[Imports] Antarctica coastline/shelf-ice import

Christoph Hormann chris_hormann at gmx.de
Tue Mar 12 11:16:01 UTC 2013


A week ago we announced the plan to import better coastline and new
ice shelf data for Antarctica on this list. We also sent mails about
these plans to tagging at openstreetmap.org, talk at openstreetmap.org, and
nzopengis at googlegroups.com. (The last is the New Zealand OSM list,
people from there had been importing official New Zealand Government
data to Antarctica and we wanted to coordinate with them.)

Please see the wiki for the current state of this proposal:

There were a few questions about our plans and some objections to the
planned removal of the Antarctica-wide glacier multipolygon and the
change of the coastline from the "grounding line" to the "calving
line". We answered all these mails and explained our reasoning.   The 
coastline closing question which has been discussed as well won't 
seriously affect the import since we can artificially close it if 
necessary although we would prefer not to.  We haven't heard any 
further arguments or objections in the last days so we believe people 
are generally happy with our plans.

If we don't see any objections in the next days, we will do the
import on the next weekend (March 16./17.). The import will be done
according to the plan outlined on the wiki page, a special user will
be used for the import and it will be done in a single or very few
changesets, so if there are any problems it is easy to revert.

One further issue: Before announcing the import we had contacted the 
SCAR (Scientific Committee on Antarctic Research) concerning possible 
use of the Antarctic Digital Database (http://www.add.scar.org/). This 
data is more detailed and mostly newer than what we plan to import now 
but it is assembled from various sources making it somewhat 
non-uniform. Last week, after our announcement here we got permission 
for using their data in OSM.

We have looked into this data but nevertheless decided to "stick to the 
plan" and import the MOA data first. Here are the reasons:

The MOA data is much better than what we currently have and it can
be imported all at once. We can test the new tagging scheme with it and
we can easily revert if there are problems. The ADD data is 
significantly more detailed and our first analysis revealed various 
inconsistencies that need to be dealt with.  This means there is more 
work involved preparing it for import and it will be necessary to do 
this in several pieces.  This will be much easier with a fairly 
accurate data set and tagging already in place.

So we want to do the MOA import as planned - although we might leave out 
those islands that require significant hand work to avoid the need to 
do this twice.  Afterwards we can work on the ADD data, but that will 
be a separate effort. Maybe it can be done as a community effort where 
people check and import pieces of the data.

Christoph Hormann

More information about the Imports mailing list