[Geodata] Matrix of land cover classifications

Martin Spott Martin.Spott at mgras.net
Tue Jan 27 12:39:34 EST 2009

Hi Alexandre,

Alexandre Leroux wrote:

> I have a few questions about your matrix and I included them in the Talk 
> page:

I don't have much of a clue how a talk page is supposed to be used .... 
therefore I take the freedom to respond at this lovely place. Regarding
your questions, I think I should add some background - hopefully not
too much so I don't bore the readers on this list  :-)

The FlightGear Scenery has a 'tradition' of mis-representing the
real-life terrain due to inaccurate data - attempts have been made to  
ameliorate the situation for example by cutting gaps into the mountains
instead of letting rivers flow uphill ....  :-)
In fact, all these attempts were based upon vague guesses, nobody had a
real clue about how our landcover data actually looked like. Therefore
I started several years ago by setting up a PostGIS DB plus a MapServer
and stuffing all our VMap0-based layers (plus different coastline sets)
into the DB - just to get an idea of what we've been dealing with. The
collection grew over the time - and finally ended up in the status
which I've been trying to summarize here:


You see, the "LandcoverDB CS Detail" page actually has its roots in the
simple attempt to get an overview over different datasets and to derive
a 'reasonable' subset of landcover layers for FlightGear.
I'm telling this story in order to make clear that, while I'm happy to
turn this landcover matrix into a reasonable tool, I have a substatial
reason not to axe its roots, thus still lending FlightGear's landcover
developers a hand.

 - While I'm open to change its name, I'd prefer to leave the page at
OSGeo - not only because I'm in no way involved with Wikipedia but also
because all the other related stuff is in some way affiliated with
OSGeo (actually it's me who had been setting up the MediaWiki
installation on the current server).
Yet I/we might have to ask one of the OSGeo Wiki administrative staff
to change the page name.

 - I've added references to CORINE simply because it affects the region
where I live  :-)  .... well, CORINE also has what I'd call a
'significant' coverage and they also provide a nice colour schema which
I'm using for the web map service. Until now I have added references to
those schemata of which I knew they exist. I'll try to get a clue about
the GlobCover classification, yet I don't know where to find their
schema definition.

 - FAO's LCCS looks like an interesting read, but it's too early for me
to draw any conclusion. While I'd certainly be intrigued to arrange
myself with other people who focus on similar topics, I'll somehow have
to fit all this into the available spare time ....  as always  :-)

Best regards,
 Unix _IS_ user friendly - it's just selective about who its friends are !

More information about the Geodata mailing list