Re: [ferret_users] stacking variables in time

["William S. Kessler" <william.s.kessler@xxxxxxxx>]

Do you mean that each dataset (x,y,z) is a timestep? In that case  
using a descriptor file would make Ferret see them as one (x,y,z,t)  
dataset, after which a lot of things become easier. But descriptor  
files are a bit cumbersome to make (see discussions in the user group  
for semi-automatic ways to do it).

Another way to do it as a one-off would be to define the variance by  
hand (mean of anomalies squared).

let mean = (o_temp[d=1] + o_temp[d=2] + ...)/10
let var = ( (o_temp[d=1]-mean)^2 + (o_temp[d=2]-mean)^2 + ...)/10
let rms = var^.5

BK

On 20 Jan 2011, at 12:20 PM, Neil Swart wrote:

> Hi,
>
> I have a variable, say ocean temp(x,y,z,t) in ten separate datasets  
> (on the same grid). I want to compute the variance in temperature  
> (in space, at one time step) across the datasets. The easiest way I  
> can
> think of would be to create a new variable, where my datasets 1...10  
> are stacked along the time axis. Then the variance could just be  
> computed with t=@var. However, I'm struggling to successfully define
> the new variable.
>
> If I was not thinking like a ferret, it would be something like:
>
> temp[l=1] = o_temp[t=5000,d=1]
> temp[l=2] = o_temp[t=5000,d=2]...
>
> I've tried save/append - but with no success either:
>
> save/file=test.nc o_temp[t=5000,d=1]
> save/append/file=test.nc o_temp[t=5000,d=2]
>
> Can anyone suggest how to achieve this? Do I need to redefine time  
> axis?
>
> Many thanks,
> Neil
>
> ---------------------------------------------------
> Neil Swart
>
> Climate Modelling Group
> School of Earth and Ocean Sciences
> University of Victoria
> Victoria, BC
> Canada
>
> http://climate.uvic.ca/

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
William S. Kessler
NOAA / Pacific Marine Environmental Laboratory
7600 Sand Point Way NE
Seattle WA 98115 USA

william.s.kessler@xxxxxxxx
Tel: 206-526-6221
Fax: 206-526-6744
Web: http://www.pmel.noaa.gov/people/kessler