[McIDAS #UXQ-357287]: IMGREMAP colocation question

Hi Maryellen,

re: can I see your IMGREMAP invocations
> Absolutely!:
> 
> image1: Spain: IMGREMAP MSG/RAW MSG/REMAP LAT=40.0 3.5 SIZE=250 400 PRO=RECT 
> RES=4
> image2: msgc: IMGREMAP MSG/RAW MSG/REMAP LAT=7 -21 SIZE=2500 3000 PRO=RECT 
> RES=4

> You should be able to recreate my results below. 

Very good.  This allowed me to create the same remapped sectors as you, and that
is what I needed to really understand your question(s).

re:
> In McIDAS I manually maneuvered the
> cursor to (TVLINE/ELEM 1 1) and (250/400) in image1 (spain) then used those 
> lat/lon
> values as input to my PC E command on the larger MSG image. Then typed AltD 
> and noticed
> the differing lat/lon values.

Ah ha!  Now I understand.  I believe that if you display _any_ image regardless
of its projection and then position the cursor to a Lat,Lon point (PC E) and 
then
interrogate McIDAS for where the cursor is located (via D or E), you will find
that the Lat,Lon values are slightly different.  The reason for this is that 
there
is precision lost when going from TV coordinates to Earth coordinates and visa
versa.  The good news is that the difference in the values is within acceptable
limits.  What I mean by acceptable is that a pixel has a measurable spatial
extent, and McIDAS assigns its location to the upper left corner of the pixel.
This means that the uncertainty in the pixel location varies directly with
the resolution of the pixel.

re:
> As another point of information, I've attached a screenshot of what I get 
> when I ask for
> lat/lon values when I put cursor at 40N, 4W (PC E 40 4, Alt-D) on both of my 
> images. The
> top portion of screen is result from image2 (2500x3000), bottom is result 
> from image1
> (250 400). Notice the lat/lon values are just slightly different and then so 
> are the IR
> values.

Yup, this is the same kind of uncertainty that I saw when I ran your test.  
This is
to be expected, and it can not be made any better.  To convince yourself of 
this, simply
display the unremapped image and run the same test (PC E followed by D or E).

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: UXQ-357287
Department: Support McIDAS
Priority: Normal
Status: Closed