Re: NGSETI error and vector plots and vcMinMagnitudeF

Hi Dave,
Thanks for calling this to our attention.
I will look into a proper fix for this problem.
  -dave

On Nov 4, 2008, at 1:11 PM, Dave Allured wrote:

> This is a low priority request to improve error messaging from NCL
> vector (wind arrow) plots under the following error condition.
>
> When all vectors are accidentally suppressed, NCL prints this
> undocumented error message, followed by several other strange
> messages, then halts:
>
> ERROR 2 IN NGSETI - Parameter value out of range for CT
>
> I recently found this difficult to debug, but finally boiled it
> down to the following. The resource vcMinMagnitudeF is used with
> vector plots to suppress small arrows and show only the larger
> ones. See attached demo script, and change vcMinMagnitudeF from 15
> to 45 to see the error.
>
> When vcMinMagnitudeF is greater than the largest vector within the
> plot frame, all arrows are suppressed. Apparently this is
> interpreted internally as a "zero field" error, rather than just a
> blank vector plot, resulting in the non-obvious messages.
>
> To further complicate diagnosis, the message "zero field" is
> plotted in large font across the plot. But a check of the vector
> data finds that it is all non-zero. In my own case, "zero field"
> was printed in white over a white background because of color map
> issues, of course rendering the message invisible.
>
> So I suggest some error message improvements along these lines:
>
> 1. In a zero field condition, print "zero field", or better "zero
> vector field" to the console, in addition to the plot overlay.
>
> 2. If vcMinMagnitudeF is greater than zero, then print a
> supplemental message similar to:
>
> "All vectors were suppressed, vcMinMagnitudeF = 45.0,
> try a smaller value."
>
> 3. Make this error use standard NCL error reporting, where the
> execution "stack" is displayed like for other NCL errors. For each
> calling level, show the function name and the error line number.
>
> My version info: ncl.ppc.5.0.1 pre-release binary
> File date = 2008-may-07 (may have been changed from earlier)
>
> uname -a = Darwin mac56.cdc.noaa.gov 9.5.0 Darwin Kernel
> Version 9.5.0: Wed Sep 3 11:31:44 PDT 2008;
> root:xnu-1228.7.58~1/RELEASE_PPC Power Macintosh
> powerpc PowerMac7,3 Darwin
>
> Thanks for your consideration.
>
> Dave Allured
> CU/CIRES Climate Diagnostics Center (CDC)
> http://cires.colorado.edu/science/centers/cdc/
> NOAA/ESRL/PSD, Climate Analysis Branch (CAB)
> http://www.cdc.noaa.gov/
> ; vector_3.ncl -- NCL example program
> ; Modified 2008-nov-04 to demo error problems when plotting.
>
> ; Data file available at: http://www.ncl.ucar.edu/Applications/Data/
>
> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_code.ncl"
> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_csm.ncl"
>
> begin
> a = addfile("uv300.nc","r") ; open data file
> u = a->U(1,:,:) ; read in zonal [u] and meridional [v]
> winds
> v = a->V(1,:,:)
>
> wks = gsn_open_wks("x11","vector") ; open plot file
> or window
>
> res = True
> res_at_vcRefMagnitudeF = 50.0 ; reference vector
> magnitude
> res_at_vcRefLengthF = 0.06 ; plot length of
> ref. vector
> res_at_vcMinDistanceF = 0.05 ; thin out the
> vectors
>
> res_at_vcMinMagnitudeF = 15.0 ; blank vectors below this
> magnitude;
> ; 45 causes zero field error
>
> plot = gsn_csm_vector_map_ce(wks,u,v,res) ; make plot
> end
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Tue Nov 04 2008 - 14:04:54 MST