What I intend to show in this post is that the raw temperature records are so incredibly bad that there is really no way to correct them. In this post I will introduce you to the state of the raw data.
The way modern science is practiced discoveries are supposed to be repeatable. Joe Blow claims that he has discovered a new phenomenon. Joe publishes his methodology and results in some journal. Sally Skeptical reads the journal and decides to check Joe out, afterall, Joe is a well known blow-hard, at least that is the scuttlebutt at all the conferences. Sally performs the experiment that Joe did and decides that Joe is full of it and then publishes her results claiming that Joe is wrong. Craig Curious is now curious as to who is correct. He repeats the experiment and finds that this one time, Joe is correct. He criticizes Sally's in correct set up. That is how it is supposed to work.
With secular measurements, defined as measurements taken once over a period of time, how can one verify the measurement? With some things, like solar sunspot numbers, one can look at Beryllium-10 (Be-10) which is created by cosmic rays. When the sunspot cycle is active, the sun's magnetic field engulfs the earth and deflects cosmic rays, meaning less Be-10. But, how does one verify the temperature taken in 1926 by a fireman at a Houston, Texas firestation? One can't travel back in time to check to see if he read the thermometer correctly, or even read it at all. So, what is one to do?
One way to double check a thermometer reading from 1926 is to compare it with a nearby town which is in the same environmental setting. Of course one can't do this for a town in a valley compared with a town on top of a mountain, no matter how close they are. So, the place to test the veracity of the temperature record is to compare nearby towns on the Great Plains at similar elevations in similar settings elsewhere. Anyone who lives on the Great Plains knows that, while the temperature might be a few degrees hotter 20 miles away for a day, on average, over the year, it isn't much different 20 miles away than it is here. I need to emphasize that the stations shown below are all from the US Historical Climate Network. What I am showing is the RAW data, not the sanitized and edited data NOAA gives easy access to.
So, let's start with two towns on the flat Texas prairie. Flatonia, Texas and Hallettsville, Texas are two towns that are only 17 miles apart. Here in Texas, it is hot as hell in the summer and mild in the winters. Two locations that close should have very, very similar temperatures. Of course they won't be exactly the same, but the yearly average should be well within a degree. Why do I say this? The average temperature gradient from equator to the pole is .0268 deg F per mile. So, even if these two towns are due north-south, the maximum annual average temperature difference should be no larger than 0.45 deg F. But, with great regularity, ,there is as much as a 4 degree F temperature difference between these two towns.
I must emphasize for the AGW hysteriacs that this is NOT a daily temperature difference. This is an ANNUAL difference. That means that the thermometers are measuring an average of 4 deg F difference throughout the ENTIRE year. This is a tremendous amount of temperature difference.
Here is a plot of the raw temperature difference between these two nearby cities on the Texas prairie.
The data is the raw, unedited data. I prefer this kind of data because there is no bias in it except what is in the raw data. I obtained this data from www.CO2science.org which offers the researcher the actual raw data, untouched by other humans who might or might not have biases.
I want to call everyone's attention to the years 1927-1929, 1934, 1940, 1948-1949, 1955, 1961, 1974-1977, and 1996-1997. In all of these years, the temperature difference between Hallettville and Flatonia were greater than 4 degrees F FOR THE ENTIRE YEAR!!!!
I want to emphasize how huge this annual temperature difference is. It is 8x the January equator to pole gradient for the Northern Hemisphere. The raw data would have us believe that such temperature differences can exist for an entire year, something that I would find ridiculous, but AGW believers might ignore.
So, what do the two stations look like? These pictures are from Anthony Watts web site source
You can see that both of these stations are rural, yet the output from them does NOT verify the accuracy of the other. The correlation coefficient between the raw data of these two stations is .37. That is worse than mere chance.
Let's look at the two stations subtracted from each other.
This kind of variability in the raw data, which reverses direction doesn't make one confident that one is getting the correct data. The other interesting features is that the data doesn't seem to get better with time. 1996-1997 involve a huge temperature difference, and our instruments are supposed to be better by now.
Lets look at the 1974-1977 period in more detail.
Both stations took a drop in 1973 but Flatonia quickly went up in 1974-1977 but Hallettsville stayed COLD. For the next 3 years there was at least a 4 deg F temperature difference. This can't be real. Such a temperature gradient lasting 4 years would cause huge thunderstorms and winds. No such phenomenon were observed. Because of this, we know that the temperature measuring system is flawed.
Now, some of you are saying. this is rare. This is the exception. I can assure you that everywhere I have looked at the raw temperature data this is the norm. Below are some more stations. I will put them into the format of station A minus station B so that the temperature difference over the short distance can be clearly seen. All of the numbers are ANNUAL temperature differences. I say this to try to get the reader to not think it is a DAILY difference. Daily temperature differences can be big, but they can't be constant over the year for nearby stations effectively at the same elevation.
From New York:
I thought this last one is worthy of looking at the temperatures, given that there is a 12 deg F yearly difference in temperature.
Note that between 1959 and 1971, Walden was tremendously colder than Westpoint just a few miles away. This is, of course, stupid, but it is, of course, the raw data and when the climatologists 'fix' this, they are merely guessing at what the temperature of one or the other town is. Guesses are not data; and one can't possibly fix this data to the level of accuracy required to know how much the earth has warmed or cooled. And, I have to emphasize again, these are yearly average temperatures, not daily temperatures.
A twelve degree difference in temperature means that a cubic meter of air at West Point has 15,600 joules more energy than a cubic meter of air at Walden--for the entire year. That is a huge amount of energy, the energy required to lift 1 kg 159 meters, or 520 feet. Energy always flows from high to low energy places. Such temperature differences, if real, would mean year long winds or year long thunderstorms. None were noted in the meteorological record.
Because of all this, we know that the raw data is crap. It isn't capable of being corrected to know what the quantitative measure temperature change.