Scenarios of long-term socio-economic and environmental development under climate stabilization

https://doi.org/10.1016/j.techfore.2006.05.026Get rights and content

Abstract

This paper presents an overview of the greenhouse gas (GHG) emissions scenarios that form the analytical backbone for other contributions to this Special Issue. We first describe the motivation behind this scenario exercise and introduce the main scenario features and characteristics, in both qualitative and quantitative terms. Altogether, we analyze three ‘baseline’ scenarios of different socio-economic and technological developments that are assumed not to include any explicit climate policies. We then impose a range of climate stabilization targets on these baseline scenarios and analyze in detail the feasibility, costs and uncertainties of meeting a range of different climate stabilization targets in accordance with Article 2 of the United Nations Framework Convention on Climate Change. The scenarios were developed by the IIASA Integrated Assessment Modeling Framework that encompasses detailed representations of the principal GHG-emitting sectors—energy, industry, agriculture, and forestry. The main analytical findings from our analysis focus on the implications of salient uncertainties (associated with scenario baselines and stabilization targets), on feasibility and costs of climate stabilization efforts, and on the choice of appropriate portfolios of emissions abatement measures. We further analyze individual technological options with regards to their aggregated cumulative contribution toward emissions mitigation during the 21st century as well as their deployment over time. Our results illustrate that the energy sector will remain by far the largest source of GHG emissions and hence remain the prime target of emissions reduction. Ultimately, this may lead to a complete restructuring of the global energy system. Climate mitigation could also significantly change the relative economics of traditional versus new, more climate friendly products and services. This is especially the case within the energy system, which accounts for the largest share of emissions reductions, but it is also the case in the agriculture and forestry sectors, where emissions reduction and sink enhancement measures are relatively more modest.

Introduction

Svante Arrhenius published his seminal classic On the Influence of Carbonic Acid in the Air upon the Temperature on the Ground [1] more than 100 years ago in 1896. This first, and today still surprisingly accurate, scientific quantification of the temperature effects of rising CO2 concentrations included a sensitivity analysis to explore the effects of rising CO2 concentrations by a factor between one to three above the then prevailing level of some 300 parts per million by volume (ppmv). While noting that the burning of some 500 million tons of coal was the anthropogenic source equivalent of a natural CO2 sink in the form of rock weathering, the likelihood of quickly reaching any of the levels of atmospheric CO2 concentrations addressed in his calculations seemed rather slim from the perspective of the day.

Today's situation is fundamentally different. Atmospheric CO2 concentrations have risen to some 380 ppmv. By simply extrapolating historical growth rates (which is widely considered bad practice not only in climate science) it becomes apparent that over the next 100 years we could approach those levels of CO2 concentrations that were considered in Arrhenius' calculations of temperature effects. That is, we could enter a regime of significant alterations of the Earth's climate characterized by the proverbial ‘doubling’ of atmospheric CO2 concentrations over pre-industrial times. Given the enormous changes over the past century and vast potential for further changes in the next, there is thus a deep interest to better understand the unfolding of future emissions paths. Such a look into the future is especially interesting because it can help:

  • anticipate magnitudes of possible climate changes;

  • assess economic, social, and ecological consequences of such changes;

  • determine if and by how much undesirable consequences can be mitigated, either in better adapting to a changing climate or in avoiding unfolding climate change as much as possible (i.e., through emissions reduction).

The above considerations constitute the prime motivation for developing scenarios, that is stories and quantifications of how possible developments could unfold that can help in our desire to anticipate the potential consequences and to plan to mitigate this large-scale planetary geophysical ‘experiment’ that we are in the midst of performing.

Ironically, despite all the progress in science and technology since the time of Arrhenius, one challenge remains as large as it was 100 years ago: the need to consider a time scale of a century (or even longer), which is dictated by the twin inertias of the coupled socio-economic and climate systems. Given our current understanding of the carbon cycle, CO2 emitted today will remain in the atmosphere many decades to come and altering future climate, whose legacy (e.g., in the form of thermal expansion of oceans and resulting sea level rise) might even take a millennium to fully unfold. Likewise, given the longevity of infrastructures and the capital stock of our energy system, many decades will pass before any initiated policy changes translate into a noticeable effect on emissions and hence avoidance of “dangerous interference in the climate system”. This is the stated objective of the UN Framework Convention on Climate Change [2], a convention ratified by most of the planet (much different to the ensuing Kyoto Protocol that only applies to industrialized countries and which the USA and Australia have refused to ratify).

The task ahead of anticipating the possible developments over a time frame as ‘ridiculously’ long as a century is wrought with difficulties. Particularly, readers of this Journal will have sympathy for the difficulties in trying to capture social and technological changes over such a long time frame. One wonders how Arrhenius' scenario of the world in 1996 would have looked, perhaps filled with just more of the same of his time—geopolitically, socially, and technologically. Would he have considered that 100 years later:

  • backward and colonially exploited China would be in the process of surpassing the UK's economic output, eventually even that of all of Europe or the USA?

  • the existence of a highly productive economy within a social welfare state in his home country Sweden would elevate the rural and urban poor to unimaginable levels of personal affluence, consumption, and free time?

  • the complete obsolescence of the dominant technology cluster of the day-coal-fired steam engines?

How he would have factored in the possibility of the emergence of new technologies, especially in view of Lord Kelvin's sobering ‘conclusion’ of 1895 that “heavier-than-air flying machines are impossible”?

We do not know, as Arrhenius, perhaps wisely, refrained from a look into the future to check over which time horizon his model calculations could become a reality. However, we do know that, as at the time of Arrhenius, a perspective of 100 years represents such a challenge that traditional (deterministic) forecasting is impossible. Instead, our ability to anticipate, to imagine, and to describe the deep uncertainties that surround a 100 year future perspective is challenged, a challenge traditionally addressed through the development of alternative scenarios, or ranges of possible futures.

As a result, the development of long-term scenarios in conjunction with climate change science and policy analysis has both a distinguished tradition and has grown almost into an industry of its own. First reviews of the resulting scenario literature date back to the early 1980s [3] and have been repeated periodically ever since [4], [5]. The latter review surveyed altogether more than 400 scenarios, which required the use of data base management tools to handle the large number of scenarios published in the literature. An update of that review for the forthcoming Fourth IPCC assessment report will include altogether over 700 scenarios [6].1 A distinguishing feature of the literature on climate change scenarios (including the present study) is a customary distinction between ‘no controls’ or ‘baseline’ scenarios and so-called ‘intervention’ or climate policy scenarios that analyze various target levels in response to the stated UNFCC objective of “stabilizing greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system” [2]. In other words, it has become customary to distinguish between two major types of uncertainties of the future:

  • uncertainties in emission drivers (population, income, technology, diets, etc.) and their resulting emissions outcomes (magnitude of projected climate change uncertainty);

  • uncertainty surrounding levels, commitment, and effectiveness of globally coordinated policy efforts to slow or halt global warming (often referred to as ‘target uncertainty’).

Readers should be aware that the two types of scenarios serve different purposes and are not always to be judged with the same qualitative yardstick typically applied to a scenario (reproducibility, plausibility, internal consistency, etc.). ‘Baseline’ scenarios can range in degree of complexity and logic from ‘blind’ trend extrapolation to sophisticated blends of qualitative and quantitative scenario ‘storylines’ that attempt to check for plausibility and internal consistency of the scenario(s) under consideration with the help of sophisticated models. Across this range they aim to ‘stand alone’ in providing a ‘narrative’, or a sequence of carefully crafted conditional ‘when if, then’ statements that, when quantified with formal models, lead to quantifications of different emission drivers, their interactions, and the resulting emissions outcomes. Conversely, ‘control’ (or ‘stabilization’) scenarios are more controlled model experiments based on (one is almost attempted to say ‘tacked on to’) given baseline scenarios for a range of climate stabilization targets. While these are technically feasible, they may not necessarily meet the same criteria of scenario plausibility and consistency as applied to the corresponding original ‘baseline’ scenarios.

The scenarios considered in this Special Issue are no exception to the above dichotomy in climate change scenarios. We also first proceed in developing and presenting a range of three ‘baseline’ scenarios aimed to elucidate the major salient uncertainties in drivers and the resulting emissions outcomes that a century-long perspective necessarily entails. These three scenarios are then used as input to a number of controlled model experiments (altogether 11 ‘stabilization scenarios’ are imposed on the three baseline scenarios). In these exogenously pre-specified climate stabilization targets (represented by their equivalent CO2 concentration levels, or more precisely by various levels of stabilization of radiative forcing of all greenhouse gases (GHGs)) are examined from a multi-gas and multi-sector perspective. In other words, the customary, almost exclusive, focus on energy-related CO2 emissions in both baseline and ‘policy’ scenarios is replaced here by a much wider analytical framework that covers all relevant GHGs and all major emitting sectors.

The scenarios presented here also do not emerge ex nihilo. Instead, they are derivatives of (a subset of) scenarios developed by the authors for the IPCC Special Report on Emissions Scenarios (SRES) [7] that were also used for a subsequent analysis of the feasibility of meeting a range of climate stabilization targets analyzed in the IPCC Third Assessment Report (TAR) [8] and within the model intercomparison research performed under the auspices of the Energy Modeling Forum (EMF) [9]. We have revised the original scenarios to reflect new information that has become available with the aim to improve also scenario consistency. The new scenarios were developed with the help of the integrated modeling and assessment framework presented in more detail below. One scenario (labeled as ‘revised SRES A2’ scenario or ‘A2r’), while maintaining its main structural and qualitative characteristics, represents a major numerical revision that reflects the most recent long-term demographic outlook with a corresponding lowering of future world population growth [10].

Next to numerical scenario revisions (particularly pronounced in the demographic and economic developments described in the A2r scenario, and to a much lesser extent also in the other two scenarios B1 and B2, are described in more detail elsewhere in this Special Issue [11]), a number of novel methodological features also characterize the scenarios presented here. Foremost, the scenarios encompass a multi-sector and multi-GHG perspective in which the integrated assessment paradigm is extended from the traditional focus on the energy sector to all other salient sectors (in particular agriculture and forestry) that emit GHGs or potentially contribute to climate change mitigation efforts through either emissions reductions or enhancements of GHG sinks [12]. By a full coupling of the corresponding models that represent the energy, agriculture, and forestry sectors, we are not only able to account consistently for all GHGs and their respective mitigation potentials across the sectors,2 but also account for important feedbacks and interdependencies (e.g., competition for land-use) between sectors. Likewise, impacts of climate change are also endogenously considered in the scenarios reported here, for example, in terms of changes in agricultural production and gross domestic product (GDP) [13] or in the corresponding changing water needs for agricultural production [14]. Finally, the scenarios also incorporate previously unexplored mitigation options, such as the use of biomass in conjunction with carbon sequestration and storage (CSS), that could result in an artificial ‘sink’ for anthropogenic CO2 emissions, in addition to traditionally considered forest sinks.

The main objective of our scenario exercise is to explore the feasibility and costs of meeting alternative climate stabilization targets under a range of salient long-term uncertainties with a limited set of scenarios. To meet this objective we have developed two contrasting scenarios, A2r and B1, that aim to bracket the upper and lower quadrants of emissions and hence magnitudes of climate change and of the possible vulnerability to climate change, respectively. These two scenarios also form the backbone of the model linkages to integrate the energy, agriculture, and forestry sectors reported in this Special Issue. The more intermediary scenario B2 (with numerically minor revisions compared to its SRES variant) serves as a benchmark from which to compare the results presented here with those of earlier work, in particular that of the IPCC SRES and TAR reports, as well the earlier scenarios (in particular the scenario IIASA–WEC ‘B’) developed in collaboration between the International Institute for Applied Systems Analysis (IIASA) and the World Energy Council (WEC) [5], [11].

The use of the terms of ‘upper quadrant’ and ‘lower quadrants’ to position the scenarios reported here in comparison to the entire scenario literature is indicative only. The scenarios developed aim to be positioned above and below the 75th and 25th percentile, respectively, of the comparable scenario literature, but without all3 their salient scenario parameters necessarily always falling within this indicative range. It is also important that the above quantitative yardstick produced from a statistical analysis of the frequency distribution of the published scenario literature is not confounded with the traditional concept of probability. Given the large number of variables and their interdependencies, we are of the opinion that it is impossible to assign objective likelihoods or probabilities to emissions scenarios. We have also not attempted to assign any subjective likelihoods to the scenarios either. The purpose of the scenarios presented in this Special Issue is, rather, to span the range of uncertainty without an assessment of likely, preferable, or desirable future developments. A subjective assessment of scenario likelihoods goes well beyond the scope of this paper and those presented in this Special Issue. Likelihoods or probabilities are therefore not assigned to any of the scenarios reported here, which does not mean that we consider all the scenarios equally likely. Indeed, we do not consider the three scenarios reported here equally likely, but simply cannot offer any scientifically rigorous way to differentiate the likelihoods across the scenarios and therefore refrain from any necessarily arbitrary, subjective ranking.

Table 1 summarizes the positioning of the three scenarios with respect to the most important uncertainties examined in this study. These include, in particular:

  • Development pathway uncertainty, which includes alternative demographic, economic, and technological developments that lead to high (A2r), intermediary (B2), or low (B1) emissions of GHGs and hence magnitude of future climate change. A more detailed discussion of the scenario's that underlie the demographic and economic trends at alternative spatial scales (regional, national, and grid-cell level) is given elsewhere in this Special Issue [11].

  • Climate impacts vulnerability uncertainty, the multiple dimensions of which include, in particular, ‘soft’ institutional and technological variables, is treated here in a simplified manner framed by the variables population density, population concentration, and per capita income, which exercise amplifying and dampening effects on climate vulnerability.4 Vulnerability ranges from high (A2r), through intermediary (B2), to low (B1) in the scenarios presented here. Discussions of the impacts and vulnerability that focuses on the implications of the scenarios for the agriculture sector, people at risk and irrigation needs induced by climate change are given elsewhere in this Special Issue [13], [14].

  • Climate stabilization target uncertainty, as mentioned above, is addressed by systematic model simulations for a range of alternative climate stabilization targets imposed on the no-policy baseline scenarios. Altogether, we performed calculations for 11 stabilization scenarios for eight comparable stabilization levels that ranged from 480 to 1390 ppmv (CO2-equivalent concentration for all GHGs taken together) by 2100. The number of stabilization scenarios analyzed is highest (five) for the high emissions scenario A2r, followed by scenario B1 (four stabilization levels analyzed) and scenario B2 (two stabilization levels). The higher the baseline emissions, such as in scenario A2r, the higher, therefore, is the range of stabilization targets and resulting emissions reduction needs (and costs) examined to fully represent target uncertainties (see Table 4 below).5,6 Finally, the lower bound of the stabilization targets analyzed is also a function of the scenario baseline emissions—with higher baseline emissions the lowest stabilization targets achievable are also higher compared to the scenarios with lower baseline emissions.7

For reasons of scenario parsimony, our set of three scenarios does not include a scenario that combines high emissions (and hence high climate change) with low vulnerability (e.g., as reflected in high per capita incomes). These were the characteristics of the scenarios within the A1 scenario family in the SRES report [7], [8], which also explored the impacts of alternative directions of technological change on future emission levels. This group of scenarios, while of considerable interest, especially for technology uncertainty analysis, is not analyzed further here.

In addition to addressing the uncertainties summarized above, the scenarios also have an additional methodological purpose. They serve as an integrative tool to link a variety of sectorial models (energy, agriculture, and forestry) under continued development at IIASA, and so help to quantify interlinkages and feedbacks between various sectors that are at the core of comprehensive (multi-gas) climate stabilization efforts. The scenarios also help to put additional sensitivity and uncertainty analyses performed within sectorial models into perspective. Therefore, all articles in this Special Issue make use of comparable common scenarios in their analysis. The significance of this feature can only be fully appreciated when we consider that the climate policy analysis literature has, to date, been ‘plagued’ by significant problems of incomparability of results because different models and analyses continue to use widely different projections and scenarios as their analytical basis.

Section snippets

An overview of scenarios

This section provides a quantitative overview of the scenarios that underlie the articles of this Special Issue. Before, however, proceeding to the customary presentation of numerous input assumptions and their resulting outcomes in terms of GHG emissions and climate consequences, it might be useful to provide some context in the form of qualitative scenario ‘narratives’ or ‘storylines’ (Box 1). Indeed, the blending of both qualitative and quantitative scenario characteristics is a

Scenario methodology and model linkages

To develop the scenarios presented in this Special Issue paper we used a set of interlinked disciplinary and sectorial models referred to as the Integrated Assessment (IA) Modeling Framework (illustrated in Fig. 10). The framework combines a careful blend of rich disciplinary models that operate at different spatial resolutions that are interlinked and integrated into an overall assessment framework. The framework covers all GHG-emitting sectors, including agriculture, forestry, energy, and

Summary of scenario results

This section summarizes the main scenario results with respect to the portfolio of mitigation measures and the contribution of individual options to achieve various levels of stabilization of atmospheric GHG concentrations. Our scenario set considers two principal dimensions of uncertainty—that with respect to the development path (baseline uncertainty) and that of the appropriate level of mitigation (stabilization level uncertainty). Each of these two dimensions has important implications for

Summary and conclusions

Through our scenario analysis we have illustrated the importance of considering the two most fundamental uncertainties that surround future efforts to mitigate against climate change:

  • uncertainty of magnitude of future emission levels as described by alternative scenario baselines;

  • uncertainty that surrounds the ultimate mitigation target (i.e., the stabilization levels).

Feasibility and costs, as well as the technological options needed to meet alternative climate stabilization goals all, depend

Acknowledgments

We gratefully acknowledge Ilkka Keppo and Shilpa Rao for their help in the development of the scenarios, as well as Peter Kolp and Alaa Al Khatib for their assistance in producing the manuscript. The research reported here is part of an institute-wide collaborative effort within IIASA's Greenhouse Gas Initiative (GGI). The interdisciplinary research effort within GGI links all the major research programs at IIASA that deal with research areas related to climate change, including population,

Keywan Riahi is the Scientific Coordinator of the Greenhouse Gas Initiative (GGI) and Research Scholar in the Transitions to New Technologies Program (TNT) at the International Institute for Applied Systems Analysis (IIASA), Austria. His main research interests are the long-term patterns of technological change and economic development and, in particular, the evolution of the energy system.

References (48)

  • J.H. Ausubel et al.

    A Review of Estimates of Future Carbon Dioxide Emissions, in Changing Climate

  • J. Alcamo et al.

    An Evaluation of the IPCC IS92 Emission Scenarios

  • N. Nakicenovic et al.

    Assessment of emissions scenarios revisited

    Environ. Econ. Policy Stud.

    (2006)
  • N. Nakicenovic et al.

    Special report on emissions scenarios

  • Climate change 2001: mitigation

  • S. Rao, K. Riahi, The Role of Non-CO2 Greenhouse Gases in Climate Change Mitigation: Long-term Scenarios for the 21st...
  • B. O'Neill et al.

    Downscaling of SRES population scenarios from regional to national level. IR-05-062

  • A. Grübler et al.

    Regional, National, and Spatially Explicit Scenarios of Demographic and Economic Change Based on SRES

    Technol. Forecast. Soc. Change

    (2006)
  • D. Rokityanskiy et al.

    Geographically Explicit Global Modeling of Land-Use Change, Carbon Sequestration, and Biomass Supply

    Technol. Forecast. Soc. Change

    (2006)
  • F.N. Tubiello et al.

    Reducing Climate Change Impacts on Agriculture: Global and Regional Effects of Mitigation 2000 to 2080. I

    Technol. Forecast. Soc. Change

    (2006)
  • G. Fischer et al.

    Climate Change Impacts on Irrigation Water Requirements: Effects of Mitigation

    Technol. Forecast. Soc. Change

    (2006)
  • United Nations Department of Economic and Social Affairs Population Division: world population prospects: the 2004 revision

  • W. Lutz et al.

    The end of world population growth

    Nature

    (2001)
  • Cited by (0)

    Keywan Riahi is the Scientific Coordinator of the Greenhouse Gas Initiative (GGI) and Research Scholar in the Transitions to New Technologies Program (TNT) at the International Institute for Applied Systems Analysis (IIASA), Austria. His main research interests are the long-term patterns of technological change and economic development and, in particular, the evolution of the energy system.

    Arnulf Grübler is a senior research scholar in the Transitions to New Technologies Program at the International Institute for Applied Systems Analysis (IIASA). He also holds a part-time appointment as Professor in the Field of Energy and Technology at the School of Forestry and Environmental Studies at Yale University, New Haven, USA.

    Nebojsa Nakicenovic is Leader of the Transitions to New Technologies Program and Greenhouse Gas Initiative at the International Institute for Applied Systems Analysis (IIASA) and Professor of Energy Economics at the Vienna University of Technology. He is also an Associate Editor of the International Journal on Technological Forecasting and Social Change, Editor of International Journal on Energy, and Climate Policy, and has served as Coordinating Lead Author for a number of Assessment Reports of the Intergovernmental Panel of Climate Change and the Millennium Ecosystem Assessment.

    View full text