Skip to content

Christopher Essex: Small Minds Don’t Understand The Importance Of Unorthodox Thinking To Science

Christopher Essex, Financial Post

Beheading, war crimes, prison, death threats. Perhaps you think I’m talking about IS. No, it’s just the routine hyperbolic language of media climate activism, and the minds of our cultural overlords.

Next December policymakers will try again to save the world from what they don’t understand by proposing policy on something else, which won’t work anyway.

In anticipation, climate effusions grow. The great cultural climate PR machine slowly stretches its enormously well funded muscles, warming up for another hysterical run at utopia. Its weird unphysical world has certified carbon-free sugar, oxygen-free carbon dioxide (i.e. carbon) and the alleged carbon in glass causing interiors of cars to be warm on sunny days. Unfortunately the dense nonsense of the previous sentence won’t send the appropriate clown-time signals that it should.

Every debate has boundaries; beyond them lies other subjects. Climate debates are unique: Their boundaries exclude their own subject. Few of the elites, journalists, or academics come close to actually discussing climate. Nearly all of them try to address physics, chemistry, computer science, and mathematics by talking anthropology instead. Try discussing anything useful with such rules.

Not everyone is an expert, but surely 25 years is time enough for any intelligent layman to advance beyond stupid clichés about greenhouses and temperature. Greenhouses don’t work by the greenhouse effect. The difference involves differential equations, and the deepest unsolved problems of modern science. Greenhouses are certain; the atmosphere and oceans aren’t. The mathematics is different; the physics is different.

The mathematics of the non-greenhouse goes to the Clay Millennium Problems, which are among the most fundamental unsolved problems in mathematics. Not just one, but two of seven pertain. Solving the Navier Stokes equations, which govern the movements of air and water, and the other (P-NP) is about using computers for such complex things.

That’s $2-million in prize money, and maybe a Fields medal or two, if you get them both. But it doesn’t stop there! There’s the closure problem of turbulence in fluids: Not only can’t we determine the flow in a pipe from first principles, we can’t even get the lowest order statistic, after 150 years of trying. There is no Clay prize, but instant fame awaits you. Good luck.

What about computers? Computers don’t have enough numbers (i.e. finite representation). The mathematics is too big to fit. Consequently computer arithmetic is different: garbage out even without garbage in. You must crack differential equations like eggs to put them onto computers. The shattered remains are an approximation, but with different physics.

They don’t usually conserve things like energy! Such differences tell when integrated over long times (i.e. climate). If you invent a computer scheme that conserves all the correct things, the computer’s solution amounts to an exact solution of the original equation. If you figure out how to do that for the infinite number of conserved quantities expected for Navier Stokes, you win $1-million!

Finite representation means that the smashed equations must be hung on a grid. Think of pixels on a computer screen. Between grid points there is nothing. Grid spacings must be smaller than anything you hope to see. Everything else is lost. Proper computation calls for spacings to be smaller than all of the wiggles in the equation’s solution. But the enormous scales and complexity in climate mean that the wiggles are much smaller than grid spacings.

Not even thunderstorms show up with model resolutions of hundreds of kilometers! If you put together a grid that could capture all turbulence, you’d need a spacing of about 1 mm—air’s Kolmogorov cutoff. On modern computers, a proper computation of a 10-year forecast for the atmosphere and oceans, the calculation would take in excess of the age of the Universe, squared. The climate problem is just way too big, and computers are way too slow to do proper computation. Empiricism (i.e. fakery) is the only way forward. There is no human fault here. That’s the best humans can do. It’s no secret. Model controversy is largely among people ignorant of what model pioneers openly stated.

Models can be empirically adjusted to approximate observations. However you cannot adjust for conditions we haven’t encountered yet, but that is precisely what climate change is: conditions that we haven’t encountered yet. So, for climate change, empiricism fails. Only extrapolation remains, making models more caricature than approximation. But there is an even bigger issue. Smashed nonlinear equations, drooping from grid points, fed faux physics, integrated for extreme long times, are notoriously computationally unstable.

There has been a long struggle to get these algorithms to settle down and stop wandering off into fantasyland. Modern versions are so stable that nothing happens unless pushed from the outside. Models have no natural variability over long times. Is this a bug or a feature? Some modelers believe the latter. Thus they contend that climate is a boundary value problem, as startup conditions no longer matter. If true, an observer living on climate timescales would experience nothing analogous to weather? Every moment would be like the last.

But what if there is “climate weather?” Current conventions would fail, because they presume all climate change is pushed from the outside. But if not, reacting dynamically to pushing makes things very different. Can we tell the difference between internal and external change? No. Climate sensitivity and attribution become problematic. There’s no way to settle this empirically. We don’t live long enough. Theoretically, things are worse. Many scientists believe that fresh thinking is needed.

Fresh thinking is inherently unorthodox. Small minds, unenlightened politicians, and activists do not understand the importance of unorthodox thinking to science, and ultimately to everyone. Research must transcend the zeitgeist. Therefore it’s forever in trouble. Scientists are political targets on climate.

Those that speak out must endure vexatious, eristic tactics rather than scientific reasoning in matters from publication, to funding, to jobs. For example, my friend Dr. Willie Soon is under assault from activist groups. An army of human bots has been released to get him fired from Harvard’s Center for Astrophysics. He’s shared scientist’s skepticism publicly. Betraying acute ignorance, media pound him with silly hit pieces. Some perpetrators believe they fight for science, but they could not be more antiscientific, if not fascistic.

If there’s any chance at a rational policy on climate, two things must happen. First, intelligent laymen must take back the debate, by pushing currently out-of-bounds science back onto center stage. They must stop letting “experts” do their thinking for them. Second, political attacks on scientists must be stopped. Those must be pushed out of bounds. Let’s begin by saving Willie!

Christopher Essex, Chairman, Permanent Monitoring Panel on Climate, World Federation of Scientists and Professor of Applied Mathematics, University of Western Ontario.

Financial Post, 26 February 2015