[CF-metadata] Representing statistics in CF

From: Jonathan Gregory <j.m.gregory>
Date: Mon, 24 Apr 2017 17:17:57 +0100

Dear Jon

> 2. Representing the variability of a quantity within a time period (e.g. expressing the annual mean plus a measure of the spread of values around the mean).
> I assume that (2) is solved by using cell methods on the time axis (e.g. cell_methods = ?t: mean? or ?t: standard_deviation? etc). Please correct me if I?m wrong!

Quite right. The time period is recorded using bounds of the time coord var.

> 1. Representing the mean and spread of an ensemble of simulations (i.e. boiling down a set of ensemble members to two variables: mean and spread).
> For (1) should I use a standard name modifier? If so, is there a reason why ?mean? and ?standard deviation? are not listed as standard name modifiers? ?Standard error? is listed, but this is different from standard deviation, and I think that what I will need is probably a standard deviation.

You can do this with cell_methods in exactly the same way, "e: mean" and
"e: standard_deviation" where e is the ensemble dimension. If you don't have
an ensemble dimension, you could put "realization: mean" and
"realization: standard_deviation", since "realization" is a standard_name
for ensemble member designators.

Best wishes

Jonathan
Received on Mon Apr 24 2017 - 10:17:57 BST