HELP

[Gerald I. Evenden]

>Date: Sun, 30 Jan 1994 15:31:58 -0500
>From: mike camann <camann at pick.uga.edu>
>To: grassu-list at max.cecer.army.mil
>Subject: HELP
>
>GRASS-netters,
>
>I hate to be a pain about this, but I really am stuck and *nobody* has
>replied to my earlier posts, so I'll try one more time.  Would anyone
>who has created GRASS databases on the hemispheric scale please contact
>me.  Here is the problem:
>
>I need to digitize a map of North and Central America that was produced
>by the USGS in 1982.  In order to do so, I have created a $LOCATION
>north_america.  The map is a transverse Mercator projection of the
>globe from 90N to 5N, 1:10,000,000 scale.  The central parallel is 45N
>and the central meridian is 100W with a scaling factor of 0.926.

I had heard about this map type and it is an embarassment to be associated
with the organization that created it---from a cartographic point of
view.  It is criminal to use Transverse Mercator for regional coverage
of this sort!  Geologists playing cartographer, indeed.

Secondly, from what I've seen so far, all this discussion about location
leads me to conclude that "$LOCATION", etc. ain't no paragon of ...
either.

Material deleted, which to me and I am sure to Camann, make no sense
whatsoever.

The basic facts seem to be TM projection, central meridian 100W and
scale factor 0.926.  The "central latitude," proj's "lat_0" only
sets an arbitary origin so that in this case (+lat_0=45N +lon_0=100W)
creates an origin (0,0).  From that point, everything in cartesian
space (that is, measured off the map) is measured from that point).
If false easting or northing (ie. x_0, y_0) were specified, the 0,0
point would be translated to those coodinates.

Thus if I were measuring in window centered about 100W,45N and 10
degrees wide and high then the coordinates of my window corners
would be (Clarke 66 ellipsoid) (x,y in meters):

-395471.78      -503200.79  SW -105,40
-331889.03       525870.58  NW -105,50
 395471.78      -503200.79  SE  -95,40
 331889.03       525870.58  NE  -95,50

Note that is a "square degree" box and if we have a rectangular meter
box *containing* the above will the following cartesian coordinate range:

-395471.78      -514307.99
 395471.78       525870.58

Take careful note how the lower y value is less than the previous
minimum y because of the downward bow of the parallel between the
end points.  Geographic coordinates of the final box corners are:

104d59'31.747"W  39d53'32.571"N  SW
105d57'10.297"W  49d57'17.576"N  NW
 95d 0'28.253"W  39d53'32.571"N  SE
 94d 2'49.703"W  49d57'17.576"N  NE

This is hard to get across without pictures, which is probably part
of the reason that the documentation you are using fails so dismally.

Any digitizing program worth its salt should only query you for the
basic projection parameters and a few digitizer coordinates with
their corresponding geographic location and the software does the
rest!

>Alternatively, am I doing something else wrong?
>
>Would ANYONE who has successfully created a $LOCATION covering North
>America please contact me so I can find out how you did it.  Is it
>*possible* that no one has ever done this (hence the silence after my
>previous requests for help)?
>
>Thanks (hopefully...).
>
>%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
>Michael Camann                          camann at dial.pick.uga.edu
>Department of Entomology                camann at phoenix.cs.uga.edu
>University of Georgia                   (706) 542-1388
>Athens, GA 30602                        (706) 542-2276
>
>%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Does that help any?

Gerald (Jerry) I. Evenden   Internet: gie at charon.er.usgs.gov
voice: (508)563-6766          Postal: P.O. Box 1027
  fax: (508)457-2310                  N.Falmouth, MA 02556-1027