Working Group III: Mitigation

Other reports in this collection

2.3 Greenhouse Gas Emissions: General Mitigation Scenarios

This chapter reviews three scenario literatures, which span a range from more quantitative scenario analysis to analysis that is based more on narrative descriptions (see Figure 2.1). At the quantitative end of the spectrum are the “general mitigation scenarios” reviewed in this section, which consist mainly of quantitative descriptions of driving forces and emission profiles.

2.3.1 Overview of General Mitigation Scenarios

More than 500 emission scenarios have already been quantified, including non-mitigation (non-intervention) scenarios and mitigation (intervention) scenarios that assume policies to mitigate climate change. These scenarios have been published in the literature or reported in conference proceedings, and many of them were collected in the IPCC SRES database (Morita & Lee, 1998a) and made available through the Internet (Morita & Lee, 1998b). Using this database, a systematic review of non-mitigation scenarios has already been reported in the SRES (Nakicenovic et al., 2000). However, several mitigation and other scenarios were missing from this database and new emission scenarios have been quantified since the SRES review. Accordingly, the missing scenarios and new scenarios were collected and the database revised for this new review of mitigation scenarios (Rana and Morita, 2000).

The current database collection, covered in this report, contains the results of a total of 519 scenarios from 188 sources. These scenarios were mainly produced after 1990. Two questionnaires were sent to representative modellers in the world, and sets of scenarios from the International Energy Workshop (IEW) and Energy Modelling Forum (EMF) comparison programmes were collected. The database is intended to include only scenarios that are based on quantitative models. Therefore, it does not include scenarios produced using other methods; for example, heuristic estimations such as Delphi.

Of the 519 scenarios, a total of 380 were global GHG emission scenarios, most of which were disaggregated into several regional emission profiles. Of these 380 global emission scenarios, a total of 150 were mitigation (climate policy) scenarios. This review focuses on mitigation scenarios that cover global emissions and also have a time horizon encompassing the coming century. Of the 150 mitigation scenarios, a total of 126 long-term scenarios that cover the next 50 to 100 years were selected for this review. 24 scenarios were excluded on the basis of their short time coverage.

Table 2.1 presents an outline of several representative scenarios in this review; these scenarios exemplify the modelling literature. Columns 1 and 2 of the table show the main identifiers of the scenarios, namely, the model name and source and the policy scenario name, as given by the modellers. The third and fourth columns show the policy scenario type and specific scenario assumptions. The remaining columns contain additional important features of the policy scenarios, including reduction time-paths and burden sharing, GHGs analyzed, policy options and approaches, and feedback. Only five studies among the selected sources of Table 2.1 have detailed policies. Most of the other scenarios assume very simple policy options such as carbon taxes and simple constraints.

Table 2.1: Overview of mitigation scenarios: the main futures of representive scenarios from 26 sources
Model name and source
Policy scenario name
Policy scenario type
Specific scenario assumptions2
Reduction time paths and burden sharing
GHGs analyzed
Sectors in which policies are introduced
Emission stab.
Based on policy scenario
CO2, CO, CH4, N2O
Detailed policy scenario
EP to M
   EPA (1990)
Emission stab.
Energy supply; Land use; End use
EP to M
Control policy (2x CO2 by 2090)
Conc. stab.
Based on policy scenario
CO2, CO, CH4, N2O
Detailed policy scenario EP to M
   IPCC (1990)
Accelerated control ( < 2x CO2)
Conc. stab.
Energy supply; Land use; End use  
Other mitigation1
CO2, CO, CH4, N2O,
Energy supply; Industrial processes EP to M
   IPCC (1992)
(CO2 emissions)
Other mitigation1
Energy supply; Industrial processes;  
   Nakicenovic et al. (1993)    
End use  
   Nordhaus (1994)
Optimal policy; Other mitigation1
Utility maximization
Energy C to M
10- yr delay of optimal policy Other mitigation1
Utility maximization
Other GHGs are
  emission stabilization Emission stab. 8BtC/ yr
(CO2 +CFCs)
Based on policy scenario
  I to M
  20% emission cut Other mitigation1 6BtC/ yr
(CO2 +CFCs)
Based on policy scenario
  C to M
  Geoengineering Other mitigation   Based on policy scenario
  I to M
  Climate stabilization Slow global temp. increase 0.2° C/ decade Based on policy scenario
6 CETA “Selfish” case Emissions cont. by OECD   Cost minimization (regional)
CO2, CO, CH4, N2O,
Energy C to M
   Peck and Tiesberg (1995) “Altruistic” case Emissions cont. by OECD   Cost minimization (global)
  “Optimal” case Emissions cont. by both   Cost minimization (global)
   (IPCC, 1996)
LESS Constructions Other mitigation1    
Energy supply  
8 Manne et al. (1995) Delayed tax; Early tax Other mitigation1 750 ppm; 540 ppm Utility maximization
CO2, CH4, N2O
Energy C to M
   MERGE Emission stab. Emission stab. 540 ppm Utility maximization
  Conc. stab. Conc. stab. 415 ppm Utility maximization
9 MESSAGE Case C Other mitigation1 430 ppm Based on policy scenario
CO2, CO, CH4, N2O,
Energy supply  
WEC (1995) Ecologically driven      
End use  
10 WBGU (1995)
     (German Adv. Council)
Tolerable temp. window Safe corridor temp. rise constraint deltaT = 1° C (upper limit) Temp. rise constraint
11 AIM/ Top- down Negotiable safe emiss. corridor Safe corridor deltaT = 1- 2° C Temp. rise constraint
Energy EP to M
     Matsuoka et al. (1996)   Temperature rise const.    
12 DICE/ RICE Cooperative RICE Other mitigation1   Global welfare optimization
Energy; Land use C to M
     Nordhaus and Yang (1996) Non- cooperative RICE Other mitigation1 Regional welf. optimization  
  I to M
13 IMAGE 2 Stab 350– 650 ppm Conc. stab. 367– 564 ppm  
CO2, CH4, N2O
Energy supply; Industrial processes;  
     Alcamo and Kreileman (1996) Stab yr 1990 Conc. stab. 354 ppm  
Land use  
  St2000-a - St2000-e Other mitigation1 633– 433 ppm  
  Safe emissions corridor Safe corridor deltaT = 1– 2° C deg Temp. rise constraint
14 MiniCAM Adv. tech Other mitigation1   Based on policy scenario
CO2, CH4, N2O, SOx, aerosols,
Energy supply EP to M
     Edmonds et al. (5 Cases using different technologies)      
     Yohe and
Wallace (1996)
Stabilization Conc. stab.   Based on policy scenario
Energy C to M
     Ha- Duong, et al.
450A- D; 550A- D; 650A Conc. stab.   Cost minimization
17 FUND 1.6 Non- cooperative optimum Other mitigation1   Regional welf. optimization
CO2, CH4, N2O
     Tol (1997) Cooperative optimum Other mitigation1   Generational
welf. optim.
18 MERGE 3.0
     Manne and Richels (1997)
Range of scenarios
350 to 750 ppm
Conc. stab. 350 to 750ppm depending on scenario Utility maximization (non- Annex I begin limit in 2030)
Energy C to M
19 SGM
     Edmonds et al. (1997)
M1990 ; M1990+ 10%; M1990– 10%; M1995 Other mitigation1    
CO2 , CO, CH2 , N2O, NOx, VOC, SOx.
Energy EP to M
     Tulpule et al. (1998)
Independent abatement; Annex B trading; Double bubble Other mitigation1 Kyoto targets  
Energy C to M
21 AIM/ Top- down
     Kainuma et al. (1998)
No trading; Annex I Trading; Global trading; Double bubble;
Annex I + Chn& Ind;
No trading 5% offset
Other mitigation1 Kyoto targets Based on policy scenario
  C to M
22 G- CUBED Annex I trading; Double bubble Other mitigation1    
Energy C to M
     McKibbin (1998) Global permit trade      
23 MARIA Case B Emission stab. 1990 level  
Energy supply; End use; Land use C to M
     Mori and
Takahashi (1998)
24 NE21 Conc. regulation Conc. stab. Below 550ppm Cost minimization
Energy C to M
     Fujii and Yamaji
25 WorldScan No Trade; Full trade; Clubs Other mitigation1 Kyoto targets  
  EP to M
     Bollen et al. (1996) Restricted trade; CDM      
26 FUND 1.6
     Tol (1999)
EMF- 14 scenarios
450/ 550/ 650+ NC/ C)
Conc. stab. Various Various

1 Other mitigation means emission reduction not necessarily leading to stabilization.
2 Specific scenario assumption indicates year 2100 level of emissions/conc. unless specified otherwise.
3 EP: Energy price; M: Macro economy; C: Cost; I: Impact.
Conc.stab. : concentration of stabilization
Emissions cont.: emissions continuing

Based on the type of mitigation, the scenarios can be classified into four categories: concentration stabilization scenarios, emission stabilization scenarios, safe emission corridor (tolerable windows/safe landing) scenarios, and other mitigation scenarios.

Scenarios for concentration stabilization account for a large proportion of the mitigation scenarios, with 47 of the 126 mitigation scenarios being classified into this type. Many scenarios of this type were quantified in the process of the EMF comparison (Weyant and Hill, 1999) where a systematic guideline was prepared for stabilization quantification. Of the 47 scenarios, two-thirds are intended to stabilize atmospheric concentrations of CO2 at 550ppmv. The concentration of 550ppmv was used as a benchmark for stabilization in the previous studies on mitigation scenarios. This number may be related to the frequent references made to it in political discussions. The adoption by the European Union of a maximum increase in global average temperature of 2°C above pre-industrial levels is roughly equivalent to a stabilization level of 550ppmv CO2 equivalent or 450ppmv CO2. It does not imply an agreed-upon desirability of stabilization at this level. In fact, environmental groups have argued for desirable levels well below 550ppmv, while other interest groups and some countries have questioned the necessity and/or feasibility of achieving 550ppmv. Scenarios with levels of concentration stabilization other than 550ppmv are contained in IPCC (1990), Manne et al. (1995), Alcamo and Kreileman (1996), Ha-Duong et al. (1997), Manne and Richels (1997), and Fujii and Yamaji (1998).

The emission stabilization scenarios account for 20 of the 126 mitigation scenarios. Most scenarios of this type are intended to stabilize at 1990 emission levels in Annex I or the Organization for Economic Co-operation and Development (OECD) countries. Some scenarios have emissions stabilizing at other levels, for example, the emissions stabilization scenario of DICE (Nordhaus, 1994) aims at a level of 8GtC/yr of CO2 and chlorinated fluorocarbons (CFCs) by 2100. Other stabilization scenarios, namely the “Safe Emissions Corridor” or “Tolerable Windows” (WBGU, 1995; Alcamo and Kreileman, 1996; Matsuoka et al., 1996) and “Climate Stabilization” (Nordhaus, 1994) scenarios, determine the upper limit of emissions based on a constraint of some natural threshold, such as global mean temperature increase rate. Only a few studies are based on such scenarios.

Other scenarios based on DICE (Nordhaus, 1994), MERGE (Manne and Richels, 1997) and MARIA (Mori and Takahashi, 1998) determine the level of emission reduction based on net benefit maximization, which is estimated as the benefit produced by climatic policy minus the policy implementation cost. In addition to the above, the low CO2-emitting energy supply system (LESS) constructions should be noted. These scenarios were developed on the basis of detailed assessments of technological potentials, and can therefore be distinguished from many other mitigation scenarios (see Box 2.2).

Box 2.2. Review of Low Carbon Dioxide Emitting Energy Supply System (LESS) Constructions from the Second Assessment Report

The LESS constructions described in the IPCC’s SAR, Working Group II (IPCC, 1996, Ch19), were probably the only constructions akin to mitigation “scenarios” taken up in SAR. They are similar to the mitigation scenarios reviewed in this chapter in that they also explore alternative paths to energy futures in order to achieve mitigation of carbon dioxide.

A number of technologies with potential for reducing CO2 emissions exist or are in a state of possible commercialization. The LESS constructions illustrate the potential for reducing emissions by using energy more efficiently and by using various combinations of low CO2-emitting energy supply technologies, including shifts to low-carbon fossil fuels, shifts to renewable and nuclear energy sources, and decarbonization of fuels. The assumed technological feasibility and costs of each of the technologies included in these variants is based on an extensive literature review.

Both bottom-up and top-down approaches were used in the LESS constructions. For the reference cases in the bottom-up analyses, the energy demand projections for the high economic growth variant of the “Accelerated Policies” scenarios developed by the Response Strategies Working Group (RSWG, 1990) were adopted.

The five variants constructed in the bottom-up analyses were (1) BI: biomass intensive, (2) NI: nuclear intensive, (3) NGI: natural gas intensive, (4) CI: coal intensive, and (5) HD: high demand. The BI variant explores the potential for using renewable electricity sources in power generation. Both intermittent renewables (wind, photovoltaics, and solar thermal-electricity technologies) and advanced biomass electricity-generating technologies (biomass-integrated gasifier and/or gas turbine technologies through 2025 and biomass-integrated gasifier and/or fuel-cell technologies through 2050 and beyond) were applied. The NI variant involves a revitalization of the nuclear energy option and deployment of nuclear electric power technology worldwide. In the NGI variant, the emphasis is on natural gas. Any natural gas in excess of that for the reference cases is used to make methanol (CH4O) and hydrogen (H2). These displace CH4O and H2 produced from plantation biomass. In the CI variant, the strategy for achieving deep reductions involves using coal and biomass for CH4O and H2 production, along with sequestration of the CO2 separated out at synthetic fuel production facilities. Finally, in the HD variant the excess demand is met by providing an extra supply of fuels with low emissions. To illustrate the possibilities, the HD variant is constructed with all of the incremental electricity provided by intermittent renewables.

A top-down exercise was carried out to test the robustness of the bottom-up energy supply analyses by incorporating performance and cost parameters for some of the key technologies in the BI variant. Six technology cases were modelled using the Edmonds–Reilly–Barns (ERB) model. The results for CO2 emissions in two cases (cases 5 and 6) were comparable to the bottom-up LESS variants, but the energy end-uses were different owing to different assumptions.

The central finding of the LESS construction exercise is that deep reductions of CO2 emissions from the energy sector are technically possible within 50 to 100 years, using alternative strategies. Global CO2 emissions could be reduced from about 6GtC in 1990 to about 2GtC in 2100, in many combinations of the options analyzed. Cumulative CO2 emissions, from 1990 to 2100, would range from about 450 to about 470GtC in the alternative LESS constructions. Higher energy efficiency is underscored in order to achieve deep reductions in CO2 emissions, increase the flexibility of supply-side combinations, and reduce overall energy system costs.

Of the remaining mitigation scenarios, a total of 50 adopt other criteria to reduce GHGs. Some of these scenarios assume the introduction of specific policies such as a constant carbon tax, while others assume the Kyoto Protocol targets for Annex I countries up to 2010 and a stabilization of their emissions thereafter at 2010 levels.

While all the scenarios deal necessarily with energy-related CO2 emissions that have the most significant influence on climate change, several models include CO2 emissions from land use changes and industrial processes (e.g., IPCC, 1992; Nakicenovic et al., 1993; Matsuoka et al., 1995; Alcamo and Kreileman, 1996). Some of them include other important GHGs in their calculations, such as methane (CH4) and nitrous oxide (N2O) (e.g., EPA, 1990; IPCC, 1990; Manne et al., 1995; Tol, 1997), and a few go even further to include sulphates, volatile organic compounds (VOCs), and halocarbons (e.g. IPCC, 1992; WEC, 1995; Edmonds et al., 1996, 1997). With respect to the policy options used in the scenario quantifications, three fields are taken into account in the reviewed studies: energy systems (including both supply and demand), industrial processes (including cement and metal production), and land use (including agriculture and forest management).

Since most of the modelling exercises have been carried out to study the CO2 emissions from human activities linked to the use of energy, energy supply and end-use are naturally the areas where policy is applied. Energy supply options include natural gas, renewable energy, and commercial biomass; introduction of new technologies; and so on. End-use options chiefly pertain to increased energy efficiency in industry, transport, and residential and/or commercial applications.

The policy instruments analyzed depend on the underlying model structure. Most of the scenarios introduce policies such as simple carbon taxes or a constraint on emissions or concentration levels for achieving the desired reduction or stabilization. How the constraint is imposed varies from scenario to scenario. Among the models with regional disaggregation, a few regional targets have been introduced (e.g., Nordhaus, 1994; Tol, 1999). Regional disaggregation also allows modellers to let the regions trade in emission permits. Permit trading is introduced in more recent work, especially just before and after the Third Conference of the Parties to the United Nations Framework Convention on Climate Change in Kyoto (December 1997). Some studies offer permit trading as a mechanism to reduce the overall costs of abatement. Much of the work done in the early 1990s led to the development of detailed scenarios for introducing such policies (EPA, 1990; IPCC, 1990, 1992). Some models employ policies of supply-side technology introduction (Nakicenovic et al., 1993; Edmonds et al., 1996; Fujii and Yamaji, 1998), while other models emphasize the introduction of efficient demand-side technology (EPA, 1990; Kainuma et al., 1999a).

The issue of burden sharing among regions is a contentious one and it was sparsely treated in the first half of the 1990s. Most discussions about burden sharing are of a qualitative and partial nature and are not related to model-based mitigation scenarios. A few studies (most notably Rose and Stevens, 1993; Enquete Commission, 1995; and Manne and Richels, 1997) present a set of burden-sharing rules in their scenarios. Of late, the EMF exercises looking at the Kyoto scenarios have treated this issue better than in the past (Weyant, 1999).

The time-paths of emission reduction are determined in three ways in the reviewed studies. First, the emission trajectories are determined by policy scenarios that have been designed in detail for regions over the time frame (EPA, 1990; IPCC, 1990; WEC, 1995; Edmonds et al., 1996; Yohe and Wallace, 1996; Kainuma et al., 1998). Second, dynamic optimization models automatically determine these reduction time-paths by global cost minimization over time (e.g., Peck and Tiesberg, 1995; Fujii and Yamaji, 1998) or economic welfare maximization (Nordhaus, 1994; Manne et al., 1995). Third, mitigation scenarios of tolerable windows/safe landing, or safe emission corridors, can fix the time series of emission reduction by introducing a specific constraint of the rate of change in natural systems including the global temperature change rate (e.g., Alcamo and Kreileman, 1996).

Finally, there are differences in the treatment of feedback to the macro-economy in the models. While most bottom-up models have no feedback from cost to the macro-economy, top-down models allow for the feedback of energy prices to the macro-economy. The MERGE (Manne et al., 1995) and CETA (Peck and Tiesberg, 1995) models also have feedback from impacts to the macro-economy.

Technological improvement is a critical element in all the general mitigation scenarios. This is apparent when the detailed policy options are studied, where such literature is available. For instance, Nakicenovic et al. (1993) (using MESSAGE) incorporated policies of dematerialization and recycling, efficiency improvements and industrial process changes, and fuel-mix changes in the industrial sector; fuel efficiency improvements, modal split changes, behavioural change, and technological change in the transport sector; and efficiency improvements of end-use conversion technologies, fuel-mix changes, and demand-side measures in the household and services sector. It should be noted that efficiency improvement through technological advancement is emphasized in all sectors. Similar policies leading to efficiency improvement were also underlined in earlier modelling studies such as EPA (1990), IPCC (1990), and IPCC (1992).

Other reports in this collection

IPCC Homepage