Weekend Open Thread – call for essays

Canadians say, booey-heh

On Computer Programs, Computer Models, the Scientific Method, and Global Warming
Anyone who has been following the current global warming science knows that there is relationship among all of the issues in this article’s title. I will attempt to make the relationships more clear.
First, my credentials. I am a 1985 graduate of West Point with a degree in computer science. I have been working in the computer field since I left the Service, which gives me more than 20 years of experience in IT. Currently I am working as an Oracle DBA, as I have been since 1999.
Through our history, as science developed and grew more complex, the need for complex equations also developed. As equations get more complex, the chance of error and the time required to calculate the answers for the equations grew. Particularly in such applications as projectile motion, the need for rapid, correct answers to complex equations kept increasing. It was this need for quick, correct answers to difficult equations that led to the development of computers and computer programs. A computer program will always produce the same answer to the same question. Note that the above description may not be the history of computers that you have read, but it contains the basics of computer development in a nutshell.
This leads us to computer models. A computer model is simply an equation or series of equations used to simulate a real world condition in the form of a computer program, sometimes mistakenly called a software program. Computer models are used because a given set of data will always result in the same answer, and it will be the correct answer if the program is correctly written. Models are used extensively today in projectile motion, orbital mechanics, and other scientific and engineering fields.
The most import single item to understand about computer programs is this: They are used to calculate an answer based on the theory of the scientist or engineer is applying. The programmer writes the program so that it complies with that theory. For example, the basic formula for calculating the distance an object will travel in a vacuum is a quadratic equation. This same formula can be used on Earth, even though it is not in vacuum, so the equation isn’t quite right, but it will still get close to the correct answer. So the initial programs were written just to resolve the quadratic equation. As more and more variables are resolved, the equation can be modified and fixed, and the answer gets closer and closer to reality. For example atmospheric pressure and air temperatures are significant factors that are figured into ballistics computers these days, but they are not part of that initial simple quadratic equation. Even the initial equation isn’t quite right because the Earth’s gravity field isn’t consistent everywhere because a planet and its moon are not perfect spheres. But the key point to this is that the scientist has a theory on how something works, and uses a program to determine what the answer would be if his theory is correct. In ballistics, the theory was a quadratic equation, which was close, but there were other items to factor in that reflected the real world. Thus a computer model is actually the mathematical version of the scientific theory, and as such it cannot prove a theory is correct. The model in effect is the theory, so at best it can provide evidence that the theory is correct. It can never prove it is correct even though it can provide very good evidence that a theory is incorrect.
Global Warming/Climate Change theory states (simply put) that as CO2 in the atmosphere increases, temperature increases. When you see the output of a computer model, the model will show that when the input of CO2 increases, the temperature increases. The output of these computer models prove nothing other than the program was written properly. It does not prove the theory is correct. It doesn’t even provide evidence for the theory unless the output closely tracks the real world temperature.
This brings us to the models used in global warming theory. When you see a graph of the predicted warming due to increasing CO2, you will see a single line. What they typically do not tell you is that this single line is almost always an average of approximately 100 different computer models, each run many different times using different sets of numbers. The simple fact that this is an average of 100 models should immediately raise a question in any rational person. If there are 100 models being used, which model is the correct one? Yet each of those models is supposed to represent the same theory. There can be only one correct model for any theory, but we are given the output of over 100 different models as evidence of Global Warming.
If the calculated distance of travel of an artillery round is 2 miles in our previous example, but the measured distance was half a mile, there is something seriously wrong with either the theory or the execution. Keep in mind that we have actual measurements of temperature dating back to the development of accurate thermometers. Thus, the first step in checking the accuracy of a computer model of temperature is the hindcasting. That is, can you feed it data from the years prior to 2001 and get accurate predictions for 2001-2013? Most of the models used today are dismal failures at this step one.
In global warming theory, only a cursory look is required to compare the output of the global warming models with the actual measurements over the last 15 years to determine that the predicted warming is between two and three times the measured warming. This is very strong evidence that, at the very least, the equations they are using in this theory are very wrong. Yet several times a month, we hear claims about ‘proven’ global warming science or ‘settled science’ or ‘97% consensus’, most recently from our own Secretary of State. In the chart below, the red line is the predictions by the climate models, and the green and blue lines are measurements.
Another problem with the theories is the ‘extreme weather’ claims. Most theories and their models claim that there will be more extreme weather events of increasing severity due to increasing global CO2. It is the analysis (or to be more exact, the lack thereof) of this piece of theory that should make the reasonable person question the competency of those espousing the theory. In the US, and most other countries, we have very good records of extreme weather, going back over 150 years. In Europe, the records are even longer. According to global warming theory and models, human generated CO2 started having an effect around 1940-1950. This means that if extreme weather was on the increase, we should have seen a gradual increase starting at about that same time. However, a check of the records (the extreme weather page at noaa.gov is a good place to look) shows that there is no trend in extreme weather since 1950 (or before). That means no change in the severity or number of hurricanes, tornadoes, floods, cyclones, droughts, blizzards, etc. So in this case, we can compare the model output with current measurements, and determine that once again there is a major problem with the model.
The larger problem here is that these scientists made the claim of increasing extreme weather due to increasing CO2, programmed it into their models, but evidently didn’t know that the evidence they needed was already available to prove or disprove this theory.
Competent scientists would have known that the evidence to disprove this part of the theory was available, and researched it before they published theories that are obviously incorrect. Yet these same scientists are sticking to this part of the theory, along with the rest of the theory, even though it has been disproven. The fact that these ‘scientists’ ignore the fact that their theory has been disproven, and even continue to espouse the theory, is enough to bring the competency of these scientists into serious question.
This leads us to the next topic, what some people have called PlayStation climatology. PlayStation climatology is the result of examining a particular piece of global warming theory and then taking it to its worst case. The drastic result of this input is then examined for real world consequences. A classic example of this is the erroneous scare tactic report that polar bears are in danger due to shrinking arctic ice. The idea is that if all the arctic ice disappears, polar bears will go extinct because they need the ice to survive. Well, first of all the arctic ice is not going to disappear. The temperatures in the arctic regions are still well below the freezing point of sea water. Second, according to global warming theory, arctic ice has been receding since approximately 1977. However, if we do a little research we discover that the polar bear population has been on the increase since approximately 1945, and is currently at an all time high. Yet this claim about polar bears was made in normally reputable s and labeled science.
As we noted above, the climate models aren’t even close on the temperature predictions, which is what they are intended to predict. How could they possibly be accurate on data that is normally used as just another parameter in the program? Other examples of PlayStation climatology include the disappearing sea turtles (they’re not, at least not due to climate change), decreasing trade winds, increasing droughts, increasing floods, and almost any alarmist prediction you have read in the last decade that is related to global warming. All of these come from nothing more than feeding data back into these same incorrect models to get alarming predictions.