Re: [ferret_users] Using @VAR

[Ansley Manke <ansley.b.manke@xxxxxxxx>]

Hi Zhen,
To answer your second question first, yes, the size of the grid 
intervals is taken into account, so @VAR computes the variance correctly 
for irregular grids.

Your syntax is correct for calculating variance in the Z direction.  The 
result should depend on time but not on Z, as for each time the result 
is the sum over all Z of the squares of the data minus mean (with the 
proper weighting factors). Is this what you're seeing?  If not, let's 
continue the discussion off-line and figure out what's happening.

Ansley



zhen.li@gsfc.nasa.gov wrote:

> Dear Ferret Users:
>
> I'd like to use @VAR function to calculate the variance along the 
> Z-axis of a variable, salt. "salt" is sized 43 by 252 in K and L. So, 
> I did  "let std_var_salt=salt[k=@VAR]^0.5". However, the variance is 
> still calculated on "t" axis even if I specify it on "k" axis.  Does 
> anyone know how to do this?
>
> A related question: since @VAR is a weighted variance, so I assume it 
> automatically takes care of the irregular grid, and I don't need to 
> regrid the variable to the regular grid. Am I right on this?
>
> Your response is highly appreciated.
>
> Zhen