled at Intermediate level

At least one objective fulfilled at Weak level

Added costs of ES and PH are comparatively low when CC is taken as an entry point

Source: McCollum et al., 2011, Nature Climate Change

An integrated approach to energy sustainability: Synergies of Multiple Objectives

Stringent

Intermediate

Weak

Fulfillment of Objective

Steven Rose srose@epri.com

LAMP 1st Meeting, Brazil January 17-19, 2012

MERGE & LAMP A Few Thoughts

30 © 2012 Electric Power Research Institute, Inc. All rights reserved.

MERGE

• Like other IAMs w/ coupled climate-economic systems

• Intertemporally optimizing

• Top-down general equilibrium representation of the economy, with nested production structure

• Bottom-up process representation of energy technologies

• Flexible regional aggregation (w/ four Central & South American regions – Brazil, Mexico, rest of middle-income, rest of low-income)

31 © 2012 Electric Power Research Institute, Inc. All rights reserved.

Interesting areas of study for LAMP

• Potential near-term development pathways

• Current energy systems and transformation constraints

• International climate policy interactions with development objectives

Thank You!