NOAA temperature record “adjustments” could account for almost all “warming” since 1973

Dr. Roy Spencer, Principal Research Scientist at the University of Alabama in Huntsville, analyzed the records of weather stations in the U.S. Historical Climatology Network (USHCN) which NOAA touts as the official U.S. climate record. He found that almost all of the reported warming since 1973 can be accounted for merely by the adjustments to the record made by NOAA.

Spencer also analyzed the British CRUTem3 and International Surface Hourly (ISH) data. His main conclusions are:

1) The linear warming trend during 1973-2012 is greatest in USHCN (+0.245 C/decade), followed by CRUTem3 (+0.198 C/decade), then my ISH population density adjusted temperatures as a distant third (+0.013 C/decade)

2) Virtually all of the USHCN warming since 1973 appears to be the result of adjustments NOAA has made to the data, mainly in the 1995-97 time frame.

3) While there seems to be some residual Urban Heat Island (UHI) effect in the U.S. Midwest, and even some spurious cooling with population density in the Southwest, for all of the 1,200 USHCN stations together there is little correlation between station temperature trends and population density.

4) Despite homogeneity adjustments in the USHCN record to increase agreement between neighboring stations, USHCN trends are actually noisier than what I get using 4x per day ISH temperatures and a simple UHI correction.

When “global warming” only shows up after the data are adjusted, one can understand why so many people are suspicious of the adjustments.

See his post on the matter here.

Spencer has also examined the surface temperature records according to the population density near the stations. He finds “clear evidence of an urban heat island effect on temperature trends in the U.S. surface station network” which produce a warming bias to station records.

See his posts here and here.

Among his findings:

1) Essentially all of the +0.20 deg. C/decade average warming trend over the U.S. in the last 40 years computed from the CRUTem3 dataset (which the IPCC relies upon for its official global warming pronouncements) evaporates after population adjustment (no claim is made for countries other than the U.S.)

2) Even without any adjustments, the ISH data have a 20% lower warming trend than the CRUTem3 data, a curious result since the CRUTem3 dataset is supposedly adjusted for urban heat island effects.

Comment: These analyses may explain, in part, why the official surface temperature record does not agree with the record from satellites. It also implies that recent warming is not nearly as dire as some would have us believe, nor does the real temperature reflect any major influence by carbon dioxide emissions.

NOAA is not the only agency “adjusting” the temperature record. NASA’s GISS has been doing that for U.S. and global temperatures also (for instance see here.)

All the adjustments tend to make older temperatures, for example, those in the 1930s, colder while making all more recent temperatures warmer. This is statistically improbable if the adjustments were to be justified by some physical conditions. This whole process is troubling because it suggests corruption within the government-funded scientific establishment whose results apparently have to be politically correct rather than scientifically correct.


See also:

A Perspective on Climate Change a tutorial

NOAA accused of fabricating temperature data

The State of our Surface Temperature Records




  1. The reporter here is not well informed on this issue.  The BEST temperature reconstruction validated the other data sets.  It’s more likely the satellite data is getting it wrong due to the fact that the satellites do a poor job of measuring polar regions where we see much more warming due to polar amplification.

    1. What the BEST result shows is that surface thermometers from the land area of the globe (about 29% of the earth’s surface) show a warming trend.  But this is not global warming.  And BEST director Professor Rich Muller explicitly disclaims that his trend results indicate a human cause.
      He also correctly points out that many of the weather stations used are badly distributed, mostly in the U.S. and western Europe, and possibly subject to local heating effects, such as urban heat islands.  He cautions that a third of his monitoring stations show a cooling, not a warming.  And that 70% of the U.S. stations are poorly situated and don’t satisfy the requirements of the U.S. Weather Service.  It is likely that stations elsewhere have similar problems.

      1. How does pointing out that twice as many stations show warming as opposed to those showing cooling, support your contention? Did you expect global climate to somehow become homogenous? What difference does the number of stations showing warming or cooling have to do with anything if you don’t also include the amount of warming or cooling?   

        When all stations are compared to just rural stations, we see an approximate 5% difference. If you consider that over two thirds of the earth is ocean, you arrive at a UHI value of about 1.7 percent. Jon seems to think that he has uncovered an important variable of which others are unaware. He hasn’t. UHI is well known, analyzed and accounted for in temperature records. Talk about grasping at straws.   JP

  2. Jon, A large portion of the adjustments you refer to were for errors on TOB’s (time of observation). The instrumentation for earlier periods was designed to measure the daily high and low temperature. Before newer digital equipment was available, these instruments couldn’t “tell” what day they were measuring. Consequently, if the person taking the measurement did so early in the morning, it was possible for the low temp reading to be the low from the previous day. In other words, the most recent overnight low was actually higher than the previous low. The older instruments could not account for this. When this problem was identified, it was corrected.

    Are you suggesting that data, when shown to be wrong, should not be adjusted to correct the problem?

    There are numerous problems with Dr. Spencer’s handling of the statistical analysis. If this work had been peer reviewed, those issues would have been settled. Your readers can see some of those problems by reviewing the discussions at Dr.Spencer’s blog or at WUWT. 

    There is however one glaring problem for Dr. Spencer. Most here recognize Dr. Spencer as the leader of one of the three satellite temperature datasets, the UAH/MSU. In that capacity he provides a subset of his global temperature data that covers just the U.S. lower 48 states. As others have pointed out, Dr. Spencer’s own data show a trend of +0.22 degrees C per decade from 1979.
    This is much closer to USHCN and CRUtem than to his population density trend. The one your post seems to support.

    So which one is wrong Jon? Is the UAH data wrong or is the data from this post wrong. Or is the tropospheric trend 15 times higher than the surface trend?   JP

    1. Your explanation of TOB would require corrections in the opposite direction of what they actually did.

      So, it is just an absolutely amazing coincidence that all the frequent adjustments lowered the 1930s warm period temps and raised the most recent temps.  Is that it?  And who decides how much of a correction to give to the numbers?  On what basis?

      Reader can also see more on NASA data manipulation here:

      1. I showed you why one of the largest adjustment was necessary. If you have information that shows why a particular adjustment was incorrect, then show it. Because something seems like an “amazing coincidence” to Jonathan DuHamel, is not a basis for making a scientific decision.

        Now, would you like to address the seeming contradiction in the Spencer UAH data and the Spencer data in your post? JP

      2. Jon, My description was ambiguous. The recorded temperature was, in instances where the measurement was taken too early in the day, actually a measurement of the previous 24 hour low. Many readings were recorded before the low temp of the most recent 24 hours was reached. Hence, the lower temp was not recorded. That needed to be corrected. Not to do so would have been knowingly inaccurate and improper.

        To answer your questions:

        1. “Who decides?” The scientists at USHCN.

        2. “On what basis?” Presumably on the nature of the error. The raw data, the adjustments and the code for the adjustments are all public and can be obtained through NASA-GISS.

        I’ve answered your questions. Will you answer mine? JP

      3. Since NOAA made all the older temps colder and all the newer temps warmer, you would have us believe that all the technicians, at all the stations continually made the same mistake over and over again.  And, I suppose, none of them would ever make the opposite error of failing to record the highest high. Right!

        But it actually doesn’t matter in a given 24 hour period because the new lower temp reading would be recorded the next day if the process was as you pose.  This is similar to recording rainfall in the network I’m in.  We record rainfall from 7am to 7am and report it to the previous day.  Temperatures recorded in the same way would just be one day off.

      4. “And, I suppose, none of them would make the opposite error of failing to record the highest high.” Errors WERE made on the high temps. And they were included in the correction algorithms. Jon, the errors were not failures by the technicians. They were failures of the system and the equipment, and how those failures interacted with the times of day when technicians were making measurements. Before you accuse people of “corruption”, maybe you should look at the nature of the errors and the methods used to correct them. 

        Your assumption that the TOB errors are analogous to rain gauge measurement errors is completely wrong. Rainfalls are cumulative measurements. Temperatures are boundary measurements. Boundary measurement errors, like the ones described, don’t simply move the whole record forward 24 hours. Incorrect measurements took the place of correct measurements. Furthermore, you example had rainfall measurements all being taken at the same time. That was the problem. Older records were NOT all taken at the same time. That’s why it’s called a Time Of Observation (TOB) error. 

        Most importantly, if you believe that the methodology used in these adjustments (or any others) were done incorrectly, then present the facts that support your conclusion. All of the information is in the public record. Don’t accuse people of being corrupt because their work seems “amazingly coincidental” to you. That is not a basis for denigrating the work of entire agencies, as you’ve done here.   

        Jon, you have still not addressed the issue of the incongruity of Dr. Spencer’s data presented in your post and Dr. Spencer’s UAH/TLT data. Would you please explain how the temperature trend of the lower troposphere could possibly be 15 to 17 times greater than the surface trend? One of those datasets is most certainly incorrect. I believe we both know which one that is.   JP

  3. As Steve Goddard speculates:
    No doubt all these changes in 2012 are just a coincidence
    HadCRUT4 – adjusted upwards to make 2010 warmer than 1998GISS – past adjusted downwards across the Arctic and many other placesGISS – present adjusted upwards across the Arctic and many other placesEnvisat sea level – adjusted massively upwards by a factor of 3NSIDC ice – adjusted away from the mean

    1. Why speculate? Why not present evidence that supports your accusations of corruption? If you fail to do that after repeated invitations, others are likely to “speculate” that you don’t have any evidence.

      Your silence on the incongruous Spencer data speaks for you. JP

    1. That’s your proof? First of all it was purported evidence not proof. When that purported evidence was examined by independent judges, it was proved to be wrong. But I understand your need to divert attention from the problematic data in your above post. Don’t feel too bad, Jon, most of the ‘skeptic’ blogosphere followed Dr. Spencer right off the cliff with you. JP

Comments are closed.