Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 19
Modeling the Economics of Greenhouse Gas Mitigation: Summary of a Workshop 5 Story Lines, Scenarios, and the Limits of Long-Term Socio-Techno-Economic Forecasting Steve Smith of the workshop planning committee provided context for the workshop’s fourth session by noting that the socioeconomic and technological characteristics of the baseline scenario are key determinants of mitigation costs. Cost estimations of greenhouse gas reductions and other mitigation efforts do not happen in a vacuum, but rather as a change from assumed baseline conditions. The development of the baseline and alternative future scenarios has been driven largely by research needs of the integrated assessment community with interaction from other user communities. However, these scenarios increasingly are becoming end products in and of themselves. Developing them is becoming more complex and more time-consuming, and the resulting scenarios are not just research tools for one community. There also is tension between the modeling community that generates most of these scenarios and the users of these scenarios regarding how scenarios are used in the policy process. Modelers view scenarios not as predictions or forecasts, but rather as alternative images of how the future might evolve. However, outside the integrated assessment modeling community, these scenarios are often and perhaps even largely viewed as predictions. Smith went on to describe how scenarios are generated. This process starts with some fundamental assumptions about factors like population and labor productivity, which are then translated into energy service demands and the technologies that are available to serve those demands. Simply doing these calculations requires a large number of assumptions about many factors, including fertility, mortality, and the availability of new technologies that may not exist today. As a result, the uncertainties multiply as one proceeds from assumptions about population to estimates of energy use and greenhouse gas emissions. Further, these scenarios typically have been produced almost from scratch for each major application, and transparency about the methods, reasoning, and assumptions is a challenge. Thus, a fundamental issue laid out by Smith is how the modeling community moves forward in the development of new scenarios in a context where the number and the variety of scenario “users” are increasing. Richard Moss of the Pacific Northwest National Laboratory discussed next-generation scenarios for climate modeling and analysis of adaptation and mitigation. He described as motivations for a new process for generating scenarios the need (1) to help address critiques of past scenarios, including the perceived overconfidence in scenario details; (2) to recognize evolving information needs, including the need for more information on adaptation to and mitigation of climatic change; (3) to include more scientific information, such as greater attention to feedbacks among elements of the human-climate system; and (4) to improve the coupling of integrated assessment modeling with impact, adaptation, and vulnerability (IAV) models. In the past, especially under the earlier Intergovernmental Panel on Climate Change process, the process of developing scenarios began with a set of detailed
OCR for page 20
Modeling the Economics of Greenhouse Gas Mitigation: Summary of a Workshop story lines laid out in a report. These story lines were then turned into various quantifications of the underlying driving socioeconomic forces. The story lines were then used to estimate emissions, atmospheric concentrations, radiative forcing, and finally, from climate models, ranges of temperature and precipitation. IAV models could then use these estimations at an aggregate level to look at impacts such as risks to species, risks from extreme climate events, distribution of impacts across societies, and aggregate economic impacts. However, this kind of activity took several years to play out so that by the time the models of impacts were producing estimates, there might have been a new generation of climate models in use. Further, Moss pointed out that there has been quite a bit of research on how people interpret these story lines and scenarios and this research indicates that there can be overconfidence in interpreting results that are simply illustrative story lines. People often begin to believe that such story lines are the most likely story lines, which is not the case. This belief can limit their thinking about alternative futures, whereas taking a broad approach is extremely important for bracketing the widest possible future conditions. Finally, Moss noted that there is not necessarily a one-to-one correspondence between the story lines generated by models and the socioeconomic quantification of those story lines. As described by Moss, a newly developed scenario process that can help to address these issues starts with radiative forcing instead of with socioeconomic story lines. The new process begins by assuming some different levels of radiative forcing (or representative concentration pathways; RCPs) and then models in parallel both the climate scenarios that result from using the RCPs and the socioeconomic scenarios that could produce those RCPs. Some of the new socioeconomic scenarios will be consistent with the levels of emissions required to produce the RCPs, and some will be independent. Moss concluded that this new scenario process, although not perfect, presents opportunities, including greater openness and flexibility, especially for socioeconomic scenario development, that could lead to increased collaboration across the distinct research communities and improved synthesis and coordination across assessments at different scales. He noted that the challenges for mitigation and impacts analysis include the need to carefully consider what projections are needed on what time scale and how that information is going to be used. He also noted the need for approaches to performing probabilistic analysis and to communicating uncertainties. Dale Jorgensen of Harvard University then discussed a modeling approach (the IGEM model) and scenarios for climate modeling based on the requirements specified in the Waxman-Markey bill. Jorgensen first discussed the major determinants of economic growth, including productivity changes, capital accumulation (investment), population, labor supply, and human capital. He went on to discuss the historical record of technical change for various industries in the United States and the modeling of technical change and substitution in the IGEM model. The use of econometrics in the IGEM model makes it possible to sort out what portion of the technology change occurring over the very extensive historical record is attributable to price changes like those experienced during the energy crisis periods and how much is due to changes separate from price. Jorgensen also described the modeling of household savings/investment and consumption and leisure. Finally, he described the demographic assumptions used in the IGEM model, given that projections of consumption and welfare depend on projections of population. This model was then benchmarked to the base case in the 2009 Energy Outlook (EIA, 2009) and was used by Jorgensen to look at nine scenarios related to the Waxman-Markey bill. The modeling results demonstrate the importance of demographic and technology assumptions. Jorgensen argued that the model of technology and technological change should include substitution (“elasticities”) and technical change (“trends”) and that the IGEM econometric model provides a unified representation. Jorgensen also concluded that standard statistical techniques, based on confidence intervals generated by the econometrics, can capture uncertainties in estimated impacts, and that this econometric approach avoids the limits on dimensionality of a Monte Carlo approach. The remainder of the session included a panel discussion and comments from the audience. Nebojsa Nakicenovic of the International Institute for Applied Systems Analysis served as the discussant, with comments that focused on story lines. His hypothesis was that the importance of story lines will increase as the RCP scenarios require additional logic and justification. He stated his belief that, because of the large numbers of parameters and complicated assumptions in coupling socioeconomic modeling to modeling of climate process, it is important to have the complementary analysis of story lines to explain the logic of how these scenarios are constructed. Finally, Nakicenovic described the story lines in the case of multiple scenarios as very helpful in explaining the logic for
OCR for page 21
Modeling the Economics of Greenhouse Gas Mitigation: Summary of a Workshop differentiating among the set of scenarios. During the question-and-comments period, David Montgomery from Charles River Associates noted that a conclusion to be drawn from some of his work is that it is almost always the existence of institutions that allows societies to sustain technology progress and income growth over time. However, the process described by Richard Moss, starting from an RCP, considerably narrows the possibilities of what might actually be explicitly explored with regard to institutions. Moss responded by noting that one of the advantages of starting with an RCP is to say that we have an RCP, but this does not mean that we have to have a single story line that produces a single RCP. Thus, to address an interest in a particular issue, for example institutions and governance, there is no reason to avoid developing a scenario or a story line that focuses on that interest. William Nordhaus commented that he found the RCP approach foreign in that the scenarios then seemed to be organized around variables endogenous to the socioeconomic models. Nordhaus stated that he thought the natural place to start was with baselines, using as inputs some given policies rather than intermediate variables such as RCPs. Moss responded by saying that it is important to think about the issue from the point of view of what the climate modeling community is interested in: the reason for not starting with the story lines is that a great deal of effort had to be expended upfront to get to what the climate people were interested in, and, from the climate-modeling perspective, it did not seem necessary to actually predetermine what those story lines were in order to get to a particular climate future.