Causation Of Climate Change, And The Scientific Method

/

Let’s have yet another go at trying to apply the scientific method to the subject of causation of climate change. This is just basic logic, and not that complicated. We can do it.

As simple and basic as this is, you will shortly see that the agglomeration of all of the world’s leading “climate scientists” can’t figure it out. They are completely lost and befuddled. Check me and see if I’m wrong.

The proposition we are addressing is the one for which you see a constant drumbeat of advocacy. It runs something like, “the climate is changing, and we are the cause.” OK, nobody denies that the climate is changing; but how about the “we are the cause” part? What is the proof?

Let’s apply the scientific method. We start with the basic maxim that “correlation does not prove causation.” Instead, causation is established by disproof of all relevant alternative (“null”) hypotheses.

Everybody knows how this works from drug testing. We can’t prove that drug A cures disease X by administering drug A a thousand times and observing that disease X almost always goes away. Disease X might have gone away for other reasons, or on its own. Even if we administer drug A a million times, and disease X almost always goes away, we have only proved correlation, not causation. To prove causation, we must disprove the null hypothesis by testing drug A against a placebo. The placebo represents the null hypothesis that something else (call it “natural factors”) is curing disease X. When drug A is significantly more effective at curing disease X than the placebo, then we have disproved the null hypothesis, and established, at least provisionally, the effectiveness of drug A.

Back to climate change. The hypothesis is “humans are causing significant climate change.” An appropriate null hypothesis would be “observed climate change can be fully explained by some combination of natural factors.” How might you test this?

The most obvious test would be to ask, in earth’s recent history, has it been warmer than the present — the present having been the subject of significant human greenhouse gas emissions? If periods in the recent past prior to human emissions have been warmer than the present, then quite obviously some combination of “natural factors” is sufficient to bring about temperatures as warm or warmer than we are experiencing.

And it doesn’t matter whether or not we know what the alternative “natural factors” might be, any more than, in the failed drug trial, it matters whether or not we know why the placebo beat the experimental drug. In the failed drug trial, it could have been the human immune system, or it could have been gut bacteria, or it could have been the weather, or anything else. The fact is that, whatever they might have been, the “natural factors” outperformed the experimental drug.

For the test of the climate hypothesis, consider a December 28 blog post from retired physicist Ralph Alexander titled “New Evidence That Ancient Climate Was Warmer Than Today’s.” Alexander summarizes the results of two recent studies:

The Margaritelli, et al., piece analyzes proxy data from “fossilized amoeba skeletons found in seabed sediments” to reconstruct Mediterranean Sea temperatures over the past 2000 years. “The ratio of magnesium to calcium in the skeletons is a measure of the seawater temperature at the time the sediment was deposited; a timeline can be established by radiocarbon dating.” Conclusion:

With the exception of the Aegean data, the results all show distinct warming during the Roman period from 0 CE to 500 CE, when temperatures were about 2 degrees Celsius (3.6 degrees Fahrenheit) higher than the average for Sicily and western Mediterranean regions in later centuries, and much higher than present-day Sicilian temperatures.

The Baraniuk study, from Norway, analyzes large new finds of ancient artifacts, including arrows, arrowheads, and clothing, that have been revealed by recent retreats of glaciers in that country. But of course, the existence of the artifacts in these areas implies that the areas were not covered in ice at the time the artifacts were deposited:

That the artifacts come from several different periods separated by hundreds or thousands of years implies that the ice and snow in the region must have expanded and receded several times over the past 6,000 years. During the Holocene Thermal Maximum, which occurred from approximately 10,000 to 6,000 years ago and preceded the period of the stunning Norwegian discoveries, global temperatures were higher yet. In upper latitudes, where the most reliable proxies are found, it was an estimated 2-3 degrees Celsius (3.6-5.4 degrees Fahrenheit) warmer than at present.

Whether 2000 years ago (the Roman Warm Period) or 6000 years ago (the Holocene Thermal Maximum), these periods clearly long preceded any significant human greenhouse gas emissions from the burning of fossil fuels. Obviously, then, some combination of “natural factors,” whatever they may be, is sufficient to cause terrestrial temperatures to increase to levels as high or higher than we are experiencing today, in the era of human use of fossil fuels.

I should mention that the two papers discussed by Alexander are just the latest of many dozens of studies giving evidence for the proposition that times in the recent geologic past — either the Medieval Warm period, or the Roman Warm Period, or the Holocene Thermal Maximum — were warmer than today. One collection of many papers, mostly focusing on the Medieval Warm Period, can be found at Craig Idso’s CO2 Science website.

You would think that mainstream climate “science” would be focused like a laser beam on trying to deal with these early periods that were warmer than today. But instead, these guys have almost entirely taken a different approach. They call their approach “detection and attribution.” Of many examples of the art, here is a major paper from 2018, sponsored by the UN’s Intergovernmental Panel on Climate Change, with the title “Detection of Climate Change and Attribution of Causes.” The authors are a who’s who of the official climate establishment, including the likes of Phil Jones, Michael Mann, Gerald North, Gabriele Hegerl, and Ben Santer.

Instead of evaluating whether available data refute either their main hypothesis (human causes) or the null hypothesis (natural factors), these guys adopt a different approach which I would describe as “we can’t think of anything else other than human greenhouse gas emissions that could be causing this, so therefore human emissions it is.” They create so-called models of what they think natural factors might cause in the way of warming, and then test those against the data. Since when does that prove anything? The article is very long and riddled with nearly impenetrable jargon that makes it near impossible to get a good quote, but here are a couple of the best:

Attribution studies have applied multi-signal techniques to address whether or not the magnitude of the observed response to a particular forcing agent is consistent with the modelled response and separable from the influence of other forcing agents. The inclusion of time-dependent signals has helped to distinguish between natural and anthropogenic forcing agents. As more response patterns are included, the problem of degeneracy (different combinations of patterns yielding near identical fits to the observations) inevitably arises. Nevertheless, even with the responses to all the major forcing factors included in the analysis, a distinct greenhouse gas signal remains detectable. Overall, the magnitude of the model-simulated temperature response to greenhouse gases is found to be consistent with the observed greenhouse response on the scales considered.

And here’s another:

To detect the response to anthropogenic or natural climate forcing in observations, we require estimates of the expected space-time pattern of the response. The influences of natural and anthro- pogenic forcing on the observed climate can be separated only if the spatial and temporal variation of each component is known. These patterns cannot be determined from the observed instru- mental record because variations due to different external forcings are superimposed on each other and on internal climate variations. Hence climate models are usually used to estimate the contribution from each factor.

Apparently, this kind of mumbo jumbo is good enough to fool pretty much all of academia, and most all journalists, not to mention gaggles of billionaires. But how about the Medieval Warm Period and the Roman Warm Period? Don’t those refute the whole thing?