G'day again,
This is my current favourite topic, temperature data. So Mike Kramer believes that everyone is admitting the world is getting warmer. How do they reach that conclusion? As Mr Blair, PM of Britain says, science is not a matter of voting for a position, it is a matter of scientific enquiry. The majority deserve to be listened to but that does not mean they are always right.
I liked that rather reasoned and rational approach.
The last couple of days I've been looking at a research paper that is all about temperatures. It shows the earth is cooling! The authors do not believe this by the way but that was the only conclusion they could reach from the data they had when they attempted to eliminate various errors in the Surface Air Temperature data. I don't believe this is going to be published in Science any time soon.
This is where it gets tricky. In order to replicate this study, you need access to the data. Hey presto, the GHCN data set is available on the Internet for you to download. I've posted this a few times, but this is an easy access version:
http://data.giss.nasa.gov/gistemp/station_data/ . Here's another link:
http://fuzzo.com/ghcn/. If you want to do anything with the GHCN data, this is as good a program as any to be able to read it.
OK, so you now have 7,000 weather station monthly averages. Dr Hansen and his co-authors are actually talking about a computer simulation of rather course structure. Way too course to take into account clouds for instance and very much out of date. But it is still a model. ie. A guess. Why would every other science, be extremely sceptical of researchers guessing the future, yet climate studies, love it so much?
Forget the "average" fiasco which may make the figures, especially from 1970 on, out by as much as 2 degrees. In order to do anything with this data, you have to eliminate urban effect. That cuts out the majority of the figures. Then you need to standardise your model. Can't have 500 points in the US and none in Africa. That would bias the results if there is a regional climate shift that is not worldwide. But you also need data that goes back a few years and relates to the same station. So now we are down to about 1% of the stations. That's 70 stations. And that's why you get global cooling. You don't have enough data.
Why not use the satellite data or the balloon data for Dr Hansen's model? Well, because it does not show global warming. It shows a marginal warming in 30 years based on two hot years. It matches sunspots where the SAT data diverges from the sunspot data markedly since 1970. I quote Dr Sami Solanski of the Max Planck Institute for Solar System Research:
"The temperature of the Earth in the past few decades does not rcorrelate with solar activity at all."
But did it in the past. Yep. Another scientist:
"A couple of years ago, I would not have said that there was any evidence for solar activity driving temperatures on Earth. Now I think there is fairly convincing evidence"
Paula Reimer, palaeoclimatologist.
And this:
"If you look back into the sun's past you find that we live in a period of abnormally high solar activity"
Scientific method suggests that if you have persuasive links between two systems then if you have records that do not match either of these two systems at all, generally, you have a look at what might be wrong with the data.
So, what is it. If global warming is man-made and if you believe satellite data is accurate for such things as ice measurements of the Arctic, then satellite data must be accepted for the world's comparison temperatures, you have a really big problem, that the satellite data does not show global warming overall, just a bit of an increase in a couple of years that really do correspond with solar activity.
This cannot be that hard to understand, really. You have a data set that any study of demonstrates just how bad it is. You have two other data sets that no one can argue are extremely accurate that do not agree with it at all. Sure you can argue a bit about shielding problems in the early 70s or shift problems with satellites but if you assume the absolute worst and correct for these, you still do not get any match with the other set of data and you still do not get global warming. You get a little variation upward for a very short time within a 30 year period.
If you want to argue with this, it is really very simple. Point to any research that suggests that the SAT data is not flawed. Suggest logic problems with the comparison of the SAT against the balloon and satellite data. Suggest some global warming reason why the satellite data will show nothing when the world really is warming such as the models suggest that global warming will cause a cooling in the upper atmosphere and that is screwing up the data. I wouldn't suggest you use this theory on the two scientist groups that produce the satellite data, however, because it really ignores how satellite data is collected. It does not collect upper atmosphere temperatures as a major component of the temperatures that are used.
So, if you really think there is global warming (and you might well be right but not based on the satellite data currently available - the 30 years does not show it and it is just not long enough), suggest data that is comparable and is not just regional and is not seriously flawed. While you're at it, suggest why the very good data that is available from the US and Canada, and New Zealand and Australia, does not correspond with the rest of the SAT. If you take them alone, taking out urban areas, you get some regional climate changes with some of the US warming, some cooling, some of Australia warming, some cooling, but overall no pattern or a cooling one.
This is not all that complicated in the end. It is really easy to obscure facts or cover for bad data.
A Couple of Long Stories About How Errors Can Occur Where Everyone Turns out to be Wrong I had to work out why a railway line sunk once. For four years the very best engineering firms in the world had been arguing amongst themselves. Each expert opinion on why the collapse occurred, brought a response, usually another hundred pages longer than the opinion, from the other four groups pointing out all the flaws and why another theory fitted better. 40,000 pages of expert opinion eventually turned up. It would have been really funny if it did not involve several hundred jobs and $120 million dollars and the railway line not being fixed (it didn't carry trains but 180 tonne specialist equipment).
All I found with all that opinion was not facts but science used to obscure facts. The opinions became progressily more densely scientific, with more complex formulas, each time.
End result? I asked each group to submit their theory in a standard form without attacking any other theory and while they did that I interviewed those involved and had another expert take samples and we both examined the scene by cutting a rather big hole under the railway line. Not one of the theories came close to being correct because they were not being produced to do that. The reputations of some very big egos rested on the results of this, not to mention who would cough up the $120 million.
It turned out that everyone made some mistake that contributed to a very simple failure in the end. The original geotech survey didn't sample enough of the area to really determine what was there. The geotech survey was fudged a bit towards a way that would cut a couple of million dollars of the building costs and quality material was assumed to be in the excavation area when the geotech survey suggested only that it might be, and it was wrong in this because it was in an area where there were too few samples.
So then the next group of experts were faced with a decision of using material that wasn't what was needed. So they changed the specs, using some calculations that would have worked except for some drainage issues and the fact that this was not a standard railway line. The definition on what "crushed" was, was altered a little bit.
One person in construction, who was very experienced in construction but did not have a great many letters after his name, actually pointed out that the material was deficient. He was told he didn't know what he was talking about.
Some pipes that needed to be specially made for drainage were not available and so a substitution was made for another type that should have been as good, but wasn't because of the other changes.
Overall, each mistake or correction was very minor. Each looked like it would not affect the outcome or impinge on the margin of safety by more than a tiny amount. That was true if the stuff was up to scratch. But it wasn't and the whole thing started to sink after a couple of years.
I had the world's top experts trying to figure out why power grid stations were blowing up (and in a couple of cases frying workers). Couldn't do it until it was found that a purchasing officer had a friend in a chemical company and swapped over the supply of some insulating material to that company. They made almost identical insulation gunk so wham was the harm? Helped his friend make a $10,000 sale. Trouble is the material had a temperature range only up to 55 degrees celcius and the material it was covering on occasion reached 58 or 59 degrees. The actual specifications required material rated to 65 degrees (although the normal working range was the same for the originally requested material and for the swapped one). This little change, that no one could figure out for months cost over $3 billion.
I've done hundreds of these types of enquiries and, while, they have nothing to do with global warming, they do have to do with why everybody just knows that global warming is irrefutable and even man-made because of the CO2 and the very real possibility that "everyone" doesn't know that they are talking about. Not saying I do, but I really would like more people to focus on the fundamental part of global warming, the data. If it cannot be trusted, then you either need to get better data or work out a way to prove global warming through other means, if it is true, which the satellite data suggests it might not be.
Regards
Richard
