[ncl-talk] Divide by zero error using "pot_vort_isobaric" function

[Kyle Griffin]

Hi Daniel,

OK, I finally read your email in all the detail I should have the first
time and found what should likely be an easy fix (this is the 3rd time I've
written this email - probably shouldn't reply to the list at this hour!).

You're reading a version in from shea_util.ncl, which is an older, out of
date, and unsupported version of pot_vort_isobaric (anything in
shea_util.ncl is unsupported, technically - and note the file and line
numbers in your error message). The safe guards for static stability that
you've likely seen Dennis and I discuss in previous posts have been fully
implemented in the contributed.ncl version that became officially supported
in the 6.3.0 release. I suggest you either comment out your load of
shea_util.ncl (if possible) or ensure that you load contributed.ncl after
you load shea_util.ncl - this should ensure the older function is redefined
when you load contributed.ncl.

Hope that is an easy fix and that your calculations will work now.


Kyle

On Wed, Mar 9, 2016 at 6:57 PM Daniel Swain <danielswain2 at gmail.com> wrote:

> Hello,
>
> I'm currently attempting to use the built-in NCL function to calculate
> potential vorticity ("pot_vort_isobaric"). Unfortunately, I am repeatedly
> getting a divide by zero error:
>
> "
> fatal:divide: Division by 0, Can't continue
> fatal:Div: operator failed, can't continue
> fatal:["Execute.c":8575]:Execute: Error occurred at or near line 3892 in
> file $NCARG_ROOT/lib/ncarg/nclscripts/csm/shea_util.ncl
> fatal:["Execute.c":8575]:Execute: Error occurred at or near line 3909 in
> file $NCARG_ROOT/lib/ncarg/nclscripts/csm/shea_util.ncl
> "
>
> I have confirmed that my original units are in hPa (thus the *100 hPa-> Pa
> conversion should be correct), and that the original latitudes were
> north->south (so the lat flip should also be correct). I'm assuming this
> occurring during the static stability calculation (as mentioned in previous
> ncl-talk discussions), but I'm not sure whether there has yet been a
> resolution.
>
> I've attempted this with data from different years, but eventually a zero
> division error always crops up. A short sample script is below, and
> representative data files have been uploaded to the FTP site with the same
> names as below.
>
> I'm currently using NCL version 6.3.0.
>
> Thanks very much for any advice!
>
> ~Daniel
>
> "
> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_code.ncl"
> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_csm.ncl"
> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl"
> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/shea_util.ncl"
>
> begin
>
> infile1 = addfile ("air.2013.nc", "r")
> infile2 = addfile ("uwnd.2013.nc", "r")
> infile3 = addfile ("vwnd.2013.nc", "r")
>
>      U   = infile2->uwnd   ; (time,lev,lat,lon)
>      V   = infile3->vwnd
>      T   = infile1->air    ; K
>
>      lev = T&level  ; hPa
>      lev = lev*100 ; convert units
>      lev at units = "Pa"
>
>      U   = U(:,:,::-1,:)    ; reorder to South -> North
>      V   = V(:,:,::-1,:)
>      T   = T(:,:,::-1,:)
>
>      lat = T&lat
>      gridType  = 0          ; Gaussian grid
>      opt = 0
>      pvort = pot_vort_isobaric(lev,U,V,T,lat, gridType, opt)
>
> outfile1=addfile("pvort.2013.nc","c")
> outfile1->pvort = pvort
>
> end
>
> "
> _______________________________________________
> 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/20160310/3fd0bcc5/attachment.html