The Uncertainty Principle for climate (and chemical) models

A recent issue of Nature had an interesting article on what seems to be a wholly paradoxical feature of models used in climate science; as the models are becoming increasingly realistic, they are also becoming less accurate and predictive because of growing uncertainties. I can only imagine this to be an excruciatingly painful fact for climate modelers who seem to be facing the equivalent of the Heisenberg uncertainty principle for their field. It's an especially worrisome time to deal with such issues since the modelers need to include their predictions in the next IPCC report on climate change which is due to be published next year.

A closer look at the models reveals that this behavior is not as paradoxical as it sounds, although it's still not clear how you would get around it. The article especially struck a chord with me I see similar problems bedeviling models used in chemical and biological research. In case of climate change, the fact is that earlier models were crude and did not account for many fine-grained factors that are now being included (such as the rate at which ice falls through clouds). In principle and even in practice there's a bewildering number of such factors (partly exemplified by the picture on top). Fortuitously, the crudeness of the models also prevented the uncertainties associated with these factors from being included in the modeling. The uncertainty remained hidden. Now that more real-world factors are being included, the uncertainties endemic in these factors reveal themselves and get tacked on to the models. You thus face an ironic tradeoff; as your models strive to mirror the real world better, they also become more uncertain. It's like swimming in quicksand; the harder you try to get out of it, the deeper you get sucked in.

This dilemma is not unheard of in the world of computational chemistry and biology. A lot of the models we currently use for predicting protein-drug interactions for instance are remarkably simple and yet accurate enough to be useful. Several reasons account for this unexpected accuracy; among them cancellation of errors (the Fermi principle), similarities of training sets to test sets and sometimes just plain luck. Error analysis is unfortunately not a priority in most of these studies, since the whole point is to publish correct results. Unless this culture changes our road to accurate prediction will be painfully slow.

But here's an example of how "more can be worse". For the last few weeks I have been using a very simple model to try to predict the diffusion of druglike molecules through cell membranes. This is an important problem in drug development since even your most stellar test-tube candidate will be worthless until it makes its way into cells. Cell membranes are hydrophobic while the water surrounding them is hydrophilic. The ease with which a potential drug transfers from the surrounding water into the membrane depends among other factors on its solvation energy, on how readily the drug can shed water molecules; the smaller the solvation energy, the easier it is for drugs to get across. This simple model which calculates the solvation energy seems to do unusually well in predicting the diffusion of drugs across real cell membranes, a process that's much more complex than just solvation-desolvation. 

One of the fundamental assumptions in the model is that the molecule exists in just one conformation in both water and the membrane. This assumption is fundamentally false since in reality, molecules are highly flexible creatures that interconvert between several conformations both in water and inside the membrane. To overcome this assumption, a recent paper explicitly calculated the conformations of the molecule in water and included this factor in the diffusion predictions. This was certainly more realistic. To their surprise, the authors found that making the calculation more realistic made the predictions worse. While the exact mix of factors responsible for this failure can be complicated to tease apart, what's likely happening is that the more realistic factors also bring more noise and uncertainty with them. This uncertainty piles up, errors which were likely canceling before no longer cancel, and the whole prediction becomes fuzzier and less useful.

I believe that this is what is partly happening in climate models. Including more real-life factors in the models does not mean that all those factors are well-understood. You are inevitably introducing some known unknowns. Ill-understood factors will introduce more uncertainty. Well-understood factors will introduce less uncertainty. Ultimately the accuracy of the models will depend on the interplay between these two kinds of factors, and currently it seems that the rate of inclusion of new factors is higher than the rate at which those factors can be accurately calculated.

The article goes on to note that in spite of this growing uncertainty the basic predictions of climate models are broadly consistent. However it also acknowledges the difficulty in explaining the growing uncertainty to a public which has become more skeptical of climate change since 2007 (when the last IPCC report was published). As a chemical modeler I can sympathize with the climate modelers. 

But the lesson to take away from this dilemma is that crude models sometimes work better than more realistic ones. Perhaps the climate modelers should remember George Box's quote that "all models are wrong, but some are useful". It is a worthy endeavor to try to make models more realistic, but it is even more important to make them useful.
Image source

The price of global warming science is eternal vigilance

John Tierney of the NYT weighs in on the hacked emails and accurately nails it

I’ve long thought that the biggest danger in climate research is the temptation for scientists to lose their skepticism and go along with the “consensus” about global warming. That’s partly because it’s easy for everyone to get caught up in “informational cascades”, and partly because there are so many psychic and financial rewards rewards for working on a problem that seems to be a crisis. We all like to think that our work is vitally useful in solving a major social problem — and the more major the problem seems, the more money society is liable to spend on it.

I’m not trying to suggest that climate change isn’t a real threat, or that scientists are deliberately hyping it. But when they look at evidence of the threat, they may be subject to the confirmation bias — seeing trends that accord with their preconceptions and desires. Given the huge stakes in this debate — the trillions of dollars that might be spent to reduce greenhouse emissions — it’s important to keep taking skeptical looks at the data. How open do you think climate scientists are to skeptical views, and to letting outsiders double-check their data and calculations?

We are all subject to the confirmation bias, and I can say from experience that we have to battle it in our research every single day as fallible human beings. But as Tierney says, when the stakes are so incredibly high, when governments and international budgets and debts and the fate of billions is going to be affected by what you say, you better fight the conformation bias ten times as much as usual.

Listen to Capt. Ramsey, son:

"Mr. Hunter, we have rules that are not open to interpretation, personal intuition, gut feelings, hairs on the back of your neck, little devils or angels sitting on your shoulders..."

The damning global warming emails; when science becomes the casualty

By now everyone and his grandmother must have heard about the hacked emails of the prestigious University of East Anglia Climate Research Unit (CRU). The emails were sent by leading climate change scientists to each other and seem to express doubts and uncertainty. More importantly they also seem to display some troubling signs of rather dishonest discourse, with scientists trying to hold dangerously unfavorable opinions of journal editors who seem to be open to publishing papers that don't seem to agree with their views, and asking each other to delete emails which might signal doubt.

There is at least one example of bad science revealed in the emails. It seems that one set of data from tree ring proxies did not show the expected rise in temperatures for a particular period and showed a decline. What was done was that just for that period, a different set of data from another method which did show the rise was grafted on to this piece of data. John Tierney of the NYT has the two graphs on his blog. Does this change the general conclusion? Probably not. Is this bad science and enough to justify a flurry of indignant questions in the minds of outsiders? Certainly so. Good science would have meant revealing all the pieces of data including those which showed a decline.

Now what is remarkable (or perhaps not remarkable at all) is the vociferous political- not scientific- reaction that has erupted in blogs all over the internet. I would point readers to my fellow blogger Derek Lowe's succinct summary of the matter. While I am not as skeptical about climate change as he is, it is disconcerting to see how much political, personal and social baggage the whole issue is carrying. Whenever a scientific issue starts carrying so much non-scientific baggage, one can be assured that we are in trouble.

The comments on most blogs range across the spectrum. There are the outright deniers who claim that the emails "disprove global warming"; they don't, and I can't see how any set of personal exchanges could say almost anything definitive about a system as complex as the climate. Phrases like "hide the decline" (in the case of the above tree ring proxy data) and "trick" have been taken out of their technical context to indicate subversion and deception. And then there are the proponents who want to act like nothing has happened. I like George Monbiot's take on it where he says that even if the science of climate change has certainly not come crashing down, the public image of climate change has been dealt a serious blow, and denying this would simply mean burying your head in the sand. After all, we are supposed to be the good guys, the ones who are supposed to honestly admit to our limitations and failings, and we are not doing this. What ramifications this will have for the important Copenhagen climate summit this month is uncertain.

However, the very fact that we have to worry as much about the public image of climate science as the science itself plainly speaks to the degree of politicization of the issue. I think the liability of this entire matter has basically become infinite and I think scientists working in the field are facing an unprecedented dilemma which few scientists have ever faced. Here's the problem; we are dealing with an extremely complex system and it is hardly surprising if the science of this system (which after all is only a hundred years or so old) keeps getting revised, reshuffled and reiterated even if the basics remain intact. That would be perfectly normal for a vast, multidisciplinary field like this. That is the way science works. One finds such revision and vigorous debate even in highly specific and recondite areas like the choice of atomic partial charges in the calculation of intermolecular energies. The climate is orders of magnitude more complicated. If the usual rules of scientific discourse were to be followed, making such debates and disagreements open would not be a problem.

But with an issue that is so exquisitely fraught with political and economic liabilities and where the stakes are so enormously high, I believe that the normal process of scientific debate, discourse and progress has broken down and is being bypassed. Scientists who would otherwise engage in lively debate and disagreements have become extremely loathe to make their doubts public. These scientists fear that they would essentially be condemned by both sides. The right wing extremists would seize upon any honest disclosure of debate as the kick that brings the entire edifice crumbling down. They would predictably try to discredit even reasonable conclusions drawn by climate change scientists. At the same time, left wing extremists would essentially disown such scientists and either declare them an anomaly or more predictably declare them to be political and corporate shills. A scientist who honestly voices his doubts would become a man without a country.

This is of course in addition to the ample scorn that establishment upholders like climate blogger Joe Romm would heap on them. Thus, if you are a scientist working in climate change today, it would be rather difficult for you to make even the normal process of science transparent. Plus, most scientists are genuinely scared that all the momentum they have built over the years would fizzle out if their right wing opponents pounce on their private doubts. Think about it. The Copenhagen summit is going to be held in a month. Scientists have faced enormous obstacles in convincing the public and governments about climate change. Your work has been crowned by grudging acknowledgement even by George W Bush and the Nobel Peace Prize for Al Gore. Would you be ready to throw away all this rightly hard-earned and hard-fought consensus for the sake of a few dissenting opinions? The simple laws of human nature dictate that you probably would not.

In my opinion, that is what seems to have happened with the scientists at the CRU. They have been so afraid of not only expressing their doubts (many of which as noted above would be valid given the science involved) but also entertaining other dissenting opinions that they have unfortunately picked the option of trying to silence open debate in a way that would be unacceptable in science in general. One can understand their motivation, but their actions still seem deplorable.

I think these emails point to a much more serious structural problem in the scientific enterprise of climate change. For good reasons and bad, whether to stand up to political hacks or ironically to defend good science, this enterprise has accumulated so much political baggage that it is now virtually impossible for it to compromise, to change, to maneuver even in the face of cogent reasons. The science of climate change has essentially bound itself into a straitjacket. My prediction is that important decisions about this science will in the future be mainly politically motivated. Public consensus not completely backed by good science will be the driving force for major decisions. The consequences of those decisions, just like the climate, are uncertain. We will have to wait and see.

But as usual, the casualty is ultimately science itself. What was good science and ineffective politics before is becoming effective politics and bad science. Whatever else happens, science never wins when it gets so overtly politicized. And hopefully about this there will be universal consensus.

A wrong kind of religion; Freeman Dyson, Superfreakonomics, and global warming

The greatest strength of science is that it tries to avoid dogma. Theories, explanations, hypotheses, everything is tentative, true only as long as the next piece of data does not invalidate it. This is how science progresses, by constantly checking and cross checking its own assumptions. The heart of this engine of scientific progress is constant skepticism and questioning. This skepticism and questioning can often be exasperating. You can enthusiastically propound your latest brainwave only to be met with hard-nosed opposition, deflating your long harbored fervor for your pet idea. Sometimes scientists can be vicious in seminars, questioning and cross questioning you as if you were a defendant in a court.

But you learn to live with this frustration. That's because in science, skepticism always means erring on the safer side. As long as skepticism does not descend into outright irrational cynicism, it is far better to be skeptical than to buy into a new idea. This is science's own way to ensure immunity to crackpot notions that can lead it astray. One of the important lessons you learn in graduate school is to make peace with your skeptics, to take them seriously, to be respectful to them in debate. This attitude keeps the flow of ideas open, giving everyone a chance to voice their opinion.

Yet the mainstay of science is also a readiness to test audacious new concepts. Sadly, whenever a paradigm of science reaches something like universal consensus, the opposite can happen. New ideas and criticism are met with so much skepticism that it borders on hostility. Bold conjectures are shot down mercilessly sometimes even without considering their possible merits. The universal consensus separates scientists into a majority who provide a vocal and even threatening wall of obduracy against new ideas. From what I have seen in recent times, this unfortunately seems to have happened to the science of global warming.

First, a disclaimer. I have always been firmly in the "Aye" camp when it comes to global warming. There is no doubt that the climate is warming due to greenhouse gases, especially CO2, and that human activities are most probably responsible for the majority of that warming. There is also very little doubt that this rate of warming has been unprecedented into the distant past. It is also true that if kept unchecked, these developments will cause dangerous and unpredictable changes in the composition of our planet and its biosphere. Yet it does not stop there. Understanding and accepting the details about climate change is one thing, proposing practical solutions for mitigating it is a whole different ball game. This ball game involves more economics than science, since any such measures will have to be adopted on a very large scale that would significantly affect the livelihood of hundreds of millions. We need vigorous discussion on solutions to climate change from all quarters, and the question is far from settled.

But even from a scientific perspective, there are a lot of details about climate change that can still be open to healthy debate. Thus, one would think that any skepticism about certain details of climate change would be met with the same kind of lively, animated argument that is the mainstay of science. Sadly, that does not seem to be happening. Probably the most recent prominent example of this occurred when the New York Times magazine ran a profile of the distinguished physicist Freeman Dyson. Dyson is a personal scientific hero of mine and I have read all of his books (except his recent very technical book on quantum mechanics). Climate change is not one of Dyson's main interests and has occupied very little of his writings, although more so recently. To me Dyson appears as a mildly interested climate change buff who has some opinions on some aspects of the science. He is by no means an expert on the subject, and he never claims to be one. However he has certain ideas, ideas which may be wrong, but which he thinks make sense (in his own words, "It is better to be wrong than to be vague"). For instance he is quite skeptical about computer models of climate change, a skepticism which I share based on my own experience with the uncertainty modeling even "simple" chemical systems. Dyson who is also well known as a "futurist" has proposed a very interesting possible solution to climate change; the breeding of special genetically engineered plants and trees with an increased capacity for capturing carbon. I think there is no reason why this possibility could not be looked into.

Now if this were the entire story, all one would expect at most would be experts in climate change respectfully debating and refuting Dyson's ideas strictly on a factual basis. But surprisingly, that's not what you got after the Times profile. There were ad hominem attacks calling him a "crackpot", "global warming denier", "pompous twit" and "faker". Now anyone who knows the first thing about Dyson would know that the man does not have a political agenda and he has always been, if anything, utterly honest about his views. Yet his opponents spared no pains in painting him with a broad denialist brush and even discrediting his other admirable work in physics to debunk his climate change views. What disturbed me immensely was not that they were attacking his facts- that is after all how science works and is perfectly reasonable- but they were attacking his character, his sanity and his general credibility. The respected climate blogger Joe Romm rained down on Dyson like a ton of bricks, and his criticism of Dyson was full of condescension and efforts to discredit Dyson's other achievements. My problem was not with Romm's expertise or his debunking of facts, but with his tone; note for instance how Romm calls Dyson a crackpot right in the title. One got the feeling that Romm wanted to portray Dyson as a senile old man who was off his rocker. Other bloggers too seized upon Romm-style condescension and dismissed Dyson as a crank. Since then Dyson has expressed regret over the way his views on global warming were overemphasized by the journalist who wrote the piece. But the fact is that it was this piece which made Freeman Dyson notorious as some great global warming contrarian, when the truth was much simpler. In a Charlie Rose interview, Dyson talked about how global warming occupies very little of his time, and his writings clearly demonstrate this. Yet his views on the topic were blown out of proportion. Sadly, such vociferous, almost violent reactions to even reasonable critics of climate change seems to be becoming commonplace. If this is how the science of global warming is looking like, then it's not a very favourable outlook for the future .

If Dyson has been Exhibit A in the list of examples of zealous reactions to unbiased critics of climate change, then the recent book "Superfreakonomics" by economists Steven Levitt and Stephen Dubner (authors of the popular "Freakonomics") would surely be Exhibit B. There is one chapter among six in their book about global warming. And yet almost every negative review on Amazon focuses on this chapter. The authors are bombarded with accusations of misrepresentation, political agendas and outright lies. Joe Romm again penned a rather propagandish and sensationalist sounding critique of the authors' arguments. Others duly followed. In response the authors wrote a couple of posts on their New York Times blog to answer these critics. One of the posts was written by Nathan Myhrvold, previously Chief Technology officer of Microsoft and now the head of a Seattle-based think tank called Intellectual Ventures. Myhrvold is one of the prominent players in the book. Just note the calm, rational, response that he pens and compare it to one of Joe Romm's posts filled with condescending personal epithets. If this is really a scientific debate, then Myhrvold surely seems to be behaving like the objective scientist in this case.

So are the statements made by Levitt and Dubner as explosive as Romm and others would make us believe? I promptly bought the book and read it, and read the chapter on climate change twice to make sure. The picture that emerged in front of me was quite different from the one that I had been exposed to until then. Firstly, the authors' style is quite matter of fact and not sensationalist or contrarian sounding at all. Secondly, they never deny climate change anywhere. Thirdly, they make the very important general point that complex problems like climate change are not beyond easy, cheap solutions and that people sometimes don't readily think of these; they cite hand washing to drastically reduce infections and seat belts to reduce fatal car crashes as two simple and cheap innovations that saved countless lives. But on to Chapter 5 on warming.

Now let me say upfront that at least some of Levitt and Dubner's research is sloppy. They unnecessarily focus on the so-called "global cooling" events of the 70s, events that by no means refute global warming. They also seem to cherry pick the words of Ken Caldeira, a leading expert on climate change. But most of their chapter is devoted to possible cheap, easy solutions to climate change. To tell this story, they focus on Nathan Myhrvold and his team at Intellectual Ventures who have come up with two extremely innovative and interesting solutions to tackle the problem. The innovations are based on the injection of sulfate aerosols in the upper atmosphere. This rationale is based on a singular event, the eruption of Mount Pinatubo in the Phillipines in 1990 which sent millions of tons of sulfates and sulfur dioxide into the atmosphere and circulated them around the planet. Sulfate aerosols serve to reflect sunlight and tend to cause cooling. Remarkably, global temperatures fell by a slight amount for a few years after that. The phenomenon was carefully and exhaustively documented. It was a key contributor to the development of ideas which fall under the rubric of "geoengineering". These ideas involve artificially modulating the atmosphere to offset the warming effects of CO2. Geoengineering is controversial and hotly debated, but it is supported by several very well known scientists, and nobody has come up with a good reason why it would not work. In the light of the seriousness of global warming, it deserves to be investigated. With this in mind, Myhrvold and his team came up with a rather crazy sounding idea; to send up a large hose connected to motors and helium balloons which would pump sulfates and sulfur dioxide into the stratosphere. Coupled with this they came up with an even crazier sounding idea; to thwart hurricanes by erecting large, balloon like structures on coastlines which would essentially suck the hot air out of the hurricanes. With their power source gone, the hurricanes would possibly quieten down.

Are these ideas audacious? Yes. Would they work? Maybe, and maybe not. Are they testable? Absolutely, at least on a prototypical, experimental basis. Throughout the history of science, science has never been fundamentally hostile to crazy ideas if they could be tested. Most importantly, the authors propose these ideas because the analysis indicates them to be much cheaper than long-term measures designed to reduce carbon emissions. Solutions to climate change need to be as cheap as they need to be scientifically viable.

So let's get this straight; the authors are not denying global warming and in fact in their own words, they are proposing a possible solution that could be cheap and relatively simple. And they are proposing this solution only to temporarily act as a gag on global warming, so that long-term measures could then be researched at relative leisure. In fact they are not even claiming that such a scheme would work, only that it deserves research attention. Exactly what part of this argument screams "global warming denial"? One would imagine that opponents of these ideas would pen objective, rational objections based on hard data and facts. And yet almost none of the vociferous critics of Levitt and Dubner seem to have engaged in such an exercise (except a few). Most exercises seem to be of the "Oh my God! Levitt and Dubner are global warming deniers!!" kind. Science simply does not progress in this manner. All we need to do here is to debate the merit of a particular set of ideas. Sure, they could turn out to be bad ideas, but we will never know until we test them. The late Nobel laureate Linus Pauling said it best; "If you want to have a good idea, first have lots of ideas, then throw the bad ones away". Especially a problem as big as climate change needs ideas flying in from all quarters, some conservative, some radical. And as the authors indicate, cheap and simple ideas ought to be especially welcome. Yet the reception to Superfreakonomics to me looked like the authors were being castigated and resented for having ideas. The last thing scientific progress needs is a vocal majority that thwarts ideas from others and encourages them to shut up.

Freeman Dyson once said that global warming sometimes looks like a province of "the secular religion of environmentalism" and sadly there seems to be some truth to this statement. It is definitely the wrong kind of religion. As I mentioned before, almost any paradigm that reaches almost universal consensus runs the risk of getting forged into a religion. At such a point it is even more important to respect critics and give them a voice. Otherwise, going by the almost violent reaction against both Dyson and the authors of Superfreakonomics, I fear that global warming science will descend to the status of biological studies of race. Any research that has to do with race is so politically sensitive and fraught with liabilities and racist overtones that even reasonable scientists who feel that there is actually something beneficial to be gained from the study of race (and there certainly is; nobody would deny that certain diseases are more common to certain ethnic minorities) feel extremely afraid to speak up, let alone apply for funding.

We cannot let such a thing happen with the extremely important issue of climate change. Scientific progress itself would be in a very sad state if critics of climate change with no axe to grind are so vilified and resented that they feel inclined to shut up. Such a situation would trample the very core principles of science underfoot.