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

[McIDAS #ZXX-621494]: New Research Approach

Hi Rod,

re:
> See my response...
> 
> So the modifications that need to be made for the batch file should be:
> 
> IMGCOPY G13/IR27APR11 LOC/AREAS.3003 TIME=20:25 LAT=30:00 86:30 SIZE=320 320 
> (PRO=MERC RES=6) Should be added?
> IMGCOPY LOC/AREAS.3003 LOC/AREAS.3103 SIZE=SAME
> IMGREMAP G13/IR27APR11 LOC/AREAS.3103 TIME=20:25
> IMGCHA LOC/AREAS.3103 BAND=4 SS=180 STYPE=VISR
> IMGDISP LOC/AREAS.3203 (Trying to first display the image to be probed using 
> IMGDISP)
> IMGFILT LOC/AREAS.3003 LOC/AREAS.3103 LOC/AREAS.3203 FILTER=MERGE 185 255 0
> IMGPROBE STAT BOX OUTLINE=YES BAND=4 EXCLUDE= 234-320 DATASET=LOC/AREAS.3203

I suggest that you run each McIDAS command individually/ examine the
result; and when you are satisfied with the result, include it in a
new BATCH file.

re: Your listing of the commands in a new BATCH file left out the creation
of the first of three images.

> my response: How can this be resolved then?

The IMGCOPY invocation I sent in the previous email creates the image.
Please try running the command from the McIDAS command line and then
examine/review the result.  What I mean by this is:

- first run the example IMGCOPY command

IMGCOPY G13/IR27APR11 LOC/AREAS.3003 TIME=20:25 LAT=30:00 86:30

  You know by now that will create an image in LOC/AREAS.3003.  Quite
  some time ago I explained that because of how LOC/AREAS is defined,
  there is a one-to-one correspondence between ADDE dataset member
  and AREA file written to disk:

  LOC/AREAS.3003  ==   AREA3000

- examine the result of the IMGCOPY

  I would do this by displaying the result:

SF 1
ERASE
IMGDISP LOC/AREAS.3003 EU=IMAGE
MAP H

re:
> I am trying to create a
> third image with the filtered IR temps below 233 K and not include the
> effective radius values in that position.

OK, you are now defining what you want... The first thing you have to do
is figure out what the BRIT values that correspond to TEMP values below
233K are.  One way of doing this is to display one of the IR images
and then use the IMGPROBE command to list out the value of individual
pixels:

ERASE
IMGDISP G13/IR27APR11 TIME=20:25 LAT=30:00 86:30
IMGPROBE

IMGPROBE run this way (i.e., without specifying MODE=N) is interactive -- it
will tell you to move the mouse cursor over the displayed satellite image and
then press the right mouse button to list out the value under the cursor.
Continually do this until you find pixels that are 233K and lower and you will
see what BRIT values these correspond to.  Those BRIT values will then be
used in an IMGFILT invocation that composites the IR and Effective radius
images into your third image.

re:
> This third filtered image will then be extracted using AXFORM.

Caution: The third image created from the combination of IR and Effective
Radius images will not have a consistent calibration (meaning that
there will not be a consistent mapping between displayed value
and TEMP).

re:
> I dont want to include effective radius values in the statistics
> calculation of where the IR temps are above 233 K.

IMGFILT can be used to create a composite of the IR and effective radius
images with the conditions you are talking about.  What can be done
with that image using AXFORM is in question for me.

re:
> Oh the region of where the box should be drawn or statistics measured
> where IR temp pixels below 233 K and to include the corresponding
> effective radius values in the statistics.

> Upper Left of the box: 34.6640211, -88.133850
> Upper Right of the box: 34.661209, -87.323639
> Lower Left of the box: 34.171467, -88.191185
> Lower Right of the box: 34.190990, -87.358696

I understand that you have figured out the region you are interested 
in and know what you want to do.  What I am concerned about is how to
included effective radius values _correctly_ in statistics for an image
that contains both IR and effective radius values.  The reason that 
I am concerned is that the mapping of BRIT to TEMP for the effective
radius pixels is likely different (I don't know this,. I am assuming it.
Dan Lindsey should be able to comment on it authoritatively) from the BRIT
to TEMP mapping for the pixels from the IR image.  A different approach
would be to do the statistics for the IR pixels in one IMGPROBE invocation
(specifying an EXCLUDE= range that eliminates the effective radius pixels)
and do the statistics for the effective radius pixels in another IMGPROBE
invocation (specifying an EXCLUDE= range that eliminates the IR pixels).

re:
> For 20:15 Sat/Effective radius APR 27 2011. For example how would this be 
> done?

You have an IMGFILT example that you can use to guide you in coming up
with an  appropriate invocation that uses the DISC= keyword.  Please be
adventurous and experiment with IMGFILT.  After you have given it a try,
I will be happy to give you my opinions on what is the right or wrong way
of doing IMGFILT image combinations.

One additional comment:  the example EFFRAD.BAT file that I sent you
included a lot of explanation for what was being done and why.  My
comment at the time was that if you work your way through the steps
in this BATCH file and understand what was being done, then you would
be better able to do the things you want/need.  Please revisit this
BATCH file and let me know what puzzles you about what is being
done and why.

Cheers,

Tom
--
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: ZXX-621494
Department: Support McIDAS
Priority: Normal
Status: Closed