Transcript prCalvin - IEW Workshop

Asia
Modeling
Exercise
The Role of Asia in Mitigating
Climate Change: Results from the
Asia Modeling Exercise
Kate Calvin, Leon Clarke, Volker Krey, Geoff Blanford, Jiang
Kejun, Mikiko Kainuma, Elmar Kriegler, Gunnar Luderer, P.R.
Shukla
International Energy Workshop 2012
Cape Town, South Africa
June 21, 2012
PNWD-SA-9886
Goals of AME
Asia
Modeling
Exercise
 Objective: to better articulate the role of Asia in
addressing climate change.
 Goal: To bring together global modelers that
commonly participate in efforts to explore international
policy architectures with regional modelers and
experts with Asia-specific knowledge, understanding,
data, and analysis.
 Method: A coordinated modeling exercise that
attempts to link these communities to provide more
effective modeling and analysis of Asia within a global
context.
Participants
Asia
Modeling
Exercise
 26 Participating Models
 Australia (GTEM)
 Canada (TIAM-World)
 China (China MARKAL, IAMC, IPAC, PECE),
 EU (GEM-E3, IMAGE, MESSAGE, POLES-IPTS, REMIND,
TIMES-VTT, WITCH)
 India (GCAM-IIM)
 Japan (AIM-CGE, AIM-Enduse, DNE21+, GRAPE, MARIA23)
 Korea (KEI-Linkages)
 Nepal (Nepal MARKAL)
 United States (EPPA, GCAM, iPETS, MERGE, Phoenix)
The Models
Asia
Modeling
Exercise
 Models differ with respect to:





Regional scope (Global, China only, Nepal Only)
Time horizon (through 2100, through 2050)
Degree of foresight (myopic, intertemporally optimizing)
Underlying structure (market-equilibrium, cost minimization)
Sectoral coverage (Energy only, Energy & Agriculture/Land-Use,
Full Economy)
 Emissions included (CO2 only, Kyoto gases only, all species)
 Climate representation (No representation, GHG concentrations
only, all radiative forcing agents)
Exercise Design
 Six Core Scenarios:
 Baseline
 3 CO2 price paths
3000
2005$/tCO2
2500
2000
These scenarios were
used to link between the
global and regional
models.
1500
1000
500
0
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Asia
Modeling
Exercise
Exercise Design
Asia
Modeling
Exercise
 Six Core Scenarios:
 Baseline
 3 CO2 price paths
 2 Stabilization paths (global models only)
 550 CO2-e stabilization (total forcing)
 450 CO2-e overshoot (total forcing)
 For models without all forcing agents, we provided exogenous
paths that they could use.
Exercise Design
 Six Core Scenarios:
 Baseline
 3 CO2 price paths
 2 Stabilization paths (global models only)
 All policies are first-best (immediate accession,
economy-wide CO2 prices/constraints)
 No harmonized variables in the core scenarios
Asia
Modeling
Exercise
Asia
Modeling
Exercise
Variation in AME Baselines
Median and Range Across Models in 2100
120
1800
1600
40
140
35
120
100
1400
1200
1000
30
100
80
20
80
ppmv
800
60
GtCO2/yr
1000
800
25
GtCO2/yr
trillion 2005$/yr
million people
1200
60
15
40
600
600
400
40
10
400
20
5
200
0
0
China Population
China GDP
0
China CO2 Emissions
20
0
Global CO2 Emissions
200
0
CO2 Concentration
Exercise Design
Asia
Modeling
Exercise
 We created several subgroups, which allowed us to
explore different aspects of the scenarios more in depth.
 Subgroup topics:
 Base Year Data
 Baseline Scenarios
 Urban/Rural development
 Technology and Technical Change
 Global and Regional Mitigation Efforts
 National Policies and Measures
 Low Carbon Societies
Exercise Design
Asia
Modeling
Exercise
 We created several subgroups, which allowed us to
explore different aspects of the scenarios more in depth.
 Subgroup topics:
 Base Year Data
 Baseline Scenarios
 Urban/Rural development
 Technology and Technical Change
 Global and Regional Mitigation Efforts
 National Policies and Measures
 Low Carbon Societies
RESULTS
Asia
Modeling
Exercise
Base Year Data
GDP
Population
12%
India
China
Japan
Korea
40%
9%
30%
6%
20%
3%
10%
0%
0%
-3%
-10%
-6%
-20%
-9%
-30%
-12%
-40%
India
China
Japan
Korea
Deviation from WB
Deviation from UN 2010
AIM-CGE
AIM-Enduse
China-MARKAL
DNE21
EPPA
GCAM
GCAM-IIM
GEM-E3
GRAPE
GTEM
IMAGE
iPETS
KEI-Linkages
MARIA-23
MERGE
MESSAGE
Phoenix
POLES-IPTS
REMIND
TIAM-WORLD
TIMES-VTT
WITCH
Asia
Modeling
Exercise
Base Year Data
Korea
IEA-2007
Japan
IEA-2010
India
WB
China
National
Stats/OECD
-6%
-4%
-2%
0%
2%
4%
Asia
Modeling
Exercise
Base Year Data
GDP
Population
12%
India
China
Japan
Korea
40%
9%
30%
6%
20%
3%
10%
0%
0%
-3%
-10%
-6%
-20%
-9%
-30%
-12%
-40%
India
China
Japan
Korea
Deviation from WB
Deviation from UN 2010
AIM-CGE
AIM-Enduse
China-MARKAL
DNE21
EPPA
GCAM
GCAM-IIM
GEM-E3
GRAPE
GTEM
IMAGE
iPETS
KEI-Linkages
MARIA-23
MERGE
MESSAGE
Phoenix
POLES-IPTS
REMIND
TIAM-WORLD
TIMES-VTT
WITCH
Asia
Modeling
Exercise
Base Year Data
Primary Energy
30%
China
India
Japan
Total CO2 Emissions
Korea
24%
India
China
Japan
Korea
18%
20%
12%
10%
6%
0%
0%
-10%
-6%
-20%
-12%
-30%
-18%
-24%
-40%
Deviation from IEA
Deviation from CDIAC
AIM-CGE
AIM-Enduse
China-MARKAL
DNE21
EPPA
GCAM
GCAM-IIM
GEM-E3
GRAPE
GTEM
IMAGE
iPETS
KEI-Linkages
MARIA-23
MERGE
MESSAGE
Phoenix
POLES-IPTS
REMIND
TIAM-WORLD
TIMES-VTT
WITCH
Base Year Data
Asia
Modeling
Exercise
 Key findings:
 There are some good reasons why base year data
differs across models. Examples include:




Differences in region definition
Differences in data sources
Differences in modeled base year
Differences in calibration method
 While differences in base year data do affect future
growth projections, differences in assumed growth
rates have a much larger impact on the future.
Baseline Scenarios
Asia
Modeling
Exercise
Asia
Modeling
Exercise
Baseline Scenarios
Average Growth Rates in China, 2005 – 2020 with comparison to Asian history
Per Capita Income Growth Rate
0%
1%
2%
3%
4%
5%
6%
7%
8%
1%
Malaysia
1979-1994
Energy Intensity Rate of Change
0%
-1%
-2%
-3%
-4%
-5%
-6%
Model projections
EIA projection
1990 – 2005 data
Korea
Japan
197719591992
1974
Taiwan 1973-1988
9%
Baseline Scenarios
Average Growth Rates in China, 2005 – 2020
Asia
Modeling
Exercise
Baseline Scenarios
Asia
Modeling
Exercise
 Key findings:
 Models differ in their projections of economic growth,
energy intensity, and carbon intensity
 Differences in underlying growth assumptions result
in a factor of 2 difference in Chinese CO2 emissions
across models in 2020
 Models with similar emissions levels may achieve
them in very different ways
 The models do not span the full uncertainty range.
This is merely the range of modelers’ “best guesses.”
Asia
Modeling
Exercise
Variation in AME Policy Case
Median and Range Across Models in 2100
1800
100
1600
90
700
30
10
600
25
80
1400
8
600
40
4
15
ppmv
50
GtCO2/yr
800
6
60
GtCO2/yr
trillion 2005$/yr
1000
500
20
70
1200
million people
35
12
10
300
5
2
0
30
200
0
400
-5
20
200
10
0
0
China Population
400
100
-2
-10
-15
-4
China GDP
China CO2
Emissions
0
Global CO2
Emissions
Assumes a carbon price of $30/tCO2 in 2020, rising at 5% p.a.
CO2 Concentration
Global & Regional Mitigation
Global Marginal Abatement Cost Curves, 2005 – 2050
Asia
Modeling
Exercise
Global & Regional Mitigation
Fossil fuel & industrial CO2 emissions in 2050, relative to baseline
Asia
Modeling
Exercise
Global & Regional Mitigation
Asia
Modeling
Exercise
Fossil fuel & industrial CO2 emissions reductions in 2050, relative to world
Oil Total
Coal w/o CCS
Gas w/ CCS
Nuclear
Solar
DNE21
60
40
20
0
EPPA
Oil w/o CCS
Coal w/ CCS
Biomass Total
Non-Bio Renewable Total
Geothermal
MERGE
POLES-IPTS
Phoenix
MESSAGE
REMIND
WITCH
TIMES-VTT
China
India
Japan
OECD90
China
India
Indonesia
Japan
Korea
OECD90
POLES-IPTS
REMIND
0.4
0.2
0.0
TIMES-VTT
WITCH
China
India
OECD90
TIAM-WORLD
Coal Total
Gas w/o CCS
Biomass w/ CCS
Wind
China
India
Indonesia
Japan
Korea
OECD90
0.0
2005
China
India
Indonesia
Japan
Korea
OECD90
0.2
IMAGE
China
India
Indonesia
Japan
OECD90
0.4
GTEM
China
India
Japan
OECD90
0.6
GRAPE
China
India
Japan
0.8
GEM-E3
China
India
Japan
OECD90
China
India
Indonesia
Japan
Korea
OECD90
Oil w/ CCS
Gas Total
Biomass w/o CCS
Hydro
Other
TIAM-WORLD
GCAM
China
India
Japan
Korea
OECD90
China
India
OECD90
EPPA
MARIA-23
PHOENIX
IMAGE
GTEM
GRAPE
GEM-E3
GCAM
DNE21+
China
India
Japan
Korea
OECD90
China
India
Indonesia
Japan
Korea
OECD90
MESSAGE
AIM-Enduse
0.6
MERGE
China
India
Indonesia
Japan
Korea
OECD90
AIM-ENDUSE
China-MARKAL
0.8
MARIA-23
China
India
OECD90
China
India
Indonesia
Japan
Korea
OECD90
AIM-CGE
AIM-CGE
1.0
China
India
Japan
OECD90
China
India
Indonesia
Japan
Korea
OECD90
1.0
China
India
Japan
OECD90
Technology
Asia
Modeling
Exercise
Electricity Generation by model, region, and fuel in 2050 in $30/tCO2 Scenario
2005
Global & Region Mitigation
Asia
Modeling
Exercise
 Key findings:
 Models differ significantly in the amount of mitigation
achieved for a particular carbon price
 Some regions show less mitigation than others,
regardless of the model considered
 Differences in mitigation are due to a variety of
factors, including:
 Differences in baseline emissions levels
 Differences in model flexibility
 Differences in technology cost and availability
Technology
Asia
Modeling
Exercise
 Key findings:
 Models show a wide variety of future energy systems
across time and scenarios.
 Variation is due to differences in assumed technology
cost, resource availability, etc.
 While there is some variation across regions within a
model due to resource constraints, many models tend
to “favor” certain technologies.
Other Analyses
Asia
Modeling
Exercise
 Urban/Rural Development:
 Analyzed the effect of urbanization on energy use and emissions
 Finding: Urbanization has an effect on solid fuel consumption,
but may not strongly influence total CO2 emissions
 National Policies & Measures:
 Compared results from the models to Copenhagen pledges and
MEF/G8 goals
 Finding: Stringency of Copenhagen pledges varies across
regions, and to a lesser extent across models
 Low Carbon Societies:
 Assessed policies and measures needed to implement 2 degree
scenarios
Summary
Asia
Modeling
Exercise
 The Asia Modeling Exercise brought together more than
20 energy-economy and integrated assessment models.
 These models ran a set of coordinated scenarios.
 We focused our analysis of the results on Asian regions.
 We analyzed results across a variety of dimensions,
including base year data, baselines, global & regional
mitigation, technology, and national policies & measures.
 We find that models differ significantly across a number
of variables, reflecting uncertainty in the future evolution
of the world’s economy and energy system. However,
there were some robust results across models.
Asia
Modeling
Exercise
THANK YOU!