A friend of mine from college emailed me the other day and expressed some skepticism about the connection between carbon dioxide emissions and global warming. It occurred to me that it would make a good topic for my next post.

So here is the reasoning that has led me to conclude that business-as-usual carbon dioxide emissions will lead to temperature increases over the next century of around 3 degrees C.

First, it has been known for over 150 years that adding carbon dioxide to the atmosphere will increase the temperature of the planet. In fact, the very small number of credible skeptics out there, such as Dick Lindzen and Pat Michaels, are on record agreeing that adding greenhouse gases to the atmosphere will warm the planet. What they argue is that the warming will be very small. More on that later.

The conclusion that emitting greenhouse gases will result in warming does not rest on the output of climate models, but is a simple physical argument that predates the invention of the computer. And if you don’t believe in physics, take a look at Venus. That planet features a thick carbon dioxide atmosphere and consequently a surface temperature hot enough to melt lead.

So we know that adding carbon dioxide is going to warm the planet. This leads us to the real question: How much warming are we going to get?

Carbon dioxide by itself will only provide somewhere around 1 degree C warming over the next century. In order to get really large warnings over the 21st century, there needs to be strong positive feedbacks to amplify the initial warming from carbon dioxide.

The strongest positive feedback is due to water vapor. The so-called water-vapor feedback refers to the process whereby an initial warming of the planet, caused for example by an increase in atmospheric carbon dioxide, causes an increase in the specific humidity of the atmosphere. Because water vapor is itself a greenhouse gas, the increase in specific humidity causes additional warming.

The water vapor feedback has long been expected to exert a powerful warming effect because of the belief that the atmosphere’s relative humidity would remain roughly constant — meaning that specific humidity would increase rapidly with surface temperature. Models reproduce this, and the water vapor feedback is the most important reason for the models’ large predicted warming for increases in greenhouse gases over the 21st century.

If you read the blogs, you’ll often see the argument made that no data exist to support the models’ strong positive feedback and that they are simply “computer-generated.” This is incorrect. I recently published a paper [PDF] estimating the magnitude of the water-vapor feedback exclusively from data. No climate model involved.

What we found was evidence that the water vapor feedback is indeed strong and positive. And there are several other observation-based analyses that agree with this conclusion (see the references in my paper).

Overall, the water vapor feedback about doubles the Earth’s response to carbon dioxide alone. If you throw in the other feedbacks (albedo, lapse rate, etc.), you’ll get about a 3 degrees C warming for doubled carbon dioxide (compared to 1 degree C for carbon dioxide alone).

While there is significant uncertainty in this number, it’s hard to believe that it will be fall outside a factor of 2 of this number.

The credible skeptics agree that the Earth will warm over the 21st century, but that the warming will be very small (less than 1 degree C). Could they be right? Possibly. The one way that climate change may not be large is if there exists, somewhere in the climate system, a large negative feedback that compensates for the positive feedback provided by water vapor. If such a negative feedback exists, it seems highly likely that it would somehow revolve around clouds.

To their credit, the few credible climate skeptics out there (Lindzen, Spencer) are indeed searching for negative feedbacks. So far, though, their arguments have been pretty weak and not convinced anyone in the scientific community.

One interpretation of the IPCC’s statement that humans are “very likely” responsible for most of the observed recent warming is that there is about a 10 percent chance that a significant negative feedback (or something equivalent) does exist in the climate system and will preclude significant warming over the century.

My personal opinion is that this is conservative, and that the actual chance we will discover a big negative feedback is actually far smaller. There is simply too much evidence in the paleoclimate record showing large swings of the climate, which tends to preclude the existence of stabilizing negative feedbacks.

This is how I reach the conclusion that warming of a few degrees Celsius is very likely unless we do something about greenhouse-gas emissions. Note that this conclusion is not derived from climate models. Rather, it is derived from fundamental physics combined with observations.