[CF-metadata] the need to store lat/lon coordinates in a CF-compliant netCDF file

From: Bentley, Philip <philip.bentley>
Date: Wed, 3 Aug 2011 15:53:05 +0100

Hi Dave,

> The problem here is that there is no analysis that does not
> require this information. While some choose to assume that
> all data uses the same datum and drop those terms of analysis
> out of their math, that does not mean the information is not there.
>
> Having personally been in this situation, it took me two
> years to track down exactly the right geographic
> transformation to apply to accurately apply radar data to the
> landscape. It is unfair to expect that a terrestrial modeler
> understand the handling of geographic data in climate and
> forecasting applications to such an extent that they are
> comfortable making such a decision.

I wasn't for one minute suggesting that geodetic datum (or any other)
information shouldn't be provided, assuming it's known. As the author of
the last significant batch of CRS-related updates to the CF
specification, I'd have no reason to want it otherwise! I was merely
opining that specifying a CF default datum doesn't necessarily help to
solve the underlying problem.

> Your argument about darwinian evolution of data use would
> cause a massive set back in interdisciplinary science.

Crikey, I'd hate to think that I was single-handedly responsible for
causing a massive setback to interdisciplinary research. That nice Mr
Gore would be breathing down my neck before you could say 'international
debt crisis' :-)

On the basis that, to date, the global community has done plenty of good
interdisciplinary research - often without knowing the arcana of
geodetic datums - then I think we can have good reason to believe that
this will continue.

Regards,
Phil
Received on Wed Aug 03 2011 - 08:53:05 BST