20000719: GINI to netCDF (cont.)

>From: David Harper <address@hidden>
>Organization: NCAR/RAP
>Keywords: 200003012325.QAA04248 McIDAS-X GINI netCDF

Dave,

>I grabbed an image from 10:00 (local) this morning.  I don't have access to
>the program we were using to look at the images, so I'll have to assume it is
>the same as the one Lauri was looking at.
>
>I have it set up in the mcidas account on 4dwx-dev.  There is a .gini file
>in the correct location.

???

>The output is a netCDF file named GINI0001.nc in
>the workdata directory.  The script I use is vis_convert.sh.  It also might
>be interesting to ingest the netCDF into your version of Mcidas to see if you
>see the same dislocation.

The only problem with this is I have no way of looking at the image in
netCDF format.  McIDAS only supports output to the netCDF form.  I have
to believe that this _should_ be OK since the code was written for use
by AWIPS (but you never know).

>Anyway it should give you a gini file straight from the NOAAport and
>the netCDF I use to go into Zebra.

OK, so Zebra is viewing the McIDAS-generated netCDF directly?

>I also have a converted netCDF file,
>but I assume mcidas won't like the version of netCDF I have to use for Zebra.

Are you converting the McIDAS-generated netCDF to some other netCDF?

>I do not have the imager.

???

I ran a couple/three quick tests with GINI GOES-West VIS images:

o loaded a GW1KVIS GINI image we have access to in McIDAS in the same way
  that you are doing the copy in vis_convert.sh:

  IMGDISP RTGINI/GW1KVIS LATLON=33.1 106

  I get the LAT,LON of the center of white sands to be approx 32:53N,
  106:18W.

o I created a local dataset in AREA file format on 4dwx-dev:

  cd workdata
  dsserve.k ADD MYDATA/IMAGES AREA 1 9999 \"Local images in AREA format

o I then copied a GW1KVIS GINI image to this local dataset:

  imgcopy.k RTGINI/GW1KVIS MYDATA/IMAGES.1234 LATLON=33.1 106 SIZE=800 800

o I FTPed this AREA file (AREA1234) back to my machine and displayed it
  in a frame:

  IMGDISP MYDATA/IMAGES.1234 REFRESH='EG;MAP H 9'

o I then interrogated this image to see where the white sands are depicted
  to be and I again got approx. 32:53N, 106:18W.

This agrees well with your follow-up email that said that the white sand
blob should be centered at about 32.9N, 106W (32:54N, 106W).

This demonstrates to me that the GINI server that I wrote is spitting
back reasonably navigated images.

The question now is whether or not the netCDF output image is somehow
screwed-up, or if the display of the image is screwed up.  We will have
to look at this more tomorrow as I have to leave now.

>From address@hidden Wed Jul 19 17:53:23 2000

>References: The white sand should be a white blob centered at about 32.9N 106W
>a dark flow lava should also appear North of that angling to the North-East.

>I really appreciate the time, it would take me many times as long to verify
>this,

Perhaps we should get together to take a look at the display you have.
Send me email tomorrow with a good time (not lunch!).

Tom