Dear Lars
Answers to some of your points:
> 1. Is there a standard name for "diurnal temperature range", or a combined standard name/cell method representation? Cf. the 2010 mail archive thread http://mailman.cgd.ucar.edu/pipermail/cf-metadata/2010/054016.html that added "range" to Appendix E.
The cell_method of range was approved in ticket 65
http://cf-trac.llnl.gov/trac/ticket/65
and will be added in CF 1.7, which is under preparation. You could use this,
in combination with time bounds, to describe a diurnal range.
> 2. 'Heating Degree Days' and similar: I take it this was resolved/decided to have standard name "integral_of_air_temperature_deficit_wrt_time" (or ..._excess_...), with canonical units K s in this post http://mailman.cgd.ucar.edu/pipermail/cf-metadata/2005/051014.html . However, as the threshold is not explicitly referred to in the standard name, how is this linkage formally established? A practical example would be helpful.
The data variable of the integral should have a scalar coordinate variable
or a size-one coordinate variable with the standard name of air_temperature_
threshold, to indicate the threshold. I agree, this should be in the
description of the integral standard names.
> 3. Several of the standard names directly targeted at climate indices (e.g. "spell_length_.....", "number_of_days_...") are dimensionless because the unit is already included in the standard name. This is of course consistent and parsimonious in terms of describing the data. But with respect of a software package producing a plot that needs a legend with units attached, it is not very enlightening to have the unit "1". For example, in a map of Frost Days the title drawn from the long name or the comment may be something like "Number of Frost Days (Tmin < 0 degC)" but the color scale should preferably have units "days" rather than "1". As this is not specific to climate indices, what is the CF accepted advice or solution to this?
The spell_length variables have canonical units of days. The number_of_days
variables have canonical units of 1, as you say. This is logical. Could you
use the standard name, including the phrase "number_of_days", to derive the
appropriate label?
Best wishes
Jonathan
Received on Thu May 12 2016 - 10:26:51 BST
