James Hansen on How Science Works

The following is excerpted from a longer report that James Hansen has just posted here. The excerpt appears under the heading "How Science Works" in the original:

How Science Works

by James Hansen

One of the books about Richard Feynman (can’t find it now) has a story something like this. When he started to work at Los Alamos on the Manhattan project he was essentially a post-doc. In an early meeting, led by one of the physics giants of the first half of the century, they were discussing possible interpretations of some data. One possibility was described and seemed to Feynman to be clearly correct, but they continued to go through many alternative suggestions and Feynman was getting more and more antsy: why didn’t the leader restate the evidence for the logical interpretation? Eventually the leader stood up and said something to the effect “o.k., it seems that so-and-so’s explanation is still best, let’s get to work.”

Analysis of complex problems depends upon being able to assimilate many imperfect sources of information. With a new idea or improved data, scientists have to reevaluate interpretations with an open mind. Sometimes this yields modification of understanding, and (very, very seldom) sometimes a dramatic change (Einstein was disgusted with media that, he said, portrayed science as if it were continually beset by revolutions, like the small southern European republics).

Being objective is not sufficient for scientific success, especially for a topic as complex and diverse as global climate. One also needs some training in core sciences such as physics, mathematics, etc., experience in examining how the real world behaves, and a mind capable of putting it together. The “laws” of science are only descriptions of reality as it has been observed, descriptions to be changed or refined as better data and knowledge develop.

Our understanding of the Earth’s climate, in particular, depends foremost on the Earth’s history: how past climate changed in response to changing boundary conditions. I rate observations of ongoing climate change and processes today, processes on the ice sheets, in the oceans, etc., as the second most important source of knowledge about climate change. Climate models rate only third, in my opinion. Models are a tool that helps us understand the other two, i.e., climate history and on-going climate phenomena. Models are a representation of reality, one that helps us combine different factors, evaluate relative importance, and try to understand interactions. As we make progress we add more processes to the models and improve representations of others.

Here is the strange part: within the many irate e-mails that I received after my 23 June talk was a common theme – the writer stating flatly that he would prove my fraudulence by exposing flaws in our climate modeling (the frequency of this theme made me wonder whether a bounty for such had been offered!). Now I am receiving messages demanding that I fess up to having made wrong projections of climate change in the 1980s. Oddly, the “errors” they point out are results I considered to be the most interesting and useful. What we have here, it seems, is a failure to understand how science works.

Their criticisms emphasize (1) the magnitude of global mean warming over the past 20-30 years, and (2) the amount of Southern Hemisphere warming, especially of the ocean around Antarctica. To understand what we learn from observed climate change, and from comparison with simulations, one must understand what the models contain.

Our 1980s climate model had sensitivity 4°C for doubled CO2. Real world climate sensitivity was estimated to be 3±1.5°C, so we realized that our model was within the high side of the likely range. That is fine – the practical approach is to put in the ‘physics’ at our disposal and then use the resulting model, being aware of its sensitivity – and it is usually possible to estimate how results would differ for an alternative sensitivity.

An original aspect in our modeling then (J. Geophys. Res., 93, 9341, 1988) was the fact that we made the first “transient” simulations, i.e., including time-dependent greenhouse gas amounts using measured or estimated greenhouse gas changes for the past and several scenarios for the future. Our simulations covered 1958-2030, which was not easy, given that we had only a single old computer, one that had been discarded by the census bureau.

We were able to make long simulations by using an atmospheric model with coarse resolution and a simple representation of the ocean, the latter including uptake of heat perturbations as a passive tracer with mixing rates calibrated by observations of ocean tracers. This model would be expected to do a good job of simulating heat uptake, and also mean surface temperature change if the model’s sensitivity is right, but no ocean dynamical feedbacks were included.

Actual global warming over the past 2-3 decades is at a rate close to that in our 1988 paper for scenario B, the scenario closest to actual climate forcings. The slightly larger warming rate in the model (Figure 2 in Proc. Natl. Acad. Sci. 103, 14288, 2006) is consistent with the fact that the model sensitivity (4°C for doubled CO2) is somewhat higher than real-world sensitivity (~3°C for doubled CO2) inferred from paleoclimate data and more recent models.

As for the relatively small observed warming of the Southern Ocean in recent decades, this characteristic is obtained in our more recent models that include ocean dynamics (e.g., Clim. Dyn. 29, 661, 2007). Observations are not good enough to be sure how large the Southern Ocean warming has been on the century time scale. However, contrary to the assertion of people sending me critical e-mails, observed temperature changes provide no basis for doubting that there will be polar amplification of warming in the Southern Hemisphere, as well as in the Northern Hemisphere – paleoclimate data make that amplification abundantly clear. And it is important that we develop the ability to understand and simulate well the unfolding Southern Ocean dynamical response, because it will affect the stability of ice shelves around Antarctica and eventually the ice sheets that determine future sea level change.

The public, however, is not presented a realistic picture of how science works on such matters. Instead public discussion of global warming is befogged by contrarians, whose opinions are given a megaphone by special interests that benefit by keeping the public confused. Some of the contrarians were once scientists, but now they behave, at least on the topic of global warming, as lawyers defending a client. Their aim is to present a case as effectively as possible, citing only evidence that supports their client, and making the story appear as favorable as possible to their client. The best, the most articulate, are sought out by special interests, and even by much of the media, because the media likes to have “balance” in its coverage of most topics – and especially this topic because special interests have influence on the media.

The barrage of e-mails that I have received from the public highlights another aspect of the global warming story: it is now very political. The people sending these messages are not generally scientists, even though in many cases they parrot “scientific” statements of contrarians. In their opinion these matters should be discussed in you-tube “debates” between scientists and contrarians. My guess is that scientists may not fare very well in such a format.

It is this situation that has created what I call a huge gap between what is understood about global warming, by the relevant scientific community, and what is known about global warming, by those who need to know, the public and policy-makers. Nobody ever asks me what I mean by “the relevant scientific community”. If they did, I would say: people who know what they are talking about (which may cause a bit of consternation, but this is no time to mince words).

Is there any way out of this situation? Continuing real world climate change and the scientific method will eventually make things clear. Unfortunately, because of inertia of the climate system and climate tipping points, it is extremely dangerous to wait for real world events to be so large that they overwhelm special interests and their contrarian lawyers.

Here is one suggestion: the next President should ask the National Academy of Sciences to provide him a prompt assessment of the situation. After all, Abraham Lincoln established our Academy for just such purpose. Interestingly, at the beginning of the current administration, in early 2001, the President asked the Academy for a (albeit limited) assessment of global warming, apparently under the belief that the Academy would be critical of the most recent report of the Intergovernmental Panel on Climate Change. Well, the Academy’s report did have some criticisms, but, with clarity and authority, it reiterated the reality of global warming, the predominant role of humans in causing the warming, and the need for a policy response to minimize climate problems. The administration was apparently so taken aback that they never asked the Academy again for any broad advice on the topic. It does not do much good to cry over that tragedy – now is the time to figure out the best way forward from this point.

• ‹ previous
• up
• next ›