Pursuant to Dave’s suggestion, check out the CF convention for netCDF metadata.
In particular see section 9 on discrete sampling geometries, and its discussion
of the timeSeries featureType at multiple locations.
Appendix H.2.1 shows example CDL notation for an case that sounds fairly
similar to what you are looking for.
http://cfconventions.org/cf-conventions/cf-conventions.html
http://cfconventions.org/cf-conventions/cf-conventions.html#discrete-sampling-geometries
http://cfconventions.org/cf-conventions/cf-conventions.html#time-series-data
rbs
> On Mar 18, 2019, at 21:41, Dave Allured - NOAA Affiliate
> <dave.allured@xxxxxxxx> wrote:
>
> Ken,
>
> I see that Gus just posted an involved gridding method. Here is an alternate
> approach.
>
> You are right, you definitely do not want to try to grid these data. The
> layout that you designed today is precise and efficient for these 2-D (time,
> ID) site data. I suggest that you need your own custom command line tool for
> spatial subsetting, and share that with the data users. A tool with limited
> functionality, just for datasets of this particular layout, should be quite
> easy in several programming languages.
>
> NetCDF is well suited for all kinds of numeric arrays, not just geospatially
> gridded. You see this emphasis on geospatial grids in netCDF discussions and
> tools, because of their distant origins and current popularity in weather and
> climate research.
>
> —Dave
>
--
Robert B. Schmunk
Webmaster / Senior Systems Programmer
NASA Goddard Institute for Space Studies
2880 Broadway, New York, NY 10025
