[Wrf-users] shift is less apparent in Ferret

Rob Cermak cermak at sfos.uaf.edu
Wed Aug 30 17:12:06 MDT 2006


I just had to follow up and see what all this stuff looks like in
Ferret to get another independent view.

Things look much better in Ferret.  You could argue there is a
slight shift going on.  Different software, different results,
different coastlines.

NCL scripts in WRFSI:
  http://137.229.40.27/data/forecasts/WRF/avc.d01.jpg

Via Mapserver/Opendap:
  http://ak.aoos.org/data/tmp/wrf-hgt-pws.png
  * We are still working out the fine details of displaying this
    so an offset is not surprising at this stage.

The NCL versions are shifted South and East, the mapserver ones are
shifted North and West.

This immediately told me that things should be looked at with maybe
another independent package: Ferret.

use
"http://137.229.40.27/dap/data/forecasts/WRF/PWS/AKPWS4/AVN/wrfout_d01_2006-08-21_12:00:00"
! 1
set region/L=1
shade HGT,xlong,xlat
go land_detail black overlay
frame/file=ferret_wrf_hgt.gif

! 2
set region/L=1
shade XLAND,xlong,xlat
go land_detail black overlay
frame/file=ferret_wrf_xland.gif

cancel data/all

! 3
use "http://137.229.40.27/dap/data/wrf/runs/Alaska/static.wrfsi.d01.nc"
set region/L=1/Z=1
shade avc,lon,lat
go land_detail black overlay
frame/file=ferret_wrf_avc.gif

Ferret/Opendap:  From the static file
  http://ak.aoos.org/data/tmp/ferret_wrf_avc.gif

Ferret/Opendap:  Plot after the run
  http://ak.aoos.org/data/tmp/ferret_wrf_hgt.gif

One could argue that there could still be a slight shift in the
XLAND plot from Ferret, but it is less noticeable than the first two
attempts.  There is a little shifty business going on between the
static file and the map from the output file.  Scales are not the
same, but the color contours from avc -> hgt  shift to the North and
East.  The scales of the graphic do shift as well so it is not an
absolutely solid example.  Maybe I need to get a stronger
prescription for my glasses.

http://ak.aoos.org/data/tmp/ferret_wrf_xland.gif

Back to it!  Bug or no bug... can't really spend anymore time on it.

Rob
-- 
Alaska Ocean Observing System
Database Manager
907-474-7948





More information about the Wrf-users mailing list