Climategate II
13 December, 2009
As expected, Climategate has done nothing to temper the enthusiasm of the climate ideologues in the funfest of Carbonhagen, even the Danish sex workers harnessed themselves to the task of ensuring a successful event by opening their hearts (and their legs) to the delegates.
Climategate has NOT taught us anything we had not known before, it merely added another level of confirmation that global warming is a fraud of gigantic proportion. I shall explain that later.
The main stream media, who finally could not ignore the story went into a “yes but” mode, glossing over the fact that the “robust science” is a work of scientists who are more preoccupied with politics s and funding that with science. I is all a sceptic stunt to undermine the Crapenhagen conference, and the global warming is still as real as ever and the current cooling is only a natural variability after a long warming and still 2008 was the fourteenth hottest year since …. etc etc etc, whoopee!
Natural variability hey? and what would you call the warming of the 20th century after the Little Ice age that ended at the end of the 19th century? If you go by Climategate that has never happened — Prof Michael Mann, the father of the Hockey Stick theory (as we saw in Climategate I) religiously prefixed the terms Medieval Warming Period and the Little Ice age with the words “so-called” as if they had never happened. And they call us “deniers”(?!)
Sure, the Hockey Stick theory is now truly buried, even by the dogmatic IPCC. I was shown to be a plain scientific hoax but the charade goes on.
The Climate Models
The global warmers tell us ad-nauseam that the climate models (there are about six versions) provide the scientific proof to global warming.
Let me explain what a model is.
A climate model is a mathematical model, or model for shot. A model is a set of mathematical (including statistical) calculations, known as algorithms, attempting to simulate the real world. A model is as good, or bad, as the rules and the data put into it, for example:
Suppose I want to model trains time table; I know the distances between stations, I know what speed the train can do, therefore I know how long it takes a train to get from one station to the next. I allow 2 or 3 minutes for every stop and bingo I have a model that simulate the run of one train. But this only a start.
My aim to provide service to the travelling public so I need more than trains, I know how many passengers I need to carry thus I can work out how many cars each train will have (consistent with the length of platforms), I space the trains and run my model to ensure that the no two trains reach a station at the same time, or meet each other in the opposite directions, if they use the same tracks on both direction.
But I also know that the vast majority of the travelling public require to travel to and from work in the morning and the evening, so I run more and longer trains during rush hours. As more people need to get on and off the trains, I need to allow more time at stain stops, meaning yet more trains — it is getting complex but manageable because ,so far, I have dealt with known factors.
The difficulties starts emerging when I deal with known unknowns; mechanical breakdowns (trains and signals), accidents, weather delays (floods, snow, heat waves). I use probabilities that I work out from past records (data) and build it into my model, I then run “what if” scenarios (called sensitivity analysis).
Eventually I will have a model simulating the whole railway network in a quite reliable MODEL, the technology to do it exists, it was done successfully many times in the past, so far so good. BUT,
How successful would be my model if I try to do it for a city in a country that has never had trains before?
Not much because I don’t have the required data to construct my model — I would need to rely on guesstimates and experiences in other places and ASSUME that it is relevant. In technical wards, I built assumptions into my mode.
As I don’t really know the size of the train travelling public, I would try and estimate it from other available data, say, buses statistic – The bus data becomes my “proxy”, that proximates my train travelling public, is the yardstick by which I guess the size of the travelling public, it is not perfect but it the best I have.
You can now see, that usefulness of such model is limited by my lack of data, my confidence in my model would be shrunk by comparison to the earlier model.
What all this has to do with climate? I hear you ask; the climate models akin models that simulate trains that have never run (as yet). They rely on data, such as tree rings, ice core samples (called boreholes), and other proxies to simulate temperatures for the times when there was no methodological collection and recoding of climate data on earth which is the whole of the planet’s 4.55 billion years history barring the last 200 years at the most.
Climate Proxies
There is a legitimate debate about how well the various proxies represent past climate data. Such debate is a matter for the science to resolve and I am not going to buy into it.
This is not a criticism on the use of proxies, science has to use what is available to it, but we must bear in mind that whilst the bus travelling public may give us a good clue as to how many train passengers it is only an indication, that may or MAY NOT come to fruition.
In my second train model I would take bus data and CALIBRATE it. I would try to run my model a number of times with different assumptions such as 50% 60% 70% of the bus passengers will travel by train whiles it start running.
The climate science does just that, so as you can clearly see, it is an educated estimate, at best, and with all the care I take, I would not stick my house on being 100% right, would you?
It may take some 10 or 15 or twenty years before the new trains, which I just modelled, will start running. Would you now, base on my MODEL, commit yourself to be at the station at 8:17 am on Monday, 2025? of course not. Yet these global warmers not only want us to commit ourselves to be in station that has been built yet but they also want us to buy the tickets, (carbon credits) NOW because the models say so.
But there is more.
Suppose I see a bump in the number of passengers between 2pm and 3:0pm (presumable caused by school kids going home) and I ignore it as a “natural variability” as it does not suit my model. This is exactly what the Hockey Stick theory does — ignore available data because it spoils the model!
In fact, it is worse than that, the global warmers goes further and tell us that the trains are already running. Yet a mere 8 years into their predictions and the trains are running indeed but in the wrong direction!!! yet they insist that the model is 95% accurate.
This bring us to the question of:
How Certain Are The Models?
We often hear that the accuracy of the models are within 95% probability. No, it is Not!!!
The warm mongers are in fact referring to the statistical term known as degree (or level) of confidence, (also Confidence Interval) which measures the accuracy of their models.
That term has little to do with probability. It is a statistical measurement of an interval , a “window” around the model result into which a certain percent of the eventual and real life events expected to falls. It is typically 95% or 97.5% but it can be any number under 100%. I’ll give an example
Taking the trains again, it means that within a certainty of, say,95% a train will arrive within a time” window” around the schedule time. In other words, 95% of trains will arrive within a certain time before or after the appointed schedule time. Naturally, it stand to reason that the larger the time window is the more trains will be “on time”.
As you can see 95% confidence interval is meaningless without knowing what is the actual interval — it is one thing if 95% of trains arrive within one minute of the time table and a completely different story if 95% of trains arrive within three hours of the schedule time and still “be on time”.
Climate is a lot more complex than my trains model example and whilst in the case of running trains we know all there to know about what affect the trains running on time , when it comes to climate science, we don’t even know that the track is reaching the next station let alone the our final destination.
The best science do is forecast the weather reasonably accurately for FOUR DAYS into the future, any further then that is an educated guess and they want to tell me that they can tell the weather in 2050?
Enjoy you cold showers in the dark if you still think that you are saving the plant.
All aboard!
© copyright Jacob Klamer 2009 — all right reserved
Tags: climate models, climategate, confidence interval, COP15, Global Warming, mathematical models