With the international political, financial and reputational stakes so high, it was only a matter of time before climate change appeared in the dock, handcuffed to its partner in prognostication, the dodgy discipline of extreme weather attribution.

Attribution, n., the art of evaluating the relative contributions of multiple causal factors to a change or an event, according to one’s prejudices.

To make sense of the climate change scene today, it is best to begin with the end game: the orthodoxy’s search for an argument, however abstruse, that will stand up in court. It needs one sufficiently “robust” to ensure developed countries—still effectively on trial in the United Nations, where a protracted “loss and damages” claim awaits resolution—and fossil fuel companies are legally liable to pay multi-billion-dollar “climate reparations” to the alleged victims of “carbon pollution”, be they in the developing world or in the path of a natural disaster.

Indeed, the credibility of the “relatively young science” of extreme weather attribution, the legitimacy of its ambition to “tease out the influence of human-caused climate change from other factors”, the whole alarmist movement and fate of the UN’s Green Climate Fund, all crucially depend on delivering such a legal argument.

How did we get to this point? When the climate change meme was planted successfully in the collective mind a decade ago as the most serious existential threat facing humankind, the orthodoxy wanted it to stay there. A sense of public anxiety had to be maintained, despite the risk of apocalypse fatigue syndrome.

So it created an Attribution of Climate-related Events (ACE) initiative. The international research agenda gradually shifted to the tricky territory of extreme weather attribution.

ACE’s first workshop was held on January 26, 2009, in Boulder, Colorado, at the Pei-designed National Center for Atmospheric Research (NCAR) Mesa Lab. Attendees included Myles Allen (Oxford University), Martin Hoerling (NOAA, USA), Peter Stott (UK Met Office, Hadley Centre), Kevin Trenberth (NCAR) and David Karoly (University of Melbourne). Its objective was to:

develop a conceptual framework for attribution activities to be elevated in priority and visibility, leading to substantial increases in resources (funds, people, computers) and both a research activity and a framework for an “operational” activity, that sets forth a goal of providing a lot more concrete information in near real time about what has happened and why in weather and climate.

ACE later released a four-paragraph statement. Its mission would be: “to provide authoritative assessments of the causes of anomalous climate conditions and EWEs” (extreme weather events), presumably for government agencies and the Intergovernmental Panel on Climate Change’s 2013/2014 Fifth Assessment Report (AR5).

But just how “robust”—one of the orthodoxy’s favourite adjectives—was the climate modelling underpinning this grand design? How could it be sold to the public, given the challenging uncertainties? ACE participants agreed they would need “increased real-time numerical experimentation activity” and something else too, a narrative that would ensure public interest.

To succeed, everyone would have to sing from the same song-sheet. There would have to be consistent use of terminology and close collaborative teamwork “to maintain an authoritative voice when explaining complex multi-factorial events such as the recent Australian bushfires” (my italics).

Three years later, Dr Peter Stott, the UK Met Office’s head of climate monitoring and attribution, again stressed the importance of reining in mavericks and having a unified voice, this time in a conference paper:

Unusual or extreme weather and climate-related events are of great public concern and interest. Yet there are often conflicting messages from scientists about whether such events can be linked to climate change.

All too often the public receives contradictory messages from reputable experts. If the public hears that a particular weather event is consistent with climate change they may conclude that it is further proof of the immediate consequences of human-induced global warming. On the other hand, if the public hears that it is not possible to attribute an individual event, they may conclude that the uncertainties are such that nothing can be said authoritatively about the effects of climate change as actually experienced.

Above all, then, ACE’s experts must not confuse the public. Imagine the furore if too many folk begin to suspect that nothing “can be said authoritatively about climate change”; other than that (unpredictable) change is what the planet’s climate (and weather) does and always has done. As for not seeing EWEs as “consistent with” an alleged human cause—namely the presence of an atmospheric trace gas vital for all organic life—few paid-up members of the Carbon Cargo Cult Club wanted to go there.

Stott did admit, however, that the initial studies “highlight many of the challenges still to be faced”, the “considerable uncertainties that remain”, the uncertainty around alleged “causal links” and “whether relevant processes are captured adequately” by models. Several years later, he was still concerned that “many questions remain as to current capabilities to robustly attribute the contribution of anthropogenic climate change to the risk of many extreme weather and climate events”.

Shifting the focus from long-term climate change—say a French-fry fate in 2100—to yesterday’s weather enabled the orthodoxy and its cheer squad—a decarbonising political elite, carbon capitalists, renewable energy aficionados, militant environmentalists and worried religious groups—to promulgate ambiguous “probability” attribution statements after every hurricane, storm, flood, drought and bushfire. Two examples:

Human influences have increased the risk of some extreme weather and climate-related events, reduced the risk of others and, for some, may not have affected the risk substantially.

A finding that human influence has not contributed substantially to the magnitude of a particular EWE may not be incompatible with a finding that human influences substantially altered the odds of such an event happening (especially a particular threshold).

For many, however, there was nothing ambiguous here. Brian Merchant, an American journalist, wrote last September that “science denial” should be treated as a form of gross negligence:

Climate change denial should be a crime. [Yet] we refuse to hold the negligent accountable. We refuse to strike back with adequate force at the toxic climate denial that corrupts our public policies.

Natural disasters that not so long ago were deemed acts of God are being rebranded as the outcome of “science-denying recklessness”. According to Merchant, it is “happening all over the country, in various guises, and many more lives are in danger”:

I hope Hurricane Harvey is a lightning rod that makes this clear: Climate change denial can and will leave people dead. It has never been more evident than now that it is not only scientifically wrongheaded but dangerously and morally abject.

There is, however, a key challenge facing folk who believe that “climate change” is responsible for everything bad: persuading a court to accept the so-called detection-and-attribution community’s probability games as compelling evidence.

Meanwhile, January’s big storm in the eastern United States not only brought record low temperatures but also confusion. The public struggled to reconcile its arrival with the media’s relentless global warming scare campaign.

Scott Detrow asked the attribution question on US National Public Radio’s Morning Edition on January 4, during a chat with science reporter Rhitu Chatterjee: “So I have to ask, as we do any time there is an extreme weather event, is climate change any kind of a factor here?” By “climate change”, Detrow—and most media folk—meant “dangerous anthropogenic climate change”, also known as global warming. Chatterjee replied:

Scientists have a much better idea of climate change’s influence with regard to hurricanes, but for these kind of winter storms, they really don’t know. The models are not quite there yet.

The orthodoxy’s reaction goes to the heart of the controversy. David Karoly insisted climate modelling showed that such “cold snaps” are becoming less common as a result of global warming. With “rapid attribution analysis”, experts apparently now know that “climate change is linked to a dangerous pattern of major weather events”. But precisely how, and by what kind of link, seems as vague and contestable as ever.

One person who commented on the January EWE was Marshall Shepherd, a director of the University of Georgia’s Atmospheric Sciences Program and past president of the American Meteorological Society. Shepherd posted two articles on Forbes, one on December 28 last year, titled: “A Response for People Using Record Cold US Weather to Refute Climate Change”; with an update on December 30: “Three Observations about Conversations on the Extreme Cold Weather This Week”:

A distracting narrative is emerging. There have certainly been some misinformed tweets, posts, and innuendo about the cold weather and the global warming narrative. One thing that I have observed is a narrative in certain circles that some people are being hypocrites by pointing out that cold weather doesn’t refute climate change but not making a similar point during the warm season.

I make that point all of the time and as recently as my aforementioned article. I also cautioned during Hurricane Harvey and other 2017 hurricanes that there may be climate change DNA in those storms, but it is important to let peer-reviewed attribution studies confirm it. 

Climate change DNA? The Australian science writer Joanne Nova responded:

Don’t cringe now, but these are not equal and symmetrical. Saying that a hot spell “may contain climate change DNA” is not the same as saying “cold spells are weather”. If a hot spell could contain DNA (we molecular biologists don’t think so) then so can a cold one. If hot DNA can spell “manmade global warming”, then cold DNA can spell “the models are wrong”.

Shepherd was one of several experts on a 2016 National Academy of Sciences (NAS) study. It provided, he said, “the most robust understanding of where the science is on the topic. I summarized the findings in my earlier Forbes article. I highly recommend giving this report a read (it is verbose but worth it) before weighing in on attribution.”

Judging by the 186-page report, Attribution of Extreme Weather Events in the Context of Climate Change, the quest for “predictive forecasts of future extreme events at lead times of days to seasons, or longer, accounting for natural variability and anthropogenic influences” still seems futile. How, for example, do we reconcile this ambitious goal with the US climatologist Judith Curry’s assessment of the key issues?

On short time scales (decade to centuries), there is no satisfactory way of sorting out forced climate variability from natural internal climate variability unless you have a really good climate model that can adequately handle the natural internal variability on the range of time scales from years to millennia. Empirical methods have yet to do this in any sensible way … Until we better understand natural internal climate variability, we simply don’t know how to infer sensitivity to greenhouse gas forcing. The issue of how climate will change over the twenty-first century is highly uncertain. Oversimplification and overconfidence on this topic have acted to the detriment of climate science. As scientists, we need to embrace the uncertainty, the complexity and the messy wickedness of the problem. We mislead policy makers with our oversimplifications and overconfidence.

One hesitates to question a document like the NAS report. Nevertheless, when the word uncertainty appears 135 times the attribution goal seems little more than a mirage. As for robust (forty-two times), it is not a scientific term. Is it being used to give an impression of precision about a phenomenon or relationship where clear causal links are either absent or undetectable?

When told the “science of extreme weather attribution is like baking cookies” in a two-minute video, and analogous to epidemiological research or DNA analysis, you might conclude that perhaps there is too much heat in the NAS kitchen.

Despite some of Curry’s concerns also appearing in peer-reviewed papers, such this one—“Causal Counterfactual Theory for the Attribution of Weather and Climate-Related Events”—the show must go on. Consider the three examples below.

While almost a decade has passed since the inaugural ACE meeting, the detection-and-attribution community is apparently yet to agree on a definition for “cause”:

Among other lacking items [in the event attribution framework], perhaps the most important one regards the absence of definition for the word cause. Several recurrent controversial arguments in the realm of event attribution may possibly be related to this lacking definition of causality: for instance, an argument often made (Trenberth, 2012) is that any single event has multiple causes, so one can never assert that CO2 emissions, nor any other factors, have actually caused the event. (A. Hannart et al, American Meteorological Society, January 2016, p. 100)

The community also has had to conjure up a “counterfactual world”, a hypothetical “control” world assumed to be impacted only by natural forcings and internal variability:

Most unfortunately, in the climate sciences, no such sample of Earth-like climate systems is accessible to natural observation and even less so to experimental testing … With such strong limitations on the natural observation side and with in situ experimentation inaccessible, we are left with the only remaining alternative: so-called in silico experimentation. (Hannart et al, p. 105)

In silico, Latin for “in silicon”, refers to experimentation performed solely on a computer or using computer modelling. Climate models clearly have flaws and arguably are not fit for purpose:

Another serious difficulty is that climate models, including the most detailed GCMs, are simplified representations of reality that are affected by both numerical and physical modeling errors. Thus, the real causal effects may differ from the model causal effects. (Hannart et al, p. 106)

Such admissions rarely, if ever, appear in the orthodoxy’s media releases or the public statements of its disciples. They are the dark secrets in the alarmist attic.

The current prognosis for Planet Earth, then, is based less on empirical science and more on computer modelling and contestable probability analyses. Frustrated by the inability of the modelling to validate its ambitions, the detection-and-attribution community has focused on developing more arcane theory.

Computer games involving counterfactual worlds, causal theory, convoluted semantics and who knows what else have superseded—some might say usurped—the traditional role of observation and falsification. Given the complexity of the task, the state of available datasets and the entrenched politicisation of climate science, this is no surprise. But for how much longer will it ensure the alarmist narrative still has a race to run?

Ironically, chapter two of the NAS report is on “framing”. It begins with this quotation from Werner Heisenberg: “What we observe is not nature itself, but nature exposed to our method of questioning.” What questions did the detection-and-attribution community ask to convince itself of the presence of an “anthropogenic forcing” in the climate system?

To answer a question scientifically, the question needs to be posed in a way that is amenable to scientific analysis. The question often asked by the public and the media—“Was this extreme event caused by anthropogenic climate change, yes or no?”—is not well posed because the word “cause” can have several different meanings.” (NAS Report, March 2016, p. 27)

There is mounting pressure on researchers to come up with arguments that will stand up in court. The following comment actually compares “the problem of climatic event attribution” to a lawsuit:

We can consider two typical perspectives: the ex post perspective of the plaintiff—or the judge or insurance contract holder—and the ex ante perspective of the planner—or the policymaker or campaigner. In the first case, the question of who is to blame for the event that occurred—with potentially many implications of its answer—is central. The problem of climatic event attribution can thus be compared to a lawsuit and actually does already appear in courts (Adam 2011); we may primarily seek to determine responsibilities for the event and its aftermaths, where responsibility is understood in a legal sense, that is, in a necessary causation sense. Event attribution thus requires the adversarial debate typical of a lawsuit in order to cautiously balance incriminating versus exonerating evidence, that is, to evaluate the main cause under scrutiny; for example, anthropogenic forcings, as well as each and every possible alternative explanation; for example, natural forcings or internal variability of the climate system. If the resulting probability of necessary causality is high enough, then human responsibility is established and a ruling in theory may follow, as it does in litigation cases. (Hannart et al, p. 104)

Is a cautiously balanced assessment of the evidence possible? The orthodoxy has insulated itself from deeper scrutiny for years with a cordon sanitaire of unaudited black-box models and “consensus” science. In any case, given the revelations earlier, is it in a position to evaluate objectively—that is to say scientifically—“each and every possible alternative explanation”? Or has it become so affected by confirmation bias that it devotes too much energy to ensuring the “probability of necessary causality” is sufficiently high to indicate human responsibility?

Two years ago, Peter Stott and his detection-and-attribution colleagues urged greater dialogue with the so-called “stakeholders” in attribution research. But is there a risk the integrity of the investigative process could be compromised, at least in some cases?

Better information about climate risks could be of potential use to the insurance industry, to regional managers developing climate adaptation strategies, to litigators, to policy makers and for disaster risk reduction. But profitable use of such information requires a dialogue between stakeholders and scientists that allows the development of trust as a way to develop the credibility, saliency, and legitimacy of scientific findings. (P. Stott et al, WIREs Climate Change, 2016; emphasis added)

When scientists and stakeholders are urged to “work together to ensure the science supports stakeholder needs”, could it lead inadvertently to a conflict of interest?

As for “adversarial debate”, it seems to be moving from the blogosphere—not the academy—to the court. But if the early signs are any guide, this could backfire on the plaintiffs.

Consider the latest case. New York City and other coastal cities are suing the world’s five largest listed oil producers, seeking to hold them responsible for present and future damage from climate change and sea-level rise. They claim that oil and gas companies knew decades ago their products were causing dangerous climate change, but concealed that knowledge and funded advocacy groups opposing climate policies.

Plaintiff lawyers are running a classic argument by analogy, based on successful litigation against big tobacco in the 1990s. They claim that oil companies should face penalties because they know their products harm the environment in the same way tobacco suppliers knew smoking causes health issues.

However, the case will probably fail because of the orthodoxy’s fatal flaw, its failure to prove (beyond reasonable doubt) either anthropogenic causation or attribution. According to Michael Burger, director of Columbia University’s Sabin Center for Climate Change Law, it will be legally difficult to trace precise amounts of causation and blame to the oil industry, let alone particular companies. Showing causation and attributing blame was much easier with cigarette companies.

One person who might be keen to testify for the defendants is Judith Curry.