Climate Science Voyage of Discovery

Post-normal science is an idea that we seldom talk about in the forest-related sciences.

The graph above shows the basic concepts, and there is a fairly explanatory Wikipedia review here. There is also a very active PNS academic community. Sometimes it’s hard to wade through the academic-ese, but if you can look through that, there are some ideas that may be helpful in the interface between science and policy.

When I worked at the Forest Service in Research and Development, I remember many discussions with the person in my group assigned to “science and management.” His model was basically what I call the “briefcase under the bridge” model..scientists come up with what the problems are, what disciplines are involved, and how the research is to be done. They get answers for what (in this case, National Forest folks) should do, and leave it under the bridge. If National Forest folks pick up the briefcase and don’t use everything, perhaps because they disagree with what the problems are or any of the other decisions, they are accused of “not following the science.” In retrospect, perhaps I should have invited some PNS experts to give seminars.

A summary of papers from the 2014 and 2015 PNS meetings can be found here.
In fact, there is a conference (PNS5) this year in Florence, Italy, for those interested in getting a taste for current research in the area. Registration is only 50 Euros for non-academics and other interested people.

The below is from a description of a paper in that collection by Konig, Borsen and Emmeche.

The ethos of Post-normal science”, gives an overview of important concepts of PNS, and investigates the norms and values of PNS through a structured literature review. The authors refer to Funtowicz and Ravetz’s (1993, 2008) development of PNS from the mid-1980s onwards, and describe the conditions characterizing a post-normal situation: Irreducible complexity, deep uncertainties, multiple legitimate perspectives, value dissent, high stakes, and urgency of decision-making. PNS seeks to cope with such situations through extended peer communities encompassing broader notions of knowledge, uncertainty management, and acknowledgement and management of multiple valid perspectives. Unlike normal science, the goal is not to attain certain knowledge. The goal of PNS is quality, a more robust ‘science for policy’. Inspired by the legacy of the Mertonian norms of CUDOS (Communalism, Universalism, Disinterestedness, and Organized Skepticism), the authors point to how politicization of science renders Merton’s norms invalid. Through their analysis of 33 norms and values found in 397 PNS-related documents, they identify an ethos for PNS which they denominate TRUST (Transparency, Robustness, Uncertainty management, Sustainability, and Transdisciplinarity), considered as a nexus for reflexivity practices. They propose that the public trust in science advice can be restored through the PNS ethos.

And a paper by Scott Bremer “Have we given up too much? On yielding climate representation to experts.” Here’s the abstract:

Our representations of climate are changing, and with them the ways we claim to know our local climates, and live according to them. Amidst this destabilising change in natural and social orders, scientific representations of climate are emerging as dominant. In this perspective piece I start from post-normal science and offer two arguments against completely yielding climate representation to experts, illustrated by my experiences in northeast Bangladesh. Descriptively, abandoning non-scientific representations and knowledge of climate results in a smaller and more fragile knowledge base for adaptation. Normatively, it is our common enterprise to rediscover the places we inhabit; it is not the responsibility of the expert community alone to reinterpret our places under a changing climate. Looking at the co-existence of modern and pre-modern representations of climate in Bangladesh, I suggest that post-normal science approaches may help bridge these different representations, with equal attention to their quality for adaptation, and the values and meanings underpinning them

(my bold).

Perhaps The Smokey Wire sometimes behaves like the “extended peer community” envisioned in the PNS literature. I’m bringing up this literature because many folks claim that climate change is the poster child of a problem that requires PNS concepts and practices, and we’ll be looking more deeply into that as the Voyage continues.

Steve Wilent pointed our attention to what is known as the 8.5/BAU (does RCP 8.5 represent Business as Usual, as many scientific papers say?) controversy a while back. I thought I’d post this Twitter roll (hopefully you can click to it) by Oliver Geden because it includes the sociology of science perspective. so often missing from climate science discussions. Also it seems like a gradual introduction to acronyms, and has links to many other papers of interest.

If you have trouble following the Thread reader, please let me know.

What is a Representative Concentration Pathway? You can read definitions, which aren’t necessarily very helpful. I like this article by Zeke Haufather because it tells the story of how the RCP’s were developed and used, leading to:

However, its position as the only non-mitigation scenario considered in the IPCC AR5 along with relatively poor communication between energy modelling and climate modelling communities led to a widespread misperception both in the media and in the academic literature that RCP8.5 was the expected “business as usual” outcome in a world without any future climate policy.

While worst-case outcomes are important to take into account, particularly given the uncertainties in the magnitude of carbon cycle feedbacks, it is important that they not be considered in isolation. Taking the range of possible baseline outcomes from 6.0 to 8.5 W/m2 forcing would provide a more realistic set of scenarios for studying climate impacts in a no-policy future.

On the other hand, though, if models and scenarios estimate so many things, how can we be sure that their estimates are accurate? From the Ritchie paper:

This paper finds climate change scenarios anticipate a transition toward coal because of systematic errors in fossil production outlooks based on total geologic assessments like the LBE model. Such blind spots have distorted uncertainty ranges for long-run primary energy since the 1970s and continue to influence the levels of future climate change selected for the SSP-RCP scenario framework. Accounting for this bias indicates RCP8.5 and other ‘business-as-usual scenarios’ consistent with high CO2 forcing from vast future coal combustion are exceptionally unlikely. Therefore, SSP5-RCP8.5 should not be a priority for future scientific research or a benchmark for policy studies.

It’s also clear that “poor communication” between energy modelling and climate modelling communities and the difficulty of communicating what they are doing to other scientific communities, including the impacts folks, as well as the media, may be a real problem.

Geden also refers to the BECCS debate, which seems to me another discipline gap – modelers are going to assume BECCS but there is no link to people who might say it’s feasible or not in a particular place (Geden’s slide 6). When he says “the STS community will have a field day.” He means the science and technology studies community, who study, among other things, how people and groups work to produce knowledge.

Roger Pielke, Jr. tied this back to the discipline of scenario planning in his article here.

Scenarios of the future have long sat at the center of discussions of climate science, impacts and adaptation and mitigation policies. Scenario planning has a long history and can be traced to the RAND Corporation during World War 2 and, later (ironically enough) Shell, a fossil fuel company. Scenarios are not intended to be forecasts of the future, but rather to serve as an alternative to forecasting. Scenarios provide a description of possible futures contingent upon various factors, only some of which might be under the control of decision makers.

The climate community got off track by forgetting the distinction between using scenarios as an exploratory tool for developing and evaluating policy options, and using scenarios as forecasts of where the world is headed.

To summarize there are three separate ideas. 1. RCP 8.5 was never feasible, as it relied on using a great deal more coal that (other scientists think) doesn’t exist. 2. Lotsa coal is not business as usual. 3. the RCPs are scenarios and are not comparable to each other.
As I recall, the idea of scenario planning was to look at a broad range of futures without likelihoods and pick things to do that make sense under a range of scenarios. Not to pick the worst ones and model the extent of further bad things. Let’s do an example. Denver Water used to do scenario planning. Suppose they had a most extreme scenario- climate change dries up supply and the population grows. What to do? The point is to think about it. Not to model and write news stories about how we won’t have any water. It’s really about what you do with the info of “what could possibly happem” not how you derive it, nor to what detail.

When we talk about why scientists disagree about different aspects and approaches to climate science, I think it’s helpful to break it down into general areas. We’ll see how this works when we move on to scientific disagreements of various kinds. So let’s start at the 30,000 foot level.

Generally Agreed Upon:

(1) Over time, climate has influenced humans, and humans have influenced climate. I think here of overgrazing, and deforestation in the historic past. Humans have adapted to changing climates with more or less difficulty, including migration from one place to another.

(2) More recently (last 50 years or so?), scientific knowledge has produced a) evidence of greenhouse gases having a role in changing climate as well as b) other land use practices, e.g. irrigation, urbanization.

***********************************
Where Scientists Diverge:

(3) Teasing out the role of different anthropogenic factors, including greenhouse gases, in specific measures of climate change is difficult for a number of reasons. They include:
a) many other changes have occurred through the same timeframe as increasing greenhouse gases, including land use practices. Correlation is not causation, so different techniques try to remove other causes of variation. Scientists may disagree about these techniques.
b) we don’t have good climate data very far back in time, so have to use proxies, and they may not all be in agreement.
c) records from historic observations by people may be hard to blend with proxies.
d) measuring techniques have changed through time and different scientists have different methods of accounting for this.
e) there just hasn’t been enough time elapsed to detect changes (say something happens once every hundred years to twice every hundred years).

(4) Scientists go ahead and make estimates about 3) and predict the future based on those estimates. But even if we know what proportion of change is due to say, carbon, we don’t know how much carbon will be in the atmosphere in the future. So we model what might happen based on different factors, including economics and technology, that have been difficult or impossible to predict in the past. Given that, how lightly to hold these predictions is an area of disagreement. In general, disciplines with a history of prediction successes and failures tend to have more GDH (generic disciplinary humility).

(5) When it comes to predicting potentially bad impacts of climate change (floods, fires, drought, and so on), though, sometimes the study does not include the work or experience of people in the business of say, managing floods, farming, running dams, or fighting fires. So there is sometimes a disconnect between (the published and believed literature) and those communities and the scientific disciplines that support them (civil engineering, agriculture, wildfire management).

(6) Even if scientists agreed on everything else, no one knows what will happen depending on the rate of slowing of carbon to the atmosphere. Would the climate ultimately “change back”? But what about all those other natural variation and human factors? Since we don’t know, is it best to assume that adaptation will need to be ongoing, and maybe climate science funds should be directed less at exploring all the things that might go wrong, to adaptation and low-carbon technologies. (Where, or does, utility of results factor in to science funding processes?)

(6) Even if scientists agreed on 1-5, they still wouldn’t agree on policy solutions. Even if we just use the example of a carbon tax versus cap and trade, reasonable people, scientist and non-scientist could disagree. If we think of scientific studies as information packets designed to inform policy makers, collectors of information such as the IPCC are very helpful in giving policy makers a summarized packet. But they can still use scientific information outside the packet, or other information entirely. At its best (or best organized), information from various scientific disciplines could give us a target (perhaps measured in global average temperatures) and compare the feasibility of various solutions. When we get to solutions, though, it’s not difficult to think scientists might disagree- for example, bat specialists might not want the landscape covered with wind turbines. And so on.

7) Finally, some scientists say “unpredictable things could happen and be even worse than models project”. This is the idea of tipping points. I’m not sure about the utility of trying to predict the unpredictable. If we take any intervention by humans (say legalizing marijuana) certainly unpredictable things happen and some are worse than we expect. We note them and try to work on them. I think the difference with climate scientists is “because it is the climate of our planet we are thinking of, we should be super-precautionary.” I tend to think these are philosophical ideas “really unpredictable things could happen” and “we need to be careful,” and not actually “science”, although some scientists write papers about it.

*******

I don’t think that this list is exhaustive, so others are welcome to contribute.

Welcome to the Climate Sciences Voyage of Discovery! This will be a weekly feature of The Smokey Wire for the foreseeable future.

I am the veteran of over 30 years of following the climate sciences and their ins and outs. This series will specifically address epistemology (how do you know what you know?), history of science, and sociology of science. That is, we will be taking several steps back from the table of climate science as it is today to investigate how those plates got to the table.

One thing I’ve noticed in online discussions is that people, including scientists, are extraordinarily mean to each other by TSW standards. So I’m going to make this place safe for folks to ask questions or make comments by having a higher level of moderation. Please feel free to invite non-TSW friends along for the Voyage.

Today we’ll start at the very beginning with some epistemology, in this write-up from New Scientist.

This weakness becomes greater as we extend the scientific method into more complex realms with more variables and so more uncertainty, such as social science or climate change. Science progresses legitimately through speculation and hypothesising, but until these speculations are tested by experiment, for a stickler any “knowledge” that emerges from them must strictly be labelled as provisional.

It is a weakness (or strength, depending on your point of view) exploited with gusto by climate-change sceptics, among others. But it points to a blunt truth: if scientific knowledge feels special to you, you are in its in-group. As we grow up, we absorb beliefs from our cultural environment. For some that means accepting scientific knowledge; for others it means “revealed” knowledge, from the Bible, say.

And here’s the thing. For all the bluster about “the evidence”, if you are a scientific believer you too are taking almost all of it on trust. “In principle everybody should be able to replicate scientific results given time, money and training,” says Brigitte Nerlich at the University of Nottingham, UK. “But not everyone has a Large Hadron Collider or a climate-modelling computer.” You are taking someone’s word for it. Like other forms of knowledge, most of science comes down to trusting the source.

Not special, then? Perhaps – except that science also provides mechanisms to justify trust in the knowledge it generates. “Authority in science is earned – at least, when a scientific community is functioning well – by success at predicting, and more generally at analysing, empirical phenomena,” says philosopher Edward Hall of Harvard University. Science’s conclusions are accepted when they fit with our experience of the physical world, and are discarded when they cease to. That makes trust in science a justified true belief – and knowledge that true science generates a cut above the rest. Just don’t take my word for it.

There are three questions in this article that we examine at each stop (topic area) on the rest of the Voyage..

(1) Are the institutions and individuals in different parts of the climate sciences worthy of that trust? We’ll go ahead and look at some scientist and institution behavior.

(2) Many climate studies predict phenomena in the future, but can’t have empirical testing of those predictions. To what extent are those actually “science” in the traditional sense of being tested against empirical phenomena? Is a series of linked assumptions, on its own, “science”? If they are not “science” what are they, and how much of the authority, if any, should they have?

(3) “Science’s conclusions are accepted when they fit with our experience of the physical world, and are discarded when they cease to.” Whose experience? That subdiscipline of the scientific community? Other disciplines? Practitioners (if they exist in a specific science world, say doctors for medical sciences)? The public? At some point, the interpersonal dynamics of the subdiscipline may say “we must be right” and may question the legitimacy of others- others who indeed have experience in the physical world. Those with experience may say that the subdiscipline is more interested in reinforcing their authority and accumulating grant funding, and can’t be objective. IMHO the practitioner/researcher discussion is the most productive for knowledge production, but the occurrence of opportunities for this interaction are highly uneven across the climate sciences.