Climategate: The Perils of Global Warming Models
Everyone readily admits that things aren’t always what they seem. But are we really applying this knowledge in our daily dealings? Are we consciously ferreting out the illusory from the reality? I think not.
For instance, despite overwhelming evidence to the contrary, we aren’t really being run by pandering politicians, self-serving lobbyists, fanatical environmentalists, and greedy Wall Street manipulators. They are the illusion.
There is another even more powerful (but much less visible) agent behind all of these puppets.
The person behind the screen is the computer programmer. And, just like in the Wizard of OZ, they do not want you to look at this real controller.
I’ll probably have to turn in my membership card, but as a computer programmer (and physicist and environmental activist) I’m here to spill the beans about the Wiz.
The first hint of trouble is spelled out in Wikipedia’s explanation about computer programmers:
The discipline differs from many other technical professions in that programmers generally do not need to be licensed or pass any standardized (or governmentally regulated) certification tests in order to call themselves “programmers” or even “software engineers.”
Hmmm.
My layperson explanation is that computer programming is all about making assumptions, and then converting these into mathematical equations.
The big picture question is this: Is it really possible to accurately convert complex real-world situations into ones and zeros? Hal may think so, but higher processing brains say no. Yet this is continuously attempted, with very limited success. Let’s pull the screen back a bit more.
We’ll start with an example about how such a model makes assumptions.
One of the computer programs I wrote was for debt collectors. A typical scenario was that a debtor was given a date to make a payment and the collection company didn’t receive it on time. What response is then appropriate?
In such a circumstance the computer program typically makes an automatic contact with the debtor. (Remember there are thousands of these debtors, and it would be prohibitively time consuming for an agency person to manually check into and follow up each case.)
So what to say in this correspondence to the debtor? Well, it comes down to the assumptions made by the computer programmer.
The programmer tries to simplify such situations into mathematical options. In this case they may decide that the question is: “Does the debtor have the money to make this payment: yes or no?” This relatively basic choice then leads to a Boolean progression within the program.
How does the programmer (model) decide on yes or no? Well, other indicators would be used (e.g., were prior payments made on time) to come up with a statistical probability.
Of course, any computer model is not one set of choices, but rather a whole series of yes/no (if/or) calculations that lead to a conclusion. In a complex situation (e.g., debt collection, climate change, or financial derivatives) there could easily be a hundred such choices to deal with.
To understand the implications of that, let’s just consider the case where there are ten such decision points — each with a “yes” or “no” answer. At the end of such a pipeline, that means that there are 210 (i.e., 1024) possible results. That’s a lot of different potential conclusions.
Unfortunately, there are actually many more possibilities! The assumption that this debtor situation could be condensed down to a “yes” or “no” answer is not accurate. There are several other real situations that fall outside of “yes” or “no.”
For instance, what if the debtor never got a notice in the first place that the amount was due by the date the agency is monitoring? Or what if the debtor sent the money and it got lost in transition? Or what if the debtor made the payment to the original person they owed, rather than the collection agency? Or what if the debtor sent in the money on time, and the collection agency incorrectly didn’t credit the debtor for the payment? Etc., etc.
For the computer program (model) to be accurate, all of these scenarios need to be able to be handled properly (legally, timely, etc.). Can you begin to see the complexity here, just with this very simple example of a payment not being received on time?
There is still another significant factor (we’re up to #4 now) not mentioned yet. What about the situation where the debtor hasn’t paid, but it’s because his child has MS, and he has no insurance? How does a computer programmer write code for more abstract concepts, like “fairness”? In other words, can ones and zeros be arranged in such a way to represent intangibles? I think not.
Back in the day there were two aphorisms which cover everything you’ve written in regard to computer modeling: 1) Garbage in, Garbage out, and 2) Figures don’t lie, but liars do figure.
This is the featured article in my ClimateGate roundup for today.
http://spinstrangenesscharm.wordpress.com/2009/12/21/climategate-2/
I make a very good living in computer modeling of vastly simpler problems, and know how fecking hard it is to get things right even there. Anybody who claims they can build a computer model that can realistically model global climate is either dishonest or delusional — or both at once.
Horoscopes aren’t accurate? Make a bet? My horoscope said I’d have to postpone my Christmas holiday (to Bermuda) because of “very cool turn of events”, and guess what.. here I am stuck at home with 4ft of snow outside. So there, astrology denier! Warmers will inherit the world – damn those cockroaches!
Excellent piece. Now if we can only get the politicians to read it.
If it looks like dung, smells like dung and tastes like dung it is probably not a good idea to step in it.
Too bad for President Obama and the 111th Congress.
They didn’t listen.
Well said sir.
We are preparing to spend trillions of dollars on an emergency program to “save the world from CO2.” This from a group of people who can’t even predict next weeks weather, yet they expect us to believe that they can predict what the temperature will be in 100 years…1000 years.
I don’t think so. Not when it’s have been shown that tempuratures from at least 4 seperate places around the world were literally changed in order to fit the “rising temp” thesis. Australia has been shown to have been manipulated. Same for Memphis, TN. The CRU only used data from 40% of Siberia, because the remaining 60% didn’t fit the rising temp criteria and was discarded. Antartic ice core temps were only used from one site because the rest disagreed with the presupposition of “warming”. This alone tells me that it is likely the remaining data has been altered.
But, we’ll never know, as most of the raw data has been discarded and all that remains is the “manipulated” data. It’s this suspicious data that we’re expected to trust. Nope. Uh-huh. I don’t buy it. If they can’t match the historic trends without “manipulating” the data, they the models are not just wrong, but are fraudulent. Any conclusions that are based upon them are fraudulent as well.
It wasn’t emphasized enough that science doesn’t know much of how the atmosphere behaves. When we don’t know the effects of clouds, the computer programmer can’t create proper rules for them. The example mentions the effect of the MS illness, so the programmer may be aware that illness is a factor but he can’t handle the effects of illness. Some illness can even increase the amount of money (such as someone in a manic phase being more productive), so merely being aware of illness is not sufficient to properly calculate the effects.
Machines have always been limited in their capacity. They are only as good as the people who design, build and use them. Your article reveals nothing new to those of us familiar with machinery.
In any situation where the premise is false (in this case, “man-made climate change”), the results will be false as well.
Standing outside,
teeth chattering, shivering and goosebumps while claiming we went to far with warming. It is burning us up.
Priceless.
Are you saying my horoscope is wrong? Oh god!, what will I do now?
Programs are tools. The first question should always be “what is the intent of this tool”.
Sadly, with CRU, the intent was to craft a lie about “climate change”.
If a model has not been proven to fully reflect reality…
As you say, a model cannot fully reflect reality.
The dynamics of weather/climate are not fully understood.
Even if they were to be fully understood (hypothetically) it is doubtful that human beings could, in fact, construct any computer model that accurately encompassed the workings of climate dynamics.
I’ve thought this for a very long time. I’ve also assumed that (some) climate modelers have built in parameters into their own models that would lead to a desired outcome, i.e., the outcome sought by providers of their grant money.
Hold on here, chum.
Forget modeling, – what about starting with a reliable data warehouse, warehousing procedures, and REAL data first?
One kind of computer program model is a simulation. I think the AGW folks use simulations. They create models that attempt to capture the actual forces in the climate. But solved simulations have to be interpreted for the real world and they are only as good as their track record. Permit me to use a real example.
Suppose you are using a system of linear equations to represent packaging plants that have known packaging capacity and known methods of shipping, mostly trucks and railroads, and you want to solve the model to find locations for new packaging capacity that will give the most efficient overall pattern of shipping in the continental USA.
There is nothing magic about the mathematics or the programming. You have to solve the model many times with many different factual assumptions in order to build a track record. Interpreting the results is a matter of learning that some aspects of solutions count as good simulations of actual phenomena and some do not. In the particular model that I managed, the most salient phenomenon turned out to be “out of pattern shipping.” In other words, if products packaged in Columbus, Ohio showed up in grocery stores down the street from the plant in St. Louis, Missouri, then we knew with certainty that the added capacity in Columbus, the hypothesis, was in the wrong place. If folks who are working with models that simulate climate phenomena cannot or will not tell you what aspects of the model are reliable, what not, and what simply unknown then they are not being candid and honest.
The AGW modelers should be able to tell us their assumptions, their track record, and their assessment of just how reliable the model is in simulating important aspects of climate. In doing so, they should admit that their model is incomplete, in that it has no information for some climate processes, and has a poor track record for some other processes. Then they could tell us with clarity just what climate phenomena it models well. Of course, the people who use a climate model are at a disadvantage because, unlike my packagaing model, they have difficulty performing experiments. The only way they can perform experiments is to solve their model using a very short future time horizon. But then they would say that the results represent weather not climate.
There has been a lot of errors in the calculations. One thing is to try to state “how much” the temperature is rising. Is it local or global? How about the Urban Heat Island Effect? How can you effectively calculate it out? Watch a little video that shows the difference between urban and rural sites. Go to YouTube and look up “Global Warming Urban Heat Effect.”
http://www.youtube.com/user/TheseData#p/a/u/1/LcsvaCPYgcI
So Simple A Sixth Grader Can Understand It!
In writing a computer program of any complexity, there are literally hundreds of assumptions made. The computer programmer does not reveal all these to his customer, for much the same reasons that an accountant does not tell his client all of the assumptions made in preparing a tax return. He goes over a few of the more basic items, and then says “sign here.”
Well, you would have been canned where I’ve worked. That is *not* how a professional analyst/programmer behaves. Not only is every ‘assumption’ documented and tested, they’re gone over with the client. Same with taxes. I went over each line with my tax preparer and every assumption was explained.
It *is*, however, exactly how the scientist/programming-dabblers have behaved. The unprofessionalism exhibited is breath-taking in its scope. Not only are human assumptions codified into ‘truths’, but the work is rife with fabrication and downright shoddy logic.
VEGANS OF THE WORLD UNITE – CO2 is plant food.
Don’t let the warmers destroy our food source.
Memorize our anthem and sing it before eating.
Beans beans the musical fruit
The more you eat the more you toot
The more you toot the better you feel
So let’s eat beans at every meal.
Incidentally producing wind power.
I am a geophysicist/geochemist by training. This is the most well written piece on climate modeling that I have read.
We have a saying in my field of chemical engineering, where we try to model the complex reactions in a refinery. All models are wrong. Some models are useful. We can sometimes design, control, and optimize the performance of the refinery using these models. But, we rely on the magic of feedback to help. What magic can these climate models use?
You make some very wise observations. It has been a long time since I did computer modeling (back in Fortran days!) but basic principles still apply. We did semi-classical quantum mechanical molecular modeling. The point was to model complex systems with simpler math. The results of the models were tested against real data to develop useful simplifications. The current crimategate crowd is doing the exact opposite. Taking real (or as we now know tricked up and phony) data and using models to create virtual data. This is not science.
Excellent insights. Here’s another real world example of how these models are deficient.
I live in Florida, where during hurricane season every names storn gets modeled by at least 7 different models to predict the path of the storm. Some site actually show the predictions of thos models plotted out, and they vary widely. This for models where data collection is massive and probably accurate, the scale is much smaller and the history much more detailed.
I’m willing to believe that these models will get better over time, as their creators learn from their mistakes. Every time a hurricane happens there are new predictions and new actualities to compare them to, and the feedback possible from all these tests should enable the modelers to do a better and better job over time, though this is far from guaranteed.
Climate change models are not tested against new data. They aren’t claimed to predict local phenomena nor very short term trends, so as time passes the actualities that occur are not seen as tests. Believing in them, then, without any proof of efficacy is ike any other faith. Ambrose Bierce said it best, I think, when he defined faith as “belief without evidence in what is told by one who speaks without knowledge, of things without parallel.”
Interesting … but what you described as programming was actually Business Analysis … you may have been forced to do that business analysis as part of your programming job but all of those if-then decisions should have been spelled out for you by an actual expert in the business you were programming for … a business requirement and functional requirement should have been developed long before you put fingers to keyboard and programmed it …
given that it appears nobody can accurately describe todays climate/weather forcings with any degree of certainty it is impossible to attempt to model it via programming …
I think the first problem is we do not have reliable data in a consistant database with valid checks and cross checks … in other words it wouldn’t matter if we had a perfect climate model since we can’t input clean data into it anyway …
The skills required of “computer programmers” are not the same as those required for statistical modeling, operations research, and all the related disciplines that should go into these climate models.
The problem is, that modeling skill set is also very different than the training of these “climatologists”.
Deliver these scientists into a year of core econometrics or statistics graduate study!
Ditto DanW!
I am a 30 year weatherman who has participated in several weather model evaluations for operational use. It is critical for meteorologists to understand that the mathematics of the model is perfect, but the output is ALWAYS WRONG because the model and the data going into it is ALWAYS INCOMPLETE. So, as a weatherman, you use the model by evaluating how wrong is it today, and in what direction, and adjusting from there. You make your money by knowing that however wrong the model is, it will be consistently wrong through its forecast period, and you go from there.
As aerospace engineers (me and the folks I work with) we design, build, test, and transition to production new complex aerospace vehicles. In early predesign phases we start off with simple / primitive models of major components to try to size all the major subsystems (e.g., airframe, propulsion, payload, wings, etc.). Then as the actual subsystem designers start to do detailed designs on their various components we have to increase the complexity / fidelity of most of these subsystem models so confidence in performance prediction can be improved (i.e., what type of propulsion, how does it work, how much fuel is required, what rate is the fuel expended, etc.). For aerodynamics (aero forces and moments) we go to wind tunnels to characterize forces and moments as a function of flight conditions and wing / fin / rudder / etc. position. We build up the first physical prototype(s) of the new vehicle and make predictions about how it will fly when completed and during its early planned flight testing phase. In the early flight testing phase we typically do simple (low risk – crawl before you start sprinting) aero exercises to validate / exercise the vehicle and flight control system to build confidence in the simulation predictions and the control system design. The model / simulation evolves to better match the flight test results and the flight control system design evolves as lessons are learned in the early flight test phase. Key subsystems may undergo extensive individual testing to improve the fidelity of these models (e.g., propulsion testing, sensor / instrument testing, etc.). Over time considerable confidence is gained in the ability of the simulation to adequately / credibly predict overall system performance (speed, range, maneuverability, etc.) because the individual subsytem models have been extensively compared to real world data and showed favorable agreement. The system simulation is then used to accurately predict / match all accomplishmed flight tests (to date) and can be used to predict performance for flight tests that are too dangerous or expensive to conduct with the actual new system. This is how we aerospace engineers typically use / evolve models / simulations as tools to design, build, test, and transition to production new complex aerospace vehicles. One could argue whether or not we qualify as scientists like the climatologists of recent fame. However, if at any point in a major development (I have personally worked 4 major developments over 24 years) we tried to massage our simulation in creative ways and convince our customers (stock holders, U.S. and foreign governments, corporations, etc.) that they should trust us and not require us to go through with all of the complex phases of model / simulation development (and validation) we would have been out of business a long time ago (and likely would have had many catastrophic flight failures along the way). If our simulation cannot adequately recreate previous flight tests (even failures or anomalies that were unexpected) then its credibility and usefulness as a design and performance prediction tool for the new system becomes highly suspect and either major new validation testing (and simulation / model updates) will be required or our program would get cancelled (its pretty cut throat for us evil Capitalists).
We can build useful software models of complex phenomena such as warehouse and financial systems only because we can constantly test the validity of the software against immediate real world events. At the very least we can spot check the predictions of the software against physical reality.
For example, to test software that controls the shipping of software, we send people out to physically check if particular packages are where the software says it or, even more importantly, if there are packages someplace the software predicts they cannot be. When an incorrect prediction is found, the software is corrected. Only after months or years of this correcting can complex software work correctly and even then it must be constantly tested and updated.
The Achilles’ heel of climate modeling software is that we have no means of checking its long term predictions. The models do not offer any hint to any phenomena we could measure today that would tell us that the model will or will not accurately predict the climate a century down the road.
By analogy, we are in the position of the shipping software predicting that a package will be delivered to a particular address ten years in the future. How could we possibly if the shipping software is accurately prediction the routing of the packages?
The climate models could be correct but given the very nature of such programming, they are most likely wrong in some significant way. We have absolutely no way of knowing how they are wrong or how to correct them. Nothing we can observe today tells us that future warming will reach unnatural levels, that such warming will be caused by human action or that the economic consequences of such warming will be a net catastrophic negative.
Government follows trends in business but delayed by a few decades. I believe that in terms of evaluating the predictive power in software, we have reached the point in government that business went through in the 1960s. Back then computer predictions were new and spiffy and often very valuable so business leaders who were naive about programming for a time treated computers like mystic oracles. As one critic of the trend said, it was not garbage-in, garbage-out but rather garbage-in, gospel-out.
It took business a good 20 years to shake near blind faith in software predictions. Let us hope the politicians learn faster.
Mathematical models of the climate sort are, by nature, simplifications of reality. They may supply insight, but do not dispense truth.
If someone told you that they built a computer climate model that made many assumptions about the earth’s carbon regulation system, “adjusted” as-found temperatures collected at sites whose condition was at best partially known, found daily high/low temperature variations of 5-20 degrees celsius, then threw in tree ring and ice core date as proxys for temperature data that they lacked, and from there told you temperature were sure to rise a few tenths of a degree per decade without limit, would you bet your prosperity on man’s ability to stop what might be a mirage? Even if that someone hid the underlying data from you so you couldn’t even assess their assumptions? Really?
Let’s say two decades before, after a review of climate data, other scientists of good repute warned of a cataclysmic nuclear winter. Would you then be willing to say that today’s scientists are really wiser than those past, or that computers make them wiser?
Let’s say that the conclusions reached lead to increased scientific grant funding, and recommended actions ceded increasing authority to certain power brokers at the expense of the masses. Would your still trust the climate model?
Is precipitous warming possible? Yes. Is precipitious cooling possible? Yes. Nearly anything is possible. Is a climate cataclysm likely? Reports of impending doom span the millenia. Global warming believers may want to take another look and add a dolop of horse-sense in the bargain. Scientists, no mater how brilliant, don’t have a monopoly on brains. Use yours.
Back in the ‘stone age’, we in engineering had to use a slide rule to solve problems. There were two things that you knew about the results. Using a slide rule will give an approximate answer accurate to maybe… three places. And the other was that to get the answer, you had to have a rough idea of how big that answer was going to be: i.e. One thousandth, one hundred, one thousand, ten thousand,…etc.
I was always amused to watch the younger engineers look at their computer answers and see that they got an exact answer out to sixteen decimal places. Therefore it must be accurate.
Our faith in computers is almost magical and having done my own programming, I know that I was just the man behind the curtain making all that smoke and mirrors.
Thank you, Dr. Droz, for a well thought out article in simple terms.
The computer models of climate are in fact excellent teaching tools. With that in mind, it should be illegal to let politicians get a hold of them and use them as fact, or to hit emotional hot buttons.
Our current problem is that the models are at the core of “evidence for global warming”, and at the core of the models is a small group of parameters including a key one for CO2′s effect on climate.
We are not getting honest updates on why these models didn’t handle the last decade of non-warming at all. We’re not hearing modelers, or those who advocate public policy based on models, say “We were wrong”. And so when the public argument becomes false, the real argument goes under the table.
Climategate. RussiaGate. Wang’s falsifying Chinese data.
Spendulus
We are attempting to model chaos and complexity. It is highly arrogant of us to believe that we understand climate dynamics when we barely understand the basics of the system. Models are simply research tools to help understand simple relationships over short term periods. The same foolish thinking contributed to the financial crisis; over reliance on financial and derivative models, the use of these models to shape critical decisions.
Fool me once, shame on you. Fool me twice, shame on me!
This reminds me of a time when I was in undergraduate school. My professor in one class asked one of my fellow students questions pertaining to the reading assignment. It became clear very quickly that he had not read the material.
The professor then stated ” Mr. D, you are a model student! –and class, if you don’t know what a model is, just look it up in the dictionary. And it will tell you that a model is a small imitation of a real thing.”
Time once was that anyone calling himself an “engineer” without a PE license was a violation of law. It’s still on the books in many states, but not enforced. Requiring PE licensure of software engineers would be a good first step, not because it proves them competent, but because it gives them something to lose.
Perhaps we need to start licensing scientists who do work that involves public policy for the same reason.
Actually, we can. We should be able to go back in time and validate the model using the 20th century. That’s not the real problem with GCMs. The GCM problem (not the same thing as the “statistical models” used with the hockey sticks) is that they’re chaotic in the broad sense of the term. That makes them mathematically unstable, and creates a whole series of issues that the modelers would prefer to not talk about.
But the real proof is, can they forecast the 20th century from 1900 without using any information from the 20th century other than the “forcing”?
When they can do that, I’ll believe that the IPCC projections to 2100 are something more than the product of an ouija board.
For those interested, one anecdote of climate modeling started with a guy named Edward Lorenz. His attempts to use a computer (analog computer, it was 1961) showed him that such modeling was chaotic (the mathematical kind of chaotic). He showed that weather prediction follows the concept of “sensitivity to initial conditions”, which today we call the “butterfly effect.” Lorenz found that weather cannot be predicted with any real accuracy beyond a short time period (like a week), because there are so many variables, and the only way to accurately predict is to know all the variables at all times, even into the past.
Read about it here: http://en.wikipedia.org/wiki/Chaos_theory
I ask every “warmer” out there: What science is there behind global warming? They always mention “scientific” studies, but when I point out that those are computer models and not true science, they always come back with faith, not science. That seems a bit scarier, as zealots often ignore reason and logic to the detriment of others.
Re:
16. Oligonicella:
In writing a computer program of any complexity, there are literally hundreds of assumptions made. The computer programmer does not reveal all these to his customer, for much the same reasons that an accountant does not tell his client all of the assumptions made in preparing a tax return. He goes over a few of the more basic items, and then says “sign here.”
Well, you would have been canned where I’ve worked. That is *not* how a professional analyst/programmer behaves. Not only is every ‘assumption’ documented and tested, they’re gone over with the client. Same with taxes. I went over each line with my tax preparer and every assumption was explained.
It *is*, however, exactly how the scientist/programming-dabblers have behaved. The unprofessionalism exhibited is breath-taking in its scope. Not only are human assumptions codified into ‘truths’, but the work is rife with fabrication and downright shoddy logic.
I have been a professional programmer/ Systems Analyst for over 30 years and Oligonicella with many others have it basically right.
In analysis we try to get as accurate description of what the customer is trying to model as possible but it is a best close and at worse not even close. Usually it is somewhere in between. In a good shop we document and try to test everything in a fairly exhaustive fashion. We still miss things because you can not program in everything or test it else you would never, ever finish.
30 years ago or so it was generally thought that 70% of all application efforts were failures because they did not meet the customers needs or were unusable. In the industry we still claim the same number, though I suspect that is more due to companies unwilling to spend the money to do it right rather than the fault of the programmers.
The final part of what Oligonicella says is key; these people are not professional programmers and designers. It is like the guy on TV: I am not a real doctor but I play one on TV. They play at being programmers, though I have to say some of those who claim to be professionbals also play at being programmers
Modelling is extremely complex and requires many rules and conditions to be understood and incorporated correctly. If the model can not replicate the past it is junked and another effort is made. Here it was; hey it does not match the match the past oh, well!
This piece betrays some fundamental misunderstandings about what programming is, and how simulations are used in science.
Good programmers do reason about all possible combinations of events. Even though one cannot enumerate all the possible ways a program can execute, logic and other tools of mathematics can be used to construct programs that provably work correctly on all possible inputs. In practice programmers do not try to prove their programs correct, and programmers do make mistakes, but the principles still pertain.
Further, there is nothing really different about what is done with computers than what is done (or was done) by scientists without computers. Computers just allow scientific models to be tested more efficiently, quickly, and thoroughly.
Of course, no simulation can perfectly model the real world. That is never the point. Scientific experiments and simulations must be _validated_ to show that they are sufficiently accurate to be predictive of the real world. The mistake of the CRU researchers, if there is one, is that their validation methodology is unconvincing to a lay person or even to a scientist outside the field. That doesn’t mean their conclusions are wrong.
The key is that their validation should be convincing to the lay person if it is to be used for political purposes, i.e., the pursuit of far-reaching and expensive global warming remedies.
Mr. Obama:
The following letter has already been mailed to the White House. PLEASE – anyone who cares about our country and reads this – PLEASE copy it and spread it as far and as wide as possible. I’m taking a stand.
I’d love to begin this letter with ‘allow me to introduce myself’, but I’m obviously not that stupid. If I had, your unionist thugs, Black Panthers, and ACORN operatives – or should we just refer to the whole collection of them as your ‘internal military force’? – would be outside my house before I’d even had the chance to swallow my corn flakes the next morning. Suffice it to say, however, that by the time you’re reading this letter, it will have begun to go viral on the internet. I just love the internet.
See, what I have begun to stridently object to is your administration’s tactic of diversion and crisis, say, like, holding a vote on Cloture for healthcare at 1 am. I’m not fooled, and neither are millions of people across this country. Whenever one particular issue is at the alleged ‘forefront’ of the news (i.e., ClimateGate), I now see that your people are already onto the next one or two issues. The American People are always one step behind. I’ve decided, therefore, to move on this letter, because, quite simply, it’s something that your people were not expecting. There’s another word for this…what is it?…oh, yes…let’s call it a ‘politically man-caused disaster’.
Like many common sense, freedom-loving Americans, I don’t necessarily enjoy paying taxes. Nobody does. That being said, however, most common-sense Americans understand the necessity of paying taxes…after all, the government (a limited one, as laid out in the Constitution) has to be able to function. Indeed, even the true revolutionaries of 1776 had no issue with paying taxes per se; they simply objected to having taxes that they had never condoned heaped upon them without being allowed even one representative in the British Parliament. Ironically, one might put forth the argument that the people once again do not have a single representative. The colonists rebelled not over taxes – they rebelled over the point of being forced to live under the dictates of a soft tyranny.
With the advent of progressivism, communism, and Nazism at the beginning of the 20th century – no doubt in ideological reaction to the effects of the Industrial Revolution – we have witnessed (around the globe, not just in America) the drive for bigger and bigger government by those who have vested interests in power and wealth. These ‘players’ attempt to ingratiate themselves to the lower classes and the portion of the middle classes that are ignorant enough to fall for the rhetoric – either polished or emotional – that lambasts the ‘wealthy’, conveniently omitting that they themselves are also a part of the very class that they’re attacking. In its own way, it’s quite intelligent. Exploiting the sensibilities of the disaffected in populist outrage has always been the key to gaining and maintaining power, because, quite simply, there’s a lot more of those people than there are very successful, wealthy people.
Which, of course, leads my discussion to the current American political administration. Despite you promises of transparency, refusing to listen to the whims of lobbyists, and making sure to appoint people that have nothing to hide, you have surrounded yourself with a collection of outrageous tax cheats, no doubt expecting that the American populace is much too dumb to pick up on any of this, and, even if they do, who cares? We’re certainly well below your superior intellect. Don’t make me laugh.
Guess again, big fella.
Despite surrounding yourself with the likes of Tim Geithner, Tom Daschle, and Charlie Rangel (WRITING OUR TAX CODE!!), you expect that the people of this country – more specifically, the common-sense, conservative middle class – to fork over more and more of the wealth for which we work to feed our families so that you can pay for your potential centerpieces – universal health care (which any educated person can see through), cap & trade (based on a fraud), and amnesty (people who will be paying zero income taxes). You’re not really this stupid, are you? I don’t know…based on everything that I’m seeing these days, maybe you are.
Consequently, President Obama – if, in fact, you’re truly the president, which would open up a whole new ‘taxation without representation’ can of worms – I’ve decided to change my income tax filing so that I claim as many exemptions as I possibly can, and when income tax time comes, you’re more than welcome to come to get those taxes from me (or at least try), because I will not be paying them until which time you finally make the decision to behave like an adult and to actually lead your people as a president would.
I’m more than willing to spend time in jail, if it ultimately be necessary. Fear not, however…I’ve already taken pre-determined measures to ensure that my stay behind bars will, indeed, be a reasonably short one. If you doubt the veracity of my resolve, then I invite you to test me. I’ve a feeling, though, that there’s going to be so many Americans that follow the same course of action that I doubt you’ll ever discover the author of this letter. Good luck.
I truly regret that you’ve brought the American People and I to this point, but actions have consequences, just as elections do. Remember this – you’re not a god; you’re only a man. I fear NO man.
They say that the pen is mightier than the sword. Well, we’re about to find out. I’m not so bold to think that I’m an especially important person, but I will personally consider this the first shot fired in the Second American Revolution.
“That whenever any form of government becomes destructive to these ends, it is the right of the people to alter or to abolish it, and to institute new government, laying its foundation on such principles and organizing its powers in such form, as to them shall seem most likely to effect their safety and happiness. Prudence, indeed, will dictate that governments long established should not be changed for light and transient causes; and accordingly all experience hath shown that mankind are more disposed to suffer, while evils are sufferable, than to right themselves by abolishing the forms to which they are accustomed. But when a long train of abuses and usurpations, pursuing invariably the same object evinces a design to reduce them under absolute despotism, it is their right, it is their duty, to throw off such government, and to provide new guards for their future security.”
Until we meet at my front door,
Yours Very Sincerely,
Anonymous
“There are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns – the ones we don’t know we don’t know.”
Donald Rumsfeld
The last is where computer models fail because all programers can do is “fiddle about” with the first two to get a result.
Part of simulation / model development and validation (for major aerospace developments) involves extensive model peer reviews (of model / code documentation and actual inspection of code). We also have peer reviews of the mathematical model by experts on the various phenomenon represented in the model (e.g., aerodynamics, communications, inertial sensors, propulsion, etc.). Our models are eventually trusted by our engineers (both overall systems and domain experts) and customers as “reasonable – good enough” (there’s no such thing as a perfect model) representations of reality for its intended use. These climatologists seemed to be pretty closed up private shops that did not want outside prying eyes. They likely did not have a seasoned software engineer doing the coding but a graduate student who knew basic coding (e.g., FORTRAN). What floors me about this is not that they had some graduate student hacking together their models. This probably happens all the time in research facilities (cost effectiveness). What floors me the most is that they did not make their models available for peer review to build the confidence of the outside community in their model integrity / fidelity / usefulness but wanted us to just trust them using their undocumented / unkown / unvalidated models and its predictions to tax the most productive nations of the world to the tune of trillions of dollars.
Nice article. Here’s my experience after 10+ years with computer models:
http://wrauny.blogspot.com/2009/02/global-warming-and-computer-models.html
Well-written and well-thought-out, BUT…
Of all these factors, the one that has caused the problem is #8, the human interpretation on the part of the “crusading” scientist and media. It allows all the other factors to run riot, which in this case is exactly what was desired, because it is the best way to lay a confusing trail of deception on the way to a pre-ordained conclusion. String enough flawed assumptions together, and you can prove anything that you set out to prove.
Sometimes (joking) I tell my younger hot shot engineers to reduce gravity if it will make our simulation performance predictions look better or change air density vs. altitude (so we can show better performance at higher alitudes) – please nobody take this seriously / I’m joking. Actually I always instruct my younger hot shot engineers to always assume there are hidden / unknown bugs or flaws in the simulation and keep checking and rechecking (never assume it’s perfect). I always worry about finding out about the one major flaw / defect / bug in a model / simulation after a major flight failure (oops – too late).
We are now in the age of world class medecin men like Gore,full of bovine manure,telling the world what to expect.He knows one thing for sure, that he is positioned to reap hoards of money.In the last few years,he has done just that with the Co2 pro
Historians tell,with actual proof,that the world goes through cycles of cold and heat and life manages to keep going.Now the PTBarnums of the world are trying to sell us that they are going to save us.The cost, just 2 or 3 sets of zeros past a trillion.
While I agree with your premise (models are imprecise), your conclusion (peer review for models) and several of your assertions, you argument contains too many proofs-by-assertion (intangibles cannot be modeled), misrepresetations (program decisions are strictly binary and uncoupled) and confusions (a model of natural processes is comparable to a model of a bureaucracy). Sorry, not at all convincing.
When it comes to supporting a “cause”:
- Belief and feeling trump facts and data.
- Logic and truth do not apply
- Morality and ethics are irrelevant.
- Cost is no object.
There are many excellent points made here, and they will only impact or reinforce the viewpoint of those who are willing to think about AGM as an open scientific question. For the True Believers in catastrophic global warming, no counter-argument will even be heard, much less acted upon or allowed to influence rigidly-held opinions. The power and money seekers will huddle with the True Believers as long as they are achieving their particular aims, then they will look for another vehicle.
That AGW is a scam is obvious. That the only people who believe it are those that stand to make millions from it and green-reds is also obvious. But I do have a question for the scientists among you. Please don’t answer unless you know the answer: We are told that CO2 is currently causing practically all the warming that it can because it absorbs logarithmically less energy as its concentration in the atmosphere increases. So, I guess a run-away greenhouse effect is not possible here on Earth. If that is the case, what is it about Venus that is different from Earth that makes the run-away greenhouse effect possible?
My comments / points about what is simulation / model development common practice in the aerospace industry are made because I have always been a denier (and evil Capitalist) since these climatologist (i.e., “real” scientists) have models that can’t even adequately predict / model weather that occurred over the past few years (or centuries). Thus, based on the requirements / standards / experiences in my industry they have zero credibility when it comes to trusting any predictions made by these models in the future (yet alone a hundred years into the future). This is common sense and practice is my industry yet these politicians call us seasoned veteran skeptics deniers / flat earthers on the payroll of big oil.
#28 Uniblogger: Excellent points, especially this:
“I was always amused to watch the younger engineers look at their computer answers and see that they got an exact answer out to sixteen decimal places. Therefore it must be accurate.”
It is a fundamental rule that any scientific calculation or prediction can be no more precise than the measure being used. With global warming, we see graphs showing historical temperature variations, based completely on proxy data, moving up or down by .5 degrees. In order for those calculations to be valid, the proxy data (tree rings and ice cores, mainly) would have to be reliable and valid to an accuracy of less than .5 degrees.
Somehow, I doubt that is the case.
#42,47
Chris the Engineer checks and rechecks the results of simulation, because Chris, or someone very like Chris, often assumes a legal responsibility when signing off on work done and predictions made. In other words, there is often more than just a risk to reputation at stake. I wonder if there is a lesson here?
This whole climate “crisis” is nothing more than a hoax. Just like the health scare fraud, the political left cares nothing for the average citizen. They only care for controlling we the people, like somehow we are too stupid to know better. I find the left highly insulting. It is beyond talk. It is time for action. Dont waste time writing letters or e-mailing your “reprersentative”. They are not listening, they listen only to ACORN,SEIU and the idiots Like Michael Mann and Phil Jones. Its time for a John Galt movement. Maybe even a revoluttion. Jefferson once said a little revolution isnt always a bad thing. Time to start proving him correct.
Great point Paul from Hamburg. Accuracy is based on the least accurate measure used. Using proxies that suggest certain temperatures are highly inaccurate. My guess would be in decrees or 10′s of degrees.
One of the nastiest arguments the AGW types make is basically that they know their sims suck, but they “may” be showing something, and that something is Bad, therefore we should destroy the world economy to deal with the possible Badness due to the “precautionary principle”.
They then argue that this is the “conservative” option, and advocating anything else is being “in denial”.
It’s a classic case of Orwell’s quote that some ideas are so incredibly dumb that only really smart people could believe in them.
Karl Popper stated it clearly and briefly: “The strength of science is the ability to test falsification”.
Anyone who resists or protests efforts to test his hypotheses for error or falsification is not a true scientist. It is the core principle of the discipline.
The postulations of the AGW are little better than the prophecies based on examination of chickens’ guts by the seers of ancient Rome.
Climategate:
The Perils of Global Warming Models
Let us ask the world’s smartest scientist,
Stephen Wolfram, about mathematical models.
(paraphrasing recklessly):
They never simulate the real-world phenomena
exactly, they often lose critical features,
some phenomena are themselves the minimal model
of their behavior, and humans are not smart
enough to abstract the relevant features of
complex phenomena, let alone translate them
into computer code.
John, you’ve hit the nail on the head. And if you read the Climategate emails themselves (click on my name to see my excerpts from them), you’ll see that the programming is a joke. For example, Phil Jones writes to Mike Mann:
“I’ll tell him [McIntyre] this, but that’s all — no [computer programs]. If I can find it, it is likely to be hundreds of lines of uncommented fortran! I recall the program did a lot more that just average the series. I know why he can’t replicate the results early on — it is because there was a variance correction for fewer series.”
Hundreds of lines of uncommented FORTRAN? Fiddling arbitrarily with the variances so that the results are irreproducible?
We left the future of the planet to these cowboys?
I’m ashamed to be a scientist right now.
Well, a few things that we DO know with absolute certainty about climate modeling and the long range predictions based thereon, at least at the current state of the art.: 1. The predictions made by the models are unassailable. 2. The modelers neither require nor desire empirical data in making their predictions. and 3. The Catastrophic Climate Change (nee Global Warming) predicted by the models has been confirmed to be anthropogenic.
56: here’s my state-of-the-art climate sim:
#!/bin/sh
for i in INPUT_DATA_FILES
do
cat $i > /dev/null
done
echo “we’re doomed!”
exit 1
# 28 and #48, This is what we call the “measure with a micrometer, mark with chalk and cut with a hatchet” mentality. When the “adjustments” to the data are larger than the trend being shown, you have created the trend. I have been saying here for months that models prove nothing. The variables in our climate can not be quantified.
Thank you for the many kind and supportive comments.
Anyone interested in a scientific perspective on our energy situation might find my online presentation of interest “EnergyPresentation.Info”. Feel free to email me questions, comments or suggestions: “aaprjohn@northnet.org.”
1.) We dont know if we are living in a warming climate. 2.) If we are living in a warming climate then what do we do about it? We cannot control even small weather events like rain storms or tornadoes or hurricanes. How do we think we are going to control the earth’s weather. 3.) It is easy to find records of the earth’s constant warm/cold swings. Hundreds of them are written in the sides of small cliffs and upthrusts. 4.) Was the last warm spell in about 1500 AD caused by exhaust fumes and CO2? 5.) CO2 is a minor greenhouse gas. Water vapor has a far stonger effect. How can we control water vapor? Maybe put a coat of oil on all of our oceans, rivers and lakes – and glasses of beer? 6.) The silly people at the US EPA want to fine those who emit CO2. Now they want us to stop breathing. 7.) What happens when CO2 levels rise? Trees grow faster. Grass too. They adjust to consume the increase.
My two cents:
It seems that it all boils down to how many degrees of temperature in the earth atmosphere a doubling of the CO2 will add to all other natural factors involved (not that some of the CO2 isn’t natural.) The added CO2 will obviously have some warming effect and it will, as far as we know, continue to increase indefinitely, or until we have burned all the fossil fuels that contribute CO2 beyond what natural effects can absorb. Added CO2 on top of a Maunder Minimum, is probably not a big deal. On top of an El Nino and sunspot maximum’s, it seems safe to say that things will be hotter than they have been since 1998, the hottest year world-wide (and an El Nino year) since temperatures have been recorded.
I think that the AGW scientists are sincere and there is a lot of impressive data out there, despite all the climategate issues. But the most honest of them will acknowledge that they don’t know for sure exactly how much a doubling of the CO2 will increase global temperatures. Most insist on a minimum of 2 degrees (some of which has already happened, of course) but others see as high as six. Obviously natural factors such as planet wobble, volcanoes, and El Nino’s/Nina’s can intervene and push things down or up, but the CO2 contribution is generally separate from that.
Personally, I think that it is time to begin the process of building a number of nuclear power plants.
Again, I have to wonder why climate models can’t be backtested, applying the data from say 4000 years ago to a predictive model, then comparing what actually happened.
Dwight,
Some of us thought we ought to be building nuclear power stations thirty or more years ago. You come to the realization late.
If the AGW scientists were sincere they would not be fanatically opposed to others using or testing their data. That’s the way science is supposed to work, in the open and with sceptical criticism.
The same people (sometimes exactly the same, like Stephen Schneider) were telling us in the 70s that we had to destroy the world economy because it was the only way to stop the new Ice Age that they were sure, certain, was coming on. Their suggested responses were much the same as now suggested for AGW. It was a crock, of course. When the new hysteria appeared I remembered the earlier panic, and that was one of my reasons for scepticism. They fooled me once, I was not going to believe again unless they proved it with data. The best data I found gave insignificant warming over a century (.07 C per decade). I knew about the Medieval Warm Period and the Little Ice Age already, so I knew there were historical climate changes, in both directions. When the “hockey stick” graph came out with neither in evidence I knew they were lying to us.
Models can be fun. Heck I used to build them. The main features of the AWG Hypothesis are quite interesting and I always supported more research into it. The real problems began when Political Types and Huge Egos (M. Mann, J. Hanson, and P. Jones) combined. The original papers had the usual Scientific Range of errors and uncertainty about the temperature record. They also displayed a huge uncertainty as to the confidence displayed by the modelers. Then the politicians at the UN IPPC got involved and said we cannot use the papers as written. We need certainty and accuracy of prediction. By this time the palyers had risen from obscurity to become rock stars in the percieved Scientific World and mountains of funding were pouring in. They allowed themselves to become seduced and there went thier objectivity.
I is sad to say but we are going to see the destruction of a few fine careers because they let thier egos and vanity get in the way of objectivity. This has almost always been the case when Politics tries and then gets results that Science really cannot give. They may have made a few mistakes at the begining. They may have made a few wrong assumtions such as High Sensitivity when we really do not know yet. The truth is they are so deep in it now they really cannot see a way out. Pride goes before the fall and the fall will be long and hard. I think the rest of the Physical Scientists thought that these were honest guys. It is well known that Carbon Dioxide is a Green House Gas. Man has contributed Carbon Dioxide to the atmosphere so Man Made Global Warming is a reality. The question that is not asked in a meaningfull way is how much warming has occured due to mans influence. Is warming really bad and how much warming is good? Where are we going to get all that Carbon Dioxide anyway in order to double the current cooncentration. Currently Carbon Dioxide makes up about 385 Parts per Million of our atmosphere. I have read that about 15% to 30% of this is caused by man. The Alarmists projectons are based on a doubleing of the Carbon Dioxide. That means in the simplest terms not taking into account any latency, because if we do it actually takes more Carbon Dioxide to get to this figure, that we need to be burning at least 6 to 10 times more hydrocarbons than at present to reach these figures. This is an assumption made by the models. Keep in mind that we would be increasing our consumption at ever increasing rates for 100 years to get to the final consumption figures. They do not take into account any new developements that a robust economy is likely to develope as oil and Natural Gas become more expensive such as a switch to Hydrogen or some other technology not even thought of.
This is where I knew it had to be pure politics driving the desired outcomes and it could not be science. One assumption piled on top of another assumption and still they do not answer a basic question. Where is all this carbon going to come from any way? If the projected consumption figures are correct there is nothing to worry about. We would have run out of Hydrocarbons by 2075 anyway.
RichB213 wrote:
Currently Carbon Dioxide makes up about 385 Parts per Million of our atmosphere. I have read that about 15% to 30% of this is caused by man. The Alarmists projectons are based on a doubleing of the Carbon Dioxide. That means in the simplest terms not taking into account any latency, because if we do it actually takes more Carbon Dioxide to get to this figure, that we need to be burning at least 6 to 10 times more hydrocarbons than at present to reach these figures. This is an assumption made by the models.
——–
Rich, Can you say more about this. What do you take as the starting point to figure the doubling? Also, I assume that there is enough carbon out there for us to find and burn to double in the atmosphere, since the CO2 rate has been that high at times in the past, right?
To #63 Michael Lonie Have you taken the time to look at the the literature and data that has been accumulated. I don’t believe the hockey stick per se, either, but there is a thread on Real Climate devoted to why the hockey stick does not matter. Basically, if we ave our current CO2 ON TOP of MWP warming, we’re screwed.
Once, I had the mind-set that if I could find a few flaws in the AGW scenario, or the people pushing it, then I could just dismiss the whole thing, and this is certainly what anyone who is philosophically opposed to taxes, government intervention of foreign aid to emerging nations to save their rain forests etc does. You can tell that more than half the comments here are purely agenda-driven. Would we agree that we need to show SOME concern about the amount of CO2 that humans are pumping into the atmosphere…or not? Our individual mileage may vary.
We know that almost all the glaciers in the northern hemisphere are in retreat and are likely to disappear soon. We may, or may not be able to do anything about that. I am not prepared to support cap and trade tomorrow, but I am also not prepared to dismiss the whole thing as a complete scam….unless or until the glaciers start coming back. Subsidies will be necessary, I’m sure to get going on nuclear; start-up costs will be huge, so we will be paying one way or another.
Hansen says that 2010 could be the year for record warmth, since there is an El Nino, but it seems likely to this schlep that the sunspots would have to return for that to happen.
Realistically, it will take such a year to offset the climategate brouha.
One more thing regarding the doubling and the “natural” CO2. People here have insisted that it is a lagging indicator, but that also indicates that increased warming also causes a build-up of the natural CO2, which would be added to the man-made (caused) contributions.
Now we know the truth.
Up yours Greenpeace!
All those Sperm whale spouts and Polar Bears are destroying the environment.
Quit feeding your dog/cat and save the world!
Brian C #62 — Again, I have to wonder why climate models can’t be backtested, applying the data from say 4000 years ago to a predictive model, then comparing what actually happened.
The author of this article doesn’t seem to understand climate models and comparing one to bill collecting is just ridiculous. No wonder you don’t understand.
The answer to your question is that models are designed to look for trends based on looking at “givens.” Trends. Not a play by play. One cannot predict when the solar output will change as it has recently; one cannot predict vulcanism; one cannot predict cosmic ray bursts which affect cloudiness (and warming/cooling as a result); one cannot predict which human culture will make a desert out of good land due to bad farming practice (the mayans for one) and affect the local climate.
Rather, models are used as a mechanism to understand the effect of CO2 and/or other GHG emission over time; the central question always revolves around whether or not CO2 increase results in rising temps and if so how much. The thrust here is to see if other factors can act as an amplifier or are other factors mitigators (negative amplifier.) Putting the model to 1000 CE and letting it run doesn’t really accomplish what the model is intended to do, nor will it give you an answer that you can understand.
Since the models are designed to examine the effects of CO2 emissions it ought to come as no surprise that the results seem to have a great deal to do with CO2.
The problem isn’t the models. It’s the politics.
If you look closely at the IPCC scenario stuff (A1 B1 etc), especially in light of the climategate data, you’ll see why these models have come under as much fire as they have. Essentially they show that a prosperous and happy mankind engaged in capitalist pursuits raises CO2 emission by one level, but dialing back CO2 emission to “sustainable” (i.e. survival rather than prospering) causes CO2 output to flatten just a bit. (This is assuming of course that a 2 or 3 degrees temp rise is “bad.” Many of us don’t buy it.)
The anti-capitalist forces and the Malthusians and the eco-terrorists who want us all to live in the stone age are using these models to have a field day. Temp rise is cast as “bad” (rather than simply not applying morals values to temps.) Then they equate this “bad” with the oil companies (and/or anything else that represents corporate success) since corporations are evil in their universe.
Meanwhile assuming models are true enough (they seem to be) they also show that the proper answer to a lot of it is to implement massive programs of nuclear power generation worldwide which will shut off massive amounts of CO2 growth (energy production.) So the plus side here is that we can assume the models are spot on and then use these to promote jobs and (nuclear) energy production; wealth and energy production are things that go hand in hand.
The problem isn’t the models. It’s the Malthusians.
#64 richB313 — Currently Carbon Dioxide makes up about 385 Parts per Million of our atmosphere. I have read that about 15% to 30% of this is caused by man.
This is the C12/C13 isotope ratio. Fossil fuel burning causes emission or appearance of one isotope; the other is natural. (And even this is not set in stone. There are exceptions.)
The interesting thing here is that we know the oceans act as sinks and emitters. The oceans cool, they soak up CO2; they warm, they spit it back out. Oceans don’t have C12/C13 detectors; they emit what they soaked up last. (They don’t emit what they soaked up first and cycle it through; they emit the last thing they got.) This gets remixed in the atmosphere so that detectors can do their magic.
That means that a good deal of the “man’s output” percentage you’re speaking of represents man’s cumulative output of fossil fuels, not some sort of ongoing rate. As such this messes with the estimates just a wee bit. I like it when the warmists point to isotopes.
Meanwhile there is no real agreement of how long CO2 stays in the atmosphere. Estimates from credible sources range from 12 to 150 years.
As I hold a PHD in Badmitten I have studied the emails and some of the techniques used I have determined that
Global Warming(Climate Change) is a FARCE!You may see some of my students in action by clicking the link below!
http://www.youtube.com/watch?v=_yKlcug6Zxg
Al Gore is now the United SNAKE OIL SALEMAN of AMERICA based upon these idiots claims that “The Science Is Settled.” which has now been determined by a “Criminal Racketeering Enterprise.” Let the RICO Prosecutions begin NOW!
This may be the article to which Rich referred, casting doubt on whether we can really put out enough CO2 into the atmosphere to achieve a doubling:
http://europe.theoildrum.com/node/5084
for the true-blue skeptics here, I will rely on you to read it more carefully than I did. (It’s long, but may be really important.)
Let’s put global atmospheric modeling into perspective: the effects from all sources of input energy are within about a single percentage point of the sun’s average insolation. In essence, researchers are working at an order of magnitude less than the standard error of the sun’s insolation with each of their major focuses on lesser effects on global temperature.
The essential problem is that any anthropological effect on global temperature is smaller than the standard error of the sun’s temperature effect. The effects of secondary and tertiary physical processes affecting global temperature are simply statistical noise within the statistical noise of the sun’s insolation.
Researches cannot get a grip on what’s up with the sun, how can they get a grip on secondary and tertiary anthropological effects?
We have no physical models that can begin to reliably reproduce (not even invoking “predict”) short term temperature variability in solar insolation without diverging to snowball earth or hell on earth. There are major deficiencies in predicting cloud cover, atmospheric humidity, and many other small effects, in modeling the impact of the sun on global temperature.
Including the supposed processes with respect to atmospheric CO2 – and let’s not forget CH4 – is a a fool’s mission since we cannot yet model the basis for earth temperature vs solar insolation.
The central issue is NOT the divergence of theory and programming, it is the deficiencies in basic theories and proofs. Programming is so far down the list as to be inconsequential.
>Droz: “Is it really possible to accurately convert complex real-world
>situations into ones and zeros”
That’s exactly what religion purports to accomplish! Cast all reality into dualism poles, ban one side and fully endorse the other.
Real world models CAN be precise….at least to machine precision. That’s for software that models basen on laws of physics. Just as there are laws of physics there are also laws governing human behavior. The founding fathers even understood this.
Regardless of what you think about weather modelling, every utensil you use, every machine, your car, the bus train or plane you fly in, the ship you consumer goods arrived in – all of them began life in computerised simulations of wind- or water-flow, of fuel-consumption, or break resistence, or drop resistance. This is the state of the art for technologically advanced societies. The new super-planes are modelled extensively to choose between thousands of variations before being committed to even scale trials. Modelling of nuclear explosions was so accurate when compared to underground testing that the testing was abandoned. Plastic compounds are modelled and their propertied examined interactively in software before vast petrochemical plants are committed to fabricating them.
Everyone sees these in action all the time in the extraordinary computer-generated animation in movies, tv shows, commercials, interactive computer games. Models of physical reality are used to permit the automatic generation of walking/dropping/burning/flying/growing/shooting. The waves in blockbuster movies such as Titanic or Independence day or 2012 or Avatar are the result of sophisticated computer modelling of waterflow.
Your computer-controlled car runs mini-models for the kinds of decisions that 30 years ago were performed by the weights and springs in the distributor plate and the jet system in the carburettor – and the point here is that they were a primitive form of modelling/predicting mechanism. Computer models take the laws of physics that have been established since the work of the ancient natural philosophers from Thales to Aryabhata (who predicted eclipses by modelling the solar system). The trajectories of ballistic projectiles were modelled by mathematics on a day-to-day basis by gunners using primitive modelling mechanisms. The accuracy of such modelling led to the decline of castle-building as a form of military defense. 3000 years of increasingly sophisticated physical mathematics lie behind computerised modelling, not ad-hoc decisions made by solitary programmers. The thing is, that the assumptions not agreed on in climate modelling are in the real-world climate mathematics, not in decisions made on a random basis by unmonitored programmers.
The picture you paint of how computers store information for processing, and of the algorithms used to process it, is also wrong. Fuzzy neural networks were developed exactly to represent complex real-world decision-making and -taking. Zahdi developed fuzzy sets for such systems in the 1970s, and they are employed in every washing machine, microwave and fridge. When you press “easy defrost, 3 kilograms” on your microwave such a process of modelling is being employed. When you use hand-writing recognition on your smart phone, the processor works on hidden Markov models and neural networks.
More importantly, the kind of self-taught computer programmer you speak of is not the kind who could get a job as a professional programmer, let alone a programmer employed to do modelling at an institution where people were applying for jobs to a selection panel. The idea that someone who is self-taught could get a job up against others with qualifications is plain wrong. As someone who has gone through the computer programming courses at a university, and know how very hard you have to work to pass the assignments and exams, I can assure you that the kinds of analysis that are done daily by working programmers are far more complicated than you show. The argument you are making is argument-by-story-telling: this is a story that might happen, therefore it is a story of how it must happen.
Go and look at the special interest groups of the ACM or IEEE who work on modelling, look at the papers for modelling in Arxiv or Medusa which are available. They are vastly more complex and sophisticated than the programmer working alone and making guesses.
There is a larger methodological question about the way computer models are used in climate science. The notion, held by some climate scientists, that computer simulations can test the validity of a hypothesis is not scientifically sound. The computer simulation IS the hypothesis, and it must be validated against real world data. Without adequate data input, the use of such models constitutes circular reasoning.
See http://scienceandpublicpolicy.org/images/stories/papers/originals/A_Glimpse_Inside_the_Global_Warming_Controversey.pdf
I have read some papers claiming to use computer modeling to “prove” aspects of global warming theory. Despite their being implemented on computers they are little more than a restatement of the modeler’s theory due to their inclusion of factors the model builders thought to be important weighted at the levels the modeler’s believed such variables should carry. These models are not (or only partially) derived from observed data (unlike many artificial intelligence computer models of scientific, engineering, or financial phenomena), nor have they been tested successfully on “out of sample” or new data. The authors seem to have the incredibly naive idea that because their results are the output of a computer they must be true – no awareness of the need for testing is shown.
As someone who has spent many years building successful computer-based models of complex (but not climate) phenomena I am very skeptical that models of noisy data where all the driving forces are arguably not known, where you have only a very small number of good years of data for “predictor” variables, and have completed zero successful out-of-sample tests are worth the cost of the CPU time it takes to run them. Indeed, I saw a paper not long ago that tested the ability of several climate models to predict temperature levels in the years since their inception and they failed miserably.
In the twentieth century we saw the fate of nations decided on whether they got scientific issues right or wrong. The Soviet Union applied Lysenko’s theory of the inheritance of acquired characteristics to its grain production. It went from the breadbasket of Europe to a nation that could not feed itself. This likely contributed to its collapse.
During WWII the US correctly perceived that a nuclear weapon was feasible and would be decisive. We established a crash program to acquire one. Nazi Germany, under the spell of its antisemitism, saw nuclear physics as “Jewish science” due to its association with Einstein and (thank goodness) did not strive actively to produce an atomic bomb.
I am afraid that the US is getting the science of the climate effects (or non-effects, as I believe) of CO2 increase in the atmosphere very wrong. Belief by many in completely untenable computer-based “climate models” is a major part of this error.
@ Boris Smalle:
Making widely encompassing statements is dangerous, and likely to be refuted by the “White Crow” argument.
I can assure you that the aircraft that I fly had no help from computer modelling in its design. It was built in 1966, on a design from Frank Cessna back in the 1930s.
Actually, the waves in “Titantic” were created by air-brushing the individual film cels. No, that’s not a misspelling.
Air-brushing was done because the computer modelling software was not up to the task.
As a self-taught programmer, I get a little annoyed with people who would judge by the number of letters behind the name on a business card.
Do you, by any chance, happen to have the name of the university where Isaac Newton acquired his degree?
Well done! I am a software engineer and understand intimately the limitations of computer models. Your reasoning is well thought out and very relevant to climate science. Again, well done.
Boris Smalle: I’ve worked many years in the auto industry and it’s only been 10 years or so that the tire to road interface (nonlinear) has been successfully modeled (yes, just that one thing). It is ludicrous to believe that extremely complicated climate models with inadequate/incomplete data and poorly modeled nonlinearities are anything more than good guesses. I’ll be more credulous when model output corresponds well with real world data.