2 plus ones
Shared publicly•View activity
View 19 previous comments
- Of course science isn't settled, that's basically the definition of science. If something is settled, it's not a scientific subject any more.
As for Lindzen and Choi 2011, it is rather worthless. They use an indirect statistical argument (about the shape of lead-lag regression curves for outgoing radiation vs SST) to claim a very low sensitivity, but models with a standard value for sensitivity show the same thing (http://www.skepticalscience.com/dessler-2011-rebuttal-revisions.html).Feb 4, 2012
- An enlightening website: http://www.klimaskeptik.cz/
(in Czech, however with many links to sources in English).
Author is Czech historian, specialized in research on influence of climate changes on human history.Feb 5, 2012
- Ronan, I read the Dessler (2011) paper, that seems to be addressing the problems in Lindzen and Choi (2011), but unfortunately my current knowledge is not enough to judge either one. If you google around for these two papers, you will find opinions going both ways (you posted a link saying Dessler is right, here is an example of a link that says he is wrong: http://motls.blogspot.com/2011/09/andrew-dessler-clouds-dont-reflect.html). If you read both these links, they both look rational and right, but their conclusion is completely the opposite... You can find more blog posts like these, going either way.
Typically, one then chooses blogs whose authors are more his favorites. That's of course legitimate, but I really don't want to rely on somebody's opinion, but rather form my own scientific opinion about this. I started with the log. relationship for CO2. Here I found nice explanation how it works: http://forecast.uchicago.edu/chapter4.pdf. So I think this dependence is undisputed. In other words, it is undisputed, that doubling the CO2 (from any levels, where this relationship holds) will cause the same amount of warming.
What is disputed is the "sensitivity parameter", in other words, how much of the warming is actually caused by doubling the CO2. For that, one has to have some model of the heat flow/balance, and that seems to me where all the disagreements are.
Some people say the increase of CO2 is driving the temperatures up. Some people say, that the data suggests that actually the temperature goes up for some natural reason, and then the CO2 goes up afterwards (by releasing it from the oceans). And there are many other questions raised. I currently have zero opinions on this.
But unfortunately I can't spend more time on this now. But its nice to at least make a progress with the log relationship. Thanks everyone for the discussion!Feb 5, 2012
- Ondrej, you say "In other words, it is undisputed, that doubling the CO2 (from any levels, where this relationship holds) will cause the same amount of warming." That is very interesting, however, I hear it reported frequently that we are right now at the point of irreversible, runaway climate change. This implies that the climate is now changing due to a feedback mechanism and no longer depends on how much CO2 we put into the atmosphere, but on how much CO2 is released into the atmosphere by the feedback mechanism (we might have to exclude the last 10 years of data if this argument is to be taken at face value though). It would seem that the industrial age has coincidentally occurred at a time in earth's history where there is a very narrow band of tolerated CO2 levels. After all (correct me if I am wrong) but we have not yet actually doubled the CO2 concentration even once.
However, my next three questions were going to be the following.
Given the successful predictions that were made in the 80s:
1) How many other predictions were made at that time? What level of concensus existed when that prediction was made, or are we simply picking one of the many possibilities in retrospect that happened to fit the subsequent data?
2) How well does this model "retrodict" earlier temperatures, i.e. how well does it fit the data for times before those originally plotted on the graph that is presented in that research? (We surely now have much better science to tell us what the CO2 and temperatures were before accurate climate data was recorded).
3) What is the scientific content of the prediction? Namely, if we remove known cyclical patterns and feedback mechanisms, how do we quantify in a mathematical way the correlation between the prediction and the measured data? In other words, how much does the prediction for our future, based on this successful model, depend on data we haven't measured yet and how much does it actually predict our future? and with what degree of certainty?
I wish I had more time to study this subject myself.Feb 5, 2012
- Ondrej, you state, "The problem with it is that the science is not settled at all, see for example the recent paper (http://www-eaps.mit.edu/faculty/lindzen/236-Lindzen-Choi-2011.pdf) which estimates the sensitivity to be around 0.7 C (with confidence interval 0.5C - 1.3 C at 99% levels)."
I think this is completely indefensible. According to the APS, "The evidence is incontrovertible: Global warming is occurring."
There's a vast difference between "incontrovertible" and "not settled". So you are making a statement which disagrees with the entire APS. You should not be suggesting in any way that it is anything less than incontrovertible, unless you wish to be branded a crackpot.
It is perfectly ok to be sceptical as a scientist. In fact, it is an essential quality of a scientific mind. But it is certainly not ok in the face of evidence which has been pronounced "incontrovertible".
If you aren't prepared to take the consensus opinion as your starting point, you aren't having a scientific discussion. As we all know, scientific consensus is an infallible source of truth and must be upheld as such.Feb 5, 2012
- That's a good point -- I believe that the question whether some science is right or wrong doesn't depend on any consensus. So I personally don't care about some "consensus", but I would love to understand the physics of it. I would actually still like to understand the CO2 log dependence better (if I had time...), preferably write some code that shows that, and that agrees with experiment, then I would fully trust it, but I feel that this is quite solid. So I think there will always be "some warming" per doubling (but of course this warming might not even be measurable and other effects might be more important).
But I would have to leave it to somebody else to answer your questions.Feb 5, 2012