NOAA's temperature control knob for the past, the present, and maybe the future – July 1936 now hottest month again

Two years ago during the scorching summer of 2012, July 1936 lost its place on the leaderboard and July 2012 became the hottest month on record in the United States. Now, as if by magic, and according to NOAA’s own data, July 1936 is now the hottest month on record again. The past, present, and future all seems to be “adjustable” in NOAA’s world. See the examples below.

Josh has been busy again and writes at Bishop Hill with a new cartoon:

The temperature adjustments story has been brewing for weeks principally due to the many posts at ‘RealScience’ but taken up by others, for example, Paul Homewood, see here and here. Judith Curry has a great post about it here, as does Anthony here.

H/t to Real Science/Steven Goddard for suggesting including Toto. Cartoons by Josh

Bruce at Sunshine Hours has been doing some unthreading, er plotting, and at my request, prepared some USHCN maps of Kansas, first May’s high temperatures.

I’ve annotated the plot, to include “zombie” weather station that have been closed for years, but still show “estimated” data from NOAA. Those marked NRF are “no report found”…typically meaning NOAA hasn’t gotten the data from the observer yet, which is often mailed in on paper B91 forms.  It is interesting to note how NOAA has been changing the data, in most cases adjusting it higher, though in a couple of cases, lower.

Bruce also plotted some other maps of Kansas, for July 1936, and for July 2012. Note how in July 1936 the Tmax temperature are almost all adjusted cooler, and in 2012, most all Tmax temperatures are adjusted warmer. Click images for larger versions.

Whatever happened to just using actual measured data? There is no justification for this.

And, NOAA can’t even keep their story straight about July 1936 temperatures. From a report I did in 2013:

NCDC’s SOTC July 2012:

http://www.ncdc.noaa.gov/sotc/national/2012/07

Screencap of the claim for CONUS Tavg temperature for July 2012 in the SOTC:

Note the 77.4°F value for July 1936. It is actually still in their SOTC for July 2012 today.

Now let’s look at some plots from NOAA’s Climate at a Glance. I just happened to have one from two years ago. It also says 77.4°F on the plot. The numbers match with the SOTC report. The annotations are mine.

Today, I ran the same plot again, and here is the NEW number for July 1936. The annotations are mine.

NOAA helpfully provided the data which I have saved as an Excel file, it has both 1936 and 2012 July data: NOAA_Tavg_Data_July_1895-2013 (.xlsx)

You can’t get any clearer proof of NOAA adjusting past temperatures.

This isn’t just some issue with gridding, or anomalies, or method, it is about NOAA not being able to present historical climate information of the United States accurately. In one report they give one number, and in another they give a different one with no explanation to the public as to why.

This is not acceptable. It is not being honest with the public. It is not scientific. It violates the Data Quality Act.

But wait, there’s more. In January 2013, I ran this story based on an article in the Wall Street Journal: July (2012) Was Hottest Month on Record

My story was:  Does NOAA’s National Climatic Data Center (NCDC) keep two separate sets of climate books for the USA?

In that essay, I revised the WSJ graphic. At that time, it looked like this based on new numbers for July 2012 that I found from NOAA:

Now, with the new numbers in the Excel File above, output from NOAA, I had to revise it again. It looks like this now:

Now, once again, July 1936 is the hottest month in the US, even if by the slimmest of margins, all thanks to post-facto adjustments of temperature data by NOAA/NCDC.

I suggest that NOAA/NCDC have another one of those meetings like where they decided to keep long dead weather stations reporting as “zombies”, like I showed with Marysville, yesterday, and work on getting their story straight.

This constant change from year to year of what is or is not the hottest month on record for the USA is not only unprofessional and embarrassing for NOAA, it’s bullshit of the highest order. It can easily be solved by NOAA stopping the unsupportable practice of adjusting temperatures of the past so that the present looks different in context with the adjusted past and stop making data for weather stations that have long since closed.

NOAA has been accused by others of “fabricating” data, and while that is a strong word that I don’t like to use, it looks to be more and more accurate.

That said, I don’t believe this is case where somebody purposely has their hand on a control knob for temperature data, I think all of this is nothing more than artifacts of a convoluted methodology and typical bureaucratic blundering. As I’ve always said, never attribute malice to what can be explained by simple incompetence.

We already showed yesterday that NOAA can’t get their output data files correct, and we are waiting on a statement and a possible correction for that. But I think the problem is even larger than that, and will require an investigation from an unbiased outside source to get to the root of the problem.