National Atlas Projections... driving me crazy!

[Matthew Perry]

Kelly,

 I assume you're refering to the US National Atlas coordinate system,
right? There is an EPSG code corresponding to this (
http://ocean.csl.co.uk/OceanBrowser/browseDetail.php?DB=epsg_coordinatereferencesystem&field=coord_ref_sys_code&code=2163
)

So any data layers that were in this national atlas projection could
use a projection block like:

PROJECTION
  "init=epsg:2193"
END

Likewise the same projection block could be used to define the
projection of the map output.

Understanding the difference between a layer's native projection and
the map's overall output projection is absolutely essential.

 Perhaps you could provide us with some more details and perhaps some
mapfile samples?

- matt

On 7/24/06, Kelly Koehn <kkoehn at aspiredillusion.com> wrote:
>
>
> I've been working with mapserver for quite some time now and I thought I'd
> made some strides in the areas of projections but I seem to have went into
> it all backwards.  I started out with the National Atlas datasets for
> counties and states.  I've come to the conclusion that the projection they
> want you to use is 'latlong', ellps is 'GRS80', and datum is 'NAD83'.  The
> names are 'statesp020' and 'countyp020' I believe.
>
> This is all fine and dandy except unless you project the map as LCC it seems
> to be vertically challenged.  For instance, the county that I live in
> (Sedgwick, in Kansas) is not a rectangle.. its square.  I've found some
> examples of the lat/long to LCC projection but most seem a bit foggy.
>
> My real question has to do with the whole 'lat/long to meters' conversion
> and how to make heads or tails of the 'cs2cs' program that supposedly
> converts the values.    Is there some magic trick to this program that I'm
> not finding?
>
> I have some other questions about exactly what the lat_0, lat_1, lat_2,
> lon_0 values should be for the LCC projection but I'll save those for
> another time.
>
> Any help would be great! I'm about ready to rip my hair out!
>
> Kelly
>


-- 
Matt Perry
perrygeo at gmail.com
http://www.perrygeo.net