[CF-metadata] A user has a set of variables they can't store: advice?
armin.rauthe-schoech at mpic.de
Mon Sep 4 13:27:39 MDT 2017
Dear Cathy Smith,
I saw your email on the CF-metadata mailinglist about how to store CF-metadata in NASA AMES files. Maybe you find the attached information useful for your users.
I worked many years in the CARIBIC flying observatory project (http://www.caribic-atmospheric.com/ ). CARIBIC uses NASA AMES text files to store the data measured during the flights. Each instrument used a somewhat different NASA AMES format. So in 2014 we decided to standardize our NASA AMES files and include some CF-compliant metadata into the files.
The intention was to create NASA AMES files that are
a) as compliant as possible to the original NASA AMES format
b) easily readable with Excel
c) include CF-metadata according to the CF-convention
I have attached our format description and an example file containing aircraft position and attitude data for one measurement flight.
You will see that to fulfill point b), we repeat some of the information like variable names and units in the last few lines of the header. If you load our NASA AMES files with Excel, it will show this information directly above the actual data columns.
Since we are working in a community with its own preferred units (like ppm, ppb, etc.) which are partly discouraged by the CF-convention, we include both our own units as well as the correct CF-units.
Dr. Armin Rauthe-Schöch
Max Planck Institute of Chemistry
Air chemistry department
From: CF-metadata <cf-metadata-bounces at cgd.ucar.edu> on behalf of Cathy Smith <cathy.smith at noaa.gov>
Sent: Friday, September 1, 2017 00:23
To: cf-metadata at cgd.ucar.edu
Subject: [CF-metadata] A user has a set of variables they can't store: advice?
A user trying to use CF standard names sent me this list of variables they were unable to find the names of. Note this is for observations at a point. It is to be stored in NASA Ames format and not netCDF. In some cases, the units was the issue (e.g. evap rate). In some cases, the data was at a specific level (e.g. 10m wind speed) and it wasn't clear how 'level' would be indicated in the Ames metadata.
The 'level' issue may be part of another one: CF was designed for gridded data but from a talk on the web, it is
"Framed as a netCDF standard, but most CF ideas relate to metadata design in general, hence can be contained in other formats such as XML"
So, can the CF standard names really apply to point collected data not in netCDF? And, can these all be considered as potential variables for CF if they don't already exist?
Here is the list.
Bulk buoyancy flux into ocean (W m-2)
Friction velocity that includes gustiness (m s-1) [aka ustar]
Wind stress (N m-2)
Temperature scaling parameter (K) [aka tstar]
Specific humidity scaling parameter (g kg-2) [aka qstar]
Thermal roughness length (m)
Moisture roughness length (m)
Wind stress transfer coefficient at zu () [aka Cd]
Sensible heat transfer coefficient at zu () [aka Ch, Stanton number]
Latent heat transfer coefficient at zu () [aka Ce, Dalton number]
Obukhov length scale L (m)
Monin-Obukhov stability parameter zu/L ()
wind_speed at 10 m (m s-1)
air_temperature at 10 m (C)
specific_humidity at 10 m (g kg-1)
relative_humidity at 10 m (%)
Neutral value of wind speed at zu (m s-1)
Neutral value of wind speed at 10 m (m s-1)
Neutral value of drag coefficient at 10 m ()
Neutral value of Stanton number at 10 m ()
Neutral value of Dalton number at 10 m ()
Cool-skin temperature depression (C)
Surface saturation specific humidity (g kg-1)
Latent heat of vaporization (J kg-1)
Evaporation rate (mm h-1)
NOAA/ESRL PSD and CU CIRES
Emails about data/webpages may get quicker responses from emailing
esrl.psd.data at noaa.gov<mailto:esrl.psd.data at noaa.gov>
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 276107 bytes
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
More information about the CF-metadata