[ncl-talk] How can I remove relaxation zone cells and calculate statistics from WRF output

[mmkamal at uwaterloo.ca]

Hi Dennis,

Please correct me if I am wrong: your code will give me an output  
variable with the same number of grid box with FillValue at the  
relaxation zone cells.

I have had a chance to do it slightly other way. Could you please  
check my code whether I have done it correctly or make some mistake:

============================================================
f  = addfile ("/scratch/mkamal/wrfout_d01_2014.nc","r")

;*********************************************************
   conv_prcp       =  f->RAINC(0,:,:)
   non_conv_prcp   =  f->RAINNC(0,:,:)

   conv_prcp       = conv_prcp      > 0.0
   non_conv_prcp   = non_conv_prcp  > 0.0
   total_prcp      = conv_prcp+non_conv_prcp
   printVarSummary(total_prcp)

   x = 224     ; originally 225 grid cell along east-west
   y = 174     ; 175 grid-box along south-north

;--------- Remove relaxation zone cells

  total_prcp_no_rz_cell = total_prcp(9:(y-11),9:(x-11))

  printVarSummary(total_prcp_no_rz_cell)

;--------- assing missing values

  total_prcp_no_rz_cell at _FillValue = -999

;--------- Compute average precipitation

  dom_avg_prcp  = avg(total_prcp_no_rz_cell)

  printVarSummary(dom_avg_prcp)

  print(dom_avg_prcp)

==============================================

Thanks
Kamal


Quoting Dennis Shea <shea at ucar.edu>:

> Try something like the following.
>
> x(NY,NX)
>  x at _FillValue
>
> nx = 10    ; allow for different 'relax boundaries'
> ny = 10
>
> x(0:ny-1,0:nx-1) = x at _FillValue
>
> nxStrt = NX-nx
> nxLast = NX-1
> nyStrt = NY-ny
> nyLast = NY-1
> x(nyStrt:nyLast,nxStrt:nxLast) = x at _FillValue
>
> Use
>
> print(x)
>
> or plot to debug
> =============
>
>
> Better: make it into a function and place into your personal libray
>
>
> On Fri, May 8, 2015 at 11:11 AM, <mmkamal at uwaterloo.ca> wrote:
>
>> Hi,
>>
>> I want to calculate statistics except the relaxation zone cells
>> (outermost 10 grid cells) from WRF output. Any help will be greatly
>> appreciated.
>>
>>
>> Thanks
>> Kamal
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> ncl-talk mailing list
>> List instructions, subscriber options, unsubscribe:
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>
>