Re: http://www.ssec.wisc.edu/~dglo/visad/ mcidas meteosat navigation

Hi Garrett-

"Campbell, Garrett" wrote:
> 
> I am attempting to use the mcidas meteosat navigation module
> 
> The documentation is obscure.

The best source for documentation is the McIDAS programmer's
reference which documents the MSAT nav module.  There is a link
from the javadocs for MSATnav.
 
> 1. Sometimes Mcidas lists longitude positive to the west.

The convention in VisAD is positive east longitude which is
opposite of McIDAS and VisAD.  The *nav routines are java
conversions of the McIDAS FORTRAN nav modules.  Since the
McIDAS transformations use positive west longitude, all
incoming longitudes are multiplied by -1 before the transforms
are done.

> 2. line and elemement numbers are interchangable in some sence since the
> Meteosat 7
>         satellite is at 0 longitude.

I don't think this is true.  The line/element numbers refer
to the file (AREA) coordinates.

> 3. Line numbers could count from south to north or north to south.
> Similarly element
>         numbers could count either increasing to the west (as the
> satellite scans) or
>         increasing to the east (as a picture would show).

In MSATnav, area ele/lin (0,0) is the first pixel in the (upperleft) 
corner of the image which I believe is the McIDAS convention.
AREACoordinateSystem has 0,0 in the lower left and sets the
flipped line coordinates flag in the nav module. 
 
> Has anyone used edu.wisc.ssec.mcidas.MSATnav?

Yes, through AREACoordinateSystem.

> What convention do you use for latitude and longitude/  line and element
> in the routines calls to toLatLon and toLinEle.

the default transforms are (element, line) <-> (Latitude, Longitude). 
Longitude is in degrees East.  line and element are
in file (AREA) coordinates not image coordinates.  You can transform
from file to image coordinates using the imageCoordToAreaCoord
and areaCoordToImageCoord methods.  

I did the conversion from McIDAS FORTRAN to VisAD Java, so
if you have a specific question, let me know.  The code in
the toLinEle and toLatLon methods is as well documented as
the corresponding FORTRAN code (which is pretty much 
non-existant).

Don
*************************************************************
Don Murray                               UCAR Unidata Program
dmurray@xxxxxxxxxxxxxxxx                        P.O. Box 3000
(303) 497-8628                              Boulder, CO 80307
http://www.unidata.ucar.edu/staff/donm
*************************************************************


  • 2001 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the visad archives: