Antarctica coastline/shelf-ice import

[Jochen Topf]

Hi!

The Antarctica coastline currently existing in OSM was imported years ago from
rather bad and outdated data. And the huge shelf-ice areas are mostly missing.
There is much better data available that can be used to replace the old data.
Christoph Hormann and I are planning an import that will improve this situation
greatly.

We have seen that there are some areas where LINZ data has been imported in
Antarctica. The planned import will leave this LINZ data intact. If somebody
is planning to do more LINZ imports in Antarctica, please contact us so that
we can coordinate our efforts.

More information about this import is at
http://wiki.openstreetmap.org/wiki/Antarctica/Import_2013

Jochen
--
Jochen Topf joc...@topf.org http://www.remote.org/jochen/ +49-721-388298

[pcr...@pcreso.com]

Hi Jochen, Sounds great. Something I will find very useful. Are your data fixes being fed back through to NSDIC, some of the datasets they provide at the URL below could perhaps be improved accordingly? FYI, NIWA is currently applying to EPSG to register a new Antarctic projection, much the same as Antarctic Polar Stereographic, but with a central meridian of 180, instead of 0. Maps using this projection will have NZ up rather than down. (North up is a bit meaningless in this context :-) Also, are you aware of these services from NSDIC. The MOA derived coastline plus a range of other layers. The WFS capabilities document (at http://nsidc.org/cgi-bin/atlas_south?service=WFS&version=1.1.0&request=GetCapabilities ) has abstracts for each available dataset and they are provided with no access constraints. http://nsidc.org/data/atlas/ogc_services.html The PolarView initiative also has several Antarctic datasets available via OGC web services, see: http://www.polarview.aq/data_ogc.php. The WFS service is light on metadata, with no access contraint info provided. Cheers,   Brent Wood --- On Wed, 3/6/13, Jochen Topf <joc...@topf.org> wrote:

-- You received this message because you are subscribed to the Google Groups "nzopengis" group. To unsubscribe from this group and stop receiving emails from it, send an email to nzopengis+unsub...@googlegroups.com. To post to this group, send email to nzop...@googlegroups.com. Visit this group at http://groups.google.com/group/nzopengis?hl=en. For more options, visit https://groups.google.com/groups/opt_out.

[Hamish]

Hi Jochen,

Jochen wrote:
> The Antarctica coastline currently existing in OSM was imported years ago from
> rather bad and outdated data. And the huge shelf-ice areas are mostly missing.

coastline != ice shelf :) please do not conflate the two, it's
wrong on all sorts of levels. There is good data coming through ground
penetrating radar/controlled source seismic for the bits of the (real)
coastline buried under lots of ice, and at least part of that (ie the
parts which will be of particular interest) may be exposed sooner
rather than later if current trends continue.. if the MOA data shows
the edge of the ice shelf (at the time if imagery acquisition), then
that is what it should be tagged as. :)

http://news.nationalgeographic.com/news/2010/01/photogalleries/100122-antarctic-ghost-mountains-under-ice-pictures/

As to the question of is the coastline the point of land meets water
at MSL if there was no ice, or where the overhanging fixed land ice
meets water deep below, I don't really know, but I think no one else
knows that answer either. New studies are showing that 1-3m thick
channels/wedges of fresh & brackish water are often shimming their way
many km inland of where we though the edge was just a couple of years
ago. It will take new work with AUV robotic subs to go where no man
can and see what's really there..

> There is much better data available that can be used to replace the old data.
> Christoph Hormann and I are planning an import that will improve this situation
> greatly.

good news,

> We have seen that there are some areas where LINZ data has been imported in
> Antarctica. The planned import will leave this LINZ data intact. If somebody
> is planning to do more LINZ imports in Antarctica, please contact us so that
> we can coordinate our efforts.

I'm the one who did the Antarctic coastline import for the continental
area to the west of Ross Island (McMurdo Sound). You'll see in our
upload-checkout and status web app that the coastline of Ross Island
has not been done:
http://linz2osm.openstreetmap.org.nz/workslices/create/layer_in_dataset/259/

but actually Ross Island has been uploaded from the same upstream
dataset, someone outside our group took the LINZ data and uploaded it
without attribution or additional tagging some time ago. :-/ That
still needs to be fixed/replaced.


fyi you can see the status of our Antarctic uploads here:
http://linz2osm.openstreetmap.org.nz/data_dict/dataset/antarctic/


As you can see there is still much high quality data to be uploaded
there. Also that shows pretty well the area of coverage for the LINZ
data still to be uploaded.

I think the most accessible way to preview what's in these data layers
is via the WFS at LINZ's data service in the QGIS data browser. Some
instructions here:
http://wiki.openstreetmap.org/wiki/New_Zealand/Imagery

I have the LINZ wms topo map for Antarctica set up as background
imagery in JOSM, it works quite well there.

> More information about this import is at
> http://wiki.openstreetmap.org/wiki/Antarctica/Import_2013

thanks!

Hamish


ps- if anyone sees any level 3 MODIS imagery floating by, half of the
Erebus ice tongue apparently broke off last week, it would be good to
trace the crack. (does JOSM have a tool to rotate a polygon around an
axis? we could track the bergy bit for a little while..)

[Andin]

(does JOSM have a tool to rotate a polygon around an
axis? we could track the bergy bit for a little while..)

 

Ctrl-Shift drag :)


 

[Hamish]

Hamish:

>> (does JOSM have a tool to rotate a polygon around an
>> axis? we could track the bergy bit for a little while..)

Andin:
> Ctrl-Shift drag :)

sweet, thanks.


H

[Jochen Topf]

Hi!

On Wed, Mar 06, 2013 at 09:29:12AM +1300, Hamish wrote:
> Jochen wrote:
> > The Antarctica coastline currently existing in OSM was imported years ago from
> > rather bad and outdated data. And the huge shelf-ice areas are mostly missing.
>
> coastline != ice shelf :) please do not conflate the two, it's
> wrong on all sorts of levels. There is good data coming through ground
> penetrating radar/controlled source seismic for the bits of the (real)
> coastline buried under lots of ice, and at least part of that (ie the
> parts which will be of particular interest) may be exposed sooner
> rather than later if current trends continue.. if the MOA data shows
> the edge of the ice shelf (at the time if imagery acquisition), then
> that is what it should be tagged as. :)
>
> http://news.nationalgeographic.com/news/2010/01/photogalleries/100122-antarctic-ghost-mountains-under-ice-pictures/
>
> As to the question of is the coastline the point of land meets water
> at MSL if there was no ice, or where the overhanging fixed land ice
> meets water deep below, I don't really know, but I think no one else
> knows that answer either. New studies are showing that 1-3m thick
> channels/wedges of fresh & brackish water are often shimming their way
> many km inland of where we though the edge was just a couple of years
> ago. It will take new work with AUV robotic subs to go where no man
> can and see what's really there..

I will not discuss this issue here, to not spread the discussion over too many
places. More people are probably interested, so it should be done on the Wiki
or the Imports list.

We are currently discussing a specific Antarctica-wide import to get the
coastline and ice shelves to some minimum standard. I am sure there are
other, better sources around for at least part of Antarctica.

If after the planned import the LINZ data is still better, somebody else can do
that (after discussing it on the imports list of course). We don't want to make
things too complicated by trying to import the best data from different
sources in one step.

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

[Jochen Topf]

On Tue, Mar 05, 2013 at 09:32:34AM -0800, pcr...@pcreso.com wrote:
> Sounds great. Something I will find very useful. Are your data fixes being fed back through to NSDIC, some of the datasets they provide at the URL below could perhaps be improved accordingly?

I don't have any insights into what NSDIC is doing. :-)

> Also, are you aware of these services from NSDIC. The MOA derived coastline plus a range of other layers. The WFS capabilities document (at http://nsidc.org/cgi-bin/atlas_south?service=WFS&version=1.1.0&request=GetCapabilities ) has abstracts for each available dataset and they are provided with no access constraints.
>
> http://nsidc.org/data/atlas/ogc_services.html
>
> The PolarView initiative also has several Antarctic datasets available via OGC web services, see: http://www.polarview.aq/data_ogc.php. The WFS service is light on metadata, with no access contraint info provided.

We are currently only looking at the MOA (and other) coastline and shelf-ice
data. Other data isn't currently on our agenda. But I have added the links to
http://wiki.openstreetmap.org/wiki/Antarctica/Data_Sources , maybe somebody
can use them later.