[fitsbits] Celsius unit in FITS keyword comments

[Rob Seaman]

Neither FITS or IVOA recognizes Celsius in any standard way:

	http://www.ivoa.net/documents/VOUnits/20140523/VOUnits-REC-1.0-20140523.pdf (p.9)
	ftp://legacy.gsfc.nasa.gov/fits_info/fits_formats/docs/general/ogip_93_001/ogip_93_001.ps

The IAU says “degree Celsius (oC) is used in specifying temperature for meteorological purposes", but I’ve never seen that myself:

	https://www.iau.org/publications/proceedings_rules/units/

(Have never seen “small-oh” instead of degree symbol in any context.)

I have seen comments mentioning both “deg C” and “degrees Celsius”. But degrees are generally reserved for angles as a way of shaking our fists at radians and pretending the Babylonians didn’t exist. I suppose we can blame Lavoisier getting his head chopped off for thermodynamic usage of the word “degree” being suppressed, as well as for there not being 400 degrees in a circle (and for that matter for it being called Celsius and not centigrade).

In the absence of recognized units a comment should probably spell things out verbosely, so perhaps something like:

	"/ [deg C] temperature in degrees Celsius”

Pedants can dispute whether it should be “degrees Celsius” or “degree Celsius”. Only a barbarian would say “Celsius degrees”, but fussy metrologists might prefer [K+273.15], but according to the IAU meteorologists never would.

Rob Seaman
LPL
—

> On Mar 3, 2016, at 3:42 AM, Richard J. Mathar <mathar at mpia-hd.mpg.de> wrote:
> 
> What is the recommended unit that goes into the bracket of the
> FITS keyword line if the datum is a temperature in Celsius?
> We all here know that K is Kelvin and C is Coulomb.
> 
> R. J. Mathar