http://hockeyschtick.blogspot.com/2014/07/wsj-confessions-of-computer-modeler-any.html
This is what I have been saying all along. The models are overparameterized. You give me 20 variables, whether they are related or not to the variable in question, and not only will I give you a good model, but I can make the model predict what I want it to predict.
People like Pegminer, have claimed that my PhD in statistics is meaningless in discussing climate modeling. This is, of course, absurd. Having modeled many times and knowing what you can do with various models informs me as to how much faith I can and should place into models.
Note that Gary F says "and that models do not make "predictions."" This seems absurd to me. Not only are the climate "scientists" making predictions 100 years out with the models, but they are also recommending changes to public policy based upon those predictions. Indeed those predictions are being used to model other things like rainfall in England.
Gary F mentions EOFs, which is really more of an implication of using EOF methods. The PDO is not an EOF. Theoretically, it is composed of EOFs, meaning that the PDO can be broken down into perpendicular signals. You can perform EOF analysis and rotate the EOFs until one looks similar to the PDO, but the PDO has a definition outside of EOF analysis.
The EOF is a fairly complex method of analysis, a PCA for time series analysis. I agree with Gary F, that the rationale for study is to garner a greater understanding of the climate. In fact, by all means, the scientists should be using differing models and testing the models.
But we all know they are not "testing" the models. They are allowing their research to be used to make predictions of 3 degrees of warming by 2100. These predictions come from the work of these modelers who are reporting on models that are entirely UNTESTED.
The potential for bias is HUGE in modeling. Even where no bias exists, when you use many variables, you pick out the variables that seem to be the most significant for modeling the past data. This in no way guarantees that reality will match match the model close enough to be useful.
We are currently seeing 95% of the model overestimating the current warming. In any other field, if 70% of the models were overestimating, I would be concerned that bias is entering the modeling process. Yet if you mention that the models appear to be biased, you are labeled an anti-science evil denier.
Gary F, is not here vehemently attacking those who make absurd claims of an apocalyptic future. A future that is as much greater than the "scientists" models, as no warming is less than their models. NO!!! A climate scientist is here taking a firm stance for ONE side, to the point of continuous insults.
The bias shown in overestimating models may not be purposeful, but I submit to you that THIS bias is easily seen. And why not? Lets talk about a cause of bias just from a human perspective.
Would you prefer to believe that you are helping to save humanity from itself and a climate apocalypse OR that you are helping with the boring stuff like what plants to crop.
That is to say, climate research is GREAT. IT helps in decision-making for crops to plants, dikes to build and numerous other decisions, that save us time, money and resources. But its not as sexy as saving the world, is it???
My problem is not that I think the scientists are in some big conspiracy with whoever to take over the world. My problem is that these many and clear signs of bias are being denied instead of addressed. Further, their research is being used to promote fear, NOT knowledge and they are doing very little about this.
Edit:
For clarification, I am not suggesting that the climate scientists are lying, so much as biased. Bias is natural. If you ask me about the drugs produced by my company, they are absolutely the best. Why would I not be biased? Our analysis is superb. If I did not believe it was, I would be working to change this. Plus I am helping to cure cancer, how noble right?
If someone were discussing my analysis being wrong, I would staunchly defend it, just as Gary F does. That being said, we know the bias exists and take EXTREME methods to avoid it affecting our data. Further, while I would not be as happy to stand up against people calling the drugs a miracle cure, that would still be my duty.
It is this debate that kept the IPCC from putting out a best estimate of global warming in the latest report. The best evidence indicated a number lower than they wanted to report, so the activists had to settle on no number at all. So even within the politicized world of IPCC, a healthy debate broke out and took away some of the more alarming projections. Now, if they could get more debate on the rest of their work, skeptics could give them more credibility. That the models have failed is pretty clear. Instead, the IPCC declared that they were OK, because a few models were close, therefore the average of all is acceptable. This would not work in cases of nutrition, or weather reports, or cars. Imagine you modeled the brake performance, and said that in 12 months, the car will be able to stop at a distance of 2 feet, 3 feet, 5 feet, 10 feet, and 180 feet. You report an average of 40 feet as your modeled estimate, and the result is 200 feet. You then say your long range models are still valid because the 200 feet is close to 180 feet.
If someone with good scientific credentials, working at a reputable university or institute, publishing papers in respected journals were to develop a model showing temperature rises by land, sea, and air consistent with the last century's-worth of observational data without the need to invoke the additional greenhouse gases we've pumped into the atmosphere, I'm pretty sure that skeptics would have no problem with computational models.
It's only the fact that the models don't predict the future you'd like that has you in a tizzy over the veracity of the models. Otherwise, my money says, you'd think they were great and be flagging them as supporting your position.
Wow, a consensus of one that people can lie. You can't beat that for Goddamn insight..
There is skeptical debate, but Deniers are not a part of it,
If you want to be skeptical of climate models, then become expert enough to understand them. Scientific skepticism requires scientific expertise. You cannot be skeptical of research unless you understand it well enough to do it yourself - and, in fact, you should have done it and published results for everyone to see, Less than that and you are just an ignorant person with an axe to grind.
>> that the lines between academia and consultancy are very blurry<<
This comment is clearly targeted at people who have a dislike of academia and "intellectuals" - and it is certainly even more effective on people also too stupid to be a consultant on anything. There is a Grand Canyon between academia and consulting - and anyone who has been to the real Grand Canyon knows that although just 10 miles wide, it takes 5 hours to drive from one rim to the other.
=====
edit --
>>, to realize the results dont match predictions<<
Minimally, you need to understand the purpose of the models, the results and their meaning - and that models do not make "predictions." You willful determination to not-learn from the numerous opportunities people have offered explaining that the models are dynamic experiments to improve our understanding of the climate system whose projections are climatic- and not weather-related, and that AGW theory is not based at all on the projections - not to mention that your so-called "knowledge" of model "predictions" and results comes exclusively from people who are either just as ignorant as you or dishonest political activists.
The fact that you cannot determine for yourself - based on your own understanding of the models - who is telling the truth - means that you are not a skeptic - and the fact that you entire motivation is to confirm your preexisting political beliefs makes you a science Denier.
>>and perhaps you have to be a nutrionist <<
You can either follow the advice of professional nutritionists or quacks (Denier choice),
>>or an auto engineer to decide which car to buy<<
You can research knowledgeable sources or ask some children (Denier choice).
>>or a meteorologist to decide when to go on vacation.<<
You can consider the weather report or call the psychic hotline (Denier choice).
============
edit --
>> they wont debate it. one side onl<<
OK, so now science is a debate and not a consensus of evidence.
And exactly why is it that the people clamoring for a debate are the ones who do not know anything about science?
And, how naive does a someone have to be to think any real knowledge can e gained within the time frame of a debate? That is exactly the same as not knowing the difference between climate and weather.
The PDO is an Empirical Orthogonal Function of SST. Do you know what an EOF is? No? Then you do not know what the PDO is and you cannot have an intelligent opinion of what it does. Let's say a debate lasts one hour. OK, you have one hour to learn enough about EOFs so you can design and conduct a study and write a professional quality report of the results.
Finally, there is debate. It takes place at universities, research laboratories, scientific conferences, and in the scientific literature. That is the game. If you are too stupid to play with the adults that is your problem. You can try debating the children in the playroom; but, be warned, they know more about science than you do.
=====
edit ---
re your Update 3:
Try it. If you tried to put it in a professional journal, you would be busted before it got published. Scientists interested in PDO forcing read everything published on the subject. Someone like Overpeck (whose tweets appear on the page) would recognize it as familiar and start looking through things he had read and calling people to ask if they had seen it.
You could not talk to anyone who knew anything about the PDO without being spotted as a fraud. You could not answer any questions about the paper that you allegedly wrote And, anyone who had ever heard you say anything about science would know that you did not write it.
======
Raisin Caine–
Not meaning to be overly picky, but the PDO is empirically defined as the leading EOF of a set of SST anomalies. Also, EOF and PCA both decompose time series data. The difference is that the EOF method includes geospatial factoring.
Here is the money quote:
"After three iterations I finally blurted out, "What number are you looking for?" He didn't miss a beat: He told me that he needed to show $2 billion of benefits to get the program renewed. I finally turned enough knobs to get the answer he wanted, and everyone was happy."
Is it really such a stretch to believe that climate scientists, also funded by the government, are willing to turn the knobs in order to keep the money flowing.
There is a truck coming right at us. Maybe we should move.
No, maybe that's not a truck.
Yes, it's a truck, coming fast right at us.
You can't be sure it is coming right at us. You don't know if it will turn.
If we don't move it will kill us.
Let's debate this. You only see with your eyes what looks like a big truck coming very rapidly right at us.
What we really need is a healthy debate about whether what you are obviously seeing is real. Let's take
another 20 minutes to debate whether that is really a truck coming at us before we decide whether to step
out of the street.
*Splat*
Oh, I guess you were right. That thing that looked exactly like a truck coming rapidly right at us really was a truck coming rapidly right at us. Let's now debate what we should do now.
"Is it really such a stretch to believe that climate scientists, also funded by the government, are willing to turn the knobs in order to keep the money flowing."
Heck, we've all heard of a university covering up the actions of a sexual predator to keep funding flowing.
That web link goes straight to a climate denier website !
The "debate" on global warming ended about a decade ago. At least 97% of scientists agree, which is about as good a consensus as you can get.
It's the denialists who are only interested in one side.
an interesting article on the perils of only accepting one side of a story.
http://hockeyschtick.blogspot.com/2014/07/wsj-confessions-of-computer-modeler-any.html
97% of scientists do NOT believe in catastrophic, man-made, global warming......Period!
By all means, we need to hear why the earth is flat
