Re: Question of interpolation on missing values.

Since you are passing 1 7 pt filter over the time dimension,
the first three and last three grids will contain all missing values.
All other times should have the gradients calculated correctly.

Goos luck

Yi-Hui Wang wrote:
> Hi all,
> I use a cutoff freqency for 7 day to get low pass filter potential
> vorticity. Then I calculate the gradient of this potential vorticity by
> using the function "gradsg". When I run my script, it shows that "warning:
> gradsg: 6 2-dimensional input array(s) contained missing values. No
> interpolation performed on these arrays." My result looks reasonable even
> though this warning exists. I try to use "linmsg" to interpolate missing
> values before computing their gradient. But the warning still existed and
> the pattern is wrong. My script is attached below.
>
> I would be very appreciated if people can give me some suggestions. Thanks
> a lot,
>
> Yi-Hui
>
> load "/usr/lib/ncarg/ncarg/nclscripts/csm/contributed.ncl"
>
> begin
>
> yr = new((/160/),float)
> yr = 0
> do n = 1,80
> yr(n-1) = n
> yr(79+n) = n + 100
> end do
> Q_sum = new((/152,64,128/),float)
> Q_sum = 0
>
> ;open files of 160 ensemble members
> do i=0,159
> if (yr(i) .lt. 10) then
> name="./ctl_PV_00" + yr(i) + ".nc"
> else
> if (yr(i) .lt. 100) then
> name="./ctl_PV_0" + yr(i) + ".nc"
> else
> name="./ctl_PV_" + yr(i) + ".nc"
> end if
> end if
> f = addfile(name,"r")
> lat = f->lat
> lon = f->lon
> PV = f->PV_total(:,:,:)
>
> ;low pass filter
> nwt = 7
> fca = 1.0/7.0
> ihp = 0
> nsigma = 1.
> wgt = filwgts_lancos(nwt,ihp,fca,-999.,nsigma)
> PVf = wgt_runave_Wrap(PV(lat|:,lon|:,time|:),wgt,0) ;64*128*152
>
> PVf!0="lat"
> PVf!1="lon"
> PVf!2="time"
> PVf_reorder = PVf(time|:,lat|:,lon|:)
> PVf_reorder!0="time"
> PVf_reorder!1="lat"
> PVf_reorder!2="lon"
>
> ;try linmsg, PVf_new replaces PVf_reorder in gradsg
> ;msg = 1e36
> ;PVf_reorder@_FillValue = msg
> ;PVf_new = linmsg(PVf_reorder,-1)
> ;PV_grad_lon = PVf_new
> ;PV_grad_lat = PVf_new
>
> ;calculate the gradient
> PV_grad_lon = PVf_reorder
> PV_grad_lat = PVf_reorder
> gradsg(PVf_reorder,PV_grad_lon,PV_grad_lat)
> PV_grad_lon_at_long_name = "PV longitudinal gradient (derivative)"
> PV_grad_lat_at_long_name = "PV latitudinal gradient (derivative)"
> Q = sqrt((PV_grad_lon(:,:,:))^2 + (PV_grad_lat(:,:,:))^2)
> Q_sum = Q_sum + Q(:,:,:)
> copy_VarCoords(Q,Q_sum)
>
> delete(PV)
> delete(PVf)
> delete(PVf_reorder)
> delete(PV_grad_lon)
> delete(PV_grad_lat)
> delete(Q)
> end do
> ;average Q from all ensemble members
> Q_avg = Q_sum/160
> copy_VarCoords(Q_sum,Q_avg)
> end
>
> Yi-Hui Wang
> Graduate Student
> 2101D Croul Hall, Dept of Earth System Science
> University of California at Irvine
> Irvine, CA 92697-3100
> yihuiw_at_uci.edu
> 949-824-8448
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>

-- 
======================================================
Dennis J. Shea                  tel: 303-497-1361    |
P.O. Box 3000                   fax: 303-497-1333    |
Climate Analysis Section                             |
Climate & Global Dynamics Div.                       |
National Center for Atmospheric Research             |
Boulder, CO  80307                                   |
USA                        email: shea 'at' ucar.edu |
======================================================
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk