Re: [ferret_users] A new data set

["william s. kessler" <william.s.kessler@xxxxxxxx>]

Hi Jean -

This is a job for @WEQ:

let zz = z[gz=temperature]			! variable for depth values
let zero_at_mld = zz - mld			! = 0 at zz=mld
let integrand_t = temperature * zero_at_mld[z=@weq]	! zero_at_mld 
[z=@weq] = 1 where z=mld, .lt.1 elsewhere
let t_at_mld = integrand_t[k=@sum]	! interpolate

See the documentation for the extremely powerful, but not immediately  
intuitive, function @WEQ.

Billy K

On 27Aug 2007, at 11:25 AM, jean oliveira wrote:

> Dear Ferret users,
> I have the following data set:
>
> yes? sho data
>     currently SET data sets:
> 1>	http:// opendap.mercator-ocean.fr/thredds/dodsC/ 
> mercatorPsy3v1R1v_sat_mean_best_estimate  (default)
>
> name      
> title                                                                I 
>              J          K             L
> TEMPERATURE   temperature                                        
> 1:601     1:453     1:43      1:14
> MLD      temperature ocean mixed layer thick                   
> 1:601     1:453     ...          1:14
>
> I want to construct a new data set where the new variable (tpc)  
> will be the temperature at the end of the MLD. I got the following  
> idea:
>
> Repeat/l=1:14 (repeat/I=1:601(repeat/J=1:453 ( let tpc = temperature 
> [z=`MLD`@itp] ;save/append/J=1:601/J=1:453 /file=tpc.nc tpc ))).
>
> But obviously, this way is very slow. Does anyone know a faster way?
>
> Thanks in advance
>
> Jean
>
> _________________________________________________________________
> Verifique já a segurança do seu PC com o Verificador de Segurança  
> do Windows Live OneCare! http://onecare.live.com/site/pt-br/ 
> default.htm