Re: [ferret_users] erratic IF masking behaviour

[Russ Fiedler <russell.fiedler@xxxxxxxx>]

Hi Lev,

The problem here is due to when the interpolation is being carried out 
due to the interpolation mode being set.

You are getting the interpolated value of HDF rather than rather than a 
value of HDF based on interpolating HDHF and HD and getting the 
appropriate value with the mask. The IF condition is 
calculated/evaluated on the original grid.

Therefore you need to define a new mask on the grid to which you are 
interpolating::

let ifhfnew = ifhf[g=newgrid]                         ! Interpolate the mask
let hdf= IF ifhfnew eq 1 then hdhf else hd     ! Get a value based on 
the new mask

! or explicitly... let hdf= IF ifhfnew eq 1 then hdhf[g=newgrid] else 
hd[g=newgrid]


Cheers,
Russ


On 11/12/13 14:35, Lev Tarasov wrote:
> Why does the IF condition not work in the
> masking below and how do I fix this:
> (ie CONDITION should be false, but hdf doesn't take on
> the ELSE value of hd)
>
> let hdf= IF ifhf eq 1 then hdhf else hd
>
> yes? list ifhf
>              VARIABLE : IF HFB GT HDG THEN 1 ELSE 0
>              FILENAME : idHhpt.nn9927
>              LONGITUDE: 69W (interpolated)
>              LATITUDE : 41.5N (interpolated)
>              T        : -20
>           0.5000
>
> yes? list hd
> ...
>              LONGITUDE: 69W (interpolated)
>              LATITUDE : 41.5N (interpolated)
>           36.75
> yes? list hdf
> ...
>              LONGITUDE: 69W (interpolated)
>              LATITUDE : 41.5N (interpolated)
>           35.50
>
> note using
>      NOAA/PMEL TMAP
>      FERRET v6.401
>      Linux(g77) 2.6.9-89.ELsmp - 12/09/09
>      11-Dec-13 00:05
>
>
> Cheers,
>
> lev
>
> Lev Tarasov -   Dept of Physics and Physical Oceanography,
>         Memorial University of Newfoundland.
>                 email: lev@xxxxxx
>                 http://www.physics.mun.ca/~lev/
>              Currently at Antarctic Research Centre,
>                 Victoria University of Wellington