20030605: IMG* use of site-developed MSG calibration module (cont.)

>From: "=?ISO-8859-1?Q?Marianne=20K=F6nig?=" <address@hidden>
>Organization:  EUMETSAT
>Keywords:  200306040725.h547PDLd003216 McIDAS-X calibration

Hi Marianne,

Sorry I didn't respond to this last week, but I failed to see your
question about BRIT until a couple of minutes ago.

re: Edgewater reservations for MUG meeting
>you should SOON make your reservation, they will be booked out
>otherwise (there is a group resevation number which is 6041).

OK.  I have never made reservations this early for a MUG meeting.  I
tend to wait until near the deadline and then book, but this time I
will follow your recommendation and book early.

>About this mcidas thing:
>My cal module is kbxmsg.dlm (which was originally included in the 2002
>distribution,and I simply modified it).

Given that kbxmsg.dlm is standard in McIDAS, it should be used when you
do your IMGPROBE.

>So it is linked into the code
>(my server uses that as well, and that is linked vs. the mcidas lib).

OK.

>Do I get your last comment right, that the cal module should also do
>the conversion to BRIT?

Yes.  The calibration module is the code that is responsible for
converting between possible values of the data.  Since the data is RAW,
this would include BRIT and, if the image is an IR one, TEMP (and
possibly other things).  When things are working correctly, an IMGPROBE
will list back all of the valid units for an image.

Here is a bit of the code at the top of the McIDAS-X v2003 version of
kbxmsg.dlm (the KBXINI intreface) that shows the units that can be
returned:

C   Calibration for Meteosat second generation (MSH)

      INTEGER FUNCTION KBXINI(CIN,COUT,IOPT)
      IMPLICIT NONE
      CHARACTER*4 CIN
      CHARACTER*4 COUT
      INTEGER IOPT(*)

      ! symbolic constants & shared data

      INCLUDE 'areaparm.inc'

      INTEGER ITYPE
      INTEGER JTYPE
      INTEGER JOPT(NUMAREAOPTIONS)
      INTEGER CALFLG
      INTEGER CALARR(128)
      COMMON/MSGCOM/ITYPE,JTYPE,JOPT,CALFLG,CALARR

      ! external functions

      ! local variables

C
      CALL MOVW(NUMAREAOPTIONS,IOPT,JOPT)
      ITYPE=0
      CALFLG = 0
      if(CIN.eq.'RAW'.and. COUT.eq.'BRIT') ITYPE=1
      if(CIN.eq.'RAW'.and. COUT.eq.'RAD ') ITYPE=2
      if(CIN.eq.'RAW'.and. COUT.eq.'REFL') ITYPE=2
      if(CIN.eq.'RAW'.and. COUT.eq.'TEMP') ITYPE=3
      if(ITYPE.eq.0) GOTO 900
      KBXINI=0
      RETURN
900   CONTINUE
      KBXINI=-1
      RETURN
      END

>I have not included that yet, but then, it does
>not do temperatures and radiances either.

So, the version of kbxmsg.dlm you are using is either the one that
you wrote, or the one in the distribution that has been signifiantly
modified.   In this case, the module may be missing the conversions
that are possible.

>I will do a few tests using the "old" AREA routines like araopt etc. to
>see what is going on.
>
>Thank you for your help.

Sorry I can't be of more help!

Tom

>From address@hidden Tue Jun 10 02:25:49 2003

Thanks. That might give me the right hints. I will follow it up.

Have a good day!

Marianne