[Met_help] MODE problem

[John Halley Gotway]

Jeff,

Sorry for the delay in getting back to you.  I was out of the office last week.

I took a look at the images you sent, and I do see the offset you're describing.  And I also looked at the composite reflectivity field for the 04Z May 8, 2009 case and see how it lines up with the
GRADS image you're producing.  I'm not sure what's going on here.  But I have a couple of thoughts...

(1) You're welcome to create a color table for MODE to use to more closely match the GRADS one you're using.  The easiest thing to is to use the default one but set min/max values for it.  By default,
MODE will scale the colortable to use the min/max data values in the field it's plotting.  So the colortable will change each time you run MODE.  You can manually set the range for the colortable in
the MODE config file using the following parameters:
   fcst_raw_plot_min, fcst_raw_plot_max, obs_raw_plot_min, obs_raw_plot_max
You're also welcome to create a new colortable and point to it using the config file parameters:
   fcst_raw_color_table, obs_raw_color_table

(2) The problem is basically that MODE and GRADS are not rendering the raw data in the same way.  Could we try plotting it with a third tool?  I'd suggest using IDV, a nice plotting tool put out by
Unidata:
http://www.unidata.ucar.edu/software/idv/
If you've never used IDV before, I'd be happy to plot it up with IDV.  But I'd need the raw data files you're using.  If you'd like to send those to me, you can post them to our anonymous FTP site:

ftp ftp.rap.ucar.edu
username = anonymous
password = "your email address"
cd incoming/irap/met_help/duda_data
put "your data files"
bye

It would be nice to take a look at the data files you're using.  It may expose a bug in the MET grid or plotting code.   Or it may be a problem in how the grid it defined in the data.

Thanks,
John

Jeffrey Duda wrote:
> John,
> I am experiencing a problem with MODE.  I am sending you several files to
> show you the problem.  Three of them are Grads images showing 1-hr
> accumulated precip from Stage IV data (observation).  The other three are
> the post script files from MODE for the same days as the three Grads
> images.  The dates are:
> 
> 08 May 2009 (ending at 04 UTC)
> 19 July 2008 (ending at 18 UTC)
> 22 May 2007 (ending at 21 UTC)
> 
> The problem is that MODE is displacing the observed precipitation objects
> some distance away from where they really are.  This can be seen by
> comparing the Grads images for the May cases against the images found on
> page 3 of the post script files.  It seems to be mostly an eastward
> displacement of around 0.5 to 1 degree longitude, and a slight northward
> displacement.  Strangely, this error does not seem to occur in the July case
> (or if it does, it is of such a small magnitude that I can't tell).  I
> believe the Grads images to be correct because I have spent time viewing
> archived reflectivity for these cases (a good source for viewing it would be
> at www.mmm.ucar.edu/imagearchive...then just adjust the parameters to the
> correct year/month/day, and view either Central Plains sector or Northern
> Plains sector for the July case to view the reflectivity), and the
> associated precipitation seems to line up pretty well with the archived
> reflectivity.  Thus, I don't think it's an error on the part of Grads.  My
> adviser Bill Gallus believes it has something to do with the way the
> projections assign various points to each other in MODE.  Please help.  If
> you need any more images or files to show you this, please let me know.
> Thanks.
> 
> Jeff Duda
> 
> 
> 
> ------------------------------------------------------------------------
> 
> 
> ------------------------------------------------------------------------
> 
> 
> ------------------------------------------------------------------------
>