projection questions

[Ed McNierney]

Jeff -

I looked at your map and I looked at some MapPoint maps.

It's hard to tell for sure, but I think Steve's guess that MapPoint is
using a transverse Mercator projection is a good one.  If it's wrong,
the results are pretty similar and you'll be pretty close.

The tricky part is that MapPoint seems to use a transverse Mercator
projection that is centered on the area viewed - the central meridian of
the projection is set to the middle longitude on the map.  You can see
this by going to MapPoint, zooming out pretty far, then scrolling east
and west.  Lines that run east-west will curve upward (in the US) from
the center, and as you move east and west the center of that curve will
remain in the middle of the map image.

Unfortunately, as the others have suggested, you do need to learn a
little about map projections if you're going to deal with them.  What
you want can very easily be done by MapServer - you add PROJECTION
blocks to each layer describing the INPUT projection of each data set
(lat/lon) and you add a PROJECTION block to the map file to describe the
OUTPUT projection of the map image (your custom transverse Mercator
projection).

     - Ed

Ed McNierney
TopoZone.com

-----Original Message-----
From: UMN MapServer Users List [mailto:MAPSERVER-USERS at LISTS.UMN.EDU] On
Behalf Of Jeff Portwine
Sent: Monday, January 24, 2005 3:13 PM
To: MAPSERVER-USERS at LISTS.UMN.EDU
Subject: Re: [UMN_MAPSERVER-USERS] projection questions

Well, I guess I didn't mean specifically "New England" but the general
area.... my map includes most of Pennsylvania, New Jersey,
Massachusetts,
Conneticut, Vermont, Rhode Island, New York, and part of Maine.

It is entirely possible that as you said i'm being misled by the Map
Point
map... and that I just have to get used to how this looks , but
regardless I
need to understand what's going on so that when the people I'm doing
this
for ask me I can explain the difference :)

I put up a quick and dirty web page:   http://www.aiedail.net/maps  so
that
you can see exactly what I'm talking about (I didn't want to attach
images
to an email for the entire list).    I hope that these images will
demonstrate what i've been trying to write in text.

I'm not actually using mappoint data to create the map, so I have no
idea
how they project the data when they generate maps...  but I can't really
seem to make my output look any different regardless of what I use for
output projection.

If there is no way to make it look the same that's probably OK, i'm just
trying to understand what's happening.

Thanks for the time, hope  the picture posting helps.

-Jeff


> ----- Original Message -----
> From: "Ed McNierney" <ed at TOPOZONE.COM>
> To: <MAPSERVER-USERS at LISTS.UMN.EDU>
> Sent: Monday, January 24, 2005 2:33 PM
> Subject: Re: [UMN_MAPSERVER-USERS] projection questions
>
>
> Jeff -
>
> You're confusing this New Englander with your definition of "New
England"
> <g>.  It doesn't include Pennsylvania, Pennsylvania doesn't have much
of a
> coast, and the coast from New York City to Eastport, Maine runs 4.2
> degrees north and 7.0 degrees east, which is quite a bit closer to
east
> than north, particularly in a lat/lon projection!
>
> I think you're being misled by your MapPoint map.  I'm going to guess
that
> it uses something like an Albers projection centered on the continent.
> This will cause the eastern portions of the US to appear to rotate
> counterclockwise.
>
> As a simple check, the northern borders of Pennsylvania, Connecticut,
> Rhode Island, and Massachusetts are all generally straight lines
running
> east-west at constant latitude.  Are they running straight right to
left
> on your Map Point map, or are they running at an angle?
>
> Stretching the extents will not reproject the image.  If your MapPoint
map
> doesn't have the Connecticut-Massachusetts border as a right-left
line, no
> amount of X and Y stretching will rotate it into the place you expect.
>
> If you want to duplicate the MapPoint map, the first thing you need to
do
> is find out what projection it uses.  It sounds like it is NOT the
> projection you think it is, and I don't think it's very productive to
try
> to guess at it.
>
>     - Ed
>
> P.S. Please remember to always send replies to the whole list -
thanks!
>
> Ed McNierney
> TopoZone.com
>
>