Weather, Chaos, and Climate Change

March 30th, 2009 by Roy W. Spencer, Ph. D.

Yesterday I received a request from a reader in Wisconsin for me to address a question I hear pretty frequently from the public: If weather can’t be predicted beyond 7 to 10 days, what then makes climate modelers think they can predict climate 50 years in advance?

For many years I gave the ‘IPCC-approved’ reason…but now I’m not so sure anymore.

First let’s review the ‘scientific consensus’ explanation of the difference. In weather forecasting, you take a snapshot of global weather patterns with weather balloon, satellite, surface, and aircraft-based measurements, and then extrapolate them out in time using a set of equations. And as Ed Lorenz demonstrated in 1963, any unmeasured weather on very small space scales can cause huge differences in the forecast the farther out in time one projects the weather. This is the classic example of the chaotic, nonlinear variability inherent to atmospheric circulation systems. Even the flap of a butterfly’s wing will eventually change global weather patterns. Mathematically speaking, this is referred to as sensitive dependence on initial conditions.

Global warming forecasting, in contrast, has been claimed to be possible because we are instead dealing with a small change in the rules by which the atmosphere operates. The extra carbon dioxide we are putting into the atmosphere, it is argued, changes the Earth’s greenhouse effect slightly, which is then expected to change average weather (climate) to a lesser or greater extent. Mathematically speaking, this is referred to as a change in boundary conditions.

But upon closer examination, I have come to realize that the two kinds of variability – weather and climate – maybe are not so different after all. The only major difference between the two is just one of time scale.

The weather today is impacted by what has happened on the Earth, in the atmosphere and on the surface, every day previous to today. In a very real sense, today’s weather retains a memory of all weather which has occurred in the past.

But climate variability is really no different. This year’s climate is a natural result of average weather and climate in previous years. For instance, the slow overturning of the ocean can bring water to the surface which hasn’t been in contact with the atmosphere for maybe hundreds of years. Therefore, the climate we are experiencing today can be related to average weather conditions which occurred hundreds of years ago.

So, one might argue that climate variability is just as good an example of chaos as weather variability. Mathematically speaking, the ‘sensitive dependence on initial conditions’ we associate with chaos might also be called ‘sensitive dependence on continuously changing boundary conditions’.

In fact, nature does not distinguish between changing initial conditions and changing boundary conditions…it is all just change. The real question is how a human source of change (e.g. carbon dioxide emissions) stacks up against all of the natural sources of change.

“But”, one might object, “Nature has been the same until humans came along and changed it!” Well, is it really valid to think of nature as unchanging? I don’t think so. Nature causes its own changes, all the time. And each of those changes forever alters the future direction of both weather and climate.

But we aren’t aware of these continuous subtle changes — only the spectacular ones. For instance, a major volcanic eruption can inject millions of tons of sulfur into the stratosphere, leading to a couple of years of global cooling. But what isn’t mentioned is that such an event will also forever change the future course of weather and climate on the Earth simply because future weather and climate retain a memory of that event.

Similarly, a chaotic change in precipitation patterns in the Pacific Ocean will change ocean salinity, which will then change the circulation of that water into the deep ocean over time, which maybe a hundred years hence will then reemerge as a change in surface waters which will, in turn, affect weather patterns once again.

So, are these changes in initial conditions, or boundary conditions? I think the question is only one of semantics…it is all just change. And change in nature is ubiquitous.

The possibility that mankind can change climate, even for hundreds of years down the road, does not impress Mother Nature. She has been changing climate all by herself since time immemorial. Millions of tons of sulfur spewed into the stratosphere by a volcano illustrates the power of nature, and the spectacular sunrises and sunsets that result vividly display nature’s beauty.

Of course, if humans did the same thing as a volcano does, it would be called the greatest pollution disaster in history. But I digress….

The climate modelers assume there is no such thing as natural climate variability…at least not on time scales beyond maybe ten years. In effect, they believe that chaos only exists in weather, not in climate. But this view is entirely arbitrary, and there is an abundance of evidence that it is just plain wrong. Chaos occurs on all time scales. Climate change happens, with or without our help.

And maybe there is one more difference between weather forecasting and climate forecasting…a difference which has allowed climate alarmism today to flourish. When the TV meteorologist blows his forecast for tomorrow’s weather, people will remember his error, and hold him responsible. But climate modelers get to forecast any kind of climate change they want, knowing full well that no one 50 or 100 years down the road is ever going to remember how good – or how bad — their forecast was.

Comments are closed.