[CF-metadata] Precipitation query

From: Jonathan Gregory <j.m.gregory>
Date: Tue, 22 Aug 2006 13:10:12 +0100

Dear Roy

> I am currently dealing with a dataset that contains rain gauge data recorded every hour as the amount of precipitation in mm that has fallen in the hour preceding the logged timestamp. Question is would the standard name for this be 'lwe_precipitation_rate' or 'lwe_thickness_of_precipitation_amount'? If it's the rate, in what circumstance would 'lwe_thickness_of_precipitation_amount' be used? Surely there is always a period of time over which the thickness has accrued.

It depends on the units. If you give it in mm (or m, inches, etc.), it is
lwe_thickness_of_precipitation_amount
and if in mm hr-1 (or mm day-1, m yr-1 etc.) it is lwe_precipitation_rate.

In both cases the bounds of the time coordinate variable ought to indicate the
starts and ends of the hours of accumulation. The rate is an intensive quantity
in time and so it should have cell_methods of mean in time, to indicate it is
a mean rate and not an instantaneous rate (which would be the default
interpretation). Personally I prefer rates because I think there is less scope
for misinterpreting them and making mistakes.

Best wishes

Jonathan
Received on Tue Aug 22 2006 - 06:10:12 BST