Re: [ferret_users] Standard Deviation

[Russ Fiedler <russell.fiedler@xxxxxxxx>]

Hi

I think Nitin is after the spatial standard deviation as a function of
time

let  std_dev=mask[x=65:95@VAR,y=5:38@VAR,l=1:610]^0.5

is probably what is required. I presume that the mask consists of ones
and missing values. If it is ones and zeroes then you need


let mask=abc[d=1,g=lsm[d=2]]*IGNORE0(lsm[d=2])

Russ



On Sat, 2012-11-03 at 06:28 +1100, Ansley Manke wrote:
> Hi,
> In the header that Ferret writes when you issue the LIST command,
> there should be information about the grid of the variable and what
> operations Ferret has done on the different axes of the grid. That
> might give you an idea what is going on.
> 
> What is the grid of variable lsm in dataset 2?  Does it have a time
> axis?  If it is defined only in XY, then you want to ask for the
> regridding only in the x and y directions.
> 
> let mask=abc[d=1,gxy=lsm[d=2]]*lsm[d=2]
> 
> If you still need help, we will be able to help better if you write
> back and show what the grids of the variables look like.  After
> opening the datasets, what is the output of 
> 
> 
>    show data
> 
> and what does the output of your list command look like?
> 
> Ansley
> 
> On 10/31/2012 10:06 PM, nitin patil wrote:
> 
> > Dear All, 
> > 
> > 
> > I am calculating standard deviation as follows:
> > 
> > 
> > 
> > 
> > cancel data/all
> > use precip_JJAS_2001-2005_new
> > use lsm_india_mask.nc
> > 
> > 
> > let abc=precip[d=1,l=1:610]*86400
> > 
> > 
> > let mask=abc[d=1,g=lsm[d=2]]*lsm[d=2]
> > 
> > 
> > let mask_value=mask[x=65:95@ave,y=5:38@ave,l=1:610]
> > 
> > 
> > !list mask_value
> > 
> > 
> > let std_dev=mask[x=65:95,y=5:38,l=1:610]
> > 
> > 
> > stat std_dev(sd[l=1])   
> > 
> > 
> > It will show standard deviation for one day but i want to list all
> > 610 days standard deviation values.
> > I am doing list std_dev but it shows only one value which is overall
> > of l=1:610.
> > 
> > 
> > 
> > -- 
> > Regards,
> > Mr. Nitin Patil 
> > PhD Research Scholar,
> > Climate Studies Program,
> > Indian Institute of Technology (IIT) Bombay,
> > Powai, Mumbai - 400076, INDIA.
> > e-mail: nupatil@xxxxxxxxxx
> > 
> > 
>