The 50-Point IQ Drop Phenomenon, Cont.: For some reason, when distinguished members of other sciences venture into meteorology, they shed IQ points like a winter fur coat in the summertime. While all scientists tend to do this when they venture out from their comfort zones, there is something about meteorology that brings out the very dumbest in the very smartest interlopers. I suspect its because there is the implicit assumption--sometimes explicit--that meteorologists are stupid and that the subject could be solved in a few minutes by a really smart [scientist from another field]. I have heard those words, or words to the same effect, uttered more than once, in fact. And so: in the latest Physics Today, not one but two scientists weigh in a bit harrumphingly at a May 2013 article entitled Chaos at fifty that focused on meteorologist Edward Lorenz. Oh, theres the well-deserved and outstanding palaver at the beginning, but both writers go out of their way to indicate how Lorenz wasnt the first to study chaos, etc. The esteemed mathematical physicist David Ruelle, the longer-winded of the harrumphers, trots out well-known and well-cited quotes from Poincaré to demonstrate that Poincaré knew about the butterfly effect 50 years before Lorenz. Well, yes, but as far I can tell Poincarés explanation was not in the form of a quantitative model. But whatever. The next sentence is the mind-blower, so much so that it has to be broken down into two pieces: Lorenzs contribution is not so much the knowledge of the Lorenz attractor... Wut, as the kids say? He created this simplified model, explored it, and taught the world things about dynamical systems with it, and the world named it after him. Id say thats a contribution, Dr. Ruelle. But, please, do go on... ...as it is the demonstration that with computers, meteorologists can progressively improve their modeling of the dynamics of the atmosphere. Thats an incorrect answer in several different directions simultaneously: 1. Sensitive dependence on initial conditions actually defines the LIMITS of modeling atmospheric dynamics accurately. Chaos theory is, to first order, all about how progressive improvement is ultimately impossible beyond a certain point. That limit, which is estimated to be approximately two weeks into the future, was devised by Lorenz in the early-to-mid 1960s, and still holds today. 2. There were those who were forging ahead in the 1940s and 1950s on a program of progressive improvement of atmospheric models. Jule Charney, later Lorenzs colleague at MIT, deserves much credit for that. Lorenz was not part of the groundbreaking work that led to the first successful numerical weather forecast in 1950. At roughly the same time as the first ENIAC forecast, Lorenz was pursuing other ideas in statistical forecasting, a very different approach than deterministic modeling of the atmosphere. 3. In fairness, perhaps Ruelle was referencing, a bit crudely, the advances in weather forecasting via ensemble prediction that have taken place since the 1990s and which are directly predicated upon the concepts of chaos theory. Lorenz said that he anticipated this development in the mid-1960s, before anyone else. This is a very significant scientific advance that is distinct from the insights emanating from the Lorenz attractor, which has had a profound impact across multiple disciplines. 4. One might sense a bit of a sneer in the phrase with computers, as if Lorenz was simply a technician who happened to stumble upon something important. Uh, no. Lorenz was a graduate student of Birkhoff, the famed mathematician called America s premier mathematician of the first half of the twentieth century, and those who are familiar with Lorenzs papers know that Lorenzs emphasis was on mathematical and physical insight rather than brute-force computation. The sneer becomes more apparent in Ruelles concluding paragraphs, in which he praises his own work in chaos theory and then trumpets the work of another on the stability of the solar system as chaos theory at its best. By inference and omission, then, Lorenzs work was *not* chaos theory at its best! And in a final swipe, Ruelle states that the solar system work provided new views on the history of climates and other important geological questions. Funny, I thought climate was a subject first and foremost for atmospheric science--Lorenzs field. Research into multiple equilibria solutions for climate were being explored in atmospheric science a decade before the paper Ruelle cited regarding the stability of the solar system. Ruelle has a lot of IQ points to give, but not so many that he doesnt sound kind of dumb here to anyone familiar with the history of numerical weather prediction, atmospheric science in general, and the career of Ed Lorenz. (I chair an American Meteorological Society committee that gives an award named after Lorenz.) Ruelle also reveals the thinking that, I suspect, prevented Lorenz from being awarded the Nobel Prize in Physics for his efforts... despite Lorenz having published his seminal work in 1963 and then living until 2008 (so there can be no excuse in terms of untimely death). And so it goes. This isnt the first time Physics Today has published something that takes the form of a back-handed compliment to the earth and atmospheric sciences. Now that AMS is part of AIP (aip.org/news/2013/aip-welcomes-american-meteorological-society-newest-member-society), however, maybe its time for that kind of stuff to go away quietly. Same team, yall.
Posted on: Thu, 13 Mar 2014 01:33:07 +0000
Trending Topics
Recently Viewed Topics
© 2015