Greenhouse caused warming: agreement or not?

A dispute between Stephen Sherwood (SS) and Willis Eschenbach (WE)
19 August 2005, Climatesceptics mailing-list


SS:
It seems to be universally acknowledged (for example, recently in comments to the media by Lindzen, Spencer, and Singer) that increased "greenhouse gases" must lead to a warmer planet than would otherwise occur (this, one presumes, based on the laws of physics and perhaps Le Chatalier's principle that if you put energy into a complex system it has to warm up, at least a little). Is this impression correct, or is there still discord on this?

WE:
The world will not necessarily heat up as greenhouse gases increase. My paper on "Greenhouse Equilibrium" (qv) shows that there is a very strong feedback mechanism involving the albedo. This feedback increases the amount of incoming sunlight as losses increase, and decreases the incoming sunlight as losses decrease, which tends to keep the temperature stable.
Does "physics predict" that the world will heat up? It does, but only in the way that "physics predicts" that if more people enter a room, that room will heat up ... unless there is an air conditioner with a thermostat (feedback) that keeps the room at the same temperature. The climate is a marvellously complex system, one which we are only beginning to understand.

SS:
Not only that, but the heat input associated with a given increase in greenhouse gases should not be in dispute and is quite significant compared to observed increases in ocean heat content. Is this in dispute?

WE:
Again, this is definitely in dispute. There are two questions here. 
1) How much forcing should result from a given change in GHGs, and 
2) How much temperature rise will occur from that forcing. 
The answers given to these questions, both from theory and from models, are all over the map; and most of these answers do not fit very well with observational evidence.


SS:
The main point of disagreement seems to be how much or how little we should trust climate model predictions (and how much future warming to expect).  Even there there would have to be agreement that many important details (e.g., soil moisture in Iowa) are untrustworthy, since model predictions don't even agree with each other. On the other hand, all models predict significant changes of one kind or another within not too many more decades.

WE:
I must say, I truly fail to understand why people trust computers to predict the climate 50 to 100 years out, when the same computers can't predict the weather 10 days out. The climate is a non-linear, coupled, driven, chaotic system with a wide range of both known and unknown feedbacks. Yes, all of the computer models "prove" that the climate will have significant changes. This is not surprising, given that any computer model which did not predict such changes would be judged a failure, and redesigned until it did predict such changes.

But the models disagree on the timing, the size, the location, and the severity of the changes. This has been countered by what is, to me, an even less explicable move. This is to use "ensembles" of computer models, and to take some kind of average of them to give an answer. To me, taking the average of unreliable models does not give us a better answer, especially when the models are all making the same assumption, that GHGs are going to radically heat the world.

Finally, there is a further point of disagreement which you did not mention. This is the question of whether warming is a good thing or a bad thing. We do have some evidence in this regard, however. The world has warmed a couple of degrees since the 1700s. No islands have disappeared. No polar bears have gone extinct, nor any other animals (despite the "scientific" paper which "proved" that a 0.8C increase would wipe out a third of the world's species). There has been no wave of infectious diseases sweeping north and south with the increasing warmth. In fact, none of the predicted bad things happened when the earth warmed in the past. So why should we expect these bad things to suddenly start happening now?

So is there agreement? Yes, on some things. In order of decreasing agreement, I would offer the following statements:

1) The earth is warming, and has been since the "Little Ice Age" in the 1700s. Most scientists would agree.

2) Humans are responsible for some of that warming. Again, most scientists would agree that humans can have some effect on the climate.

3) The human contribution is responsible for most of that warming. Here the disagreement starts. Although even the IPCC estimated that half of the 20th century warming was not caused by humans, many scientists feel that the natural causes far outweigh the human contribution.

4) We know enough about how the climate works to accurately model the evolution of the climate for the next 100 years. Very, very little agreement on this question. My own feeling? Ho, ho, ho ...

5) The bad effects of warming will far outweigh the good effects. Again, very little agreement, especially from people living in cooler climates.

6) We can do something about the human caused warming. While there has been little agreement about this question, the obvious failure of the Kyoto Protocol to "do something about warming" has brought about a shift towards the idea that we should concern ourselves with mitigation rather than prevention.

So unfortunately, there is no consensus, scientific or otherwise, about many questions related to climate.