Hardly anyone knows that two thirds of the warming predicted by the climate models comes from assumed changes in humidity and clouds, and only one third comes directly from the extra carbon dioxide (CO₂) and other greenhouse gases.

Some awareness of the calculations is essential, because there is no direct observational evidence that rising CO₂ caused the bulk of the recent warming.  If there were, don't you suppose we would have heard all about it by now?  The world has spent over $60B since 1990 looking for that evidence.  The ice cores shown by Al Gore in his movie were once supporting evidence, but higher-resolution ice cores subsequently showed that the temperature changes occurred a few hundred years before the corresponding changes in CO₂ and thus cannot have been caused by them.

Instead, the theory of man-made global warming relies on calculating the predicted temperature rise.  First, the direct warming of the extra CO₂ is reckoned, using principles known from laboratory work for over a century.  Then the warming or cooling effect of the changes in humidity and clouds caused by the direct CO₂ warming (the "feedbacks") are estimated.  The first part is sound; the second is where the real debate lies.

The theory of man-made global warming assumes that humidity and clouds amplify the warming directly due to CO₂ by a factor of three: extra CO₂ warms the ocean surface, causing more evaporation and extra humidity.  Water vapor, or humidity, is the main greenhouse gas, so this causes even more surface warming.

This is the most important feature of the debate, namely because it is responsible for most of the warming, but more crucially because it is based on an assumption.  It also helps explain why warmists and skeptics both insist that they are right.

The warmists are correct that CO₂ is a greenhouse gas and that it causes warming, that CO₂ levels have been rising, and that the atmosphere has been warming.  For many in the public, this is the whole argument.

Skeptical scientists agree with all those points, but note that they do not prove that something else isn't causing most of the warming.  As a thought experiment, suppose that the main cause of warming was actually Venusians with ray guns.  Then all those points would still be true!

Skeptics point out that the extra humidity caused by the extra CO₂ might cause more clouds.  Clouds reflect incoming sunlight and so cool the planet, which would counteract the direct CO₂ warming.  Skeptics say the changes in humidity and clouds, instead of amplifying the direct CO₂ warming by three, probably dampen it instead, maybe halving it.

Global cloud cover was observed to decrease during the warming of the 1980s and 1990s.  But the warmists say there are fewer clouds because it warmed.  So we are left wondering, did fewer clouds cause the warming, or the other way around?

The skeptic's main suspect is the sun.  While the sun's radiation is roughly constant, its magnetic field varies considerably.  This field shields the earth from cosmic rays that, according to recent experiments at the world's premier atom-smasher CERN, create very small particles in the air.  These in turn might lead to bigger particles that seed clouds.  Clouds cool the planet, so the warming over the last century might have been due to the sun's increasing magnetic field strength.

We scientists can calculate how much warming results directly from an increase in CO₂ levels.  We know how much CO₂ levels and temperature have risen since pre-industrial times, but the warming directly due to CO₂ is only a third of the observed warming.  The theory assumes that no other major influence on temperature changed, so the effect of the CO₂ must have been amplified threefold, presumably by changes in the atmosphere due to humidity and clouds.  That is the crucial warmist assumption.

There is no observational evidence for this amplification, but it is nonetheless implicitly built into all the models.

If the CO₂ theory of global warming is right, the climate models should predict the climate fairly well.  If the CO₂ theory is wrong, because there is another, larger driver of the temperature, then the climate models will perform indifferently.

According to the latest data from mankind's best and latest instruments, from impeccable sources, the climate models are doing poorly: they have got all their major predictions wrong since 1990.

Government climate scientists tend to excuse away these failings, often blaming unmeasured aerosols whose effects are only dimly understood.  These excuses wear ever thinner as the CO₂ level continues to rise but the temperature plateau of the last 12 years persists.

Supporters of "action on climate change" may soon have to face up to the fact that they have been unwittingly relying on assumed amplification by humidity for most of the predicted temperature increases, and that the amplification is not there in reality.

What distinguishes science is that evidence trumps the opinion of any human.  Since at least the Enlightenment, the ultimate authority in matters scientific is universally recognized to be observational data.

But who wants to look at data?  It is so much easier to just ask a recognized expert, and looking at data all seems so complicated and disputatious.  Well, we all understand graphs of temperature over time, so the data is relatively easy to interpret.  The data mentioned above is all in the public domain, and downloadable by anyone.  Ask a climate scientist about it and he'll often say something complicated and confusing, then assert that his theory is right -- basically, "trust us."  Citizens and the media now have to decide whether their highest authority is the data or the climate scientists, because these two entities are in conflict.

This issue illustrates a wider problem.  What if most experts in a field decide to promote some idea, acting on self-interest rather than the public good?  Perhaps the group decision is arrived at innocently, maybe through serendipitous evolution or maybe by failing to adapt a dominant paradigm to changed circumstances or data.  Perhaps it is reinforced by confirmation bias and by selecting newcomers as experts only if they agree with the idea.  Should we just listen to the experts?  Would we jump off an economic cliff if they told us to?

Surely the solution is to require any group of experts to be able to sufficiently explain their decision at several levels of complexity, right down to the very simple.  If you cannot teach your idea simply, maybe you cannot really defend it.  In particular, the experts must be able to explain fully without omissions, because omitting contradictory or anomalous evidence or ideas is the easiest way for an expert to convince a lay person, who is usually unaware that the confounding material exists.

Experience in financial and legal matters shows that we need some combination of auditing, checking, red teams, and professional defenders.  Why should climate be any different?

Dr. David Evans modeled carbon in Australia's biosphere for the Australian Greenhouse Office.  His wife runs the world's fourth-biggest skeptic climate site.