[Proj] epsg:2855 projection problem

[Frank Warmerdam]

Thomas Chesky wrote:
> Hello,
>  
> I have a King County, WA dataset (shapefile) which I am trying to use 
> with Mapserver.
> It is in " NAD83(HARN) / Washington North (ftUS) " projection (epsg:2855)
> but is translated to an invalid lat/lon extent:
> 
> minx="-111.388" miny="46.3886" maxx="-106.113" maxy="49.1724"
> 
> without being a GIS expert (or anything even close) I can only imagine 
> that the PROJ.4 library used by the Mapserver causes it. Is there any 
> known issue with the epsg:2855 definition of PROJ.4 ?
> 
> Can somebody provide me the Mapserver PROJECTION definition for the same 
> but not in the "init=epsg:2855" format. Maybe the problem is with the 
> epsg lookup table and if defined by the full projection syntax, I got 
> the right extent, but I am not sure how to define the proper PROJECTION 
> syntax (other than the epsg one)

Thomas,

First, note that epsg:2855 is a meter based projection, not feet based as you
seem to assume.  When I inspect the definition it looks like this:

  +proj=lcc +lat_1=48.73333333333333 +lat_2=47.5 +lat_0=47 +lon_0=-120.8333333333333
            +x_0=500000 +y_0=0 +ellps=GRS80 +units=m +no_defs

I tried converting one of your sample points (-111.388,46.3886) to WA north
meters and got 1224988.70,-23436.36. I don't know if that is correct though it
seems plausible to me.

If you want to work in feet yo might want to use a definition:

  +proj=lcc +lat_1=48.73333333333333 +lat_2=47.5 +lat_0=47 +lon_0=-120.8333333333333
            +x_0=500000 +y_0=0 +ellps=GRS80 +units=us-ft +no_defs

Finally, PROJ.4 is ignoring the HARN/datum aspect as we don't have any
definition files for HARN (nor I have thought about it carefully).  This is
fine for converting between lat/long GRS80 and epsg:3855 on GRS80 but if you
need to change datums or ellipsoids the datum shifting issue will become
significant.

You mention "but is translated to an invalid lat/lon extent".  Do you mean that
your data region is being reprojected to the named lat/long bounds, but you
believe it is incorrect?  If so, the most likely reason is the ft/m issue.
If not, please provide more details (the projected coordinates) and why you
think the result is wrong.

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent