International Institute for Applied Systems Analysis (IIASA)

Report Part Title: A model intercomparison

Report Title: GHG mitigation potentials in Annex I countries Comparison of model
estimates for 2020
Report Author(s): Markus Amann, Peter Rafaj and Niklas Höhne
International Institute for Applied Systems Analysis (IIASA) (2009)

Stable URL: http://www.jstor.com/stable/resrep15767.4

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

International Institute for Applied Systems Analysis (IIASA) is collaborating with JSTOR to
digitize, preserve and extend access to this content.

This content downloaded from

84.125.71.194 on Sat, 10 Aug 2024 10:04:24 UTC
All use subject to https://about.jstor.org/terms
2 A model intercomparison
In March 2009, IIASA invited key modelling teams that have provided estimates of GHG
mitigation potentials and costs to participate in the study and to submit data for the comparison.
A meeting was held at IIASA on May 28-29, 2009 to review the results from different models
and identify factors that explain differences in model estimates.

2.1 Participating models

Eight modelling teams have provided quantitative results to the intercomparison exercise (Table
2.1).
Table 2.1: Participating models

Model Organization Model type Main reference

AIM NIES, Japan Bottom up model Kainuma M. et al., 2007

DNE21+ RITE, Japan Bottom-up model RITE, 2009
GAINS IIASA, Austria Bottom-up model Amann et al., 2008
GTEM Treasury, Australia Computable general Australian Treasury, 2008
equilibrium model
IMAGE PBL, Netherlands Bottom-up integrated MNP, 2006
assessment model
McKinsey McKinsey Bottom-up cost curves McKinsey & Company,
2009
OECD ENV- OECD Computable general OECD, 2009
LINKAGES equilibrium
POLES IPTS Linked bottom-up/top down Russ et al., 2009

2.1.1 AIM (NIES, Japan)

The AIM model, developed by the National Institute for Environmental Studies (NIES), Japan,
comprises three main models - the greenhouse gas emission model (AIM/emission), the global
climate change model (AIM/climate), and the climate change impact model (AIM/impact). The
AIM/emission model estimates greenhouse gas emissions and assesses policy options to reduce
them. The AIM model has several distinct characteristics. It integrates emission, climate and
impact models, contains country modules for detailed evaluations at the national level
and global modules to ensure consistency across individual modules, integrates bottom-
up national modules with top-down global modules, and is designed to assess
alternative policies. AIM contains a very detailed technology selection module to
evaluate the effect of introducing advanced technologies and uses information from a

This content downloaded from

84.125.71.194 on Sat, 10 Aug 2024 10:04:24 UTC
All use subject to https://about.jstor.org/terms
detailed Geographic Information System to evaluate and represent the distribution of
impacts at the local level. More detail is provided in Kainuma M. et al., 2007 and at
http://www-iam.nies.go.jp/aim/infomation.htm.

2.1.2 DNE-21+ (RITE, Japan)

The Dynamic New Earth 21 plus (DNE21+) model has been developed by the Research
Institute of Innovative Technology for the Earth (RITE), Japan. The model covers the entire
world divided over 50 regions. The energy systems model is a bottom-up linear programming
model minimizing world total costs of energy systems. DNE21+ also treats energy-unrelated
CO2 and five kinds of non-CO2 GHG emissions. The non-CO2 GHG model is a proxy model
using elasticities that represent bottom-up assessments of mitigation technologies performed by
USEPA. More information is provided in RITE, 2009.

2.1.3 GAINS (IIASA, Austria)

The Greenhouse gas – Air pollution Interactions and Synergies (GAINS) model has been
developed by the International Institute for Applied Systems Analysis (IIASA), Austria. It uses
a bottom-up approach for quantifying GHG mitigation potentials and costs for the major
Annex I countries, and estimates co-benefits on air pollution. GAINS employs exogenous
activity projections, currently those of the IEA World Energy Outlooks 2007 and 2008 (IEA,
2007, IEA, 2008) . More information is provided in Amann et al., 2008. An interactive version
of GAINS is accessible on the Internet (http://gains.iiasa.ac.at/).

2.1.4 GTEM/MMRF (Australia)

GTEM is a recursively dynamic general equilibrium model developed by the Australian Bureau
of Agricultural and Resource Economics (ABARE) to address policy issues with long-term
global dimensions, such as climate change mitigation costs.
The MMRF model is a detailed model of the Australian economy developed by the Centre of
Policy Studies at Monash University. It is a dynamic model which employs a recursive
mechanism to explain investment and sluggish adjustment in factor markets.
The marginal GHG abatement cost curves for the GTEM and MMRF models are not produced
or derived internally by the models. The information provided by GTEM and MMRF are
abatement curves, which shows the amount of abatement that occurs at the average carbon
price. An abatement curve can differ from a marginal abatement cost curve, due to different
assumptions, environmental targets and emission trajectories.

2.1.5 IMAGE (PBL, Netherlands)

The IMAGE 2.4 Integrated Assessment model (MNP, 2006) (www.mnp.nl/image) consists of a
set of linked and integrated models that together describe important elements of the long-term
dynamics of global environmental change, such as air pollution, climate change, and land-use
change. The global energy model that forms part of this framework, TIMER (van Vuuren et al.,

This content downloaded from

84.125.71.194 on Sat, 10 Aug 2024 10:04:24 UTC
All use subject to https://about.jstor.org/terms
2007), describes the demand and production of primary and secondary energy and the related
emissions of GHGs and regional air pollutants. The FAIR-SiMCaP 2.0 model is a combination
of the abatement costs model of FAIR and the SiMCaP model (den Elzen et al., 2007). The land
and climate modules of IMAGE describe the dynamics of agriculture and natural vegetation,
and, together with input from TIMER and FAIR, resulting climate change.

2.1.6 McKinsey
The global McKinsey GHG abatement cost curve was developed since 2006 and results in this
paper are based on the second version of the global GHG abatement cost curve (McKinsey
2009). The model is mainly based on external baseline sources IEA WEO, US EPA and
Houghton and assesses bottom-up the abatement potential and cost of over 200 abatement levers
in 21 world regions. More information and the online version Climate Desk is accessible on the
Internet (http://solutions.mckinsey.com/climatedesk).

2.1.7 OECD ENV LINKAGES (OECD)

ENV-Linkages is a top-down model (CGE type). This model is still in development, the version
used for the paper is the version 2.1. The ENV-Linkages model is a recursive dynamic neo-
classical general equilibrium model, with a standard time horizon from 2005 to 2050. It is a
global economic model built primarily on a database of national economies.
The model version used for this study represents the world economy in 12 countries/regions,
each with 25 economic sectors (eight energy production sectors), and three representative
agents. Six greenhouse gases are modeled; land use and land cover change emissions are not yet
taken into account. Capital accumulation is modeled as in traditional Solow/Swan neo-classical
growth models.
All production in ENV-Linkages is assumed to operate under cost minimization with an
assumption of perfect markets and the CRS technology. The production technology is specified
as nested CES production functions in a branching hierarchy. Total output for a sector is
actually the sum of two different production streams: resulting from the distinction between
production with an ‘‘old’’ capital vintage, and production with a ‘‘new’’ capital vintage. The
substitution possibilities among factors are assumed to be higher with new capital than with old
capital. International trade flows and prices are fully endogenous and modeled using a
Armington specification. Energy efficiency is partly exogenous, as the autonomous energy
efficiency (AEE) factor is calibrated to match IEA’s projections on energy demand published in
the World Energy Outlook), and partly endogenous with substitution possibility between factors
and goods resulting from prices changes and optimizing behavior of agents. For each year the
government budget is balanced through the income tax, revenues of the carbon tax are then
indirectly rebated to the household, in a lump-sump way since labor supply is exogenous.

2.1.8 POLES (JRC-IPTS, EU)

POLES is a global simulation model of the energy system. The dynamics of the model is based
on a recursive simulation process of energy demand and supply with lagged adjustments to

This content downloaded from

84.125.71.194 on Sat, 10 Aug 2024 10:04:24 UTC
All use subject to https://about.jstor.org/terms
prices and a feedback loop through the international energy price. The model is developed in the
framework of a hierarchical structure of interconnected modules at the international, regional,
and national levels. It contains technologically-detailed modules for energy-intensive sectors,
including power generation, iron and steel, the chemical sector, aluminum production, cement
making, non-ferrous minerals and modal transport sectors (including aviation and maritime
transport). All energy prices are determined endogenously. Oil prices in the long-term depend
primarily on the relative scarcity of oil reserves. The world is broken down into 47 regions, for
which the model delivers detailed energy balances. The model is continuously being enhanced
in both detail and in the degree of regional disaggregation. Recent modifications include the
addition of detailed modules for energy-intensive sectors and an extension to cover non-CO2
greenhouse gases (GHG).

2.2 Data provided for the model intercomparison

As an input for the quantitative model intercomparison, modelling teams provided a set of data
to IIASA that describe sectoral GHG emissions that emerge for a range of carbon prices (i.e.,
for the base year 2005, for the baseline case in 2020, and for 2020 with carbon prices of 0, 20,
50, 100 and >100 US-$/t CO2, respectively.) Such data were delivered for individual Annex I
parties and for Annex I in total.
It is important to note that only the GTEM model provided data for the LULUCF sector.
As not all models cover all countries, not all teams provided a complete set of data:

• AIM: No data have been provided for Canada and Australia.

• IMAGE: Australia and New Zealand have been aggregated into one region, and sectoral
emissions are not included in the provided data. The IMAGE emissions data are not
harmonised with the UNFCCC emissions, and comes directly from the different
IMAGE submodels, which are calibrated for the year 2000. In policy applications with
the IMAGE and FAIR model harmonised data is used.

• OECD: Australia and New Zealand have been aggregated into one region.

• McKinsey: No data have been provided for Australia separately.

• GTEM calculations include LULUCF emissions; for Australia, results of the MMRF
model were provided as well.

• POLES did not provide data for the Australia, New Zealand and Canada. Furthermore,
POLES data do not include emissions from agriculture.

• GAINS and POLES data were recalculated from € to US-$.

This content downloaded from

84.125.71.194 on Sat, 10 Aug 2024 10:04:24 UTC
All use subject to https://about.jstor.org/terms