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

[McIDAS #OKG-462913]: Color Bar and Units when channel differencing



Hi Greg,

I finally got some time to revisit the color bar labeling issue for "Fog"
products created using:

Fog = 150 + 10(T10.7 - T3.9)

Or, in IMGOPER terms:

IMGOPER RTIMAGES/GE-IR RTIMAGES/GE-39 MYDATA/IMAGES.3000 FUNC=ADD COEF=1 -1 
ACON=150 MCON=10 UNIT=TEMP

where:

RTIMAGES/GE-IR -- GOES-EAST 10.7um infrared
RTIMAGES/GE-39 -- GOES-East  3.9um infrared

Since IMGOPER is not producing a correct calibration block for this channel
difference, I performed "calibrations" using sets of images for diverse times
of the day (e.g., 20081026.1945, 20071027.0945, 20071027 1415):

- chose points in the difference product that represented large ranges in
  brightness

- interrogated the original 10.7um and 3.9um images to get the temperatures at
  the same geographical locations as in the difference product

The three images I chose this morning showed the _exact_ same equivalence
between output brighness in the difference product and temperature difference
in the input products.  Representative numbers (for 20071027.1415) are:

  10.7um         3.9 um        Difference
BRIT  TEMP     BRIT  TEMP     BRIT  del(T)
----  -----    ----  -----    ----  ------
 183  235.0     170  245.0      50   -10.0
  76  292.0      69  295.5     115    -3.5
  75  292.5      60  295.5     120    -3.0
  75  292.5      70  295.0     125    -2.5
  94  283.0      91  284.5     135    -1.5
  69  295.5      69  295.0     150     0.0
 147  256.5     151  254.5     170    +2.0
 105  277.5     115  272.5     200    +5.0 

This relationship is represented by the following formula:

del(T) = (5/50) BRIT - 15

You can generate a McIDAS stretch table that reflects this relationship
as follows:

<in a McIDAS session>

SU INI IRDELT PRD TEMP
SU MAKE IRDELT -10 5 50 200

You would use this stretch to annotate your resultant image as follows:

SF 1
ERASE
IMGDISP MYDATA/IMAGES.3000 LAT=0 72 MAG=-4 EU=IMAGE
BAR SU=IRDELT LINT=1

NOTEs:

- it is unknown to me if this relationship is valid for platforms other
  than GOES-East (GOES-12).

  My gut feeling is that it should be for GOES-West (GOES-11) and GOES-South
  (GOES-10) since the calibrations for the IR channels are identical for these
  three platforms.

- I will be looking at the source code for IMGPROBE (imgprobe.pgm) to see
  if I can figure out why a usable/correct PRD calibration is not being
  created.

Please let me know if you have questions on anything listed above.

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: OKG-462913
Department: Support McIDAS
Priority: Normal
Status: Closed