Monday, February 19, 2007
Sad responses to the Australian’s Editorial
Well, since the Australian's shameful anti-science editorial, there have been a total of three letters published, all of them supporting the editorial. Now, I find it hard to believe that every Australian letter writer supported the editorial, but the letters that were published are rather sad.
The first letter was concerned that the IPCC consensus discourages critical examination. The author writes
First, science advances through critical examination (some would say scepticism) of theories to see whether they really do adequately explain observable facts and events in the real world.
Did the author miss this section of the IPCC report?
Scientific progress since the TAR is based upon large amounts of new and more comprehensive data, more sophisticated analyses of data, improvements in understanding of processes and their simulation in models, and more extensive exploration of uncertainty ranges.Climate science has advanced to this new, firm consensus by doing science, gathering and testing data. This is the result of 600 scientists distilling the latest published research over the past 6 years. At the very heart of this process is critical examination. Getting researchers to agree on something is like herding cats, and if they do agree, it is after long and exhaustive examination of the data and theories. Anyone can do a quick PubMed search to find the latest data in pre-eminent journals, it’s not like it is hidden.
But more depressing was the article in Fridays (16 Feb) Australian, apparently quoting Liberal MP Dennis Jensen in parliament. It is all the more sad as Jensen holds a degree in Applied Physics, a PhD and worked as a researcher at DSTO (and otherwise says quite sensible things).
[IPCC 90% certainty of anthropogenic component] Sounds good, but that is nowhere near certain in scientific terms.Nothing is ever certain in scientific terms. All we can say is that the balance of probabilities is in a hypotheses favour. While Jensen, as a physicist, may be disparaging of 90% certainty, many scientific fields would be envious of a 90% certainty. We have less certainty that prions are responsible for the variant Jacob Creutzfeldt disease (vCJD) that has killed people in the UK, but massive culls of animals, altered animal carcass handling and altered dietary habits have been put in place on the basis of the evidence we did have.
If you think 90% is beyond doubt, think about the Australian Cricket team. The odds of them winning the recent one-day cricket series would be have been extremely high…Aside from the issue that no scientific conclusion is ever beyond doubt (that is what makes science science, after all), he is conflating statistical inference of cause with statistical prediction on the basis of past events. This might sound a bit pedantic, but it alters how we approach the conclusions significantly (the question is not, given past performance, will Australia win further matches, but that in the test series where they won lots of matches, was that winning streak was due to then being more skilful than the English or some other factor. I will not even attempt to address Bayesian priors, at the risk of exploding reader’s heads).
The conclusion that anthropogenic causes are responsible for the major part of recent global warming is due to a process of inference. In the same way that we infer that Joe Bloggs broke into your house, by a process of inference from the fact he is a known break and enter felon, he was seen near your house at the estimated time of the break-in, foot prints that match his shoes were found at the scene, fragments of glass with the same chemical signature as the glass in your broken window are found in his clothes. Now, you might like more evidence (he may have been passing by innocently to the 7-11 for some putty to fix a broken window pane that co-incidentally has the same chemical signature as yours as it comes from the same manufacturer), but in civil trials 90% certainty will get you convicted. Similarly, we can say on the basis of the evidence seen in test series that Australia won its matches due to being more skilful than the English, not due to a voodoo curse on the English (although the English might not feel that way).
In the case of anthropogenic causes of global warming, we know that carbon dioxide and methane etc. are greenhouse gasses, the relentless logic of Arrhenius' greenhouse law (which is why Earth isn’t a frozen lump) means that temperature must rise if their concentration increases. We have humans pumping unprecedented levels of greenhouse gasses into the atmosphere. The observed rise correlates strongly with the input of our greenhouse gasses, and not with any other cause. There is lots more evidence of course (See the IPCC reports and real climate for further details), and there is a lot research minutely examining all possible causes of temperature rise. Also, keep in mind the IPCC took a very conservative approach to estimating likelihood, so the links are actually underplayed.
Again, none of this inferential process is 100% certain, science never is. With any scientific process we must always operate at less than 100% certainty.
Now the flip side of this is the prediction aspect, which is what Jensen actually gives and example of. Given the success of the Australian team in the test, you would be reasonably confident that Australia would win the one-day series (actually, calculating the Bayesian priors for this is fairly tricky, and the outcome is far less certain than you would think, but you would not be amiss in wagering a small sum on the outcome). They didn’t, and the only loss was a bit of pride and some bruised egos.
On the other hand, consider this scenario, you are the Mayor of a coastal city, there is a category 5 cyclone coming. At the failsafe point, the last time when you can successfully organise an evacuation of the city, there is a 90% probability that the cyclone will pass directly over the city. What do you do, do you ignore it, because 90% isn’t certain? Or do you evacuate, with the attendant problems that will cause?
Most people will evacuate, because the harm caused by being right (having your city flattened by a cyclone) greatly outweighs the harm caused by being wrong (the chaos of evacuation when the cyclone passes “harmlessly” to one side). The British beef industry was turned upside down with a smaller degree of certainty than the evidence for anthropogenic inputs to global warming, because the public health was in clear and present danger despite the small number of actual and predicted deaths (and subsequent research shows they were right to do so). A recent South Australian example where policy was implemented on far less than 100% certainty was the case of tracing the source of bacterial infection to a meat supplier.
In terms of climate change, we are at or near the failsafe point now, the greenhouse gasses we emit now will affect the climate for decades to come. Unlike vCJD or toxic E. coli, the major harms lie in the futures of our children and grandchildren (between 2040 to 2060), so we are somewhat less exercised about them. However, it is not the case that we have the luxury of waiting until we are 95% certain (how would you feel if the Health authorities stated, “we are only 90% certain the bacterial contamination came from this source, so we will wait until we are 95% certain”, probably pretty mad). The lags that happen in the climate means we have to slow down greenhouse gasses now.
Jensen also writes:
Vice Chairman Yuri Izrael, commented that "There is no proven link between human activity and global warming."Apart from the fact that proof is for whiskey and mathematics (remember, science does inference to the best explanation on balance of probabilities), why Jensen feels a quote from 2005, 2 years before the new consensus document, has any meaning is puzzling. Especially when the good professor also says “Natural factors and the impact of man seem to be interlinked.” Which nobody at the IPCC would disagree with. He also says "Global temperatures will likely rise by 1.4-5.8 degrees during the next 100 years." Again, this is not good news for global warming deniers, and of course has been superseded by the 2007 consensus. He also says “In my opinion, academics, politicians and governments should assess maximum permissible temperatures and carbon-dioxide levels. Quite possibly, the world would have to sacrifice something in the face of a common threat”. Again, this is two years before the current data, but still indicates that global action will needed to be taken to combat global warming. Not exactly the picture Jensen wants to promote.
Jensen also writes:
”Michael Crichton has stated that the last time he can remember when there was so much pressure for scientists, and the population at large, to accept a certain view as gospel was the issue of eugenics.”What is it with Michael Crichton and these people; he’s a novelist for Zarquon’s sake! Why would one accept the word of a novelist, one that has made fundamental mistakes in portraying climate and climate science, over the actual scientists that work in the climate field? Michael Crichton writes a cracking good novel, but he is not a scientific researcher (even if he was a medical doctor around 30 years ago). Tim Flannery is a bona-fide researcher, with a strong track record of research behind him, who has written some cracking non-fiction books. Yet when he puts his spoke in, he gets vilified by the global warming deniers. Crichton, a non-scientist who makes appalling egregious mistakes when he writes on climate, gets elevated to near saint hood. And still they don’t actually talk to climate research scientists. You know, the ones who actually do all that science stuff with data and what not.
Now, Crichton is a fair bit older than me (I read “The Andromeda Strain” in high school and was fascinated by it), but unless he is a really well preserved 90 year old, he can’t have been around during the eugenics discussion of the 20’s and 30’s (and remember them, a precocious 10 year old might just remember eugenics discussions in the 30’s). Eugenics was, by our standards today, rather bizarre, but it was a confluence of several themes around public health. Eugenics was supported by a broad coalition of conservative and liberal politicians, and the general public, based on pre-genetics concepts of “breeding” and a melding with the new science of genetics. There was no “pressure” on scientists to accept it (although some scientists promoted it, Galton being a major champion of encouraging “good” couples to breed). The acceptance and practice of eugenics legislation was complex (and the history of eugenics veru complex see Kevles, D. J. (1995), In the Name of Eugenics: Genetics and the uses of human heredity. Revised ed. Cambridge, Mass: Harvard University Press.) , worked differently in different continents and states, and in many cases was a shameful farce.
In the climate debate, there is no pressure on scientists to accept the scientific consensus as the gospel truth. It is the accumulated knowledge and data, generated by over 600 working scientists, that is driving scientists to the consensus. One may was well rail against physicists pushing relativity on the basis of their concensus. Crichton is a poor source for information about science. I doubt that physicist Jensen would cite Crichton for information on nanotechnology (despite "Prey" being a good airport novel, the nanotechnogy stuff is well, farfetched shall we say), so why quote him on climate science?
What is sad is that people who should know better (and Jensen is a person who I would largely agree with on other science topics, if his speaches are anytihing to go by) misrepresent science, and listen to novelists, rather than try and get to grips with the science and this issues that confront us in the coming decades.
About a year and a half ago I did talk to some meteorologists (essentially physicists) who work at the UK meteorological office/University of Reading. They seemed to have pretty good handles on the sources of many of the fluctuations in the global climate.
The current rise in global temperature amounts to a huge injection of energy in a very short time. That has to come from somewhere and the Sun hasn't increased in brightness enough to account for it.
An interesting aside is that the global CO2 content varies slightly during the course of the year due to the influence of plant life and the seasons. I hadn't appreciated that the difference in land mass (and therefore space for plants) between the northern and southern hemispheres would have such an observable effect when you 'fold' several years of data.