Correction of canonical unit for `product_of_air_temperature_and_specific_humidity`

[larsbarring]

In standard name table v. 82 (and earlier) product_of_air_temperature_and_specific_humidity has canonical unit K, which would imply that it is a temperature quantity, which it is not. As it in fact is a product I believe the correct canonical unit is K 1. Note that the 1 is a unit in itself, not the integer 1, which can be removed. This becomes evident if one considers that the specific humidity could be reported in units g kg-1 or kg kg-1 (and more), instead of the canonical unit 1 .

[github-actions[bot]]

Thank you for your proposal. These terms will be added to the cfeditor (http://cfeditor.ceda.ac.uk/proposals/1) shortly. Your proposal will then be reviewed and commented on by the community and Standard Names moderator.

[davidhassell]

I support this change, noting that the canonical units of specific_humidity in the table are also 1.

[JonathanGregory]

Dear @larsbarring

I don't agree with this proposal, because K 1 is exactly equivalent to K in both SI and UDUNITS, and so is 1 K. The canonical unit is K. Quantities with this unit are not necessarily temperatures. For example, the empirical inversion strength of the troposphere is a quantity in K, but it's not a temperature. It has that unit because it's derived from a temperature difference.

You've pointed to a pitfall for units conversion of product_of_air_temperature_and_specific_humidity, which I suggest should be stated in the definition of this quantity and of specific_humidity. Specific humidity is a mass fraction. It is not explicitly stated, and probably it should be, that it's the ratio of one mass to another when they are in the same units. If it is g kg-1 rather than kg kg-1, it should have units="1e-3" for instance, and that would carry over to product_of_air_temperature_and_specific_humidity.

Best wishes

Jonathan