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

20001207: IMGREMAP: MAG= vs RES= (cont.)



>From: Jason Allard <address@hidden>
>Organization: PSU
>Keywords: 200010240345.e9O3jC425238 McIDAS-X IMGREMAP MAG RES

Jason,

Sorry it took so long for me to get some time to address your questions.

re: change magnifiation

>Ahh... okay. How about this... the original image has a resolution of 8km
>for the latitude and 16km for the longitude.  If there a way to change the
>longitude resolution to 8km also?

You have to be careful about interpreting resolutions from the McIDAS
IMGLIST command.  The values listed under the Res (km) columns are
actually the multiplicative factors for the base resolution of the
instrument doing the scanning.  This is and has always been confusing
for those getting their feet wet in using McIDAS, and SSEC finally
realized it.  It is slated to be changed to be more like what folks are
looking for.

In GOES-8, -9, -10, and -11, the primary resolution of the base
instrument is 1 km at the equator along a line of longitude (i.e., in
the latitude direction) and 0.57 of that at the equator along a line of
latitude.  The instrument oversamples in the longitude directory (along
a line of constant latitude) by almost a factor of 2 (1/0.57).

The Res numbers that typically come back for images sent in the
Unidata-Wisconsin datastream will indicate something like:

IMGLIST RTIMAGES/GE-IR FORM=EXP
Image file directory listing for:RTIMAGES/GE-IR
 Pos Satellite/         Date       Time      Center      Res (km)   Image_Size
     sensor                                 Lat  Lon    Lat   Lon
 --- -------------  ------------  --------  ---- ----  ----- ----- ------------
   6  G-8 IMG       12 DEC 00347  02:15:00    23   71
    Band: 4  10.7 um Surface temp; longwave window       4.0   8.0  1400 x 1728
     proj:    0 created: 2000347  23421  memo: RT GVAR
     type:VISR     cal type:BRIT
     offsets:  data=    2816 navigation=  256 calibration=    0 auxillary=    0
     doc length:   0   cal length:   0   lev length:   0 PREFIX=   0
     valcod:          0 zcor:  0 avg-smp: N
     start yyddd: 2000347  start time: 21515  start scan:  351
     lcor: 2785  ecor:  9141  bytes per pixel: 1  ss: 70
     Image Center Point Res (derived)  Lat:   4.59 (km)  Lon:   4.67 (km)

IMGLIST: done

The reason for this is that they are created with an IMGCOPY command that
specifies a MAG= keyword sequence of MAG=1 -2:

IMGLIST RTIMAGES/GE-IR FORM=AUDIT
Image file directory listing for:RTIMAGES/GE-IR
 Pos Satellite/         Date       Time      Center   Band(s)
     sensor                                 Lat  Lon
 --- -------------  ------------  --------  ---- ---- ------------
   6  G-8 IMG       12 DEC 00347  02:15:00    23   71 4
Date   Time         Audit trail
----- ------ ----------------------------
00347  23421 IMGCOPY EASTL/NH UNIDATA/UI BAND=4 SIZE=1400 1728 MAG=X -2 STYPE=VI
             SR CYCLE=1:00 BAND=4 LATLON=23 71 PLACE=CENTER DAY=2000347 TIME=2:1
             4 2:17

IMGLIST: done

This might lead one to believe that the resolution was different in the
Lat and Lon directions when they are actually very close to being the
same (i.e., 8 * 0.57 ~= 4).

If you had access to an original, unsectorized GOES-8, -9, -10, or -11
image and displayed it, you would think that the display looked funny
since it would look elongaged in the longitude direction.  This is a
result of there being more pixels in the longitude direction than in
the latitude direction: the instrument is oversampling in the longitude
direction by almost a factor of 2.

The entire issue gets harder to grasp when you realize that the areal
coverage of a pixel on a display is fixed.  When McIDAS goes to blow-up
an image for display, it simply shows more and more of the original
pixels until there are no more in the image.  The default display of a
McIDAS image is a one-to-one mapping of pixels from the original
image to the display space (frame).  Blow-ups past this point result
in replication of values for display.

Similarly, when McIDAS goes to blown-down an image, it samples pixels
for display from the image.

The blow-up/blow-down of an image is always controlled by the MAG=
keyword.  The use of MAG= in IMGCOPY means the same thing as a MAG= in
IMGDISP.

re: use of IMGCOPY

>Using this command is how I found out the resolutions above... it's for a
>OGES 8 image from 1996.  I'd like to work with 8 by 8 km resolution images
>instead of 8 by 16 km resolution images (for when I extract the information
>with IMGPROBE and bring it into another software package).

With an IMGLIST listing of Res (km) showing 8 x 16, you are already where
you want to be for the GOES-8, -9, -10, and -11 satellites.

To get a flavor for how the coverage stuff (i.e., resolution), display
one of your images and then draw a circle over it (use CIRCLE) so that
the center of the circle is in the center of the display.  You can do
this by finding out a station at/near the center and then specify STA=
to CIRCLE, or by finding the LAT,LON of the center and then specifying
the LATLON= keyword in CIRCLE.

>>From address@hidden Thu Dec  7 21:14:51 2000
>
>Tom,
>
>I realized after the last e-mail one way to change the resolution as I
>indicated in the previous e-mail.  I had been looking at IMGREMAP to
>find out how to change the resolution (line and element) and it wasn't
>there... but when I realized that you can do what I want (what I think
>I want) with the IMGCOPY command with MAG=1 2.

Right.

>So, I had a GOES 8 image with lat res=8 and lon res=16... to get the
>image the correct size and placement, and now correct resolution, this
>is the command:
>
>IMGCOPY BWF/083096VIS.9 SUB/TEMPLATE.5 LAT=54 107 PLA=ULEFT SIZ=160 180 MAG=1 2
>
>the FORM=EXP with the IMGLIST command confirms that it's 8 by 8 km.

The original was most likely what you were looking for.

>This resultant image, however, looks likes like the original image,
>except that it is stretched horizontally.  Is the MAG=1 2 merely
>doubling up each pixel in the horizontal?

Yes.

>Ideally, I was hoping that
>the new image would look like the original, except that the resolution
>would be 8 by 8 km.  Is this the only way to go about changing the
>resolution?

The original GOES-8, -9, -10, and -11 images already have the
resolution that you are looing to use.  NOTE:  the GOES-7, -6, -5, etc.
satellites did not do oversampling in the element
(horizontal/longitudnal) direction, so their Res (km) numbers came out
looking the same.

>I guess understanding this is important for another portion of my
>analysis (after using the new algorithm to detect cumulus)... in the
>next portion I'm going to have to extract the actual values from the
>images for specific locations in the midwest.  This, I'm sure will get
>confusing since I have the locations I want the data for in a different
>projection than the satellite image... but that's another can of worms
>that I need to explore before asking about.

McIDAS is good at listing values at user specified positions.  You can
specify these positions in a variety of coordinate systems.  For instance,
you can:

IMGDISP an image
move your cursur to a point
use the ALT-E key sequence to find out where you are (ALT-E means hold down
the ALT key while you press the E key)
use the ALT-D key sequence to find out the value of the pixel under your
cursor

You can also use IMGPROBE to list out blocks of values or countour the
values in a particular region in any unit supported by the image.  Projection
in this case does not matter since McIDAS takes this into account.

For information on using IMGPROBE, you should review the Satellite Imagery 
section of the online McIDAS-X Learning Guide:

http://www.unidata.ucar.edu/packages/mcidas/770/mclearn/learn_guide.html

>Thanks,
>Jason

Tom