Guest Post by Willis Eschenbach
I see that the usual gloomy hype about arctic sea ice continues unabated. This has been going on for a while. Here’s the dean of failed serial doomcasting, James Hansen, pontificating on the subject back in 2008.
Figure 1. James Hansen’s 2008 crashed and burned prediction of an arctic ice-free summer by 2018.
The latest entry in the prediction sweepstakes is described in the climate alarmists’ favorite newspaper, the Guardian, with the obligatory tear-jerking polar bear photo:
Ice-free summers in Arctic possible within next decade, scientists say
Home of polar bears, seals and walruses could be mostly water for months as early as 2035 due to fossil fuel emissions
The Guardian hype refers to an un-paywalled study entitled Projections of an ice-free Arctic Ocean in Nature Magazine.
Now, you’ve got to be careful to watch the pea under the walnut shell. Well down in the scientific study they say:
The definition of an ‘ice-free Arctic’ has varied over time. Early on, it referred to the nearly complete disappearance of all sea ice, or zero SIE [sea ice extent]. However, as thick sea ice remains north of Greenland and the Canadian Arctic Archipelago more than a decade after the rest of the Arctic Ocean becomes ice free in September a SIE threshold of 1 million km2 [386,000 square miles] became commonplace.
So we’re not really talking about zero sea ice extent. We’re talking about getting down to a million square kilometers of ice, more than a third of a million square miles. That’s a very different question.
Next, they say:
Statistical methods have also been used to provide predictions of an ice-free Arctic. Most of these predictions are based on observed linear relationships between global or Arctic temperature and sea ice cover.
I saw that claim of a linear relationship between temperature and Arctic sea ice extent and said “Hmmm” … let’s start with what’s actually happened to sea ice since the start of the satellite era in 1979. First, the changes up until 2012.
Figure 2. Annual minimum Arctic summer sea ice extent, 1979 to 2012. Yikes! Dropping fast.
Now, this is curious. Minimum Arctic sea ice extent decreased slowly from the start of the satellite record up to the year 2000. From there it started dropping faster and faster, up to 2012.
And that’s why the scientific cognoscenti were so sure it was going to crash. I mean, in 2012, any sane person could see the inevitable. After dropping from 6 million to 4 million square kilometers since the turn of the century, 1 million square kilometers (AKA “ice free”) was obviously just around the corner. That’s why even back in 2008 James Hansen was so sure of an ice-free Arctic in the near future.
However, a funny thing happened on the way to Thermageddon™. Here’s the rest of the Arctic ice extent record.
Figure 3. Annual minimum Arctic summer sea ice extent, 1979 to 2023
Arctic sea ice extent went flat in 2012 and has stayed relatively stable since. I’m sure this made Jim Hansen tear his hair out. And it’s an excellent example of the limitations of climate models.
As far as I know, not one climate model and not one climate scientist predicted that around 2012, the strong downward trend of Arctic ice extent would go flat, and stay that way for a decade. It’s a problem with iterative climate models of chaotic systems. It’s also a problem with humans. Both humans and models tend to calculate that a trend will continue. Neither humans nor models are very good at predicting U-turns or regime shifts in chaotic systems.
So which way will it go from here? Unknown. For example, one of the oddities is that a warmer world is a wetter world, and a wetter world means more snow. Snow on top of ice insulates the ice, making it last longer. It’s a fine example of what I modestly call Willis’s First Law of Climate, which says:
It’s true. The climate has six main subsystems—atmosphere, hydrosphere, lithosphere, biosphere, cryosphere, and electrosphere. Each of these subsystems is constantly exchanging matter and/or energy with all the other subsystems. Each subsystem contains relevant phenomena at all time scales from nanoseconds to millions of years, and at all spatial scales from nanometers to planetwide.
In addition, each of these subsystems has its own chaotic internal resonances, cycles, and regime shifts, which in turn affect all of the other subsystems. Climate is a system of almost unimaginable complexity that we’re only beginning to understand. As a result, claiming that we can model it with current computers is … well … let me call it hubris of the highest order.
But I digress. I started to test one of their central claims, that Arctic ice extent has a “linear relationship” with temperature. So I took the Arctic ice extent data shown in Figure 4 and compared it to a variety of temperature records. To show things clearly, I used just the smoothed versions of each dataset, and set them all to the same range from maximum to minimum.
Figure 4. A view of what the paper erroneously describes as “observed linear relationships between global or Arctic temperature and sea ice cover”. Temperature datasets are inverted because greater temperatures should yield less Arctic ice.
Not only is there no “linear relationship” between Arctic ice extent and temperature, there is very little relationship at all. Yes, in very general terms, warmer temperatures are correlated with less Arctic sea ice extent.
But none of the temperature datasets show the recent leveling out of the Arctic sea ice extent. The closest, as you might expect, is the Reynolds OI sea surface temperature north of the Arctic Circle … but even that one diverges wildly in the early part of the record and matches poorly in the recent part.
So I’m gonna say that their claim of “observed linear relationships between global or Arctic temperature and sea ice cover” is simply not true.
Finally, to return to the topic of the study, the “projections” of the year when we’ll see the first ice-free Arctic are pretty hilarious. They are so broad that if ice-free Arctic conditions were to occur at any time between now and 2150, someone’s model could claim credit for it. Below I show the nine different models and model averages reported in the study.
The first thing to notice is that contrary to the claims of an imminent ice-free summer, in fact we’re already past the claimed earliest ice-free date of five of the nine models.
Figure 5. Earliest and latest ice-free dates of the nine models and model groups. Horizontal lines connect the names of each model with the boxes whose left and right edges show the earliest and latest ice-free dates according to that model.
The dark blue region around 2035 to 2045 shows what the models say is the most likely time when we’ll see an ice-free Arctic. However, given their accuracy to date, that should be taken, not with a grain of sale, but with a kilo of salt …
Finally, the authors have used only the most extreme climate scenarios. The current general agreement among mainstream climate scientists is that these extreme scenarios (SSP5-8.5, RCP8.5, and A1B) are all highly improbable, and are not recommended for use because they lead to very unlikely projections. Despite that, the authors have selected them, presumable because it jacks up the fear and anxiety in the public … which of course will guarantee that these authors continue to receive funding in the future.
CONCLUSIONS
• There is no simple linear relationship between either global or Arctic temperature and Arctic sea ice extent.
• Models are just a reification of the understandings and misunderstandings of the programmers.
• They are supposed to be “physics based”, but if they truly were, there wouldn’t be such huge disagreements between models.
• The use of the most extreme scenarios is clear evidence of the alarmist views of the authors of this study.
And a final thought. The world of climate science would be well served to declare a moratorium on these endless failed serial doomcasts, and study the climate of the past instead. The models are a joke in that regard. Consider that the models give climate sensitivities that range from about 1.5° C to 6.5° C per doubling of CO2. Despite that, they all do a reasonable job of emulating the historical temperature record … and if they are “physics-based” as the modelers claim, that’s physically not possible. I discuss this in my post Dr. Kiehl’s Paradox That is clear evidence that they are merely tuned to match the past, and thus they have no credibility in predicting the future.
Here, it’s nighttime. I walk outside, and the redwood forest surrounding our house is perfectly silent, not a breath of wind. The constellation Orion burns low in the western sky, and Leo is right overhead. The crisp night air is redolent of springtime, of life bursting forth on all sides.
Ah, dear friends, what a wonderfully mysterious universe, with far more questions than we will ever have answers.
My best to you and yours,
w.
Related