[Tagging] Mapping deforestation wikipage

Joseph Eisenberg joseph.eisenberg at gmail.com
Fri Mar 15 07:32:57 UTC 2019

> > “The idea is to have mapathon in
> different time, when new imagery are
> available and after check what
> changed searching in the database

In this case you only need to map the area of woodland or forest now, and
it’s no problem to leave other landuse and natural areas unmapped.

But it may be difficult to use the OSM database for checking changes over
time, if you would like scientific, publishable results.

To properly compare the current woodland area with the area in a couple of
years, you will need to make sure that the aerial imagery that you use now
is all from the same year.

Then, when you repeat the project in 1 or 2 years, you need to have new
imagery from around the same time, and you should map the woodland or
forested areas using the same standard and methods as they first time. This
will be easiest if you remap everything from scratch the second time as

If you try to use the OSM database both times, you may find that other
mappers have changed things in the meantime. Perhaps they have changed some
of your landuse=forest to natural=wetland wetland=mangrove, because they
know the area better. Perhaps they have mapped a new palm
oil plantation based on local information, but it isn’t visible in the
aerial imagery. How will you deal with such changes?

So it is great to add the forest or wood areas to OSM right now based on
the latest aerial imager.

But you should plan to keep your own copy of the database to compare with
and edit in the future. If you do this right it could be
professionally-quality data for environmental research, but that means you
need to keep a copy of the data that won’t change, and use that for
comparison in the future.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20190315/752dfa68/attachment.html>

More information about the Tagging mailing list