Guest Post by Willis Eschenbach

The new UN IPCC Assessment Report 6 (AR6) is out, available here. They make it quite clear that a good chunk of what is in the report is not science. Instead, it is the opinions of scientists. They describe what they are using, for example, as:

… structured expert judgement (i.e., a formal, calibrated method of combining quantified expert
assessments that incorporate all potential processes)

First off, there’s no way to know if they’ve included “all potential processes”. We don’t know that much about the climate, and new discoveries are made monthly. Next, what is a “quantified expert assessment” when it’s at home? A numerical guess that they’ve thought a lot about?

And what is the “formal, calibrated method” for combining a bunch of numerical guesses made by “experts”?

Here’s the description of how they assess the likelihood of something, as well as how much confidence they have in that assessment of the likelihood (emphasis mine).

Throughout this Technical Summary, key assessment findings are reported using the IPCC calibrated uncertainty language (Chapter 1, Box 1.1). Two calibrated approaches are used to communicate the degree of certainty in key findings, which are based on author teams’ evaluations of underlying scientific understanding:


(1) Confidence is a qualitative measure of the validity of a finding, based on the type, amount, quality and consistency of evidence (e.g., data, mechanistic understanding, theory, models, expert judgment) and the degree of agreement; and


(2) Likelihood provides a quantified measure of confidence in a finding expressed probabilistically (e.g., based on statistical analysis of observations or model results, or both, and expert judgement by the author team or from a formal quantitative survey of expert views, or both.

A few notes on this quote. First, “evidence” in their world is not just data, observations, and mechanistic and theoretical understanding. “Evidence”, for them, also includes models and expert judgment. As a man who has programmed computer models of a host of systems, I can assure you that model output is “evidence” only in the very simplest of systems. That’s why Boeing and Airbus use wind tunnels to test physical scale models of proposed airplanes whose design is based on computer model outputs … because model outputs aren’t evidence.

And “expert judgment”, whether it is from one expert or “expert judgment by the author team or from a formal quantitative survey of expert views”, is not evidence in any sense. It’s valuable, to be sure, but a hundred years ago “expert judgment” said malaria was caused by lack of hygiene and fresh air, said ulcers were caused by stress, and said that continental plates couldn’t move … was that “evidence”?

It is hubris of the highest order to think that is not happening now in a variety of fields.

In any case, I thought I’d take a look to see just how good their “expert judgment” might be. I noted that they have a new “Sea Level Projection Tool” to give us their expert judgment on what sea-level rise might be in various areas around the planet.

Figure 1. Screenshot of the UN IPCC Sea Level Projection Tool. For a number of locations (blue dots), it gives both future levels and future rates of rise by decade, starting in the 2020s.

Now, I’ve written about sea level before, including discussing one of the best and longest records in the world. This is the San Francisco record, measured about an hour and a half south of where I write this. Here is that record.

Figure 2. San Francisco sea-level record.

In common with about 80% of the long-term sea-level records, there’s no sign of any acceleration in the rate of sea-level rise in San Francisco, either overall or during the last half-century. The sea-level rise has been stable for a century and a half at 2 mm per year, which is just under 8 inches per century.

So with that as prologue, what would the simplest prognostication be for the future San Francisco sea level rise? Me, I’d say given that there’s been a steady 2 mm rise for 170 years, the first guess would be not much different from 2 mm per year … particularly in the current decade, the 2020s.

And what do the UN IPCC models and the “expert judgment” tell us about the future sea-level rise in San Francisco? It depends on the “Scenario”. The UN IPCC uses five different scenarios. In order of increasing CO2 and other greenhouse gas emissions, and thus in order of increasing theoretical temperature rise, they are called the “1.9”, “2.6”, “4.5”, “7.0”, and “8.5” scenarios. In addition, for sea-level rise there are two “low confidence” scenarios. They say:

Two low-confidence scenarios, indicating the potential effect of low-likelihood, high-impact ice sheet processes that cannot be ruled out, are also provided. … Global mean sea level rise above the likely range – approaching 2 m by 2100 and 5 m by 2150 under a very high GHG emissions scenario (SSP5-8.5) (low confidence) – cannot be ruled out due to deep uncertainty in ice sheet processes.

With that as prologue, here are their median (50% quantile) projections of the rate of rise of future San Francisco sea levels, by decade, for those seven different scenarios.

Figure 3. UN IPCC projected rates of sea-level rise by decade. These are the median values.

(Let me note that this reveals one of the huge benefits of this kind of analysis for their “experts”—almost regardless of what the sea level does in the future, they can truthfully say “See, we projected that!”. But I digress …)

However, there’s a deeper and much more serious problem. To highlight it, here are the four least extreme scenarios, 1.9 through 7.0.

Figure 4. Same as in Figure 3, but for the four least extreme scenarios. Again, these are the median values.

I’m sure that you can see the problem. In their “expert judgment” of the model results, the median result (50% quantile) of the models for San Francisco sea-level rise for the current decade is 4 mm per year … say what? It’s been half of that for 170 years, and it’s suddenly gonna double this decade?

Now, at present we’re 2 years into the decade of the 2020s … so for the entire decade to average 4 mm per year, the rate would have to start accelerating today and continue accelerating to the point where it would hit about 7.5 mm per year by 2029. Only then would the decade average 4 mm per year.

It gets worse. The high estimates of sea-level rise (the 95% quantile) give San Francisco rates of rise ranging from 6.4 to 11.6 mm per year … for the current decade.

Sorry, but this is not science in any form. This is a joke. There’s no way on this earth that during the 2020s the average San Francisco sea level rise will average either 4 mm per year or 8 mm per year.

Bear in mind that this is the result of “a formal, calibrated method of combining quantified expert assessments that incorporate all potential processes”. Doesn’t that make you feel all warm and fuzzy about the rest of the UN IPCC AR6 claims?

So remember this monumental sea-level rise madness whenever someone points out that “the IPCC says” something about the future … their “expert judgement by the author team or from a formal quantitative survey of expert views” may not be worth a bucket of warm spit.

My very best to everyone—even in these parlous, fractious times, life is good.

w.

My Invariable Request: When you comment, please quote the exact words you are referring to. It avoids endless misunderstandings.

via Watts Up With That?

https://ift.tt/2VOvNej

August 11, 2021