[ncl-talk] Vector subscripting (Resending this email. Please respond)

[Dennis Shea]

[1] There is no indication that you sent any previous email with this topic.

[2]  I am not clear on what is desired. Maybe:


  ntim = 10000
  klev = 50
  nrad = 75

  data = random_uniform(0, 500, (/ntim,klev,nrad/) )   ; dimension #: (0,
1, 2)
  data at _FillValue = -999.0

  data!0 = "time"
  data!1 = "level"
  data!2 = "radii"

  rmin  = 100.
  rmax  = 250.

  data_rmin  = where(data.lt.rmin, data, data at _FillValue)
  data_rmax  = where(data.gt.rmax, data, data at _FillValue)
  data_rmnx  = where(data.lt.rmin .or. data.gt.rmax, data at _FillValue, data)

  copy_VarMeta(data, data_rmin)
  copy_VarMeta(data, data_rmax)
  copy_VarMeta(data, data_rmnx)

  ndim  = 2  ; radii dimension

  data_avg_rmin = dim_avg_n_Wrap(data_rmin, ndim)
  data_avg_rmin at long_name = "Average of all radii < "+rmin
  printVarSummary(data_avg_rmin)
  printMinMax(data_avg_rmin,0)
  print("---")

  data_avg_rmax = dim_avg_n_Wrap(data_rmax, ndim)
  data_avg_rmax at long_name = "Average of all radii > "+rmax
  printVarSummary(data_avg_rmax)
  printMinMax(data_avg_rmax,0)
  print("---")

  data_avg_rmnx = dim_avg_n_Wrap(data_rmnx, ndim)
  data_avg_rmnx at long_name = "Average of all values between rmin= "+rmin+"
and  rmax="+rmax
  printVarSummary(data_avg_rmnx)
  printMinMax(data_avg_rmnx,0)
  print("---")

++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++

Note that *clarity is best* but you could avoid the creation of a temporary
array via:

data_avg_rmin = dim_avg_n_Wrap(where(data.lt.rmin, data, data at _FillValue),
ndim)

etc

On Wed, May 9, 2018 at 8:23 AM, Prashanth Bhalachandran <
prashanth.bhalachandran at gmail.com> wrote:

> Dear NCL’ers,
> I have a 3 dimensional variable say M (time x level x radius). Now, I know
> the certain radius RMW at each level (RMW is a function of level) . For
> each time, I’d to average the variable within the radius RMW at each
> height. I want to avoid loops, so I need to use vector subscripting. How do
> I do this averaging?
>
> If RMW was a fixed number like 30, then I’d have simply done a
> dim_avg_n_Wrap( M (:,:,1:30), 2).
>
> Can someone help me out with this?
>
> Thank you,
> Prashanth
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20180509/408b9694/attachment.html>