[OSM-dev] Coastline changes Antarctica

Paul Norman penorman at mac.com
Mon Mar 11 18:31:27 UTC 2013

> From: Christoph Hormann [mailto:chris_hormann at gmx.de]
> Sent: Monday, March 11, 2013 10:50 AM
> Subject: Re: [OSM-dev] Coastline changes Antarctica
> On Monday 11 March 2013, Jochen Topf wrote:
> >
> > I have described the details in this blog post:
> > http://blog.jochentopf.com/2013-03-11-state-of-the-osm-coastline.html
> > Have a look at the chapter about Antarctica. (The other stuff might be
> > interesting, too, though.)
> Concerning the 180-degree-line - for that a solution that would allow to
> process the coastline in a pacific centered coordinate system (or with
> any other cutting meridian) would be best.  And the obvious way for that
> is to simply draw the coastline ways indiscriminately across the line.
> I think the API and database would have no problem with that but
> OSMCoastline would need to be able to close the coastline in arbitrary
> configurations at the edges and the editors would need to support it.
> I suppose especially the latter means it won't happen but still it would
> be the most elegant solution i think.

OSM data is in WGS84 and the line going from -180 to +180 goes across the
entire world in this projection. This is how osm2pgsql interprets it and I
believe most other tools. There is no good way to deal with 180 problem.
This impacts roads in Russia that cross the boundary too.

More information about the dev mailing list