[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20050328: netCDF image file created by AREA2netCDF



>From: "iliana Genkova" <address@hidden>
>Organization: UIUC
>Keywords: 200503281824.j2SIOWv2028491 AREA2netCDF GOES AREA netCDF

Hi Iliana,

>I recently ordered GOES Sounder cloud top pressure data  (see:
>http://cimss.ssec.wisc.edu/goes/realtime/  ) from NCDC.
>
>Since I am not a MCIDAS user, I used an AREA2netCDF converter
>(http://www.ssec.wisc.edu/visit/netcdf/ ) to read the data.

OK.  The author of the AREA2netCDF Java application is:

Tom Whittaker <address@hidden>

He would be the best contact for information on the netCDF files
created by it; I have CCed him on this reply.

>I ended up with data variables as follow:
>
> version
> sensor_ID    
> image_date    
> image_time    
> start_line    
> start_elem  
> num_linee    
> num_elems    
> data_width    
> line_resolution    
> elem_resolution    
> prefix_size    
> project_number    
> creation_date    
> creation_time    
> bands    
> image    
> 
>
>Could you tell me how to navigate the images - how to calculate lat,lon for
>each pixel given the start_line, start_elem, line_resolution,
>elem_resolution?

I was under the impression that the netCDF file that was created by
AREA2netCDF contained lat,lon values for each image point.  Your
listing of the variables seen in the netCDF, however, says otherwise.

If AREA2netCDF does not write the lat,lon values for image pixels in
the output netCDF, the work needed to do the navigation will be long
and involved since the GVAR navigation model is quite complicated.  The
best advise I can offer you on how to approach this task -- assuming
that the output netCDF does not contain the lat,lon values you need --
is to review the documentation on GVAR navigation contined in the
McIDAS Programmers Reference Manual:

http://www.ssec.wisc.edu/mcidas/doc/prog_man/

>I assume the original area format files contains that info, however the
>netCDF doesn't.

The AREA file contains the information needed for the McIDAS navigation
routines to map image line,element to earth lat,lon locations.  I
will let Tom comment on if the same information is available in
the netCDF output of AREA2netCDF.

>Also, how to interpret the "1"  and  "2346" in the file names?  The rest
>seems to be _yyyy_day_hhmm.area , correct?

The files you attached are not GOES Sounder cloud top pressure data:

goes12_1_2005_001_0015.area
goes12_2346_2005_001_0015.area

I just looked at these in McIDAS and see that the first is a GOES-12
VIS (0.65 um) image.  The '1' in the file name, therefore, represents
the 'band' or channel of the image, and band '1' for GOES-12 is 0.65
um.

I note that this image has no discernable features.  I assume that it
has no features since it is a visible image taken of a nighttime
scene.

The second image contains multiple GOES-12 bands (image channels):

Band: 2  -  3.9 um IR Low Cloud/Fog,Fire Detection
Band: 3  -  6.5 um IR Mid-level Water Vapor
Band: 4  -  10.7 um IR Surface/Cloud-top Temp
Band: 6  -  13.3 um IR CO2, Cloud Heights

The '2346', therefore, indicate that the image contains channels
2, 3, 4 and 6.

I took a look at all of the channels in the multibanded image, and they
all look good.  The combination of the first and second images
represents all of the imager channels that are available from GOES-12.

>Thanks in advance!

Sorry I can not be of more assistance.  Tom Whittaker should be able
to help you.

Cheers,

Tom Yoksas

>---------------
>Iliana Genkova
>Research Associate 
>Department of Atmospheric Sciences
>University of Illinois at Urbana-Champaign
>105 South Gregory Street
>Urbana, IL 61801-3070 USA
>Phone: (217) 333.6349
>Fax: (217) 244.4393
>Email: address@hidden
>http://www.geocities.com/iliana_genkova/index.html
--
NOTE: All email exchanges with Unidata User Support are recorded in the
Unidata inquiry tracking system and then made publicly available
through the web.  If you do not want to have your interactions made
available in this way, you must let us know in each email you send to us.

>From address@hidden  Wed Mar 30 13:37:35 2005

Iliana:

Regarding your question, the lat/lon at each point should have been put into 
the output netcdf file if you specified no command line options (the -AWIPS 
and the -NOLL options will supress this).  Were there any ouptut error 
messages in the console that might have indicated why this was not done?

I cannot get the 'area' files you attached, but would be willing to have a 
look if you would please FTP them to either your server or ours 
(ftp.ssec.wisc.edu/pub/incoming/) and tell me where to pick them up.

In additiona to what Tom Yoksas said, the code for the GVAR navigation is 
readily available in a Java library (the same one Area2NetCDF uses, actually), 
so if your application is in Java, or is Java-enabled (like IDL or Matlab) you 
can use this directly with the AREA file.

tom

--------------------------------------------------
Hi Iliana,

>From address@hidden  Wed Mar 30 14:42:31 2005

Iliana:

Thanks for sending the files.  The problem is that the version of AreaToNetCDF 
  you were using only had navigation through GOES-10.  I have modified the 
code and uploaded the corrected JAR file to the main web page:

<http://www.ssec.wisc.edu/visit/netcdf/AreaToNetCDF.jar>

When you run with it, you will notice that the console output that says 
"Fetching, ..." will now also say "navigating".

Hope this takes care of your problems.  I must add, however, I do not know 
about the multibanded images, since no one else who has used this (to my 
knowledge) has used anything other than single-banded images...

Good luck,

tom

iliana Genkova wrote:

> Hi Tom,
> 
> I run your converter from the MS DOS Prompt with no command line options:
>  
> java -cp AreaToNetCDF.jar AreaToNetCDF goes12_1_2005_030_1815.area
> newtest.nc 
> 
> and I don't get any error messages it seems to me:
> 
> input filename: goes12_1_2005_030_1815.area
> output filename: newtest.nc
> making Gvar nav object...
> making default cal object...
> data present from channel: 1
> determined input cal type is RAW
> Image date/time = 105030 / 181500
> Fetching, calibrating, storing data AREA lines/elements=837 / 1880
> ....................................
> ........................................................................
> ........................................................................
> ........................................................................
> ........................................................................
> ........................................................................
> ........................................................................
> ........................................................................
> ........................................................................
> ........................................................................
> ........................................................................
> ........................................................................
> .........
> NetCDF encoded on 30 Mar 2005 20:58:21 GMT using AreaToNetCDF version 0.96
> 
> 
> I put the area and the netCDF files ( goes12_1_2005_030_1815.area newtest.nc
> )  on ftp.ssec.wisc.edu/pub/incoming/  .
> 
> 
> Thanks,
> Iliana
> 

NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.