funsec mailing list archives

Re: Was the ClimateGate Hacker Justified? Join the Debate!


From: Valdis.Kletnieks () vt edu
Date: Sat, 05 Dec 2009 04:52:23 -0500

On Thu, 03 Dec 2009 10:56:03 EST, Wes Deviers said:

I think he was referring to the idea that previously rural senors are becoming 
urban without necessarily taking that into account when you look at the data.

For instance, say you have 30 weather stations along the stretch of 81 between 
Blacksburg and Roanoke that were put in place in the 1950s and have been
continually monitored.  As Roanoke and the Blacksburg/Christiansburg Greater
Metropolitan Area of Traffic Growth have expanded towards each other, the heat-
island effect would have expanded with it.  So assume 10 of the 30 sensors have
shown continuous temperature increases since 1950.  How much of that is due to 
global warming, and how much of that is due to urban expansion?

A climatologist at VT would take that into account in their localized studies.  
Somebody at NOAA, viewing raw numbers in a text file, has no way to control for 
that.  It's not that cities are hotter, is that the hot areas around cities
expand with the cities, and national or international datasets cannot account
for it.

If just 5% of your weather data points from 1960 were rural but in 2009 are
urban or suburban, how much does that skew the entire set?

Here's a nice thread about how they deal with this stuff.  And apparently,
the raw numbers *do* include a tag for this, so "somebody at NOAA" can do the
applicable controls. People start posting actual stuff around reply #5:

http://www.physicsforums.com/showthread.php?t=359423

Of course, it's exactly these sort of corrections to the data that have the
climate-change deniers jumping up and down yelling 'ZOMG! Cooked Data!', and
why many of the scientists advocated not releasing the raw uncorrected data.

Attachment: _bin
Description:

_______________________________________________
Fun and Misc security discussion for OT posts.
https://linuxbox.org/cgi-bin/mailman/listinfo/funsec
Note: funsec is a public and open mailing list.

Current thread: