Thursday, December 10, 2009

There's something seriously odd about "NAPIER NELSON PK"

If you are messing with the Met Office temperature data there's one file which is really odd: 93/933710. It has temperature data for NAPIER NELSON PK at (-39.5, -176.8) There is actually a Napier Nelson Park in New Zealand but not quite at those coordinates. And I can't find a reference to a WMO station with the number 933710.

But leave that aside.

The really odd thing is the data in the file. There's a mismatch between the 1961-1990 data (which is present and complete) and the Normals (i.e. the averages for 1961-1990). According to the file the Normals come from the data itself, but they are off: almost consistently by 1C.

Here are the differences (Normals as is reported; Calculated is my calculation from the file):

Month Normals Calculated Difference

Jan 19.5 18.31 -1.19
Feb 19.0 17.94 -1.06
Mar 17.8 16.68 -1.12
Apr 14.8 13.85 -0.95
May 12.0 10.87 -1.13
Jun 9.7 8.55 -1.15
Jul 9.0 8.05 -0.95
Sep 11.9 10.88 -1.02
Oct 14.1 13.04 -1.06
Nov 16.2 15.05 -1.15
Dec 18.2 16.94 -1.26

So the normal temperature for this station is 1C higher than the observations indicate. That would end up meaning that any trend is actually lower than the real figures.

Very odd. And very odd that's almost exactly 1C off. Why? Some systematic error? Human intervention?

Labels: ,

If you enjoyed this blog post, you might enjoy my travel book for people interested in science and technology: The Geek Atlas. Signed copies of The Geek Atlas are available.


<$BlogCommentDateTime$> <$BlogCommentDeleteIcon$>

Post a Comment

Links to this post:

<$BlogBacklinkControl$> <$BlogBacklinkTitle$> <$BlogBacklinkDeleteIcon$>
Create a Link

<< Home