Science Is Self-Perpetuating, Not Self-Correcting: Sea Ice Case Study

Science Is Self-Perpetuating, Not Self-Correcting: Sea Ice Case Study

Headline from 2014: “Expert predicts ice-free Arctic by 2020 as UN releases climate report“.

Here from NASA is a picture by year of the estimated minimum sea-ice extent, which occurs around September each year.

The first thing to note is the short, the very short, time span. The second is to recall the 1970s in the northern hemisphere were cold. Which is why back then the predictions were for mass starvations, etc. etc., due to global cooling.

The graph would have been more use had it began as far back as possible. Alas, measures were not really taken with any regularity or uniformity until the “satellite age”, which commenced in the late 1970s. Which means nobody has a robust, uncertainty-free estimate of minimum sea-ice extent before satellites.

Which itself means we have to watch our tendency to extend the trend we see at the beginning of this series, and guess the values were even higher prior to 1979. They might have been; indeed, they likely were, since the period of cold was from the mid 1940s through the 1970s. But before the 1940s it was warmer.

Anyway, what we do see is a declining trend, which leveled out, more or less, around 2007, with a minimum in 2012. It is therefore not surprising somebody would make prediction extrapolating only the available data that the September minimum would be zero (0) around 2020.

If only Nature would have cooperated. Which it did not.

Here’s the article opening:

[Ice free by 2020 is] two decades sooner than various models for climatic warming have indicated the Arctic might fully open.

“No models here,” Peter Wadhams, professor of applied mathematics and theoretical physics at the University of Cambridge in England, told the Arctic Circle Assembly in Reykjavik, Iceland, on Sunday. “This is data.”

Wadhams has access to data not only on the extent of ice covering the Arctic, but on the thickness of that ice. The latter comes from submarines that have been beneath the ice collecting measurements every year since 1979.

This data shows ice volume “is accelerating downward,” Wadhams said. “There doesn’t seem to be anything to stop it from going down to zero.

There’s that 1979 again.

“No models here,” said the scientist, which is false. His prediction is a model. The observations are even a model. They are not a direct measurement, but an interpolation from satellite images: they are not raw data. And neither can submarine data, which is periodic and spotty, be sufficient for the whole Arctic. But let that pass.

In any case, a prediction necessarily involves a model. This is inescapable. The model does not have to be formal to be a model. It does not have to exist in computer code. Even brute crude guesses are models.

And this one has been proven false. It has been falsified. It, the model, which is synonymous with the mind of Wadhams, said a thing would happen that did not happen. The model should therefore be abandoned. Science should self-correct—though just who the “self” is, is entirely ambiguous.

Now our original headline came from 2014, as stated. Here is another headline (and subheadline), this one from two years earlier, 2012: “Arctic expert predicts final collapse of sea ice within four years: As sea ice shrinks to record lows, Prof Peter Wadhams warns a ‘global disaster’ is now unfolding in northern latitudes.”

Simple math shows 2012 + 4 = 2016. But it also shows that two years after 2012, Wadhams abandoned his earlier model, which is why he said the end would come in 2020, and no longer 2016. You could say he self-corrected. But his new model has also been proven wrong, as we saw.

Yet here is Wadhams, in 2016, generating another headline, this time at the World Economic Forum: ‘Ships will sail over the North Pole by 2017‘. This suggests that two years after his 2014 forecast that the end would come by 2020, he moved the date up by three years. If you follow me. For in 2016 he stated, ““Next year or the year after, the Arctic will be free of ice”.

Now let’s pop up to 2021, where we have an interview with “world-recognised authority on sea ice and climate change in the polar regions” Peter Wadhams. Who is author of the book A Farewell to Ice. He offers no new specific prediction in that interview.

Our final stop is February of this year. We have an article on the horrors that await when the Arctic is ice free (“he following are already in motion”).

As further stated by Dr. McPherson, there are name-recognition scientists that believe the Arctic will be ice-free very, very soon, as early as 2023…According to Dr. McPherson: September 2023 is the likely date for an ice-free Arctic.

“Then we can expect the full effects next year (2024).” For a detailed analysis of the Arctic: A Farewell To Ice (Oxford University Press) by one of the world’s leading experts on polar ice, Peter Wadhams.

Wadhams is now emeritus, which may be why he no longer making predictions. Before he retired, he received The Polar Medal, “presented by H.M. The Queen”.

Meanwhile, we have this (among many, many) headlines: “Abrupt Ice Retreat Could Produce Ice-Free Arctic Summers by 2040”.

UPDATE Here’s a post I wrote in 2013, about a 2007 prediction that sea ice would be gone by 2013. Bonus: Don Ho!

Subscribe or donate to support this site and its wholly independent host using credit card click here. Cash App: $WilliamMBriggs. For Zelle, use my email:, and please include yours so I know who to thank.


  1. Genpop are dumber than even the typical scientist. Scientists can say anything and they’ll be believed although mostly because genpop don’t actually care. Those that do care about modern science are religious nutters and dogmatists and by and large are over-educated midwits and/or have skin in the game in the form of academia or politics.

  2. john b()

    This year’s Nenana Ice Classic ended on May 8th @4:01 PM

    There have only been 32 other ice outs as late or later than May 8.

    Speaking of 2012 and the above data, the year after [2013], was one of two ice outs on the very late date of May 20.

  3. Clamcake

    Sounds like the Jehovah Witness Effect

  4. That’s an excellent title – and absolutely true.

    Here is a report written on the way that a worthless bit of scaremongering pseudo-epidemiology on TV in 1983 led to *33 years* of well-funded research and careerism, here summarized in 2016 as being without foundations.

    (I knew it was nonsense from when I was a lecturer in epidemiology in 1993)

    Yet still, after 33 years of negative results and refutations of an obviously false association, comes the inevitable conclusion…

    “further research is still needed… research into these possible explanations must continue. Because the people directly, or indirectly, affected by these cancers, and those still living in the area, deserve answers.”

    Of course “the people” deserve answers – and professional researchers deserve to be paid Not to provide answers.

  5. PaulH

    Oh, there’s that pesky reality showing up again at the party, ruining all the fun. No wonder the “Experts” need to redefine reality.

  6. brad tittle

    All measurement is a model. If I were to cover the entire ground with rain gauges in order to assess the rainfall, I would still be imputing a model. No matter how good I make that carpet of gauges, that carpet will be wrong in someway.

    Dear Host. If you manage to just 5% of the people you meet to start to grasp that concept of model, you might save the world…

    I am trying to help. It is not easy though to move people to see the world a little differently. We need to thank god for that. As always it is a blessing and a curse. The blessing is the help it provides in predicting. The curse is that very smart people are capable of not seeing reality in front of their face.

  7. brad tittle

    @Bruce C Charlton — It sounds like you might not be offended when I suggest that the only good epidemiologist is the one who can tell you why his field is worthless.

    i.e. you just did.

    They like to tell us that Covid killed a lot of unhealthy people. I am just amazed at how many unhealthy people without getting vaccinated did not die.

    I am amazed every day at how many accidents DO NOT happen on the roads despite the insanity of the people driving.

    I am amazed at the 80 year old man on my street that still smokes a pack a day continues to be ambulatory and active..

  8. spaceranger

    What is a “name recognition scientist”

  9. Cary D Cotterman

    Bruce G Charlton, what was the “false association”? You’ve left us hanging!

  10. Uncle Mike

    Yes, it’s a clown show. Fakirs in robes spouting Doomsday. But if you think about it, what difference does it make if Arctic sea ice melts away?

    It won’t change sea levels because floating ice displaces exactly the volume of water it would if it melted. Physics, people! Try this experiment: float some ice in a glass of water. Mark the water level. Wait for the ice to melt, and check the water level again. It will be in the same place.

    Arctic sea ice is a feature of our current Ice Age. Before the Pleistocene (1.8 million years long and counting) there was no Arctic sea ice. The last time polar oceans froze was 635 million years in the Cryogenian Period of the Neo-proterozoic Era (so-called Snowball Earth) when life consisted of red and green algae, dinoflagellates, and amoeba.

    So Big Whoop. The normative condition of Planet Earth is no Arctic sea ice all year, not merely in September — for the last 635,000,000 years or there abouts, excluding our Ice Age. The Earth and Life Itself did fine without sea ice. The Doomsayers should be flagellated.

  11. Tom Welsh

    A useful and badly-needed field for scientists to study is why so many scientists “follow the money” instead of earnestly seeking the truth.

    Stated so baldly, of course, the question answers itself to a firxt approximation. People, generally speaking, like money and prestige. And they will satisfy their needs with the minimum of effort, if they can.

    Maybe the next, slightly deeper question, is why the objective truth has fallen so far out of favour in our times. It seems that people are encouraged to think only of their own well-being, without much concern for communities, nations, or the whole human species.

    Could that be simply because there are so many of us that it increasingly feels impossible to care about everyone? Or is it a side-effect of the decline of religion?

  12. I do not have the numbers, but I’d bet $10 that the world’s total ice area is roughly constant over some cycle like 7 or 22 years – so more open lanes in the Bering Sea usually means more ice in Greenland and/or Antarctica.

    One of the characteristics of the climate change mess is that climate is local and the fear mongers tend to cite local change (almost every area on earth has been heralded as warming than everywhere else), but to see claim worldwide effects.

  13. Rudolph Harrier

    Part of the problem is that academia is so enamored with models and methods that even when they not following a narrative they often will only discover the bleedingly obvious, and then pretend like it was only their enlightened investigative techniques that allowed them to discover it.

    I once went to presentation from some statisticians who analyzed cell phone data to determine what the best retention techniques were. Their conclusion: when customers were in contracts that gave big penalties for early cancellation, they tended not to cancel. Of course, this statement only lasted about two minutes of the hour presentation. The remainder was devoted to discussions of wee p-values, gloating about their models, discussions about how much statistics has allowed us to “finally discover”, etc.

    The point is that even when academics aren’t TRYING to fool their audience, they still do not talk much about reality. So it’s very easy to slip in some absolute nonsense.

Leave a Reply

Your email address will not be published. Required fields are marked *