On Mike Brakey’s recent post on the NOAA’s 151 degrees of fudging of the temperature datasets for the state of Maine, one reader was “so incensed” that he e-mailed the NOAA and his congressman.
Well, he got a reply from Derek Arndt at NOAA, which he sent to Mike Brakey, who in turn sent me Arndt’s reply – which I post as follows:
Hi Mr. XXXXXXXXX
In early 2014, we changed to a new version of the dataset upon which our US temperatures are drawn. The new dataset took advantage of a lot of older data that hadn’t been digitized (from paper) when the old dataset was constructed. It also took advantage of advancements in quality assurance that detect station moves, changes in observing practices, etc.
We began sharing with the community these upcoming changes as early as 2011: ftp://ftp.ncdc.noaa.gov/pub/data/cmb/GrDD-Transition.pdf In early 2014 we published a more complete methodological paper:
Vose, R.S., Applequist, S., Durre, I., Menne, M.J., Williams, C.N., Fenimore, C., Gleason, K., Arndt, D. 2014: Improved Historical Temperature and Precipitation Time Series For U.S. Climate Divisions Journal of Applied Meteorology and Climatology. DOI: http://dx.doi.org/10.1175/JAMC-D-13-0248.1
Maine was one of the states that saw the biggest differences in temperature. This is probably why blogs focus in on it. In addition to the general reasons for changes that other states witnessed:
- The new method used stations in neighboring Canada to inform estimates for data-sparse areas within Maine (a great improvement)
- In the old dataset, the year 1913 was particularly problematic, resulting from a keying (transcription) error from many years ago that is now corrected. 1913 is often held up as evidence of “tampering” when in fact it is probably one of the biggest improvements in the dataset, and brings our value much more in line with what was observed at the time.
Thanks for contacting us. It is a privilege to serve you.