[OSM-talk] Antarctica coastline/shelf-ice import

[Jochen Topf]

On Tue, Mar 05, 2013 at 04:48:43PM +0100, Maarten Deen wrote:
> On 2013-03-05 16:22, Jochen Topf wrote:
> >On Tue, Mar 05, 2013 at 09:57:31AM -0500, Serge Wroclawski wrote:
> >>I worked on an Ozone monitoring instrument briefly in my career, and
> >>my understanding is that the polar ice caps change often.
> >
> >No, they don't change all that much. Of course they change a
> >little bit all the
> >time, the ice cap moves about 10m a year at the South Pole for
> >instance, and
> >sometimes large icebergs break off from the shelf-ice, but for the
> >resolution
> >we are talking about here, the resolution interesting for OSM,
> >they are
> >basically static.
> >
> >>Are you proposing that we change them every N months?
> >
> >No.
> >
> >>If so, I think we have better solutions at our disposal than
> >>trying to
> >>delete/reimport huge areas like this.
> >
> >This is a misunderstanding. The reason the old data is bad is not that
> >Antarctica changed so much between then and now. I don't know why
> >the old
> >data is so bad, maybe it is because satellite images etc. have
> >improved
> >since then so better data could be derived.
> 
> In the beginning the coastline was very coarse. So much that the
> then current algorithm of generating tiles (especially for T at H)
> sometimes didn't find data to work with (because nodes were so far
> apart) so it rendered an empty tile.
> I did a lot of work making it more precise according to the aerials
> available. I haven't checked in recent years how it correlates to
> current aerials.

Bing has rather good coverage now which also fits well to the data we are
proposing to import.

It is pretty easy to take a look youself. Download current Geofabrik Antarctica
extract, download proposed import
(http://www.imagico.de/files/moa_proc.osm.bz2), load both into JOSM in different
layers, add Bing images. Look around and compare.

Jochen
-- 
Jochen Topf  jochen at remote.org  http://www.remote.org/jochen/  +49-721-388298