worst case paper final (1)]
A few words on the intended audience and motivation for writing this:
First and foremost, this is written for the clients of Climate Forecast Applications Network who are interested in scenarios of future climate change [link]
Second, this paper is written as a contribution to my series of academic papers on the topic of uncertainty in climate science:
Third, the paper is written to inform the public debate on climate change and policy makers. I am ever hopeful that some sanity can be interjected into all this.
This paper is particularly relevant in light on the preceding post on consensus
, and Gavin’s desire for a better way to treat the extreme tails.
Overview of contents
I’m reproducing the Abstract, Introduction and Conclusions in this blog post, I encourage you to read the entire paper.
The objective of this paper is to provide a broader framing for how we assess and reason about possible worst-case outcomes for 21st
century climate change. A possibilistic approach is proposed as a framework for summarizing our knowledge about projections of 21st
century climate outcomes. Different methods for generating and justifying scenarios of future outcomes are described. Consideration of atmospheric emissions/concentration scenarios, equilibrium climate sensitivity, and sea-level rise projections illustrate different types of constraints and uncertainties in assessing worst-case outcomes. A rationale is provided for distinguishing between the conceivable worst case, the possible worst case and the plausible worst case, each of which plays different roles in scientific research versus risk management.
The concern over climate change is not so much about the warming that has occurred over the past century. Rather, the concern is about projections of 21st
century climate change based on climate model simulations of human-caused global warming, particularly those driven by the RCP8.5 greenhouse gas concentration scenario.
The Intergovernmental Panel on Climate Change (IPCC) Assessment Reports have focused on assessing a likely
range (>66% probability) for projections in response to different emissions concentration pathways. Oppenheimer et al. (2007) contends that the emphasis on consensus in IPCC reports has been on expected outcomes, which then become anchored via numerical estimates in the minds of policy makers. Thus, the tails of the distribution of climate impacts, where experts may disagree on likelihood or where understanding is limited, are often understated in the assessment process, and then exaggerated in public discourse on climate change.
In an influential paper, Weitzman (2009) argued that climate policy should be directed at reducing the risks of worst-case outcomes, not at balancing the most likely values of costs and benefits. Ackerman (2017) has argued that policy should be based on the credible worst-case outcome. Worst-case scenarios of 21st
century sea level rise are becoming anchored as outcomes that are driving local adaptation plans (e.g. Katsman et al. 2011). Projections of future extreme weather/climate events driven by the worst-case RCP8.5 scenario are highly influential in the public discourse on climate change (e.g. Wallace-Wells, 2019).
The risk management literature has discussed the need for a broad range of scenarios of future climate outcomes (e.g., Trutnevyte et al. 2016). Reporting the full range of plausible and possible outcomes, even if unlikely, controversial or poorly understood, is essential for scientific assessments for policy making. The challenge is to articulate an appropriately broad range of future scenarios, including worst-case scenarios, while rejecting impossible scenarios.
How to rationally make judgments about the plausibility of extreme scenarios and outcomes remains a topic that has received too little attention. Are all of the ‘worst-case’ climate outcomes described in assessment reports, journal publications and the media, actually plausible? Are some of these outcomes impossible? On the other hand, are there unexplored worst-case scenarios that we have missed, that could turn out to be real outcomes? Are there too many unknowns for us to have confidence that we have credibly identified the worst case? What threshold of plausibility or credibility should be used when assessing these extreme scenarios for policy making and risk management?
This paper explores these questions by integrating climate science with perspectives from the philosophy of science and risk management. The objective is to provide a broader framing of the 21st
century climate change problem in context of how we assess and reason about worst-case climate outcomes. A possibilistic framework is articulated for organizing our knowledge about 21st
century projections, including how we extend partial positions in identifying plausible worst-case scenarios of 21st
climate change. Consideration of atmospheric emissions/concentration scenarios, equilibrium climate sensitivity, and sea-level rise illustrate different types of constraints and uncertainties in assessing worst-case outcomes. This approach provides a rationale for distinguishing between the conceivable worst case, the possible worst case and the plausible worst case, each of which plays different roles in scientific research versus risk management.
2. Possibilistic framework
3. Scenarios of future outcomes
3.1 Scenario justification
3.2 Worst-case classification
3.3 Alternative scenarios
4. Is RCP8.5 plausible?
5. Climate sensitivity
6. Sea level rise
6.1 Worst-case scenarios
6.2 Possibility distribution
6.3 Alternative scenarios
The purpose of generating scenarios of future outcomes is that we should not be too surprised when the future eventually arrives. Projections of 21st
century climate change and sea level rise are associated with deep uncertainty and rapidly advancing knowledge frontiers. The objective of this paper has been to articulate a strategy for portraying scientific understanding of the full range of possible scenarios of 21st
century climate change and sea level rise in context of a rapidly expanding knowledge base, with a focus on worst-case scenarios.
A classification of future scenarios is presented, based on relative immunity to rejection relative to our current background knowledge and assessments of the knowledge frontier. The logic of partial positions allows for clarifying what we actually know with confidence, versus what is more speculative and uncertain or impossible. To avoid the Alice in Wonderland syndrome of scenarios that include too many implausible assumptions, published worst-case scenarios are assessed using the plausibility criterion of including only one borderline implausible assumption (where experts disagree on plausibility).
The possibilistic framework presented here provides a more nuanced way for articulating our foreknowledge than either by attempting, on the one hand, to construct probabilities of future outcomes, or on the other hand simply by labeling some statements about the future as possible. The possibilistic classification also avoids ignoring scenarios or classifying them as extremely unlikely
if they are driven by processes that are poorly understood or not easily quantified.
The concepts of the possibility distribution, worst-case scenarios and partial positions are relevant to decision making under deep uncertainty (e.g. Walker et al. 2016), where precautionary and robust approaches are appropriate. Consideration of worst-case scenarios is an essential feature of precaution. A robust policy is defined as yielding outcomes that are deemed to be satisfactory across a wide range of plausible future outcomes. Robust policy making interfaces well with possibilistic approaches that generate a range of possible futures (e.g. Lempert et al. 2012). Partial positions are of relevance to flexible defense measures in the face of deep uncertainty in future projections (e.g. Oppenheimer and Alley, 2017).
Returning to Ackerman’s (2017) argument that policy should be based on the credible worst-case outcome, the issue then becomes how to judge what is ‘credible.’ It has been argued here that a useful criterion for a plausible (credible) worst-case climate outcome is that at most one borderline implausible assumption – defined as an assumption where experts disagree as to whether or not it is plausible – is included in developing the scenario. Using this criterion, the following summarizes my assessment of the plausible (credible) worst-case climate outcomes, based upon our current background knowledge:
- The largest rates of warming that are often cited in impact assessment analyses (e.g. 4.5 or 5 oC) rely on climate models being driven by a borderline implausible concentration/emission scenarios (RCP8.5).
- The IPCC AR5 (2013) likely range of warming at the end of the 21st century has a top-range value of 3.1 oC, if the RCP8.5-derived values are eliminated. Even the more moderate amount of warming of 3.1oC relies on climate models with values of the equilibrium climate sensitivity that are larger than can be defended based on analysis of historical climate change. Further, these rates of warming explicitly assume that the climate of the 21st century will be driven solely by anthropogenic changes to the atmospheric concentration, neglecting 21st century variations in the sun and solar indirect effects, volcanic eruptions, and multi-decadal to millennial scale ocean oscillations. Natural processes have the potential to counteract or amplify the impacts of any manmade warming.
- Estimates of 21st century sea level rise exceeding 1 m require at least one borderline implausible or very weakly justified assumption. Allowing for one borderline implausible assumption in the sea level rise projection produces high-end estimates of sea level rise of 1.1 to 1.6 m. Higher estimates are produced using multiple borderline implausible or very weakly justified assumptions. The most extreme of the published worst-case scenarios require a cascade of events, each of which are extremely unlikely to borderline impossible based on our current knowledge base. However, given the substantial uncertainties and unknowns surrounding ice sheet dynamics, these scenarios should not be rejected as impossible.
The approach presented here is very different from the practice of the IPCC assessments and their focus on determining a likely
range driven by human-caused warming. In climate science there has been a tension between the drive towards consensus to support policy making versus exploratory speculation and research that pushes forward the knowledge frontier (e.g. Curry and Webster, 2013). The possibility analysis presented here integrates both approaches by providing a useful framework for integrating expert speculation and model simulations with more firmly established theory and observations. This approach demonstrates a way of stratifying the current knowledge base that is consistent with deep uncertainty, disagreement among experts and a rapidly evolving knowledge base. Consideration of a more extensive range of future scenarios of climate outcomes can stimulate climate research as well as provide a better foundation for robust decision making under conditions of deep uncertainty.
Since I resigned my faculty position, there has been little motivation for me to publish in peer reviewed journals. And I don’t miss the little ‘games’ of the peer review process, not to mention the hostility and nastiness of editors and reviewers who have an agenda.
However, one of my clients wants me to publish more journal articles. This client particularly encouraged me to publish something related to my Special Report on Sea Level and Climate Change
. I submitted a shorter version of this paper, in a more academic style, for publication in a climate journal. It was rejected. Here is my ‘favorite’ comment from one of the reviewers:
“Overall, there is the danger that the paper is used by unscrupulous people to create confusion or to discredit climate or sea-level science. Hence, I suggest that the author reconsiders the essence of its contribution to the scientific debate on climate and sea-level science.”
You get the picture. I can certainly get some version of this published somewhere, but this review reminded me why I shouldn’t bother with official ‘peer review.’ Publishing my research on Climate Etc. and as Reports ‘published’ by my company allows me to write my papers in a longer format, including as many references as I want. I can also ‘editorialize’ as I deem appropriate. In summary, I can write what I want, without worrying about the norms and agendas of the ‘establishment.’ Most of my readers want to read MY judgments, rather than something I think I can get past ‘peer reviewers.’
This particular paper is titled as a ‘Working Paper’, in the tradition often used by economists and legal scholars in issuing their reports. It is publicly available for discussion, and I can revise it when appropriate. I hope it will stimulate people to actually think about these issues and discuss them. I look forward to a lively review of this paper.
And finally, it is difficult to see how this paper could be categorized at ‘contrarian.’ It is not even ‘lukewarm.’ It discusses worst-case scenarios, and how to think about their plausibility. In fact, in one of the threads at WUWT discussing one of my previous ‘worst-case’ posts, commenters thought that this was way too ‘alarmist’ to be posted at WUWT.
Bottom line: we need to think harder and differently about climate change. This paper helps provide a framework for stepping beyond the little box that we are currently caught in.
This article appeared on the Climate Etc. website at https://judithcurry.com/2019/08/22/climate-change-whats-the-worst-case/#comments