by the mad LOLscientist
Old News:
[Via Open Mind]
The denialosphere is atwitter about a new paper by McLean, Freitas, and Carter (2009, J. Geophysical Res. 114, D14104). They seek to relate variations in tropospheric temperature to the southern oscillation index (SOI). The concluding sentence is:
Finally, this study has shown that natural climate forcing associated with ENSO is a major contributor to variability and perhaps recent trends in global temperature, a relationship that is not included in current global climate models.
[More]
A very nice dissection of how bad math in a recent paper removes any increase in temperature, allowing the authors to say that there is no effect of anthropogenic greenhouse gases on temperature. This a clear demonstration of how ‘climate skeptics’ can get published – use math to get rid of the thing they do not want to see and then conclude that it does not exist.
As the post demonstrates, a well defined increase in temperature can be added to the data and this increase will be completely lost when analyzed using the methods in the paper. This is a classic ‘control’ on any analytical method – add in the quantity you are discussing and see if the system can register it.
In this case, the analytical method can not, so when the authors, or any denialist, says that this paper demonstrates that there is little or no trend in temperature increase due to humans, know that it has not proven any such thing. Maybe they are right and there is no trend from human CO2 release but that can not be known from the analysis of the data that they present in the paper.
This paper used a poorly designed analysis to reach a conclusion that the authors apparently wanted to reach. Peer review does not always work, as has been shown by several incidents already recorded by some of the very authors involved in this paper. But the self-correcting nature of science means that even after publication, a paper’s weaknesses will be rapidly exposed.
One poorly designed paper, easily examined and corrected, does not change the entire system. The responses of scientists are to find what is true, how strong the author’s arguments are, to see if the conclusions really match the data and to poke any holes in the argument they can. Sometimes, as in this case, they pretty much destroy the argument.
In other cases, the structure of the argument may be so strong that it leads to much more important work.
Scientists do not just explore papers and research that only agrees with them, with their own logical arguments. We can disagree strongly with someone’s conclusions yet still recognize that the logical argument they put forward is not only rational but strong. So we either have to do more research to find better ways to poke holes in their arguments or we might just have to recognize that they have a better handle on Natural Truths than we do.
A great example of this is in another post by the same author. This is a discussion from another paper on whether temperature increases have halted and will remain that way for several more years until starting up again. The paper postulates that the synchronization of some regular climate cycles is responsible for this temporary plateau.
Tamino does not really agree with much in the paper and does not think that the conclusions are necessarily valid. There are many things about the paper that raises a skeptical eye.
After setting things up, the post gives us this:
I have two overriding opinions of this work. My first overall opinion is that I don’t believe it’s correct, for several reasons. One is that the theoretical basis (as outlined in Tsonis et al. 2007) involves an extremely intricate framework of suppositions which I don’t find convincing, and the statistics used to support that work likewise don’t convince me. I don’t believe we have sufficient understanding of the behavior of chaotic dynamical systems to make the transition from the present theoretical framework to a practical real-world analysis of the climate system. Also, the hypothesis uses certain aspects of the HadCRU temperature time series which aren’t shared by the GISS or NCDC time series, so at least in part it’s dependent on the use of a particular data set.
After providing more skeptical detail, the post ends with this:
My second overriding opinion is that this work is tremendously important. I said earlier that I don’t believe we have sufficient understanding of the behavior of chaotic dynamical systems to make the transition from the present theoretical framework to a practical real-world analysis of the climate system. But without work like this, we never will. If the central hypothesis turns out to be correct, or even if it doesn’t, we may have made great progress toward extending our knowledge of how climate dynamics works. Much of the theoretical underpinning is very appealing, and may be key to future breakthroughs — which almost never happen without someone laying a foundation upon which they’re built. Even if the present work ends up being entirely cast aside, many of its themes may persist and may be an integral part of the next generation of climate dynamics.
Spoken like a true scientist, one who is willing to drop their own logical framework if shown that another one is better, and one who, while viewing a paper as flawed, can still be excited by the implications of the work, where this will lead and what sorts of questions it can answer.
That is something that many people seem unable to understand, as they can only seem to appreciate those that agree with them and not deal at all with logical, rational arguments that reach different conclusions. As a scientist, I deal with this all the time.
Because we are all trying to solve difficult natural problems and ANYTHING that will help us solve them is important, even if it might ruin our own argument. Of course, we are only human and may have to be dragged kicking and screaming into a world where our arguments are wrong.
After all, you have to possess a very large and healthy ego to do science. So we won’t give up for just any old thing. And we kind of like being the really smart guy who finds the flaw in someone else’s model. Even a scientist likes a good ego boost every so often.
But even the most stubborn researchers have to eventually recognize when someone else’s model does a better job of explaining things. As others gravitate to the better model, as that model becomes stronger and stronger, we are left with two choices – futilely trying to sustain our own model through more an more rarified research, watching as our colleagues begin to ignore us or coming to the conclusion that the other model is more ‘right’ than ours and embracing it. (There is sometimes a third possibility – coming up with a completely new model that does an even better job than any of the other ones. Rare but it can happen.)
It is often this hard and sustained attack that creates a model that is incredibly strong. Our current view of climate change has been under such an onslaught for the last 20 years or so, mostly by the very researchers involved in the work. That the main attack on this structure that we see today derives from faulty math and conclusions not supported by the data suggests that there are few strong, rational arguments left to deal with anthropogenic reasons for climate change. The chance of the third possibility grows smaller all the time.
This mainly leaves irrational arguments. And there certainly are no shortage of those today, many of which will get someone a nice spot on a cable news network or two.
Technorati Tags: Entertainment, Science
