Sunday, August 30, 2009

How "'Correcting" the Data Heats the Earth

The above chart, taken from a NOAA website says some rather disturbing things about the veracity of the claim that the earth has warmed. The above picture has been bandied around the internet but I have never seen someone follow the logic of what it means.

First, if the editing of the data set requires the editors to warm the present vs the past, it means that they must beleive that our thermometers today measure too low of a temperature.

Secondly, that implies they think today's thermometers are worse than those used in the past because those used in the past are uncorrected but today's crappy thermometers must be 'corrected' before they will give the 'true' temperature.

Thirdly, the logic following from this is that the NOAA crowd thinks that technology has gotten worse over the 20th century requiring the 'correction' seen in the above picture. I have a solution for them. REPLACE ALL THESE MODERN MMTS THERMOMETERS WITH THE SYSTEM USED IN 1900. Then they won't have to go to the work of fixing the bad output of the technologically horrible modern thermometers. That conclusion seems to follow directly from what they do. If the modern MMTS thermometers were more accurate, more reliable, they should, one would think, require less 'correction' than the technologically primitive thermometers of 1900. Nostalgia is a grand thing.

I could not find the difference series in tabular form. I would love to have it, but, it isn't necessary as I can use the chart to create an approximation curve to the one above. Thus, I took the chart, and read the values off for every 5 years and then interpolated in the in between years. This will introduce a wee bit of error, but won't affect the general conclusions of what I am going to show below. I am interested in what the USHCN temperature anomaly would look like if I removed the editorial bias--the after the fact changes to the observed temperature.

After creating the approximate bias curve, I then downloaded the USHCN temperature anomalies from a Nasa site.The anomaly data is taken from which goes with the figure D described as:

Annual Mean Temperature Change in the United States
Annual and five-year running mean surface air temperature in the contiguous 48 United States (1.6% of the Earth's surface) relative to the 1951-1980 mean.

[This is an update of Figure 6 in Hansen et al. (1999).]

Also available as large GIF, PDF, or Postscript. Also available are tabular data.

(Last modified: 2009-01-09)


I then subtracted the false warming inserted into the data by the GISS editors and the curve would look like this--so this is approximately what the raw data shows. I used the 5 year running average curve, which is the one most people see.

Note that the 'warming' of the past 30 years does not take us beyond the natural range seen in the 1930s. If the warmth is being inserted into the data by the editors, who effectively are telling us that the modern thermometers are not as good as those used 100 years ago and thus need to be corrected towards warmer temperature, then why should we beleive their bunk?

The last picture shows all three curves on the same graph. Only after the editors working for Hansen get through with the editing of the raw data does global warming happen. Before they get their hands on the data, the world isn't warming.

I am reminded of the butler in the original movie version of The Shining. I loved the way he used the word, 'correct'. Yes, the Goddard Institute for Space Studies, 'corrects' the temperature record, and by doing so, they make the world warm up.


  1. great post Glenn! Surely the great minds at NOAA have some sort of rationale for "correcting" the modern temperature data?

  2. Yes, they believe that the world is warming and any station that doesn't show that warming is a bad station by definition. See

  3. Is there anyway to extend the raw vs fake - um... err... ahem... - raw vs GISS final into the present decade?

  4. Papertiger, yes, I can assume that the amount of editing is constant for this decade and then simply tack on the rest of the data. That may or may not be a correct assumption. YOu can see it at