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

20020410: using McIDAS to make METEOSAT/GOES-8 composite

>From: "Robert Mullenax" <address@hidden>
>Organization: NSBF/NMSU
>Keywords: 200204102048.g3AKmta26322 McIDAS IMGREMAP


>I am needing to create a METEOSAT/GOES-8 composite over the SH,
>although I would like to do NH first as a test.

OK, this shouldn't be too hard.

>I wonder if
>you might could give me a list of steps to accomplish this.

The procedure that you will need to follow are basicicaly the same
as are used in the McBASI scripts GEWCOMP.MCB and GEWCOMP1.MCB that
are included in the Unidata McIDAS distribution.

GEWCOMP.MCB used the MERGE option of IMGREMAP to do the compositing.
This has the benefit of being quick and easy, but the drawback
that image values from the second image will be used anywhere
they are available.

GEWCOMP1.MCB, on the other hand, does its compositing by using only
GOES-West images west of a certain longitude, and GOES-East images
east of the same longitude.  This has the effect of gluing together
two remapped displays, each of which is half of the final composite.

>I have GOES-8 full disk and Meteosat Full disk imagery.


>I would like the result to cover the area from around 85W to 20E

This would mean that the center point of the combined image should
be at about 28W.

>from 30N to 60N in a lambert conformal projection.

So this is for the NH test?  Otherwise your longitudes should be 30S to

You have two ways you could do this:

o composite images from both satellites and then reproject them
  into the desired sector

o create the desired sector and then use it for a remap target

Right off of the top of my head, I would say that the second
approach should yield an image that is more representative of the
original images than the first techhnique.  The reason for this
is that the first technique will involve remapping image values
twice while the second will involve remapping only once.

>Just a general outline is fine, I don't need all of the command
>specifics.  I am familiar with the basic process as I have created
>the Goes E-W composites before.

OK, here goes:

1) make a copy of your METEOSAT image

2) change the subsatellite longitude in the copy to 28W.  Header
   word 70 in METEOSAT images in AREA file format contains the
   subsatellite longitude*10000.  Use the McIDAS command LWU
   to change this value:

   LWU POKE AREA1234 280000 70     <- AREA1234 is the METEOSAT image copy

   NB: West longitudes in McIDAS are positive!

3) now that you have an image centered on 28W, create the sector you
   want using IMGREMAP.  This will look something like:


   You will have to play with the the size (SIZE=l e, where l == #lines
   and e == #elements) and the resolution (RES=r where r is the
   resolution) of the output image to get what you want.  My approach
   is to fix the output size and then play with the resolution to
   refine the image.

   Once you have the sector you want, make a copy of it:


4) probably the best way for you to see if the sector you have created is
   what you want is to remap one of the input images into that projection
   and then look at it:

   If this is not what you like, go back to 3) and try again

After getting the output sector you are looking for, you should then
copy GEWCOMP1.MCB into a new McBASI script, GWMETCMP.MCB, and then
edit the script to change entries to match your project.  The
big thing to look out for is the IMGFILT invocations as they are
setup to create filter masks that of values of 0 (zero) for one
half of the image and 255 for the other half.  These filter mask
images are then used to mask the remap copies of the GOES-West and
GOES-East images to "blank out" the portion of the image East or
West of a particular longitude, respectively.

I think that if you play with GEWCOMP1.MCB, you will get a feel for
what it is doing -- especially if you comment out the deletion of
intermediate images.  Here is the GEWCOMP1.MCB script that I ran
to refamiliarize myself with the East/West compositing procedure:


Please let me know if anything above is too big of a mystery.


>From address@hidden Wed Apr 10 16:17:34 2002
>Subject: Re:20020410: using McIDAS to make METEOSAT/GOES-8 composite  

I will digest it once I get home, but I think I understand.
I am going to do the NH as a test first as I know what
that should "look" like.  

Thanks as always for the help.


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.