Correlation, Causation and Catastrophists

It was World Environment Day on June 5 and UN Secretary-General António Guterres marked the occasion with a speech at the American Museum of Natural History. The “moment of truth” for climate action, he said, is now:

We are playing Russian roulette with our planet. We need an exit ramp off the highway to climate hell.  And the good news is that we have control of the wheel.  The battle to limit temperature rise to 1.5 degrees will be won or lost in the 2020s – under the watch of leaders today.

We have what we need to save ourselves. Our forests, our wetlands, and our oceans absorb carbon from the atmosphere. They are vital to keeping 1.5C alive, or pulling us back if we do overshoot that limit. We must protect them. 

It was déjà vu all over again. Guterres gave the same warning in November 2022 at the Conference of the Parties (COP27) in the Red Sea resort of Sharm el-Sheikh: nations must cooperate or face “collective suicide” from climate change:

Humanity has a choice: cooperate or perish. It is either a climate solidarity pact or a collective suicide pact. We are on a highway to climate hell with our foot still on the accelerator.

The Secretary-General’s concern was based on a new report from the UN World Meteorological Organisation (WMO) and data from the EU’s Copernicus Climate Change Service (CCCS). The WMO predicts there will be more record temperatures in the next five years. 

There is an 80 percent likelihood that the annual average global temperature will temporarily exceed 1.5°C above pre-industrial levels for at least one of the next five years. This is a stark warning that we are getting ever closer to the goals set in the Paris Agreement on climate change, which refers to long-term temperature increases over decades, not over one to five years. Key messages

♦ 80% likelihood of at least one year temporarily exceeding 1.5°C between 2024-2028

♦ Short-term (annual) warming does not equate to a permanent breach of the lower 1.5°C Paris Agreement goal

♦ Likely that at least one of next five years will be the warmest on record, beating 2023

♦ Report highlights urgency of climate action

The global mean near-surface temperature for each year between 2024 and 2028 is predicted to be between 1.1°C and 1.9°C higher than the 1850-1900 baseline, according to the WMO report. It says that it is likely (86%) that at least one of these years will set a new temperature record, beating 2023 which is currently the warmest year.  — WMO media release, June 5, 2024

The WMO clearly expects the planet to behave in a way that supports its alarmism. Such precision too! While the “global temperature” – whatever that is — is predicted to exceed 1.5°C above the “pre-industrial level” in the next five years, it will do so only “temporarily”. Indeed, “temporarily” is mentioned three times in the media release.

Gaia forbid, but if anticipated warming becomes a “permanent breach” of the Paris Agreement goal – a number conjured up by UN climate activists — for the next five years and beyond, presumably it would be game over: no point pretending  we can do anything about it, no “urgency for climate action” and. mercifully. no more Net Zero targets. 

Every religion has to preach a message of redemption, especially a new one. If the WMO’s predicted temperature rise is only temporary, presumably salvation is still possible: but only if we all sing more loudly from the UN song-sheet, commit to the faltering energy “transformation” with renewed vigour, deepen our belief in a future RE utopia, and transfer many more billions to the developing world.

Consider, dear reader, an alternative moment of truth: that the Secretary-General’s “highway to climate hell” is not only paved with good intentions. Imagine it is also paved with hyperbole, projections masquerading as predictions, confusion about causation, and  money being squandered pursuing a delusion: that human control of the planet’s climate, on any and every scale, is somehow possible.

How can we distinguish a projection from a prediction? At the risk of offending modellers and ministers of climate change everywhere, the former is speculation in the same category as astrology, with or without computers and AI. As for the latter, they are invariably based on laws derived from nature and validated by observation. There are no laws of climate change, no Newton or Einstein, but plenty of simulations of “counterfactual worlds”.

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 [computer modelling.] — A Hannart, et al., American Meteorological Society, 2016

Fortunately, The Devil’s Dictionary of Climate Change defines them for us. It also gives three intriguing statements from the UN Intergovernmental Panel on Climate Change (IPCC). No longer easily accessible, perhaps they are another casualty of agency anxiety about “climate change”:

1/ A projection of the future response of the climate system to emission scenarios of greenhouse gases and aerosols, or radiative forcing scenarios, based on hunches, guesses, gut feeling, confirmation bias, etc., and model simulations.

2/ A semantic tactic to ensure that when an outcome diverges from reality, the orthodoxy can claim it was not a prediction.

“Climate projections are not predictions,” explains the IPCC ‘s glossary from 1997, “as they depend upon the emission/concentration/ radiative forcing scenario used, which are based on many assumptions, concerning, e.g., future socio-economic and technological developments, that may or may not be realised, and are therefore subject to substantial uncertainty.” It continues:

A process is called non-linear when there is no simple proportional relation between cause and effect. The climate system contains many such non-linear processes, resulting in a system with a potentially very complex behaviour. 

Climate prediction, n., a futile activity or ritual claiming to produce a “most likely” description of the Earth’s future climate, generally sufficiently distant to be incapable of testing in a human half-life time.

 In climate research and modelling, we should recognize that we are dealing with a coupled non-linear chaotic system, and therefore that long-term prediction of future climate states is not possible. — IPCC 3rd Assessment Report, 2001

For a theory or a model to remain credible it has to predict something and that prediction must be compared to reality. Every scientist knows the remark that ‘many a beautiful theory was killed by an ugly fact.’….To my mind the most powerful and convincing proof of a theory or model is what it says about the future… Nature has a way of humbling even the best “state-of-the-art” predictions. — Dr D Whitehouse, astrophysicist, 2016

Causation is another tricky concept. The rooster crows every morning. The sun comes up without fail. Yet the crowing rooster did not cause the dawn. Correlation is not causation.

The concept still causes anxiety in climate researchers, who tend to replace it with “causal inference”, “is consistent with”, “can be attributed to”, “is in keeping with”, or a similar phrase. As A Hannart, et al., suggest below, the “emergence of clear semantics for causal claims and of a sound logic for causal reasoning is relatively recent”.

Among other lacking items, perhaps the most important one regards the absence of definition for the word cause. Several recurrent controversial arguments in the realm of [climate] 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

A causal inference asserts that one factor has influence over another. Note, however, the influence must be inferred because it cannot be observed directly, as philosopher David Hume suggested almost three centuries ago.

When we look about us towards external objects, and consider the operation of causes, we are never able, in a single instance, to discover any power or necessary connexion; any quality, which binds the effect to the cause, and renders the one an infallible consequence of the other…But I keep my mind still open to instruction. — David Hume, 1737

Reasoning often proceeds from cause to effect, as in a “climate prediction”; or from effect to cause, as in explanations for changes in the weather.

So a causal inference is that doing something will lead to something else. Keeping our minds open to instruction, consider the case of climate change. Any claim about the impact of carbon dioxide depends crucially on whether all other possible influences can be identified and quantified, presumably still a huge challenge despite agency and expert claims that it is being or has been done.

So a valid causal inference must satisfy several tests. Has a correlation been confused with a causal relationship? Has an explanation been missed because of a spurious cause-effect relationship? Has temporality been confused with causality, the post hoc fallacy, as in the rooster case above? While a cause must precede an effect, preceding it is not a sufficient condition for something to be a cause. And so on and so forth.

For me, the more complex a problem the more daunting the task of finding an unambiguous solution not corrupted by confirmation bias, especially in manufacturers of “counterfactual worlds”. The sceptical response: “climate models are not evidence, just calculations.”

Under the auspices of the World Meteorological Organization, forecasts of the near-term climate (1-5 years lead-time) are now routinely produced by several international operational centres. CSIRO is a leader in this area having generated a massive data resource of the climate over the past six decades comprising close to one hundred replicant “digital” earths generated through assimilation of a comprehensive set of atmospheric, ocean and sea-ice observations and a state-of-the-art global climate model. — CSIRO media release,2022

Consider this recent paper: Inferring causation from time series in Earth systems science. Detection and attribution researchers are attempting to “quantify the evidence for a causal link between external drivers of climate change and long-term changes in climatic variables. The goal is to first detect a change and then attribute this change to the contributions of multiple anthropogenic and natural forcings, and from internal variability. Such research questions require counterfactual worlds, which can only be constructed with climate models that are then statistically analysed.”

Good luck with solving such riddles, surely a task comparable to finding a tortoise with a moustache. That said, the CSIRO is attempting to do something even more ambitious in an AI for Missions Program. Its Causal inference and prediction in complex multiscale systems project “seeks to identify robust relationships between climate and socio-economic impacts. The challenge is looking at all the noisy data and identifying features in time that are evolving and causally linked to the climate change signal.”

This project aspires to deliver a world-first ML software platform for the effective estimation of climate risk via early detection and predictions of systematic changes to the physical system integrated with socio-economic, eHealth and financial databases for the delivery of impactful climate services.— CSIRO media release, 2022

Good luck with that exercise too. At least the CSIRO admits “the identification of climatic regimes and attribution of the factors driving transitions between periods of drought or increased rain associated with persistent synoptic patterns embedded within our vast data sets represents a huge challenge.”

Somewhere a rooster crowed: “correlation is not causation”. Few folk with serious skin in the renewable-energy game heard it.  So Net Zero continues to accelerate down the highway to Pandemonium and oblivion, especially in countries like this one. As Shakespeare had Edmund say in Act I, Scene 2 of King Lear:

This is the excellent foppery of the world, that, when we are sick in mind – often the surfeits of our own imaginings – we make guilty of our disasters the sun, the climate and the carbon atom: as if we were villains on necessity; fools by a diabolical compulsion; knaves, thieves, prognosticators and promulgators of false facts by spherical predominance; drunkards, liars, cheats, dissimulators, proselytisers, slime balls and all that we are evil in, by an enforc’d obedience of atmospheric influence: an admirable evasion of whoremaster man, to blame his earthly afflictions on a molecule in the air.