[GRASS-user] Creation and managing Location/Mapset

[Luisa Peña]

Hi Nikos and Luis
Thanks for your replies
I still have a few issues regarding this:
1- How can a region not to be a Square? If I have to set
North/South/East/West boundaries?
2- In my case, my mask is a Vectorial. Can I use a vectorial as a mask?

Thank you once again

No dia 17 de Maio de 2010 23:44, Nikos Alexandris <
nikos.alexandris at felis.uni-freiburg.de> escreveu:

> Luisa Peña:
> > > 1- Should I create 1 Location with 4 different mapsets (for each of
> > > the considered scales)? Or should I use 1 location, one mapset and 4
> > > different regions?
>
> Luís Ferreira wrote:
> > one location if all the analysis will be done with one coordinate system
> > one mapset = one region, one resolution
>
> Of course you can work with material of different geospatial
> extent/resolution
> on the same mapset. It's just that you have to pay attention when it comes
> to
> raster computations.
>
> Note also that most vector modules are _not_ limited by the
> extent/resolution
> of the _computational_ region.
>
> > so, perhaps, is better you to use four different mapsets (four levels)
>
> I would think of using several map sets based on "what" is the question to
> be
> answered. Something like a thematic sorting. For example an ndvi mapset.
> Combine your "questions to be answered" with your "scale" levels (if you
> want
> to _not_ think all the time about the extent and the resolution) and you
> could
> go for, let's say "ndvi_local" , "ndvi_national", "indexA_local",
> "indexA_national", etc.
>
> Finally an export mapset could be useful to drag-in all the maps you want
> to
> export (at a single step using some script for example?).
>
> > > 2- How can I define a region that is not a perfect square' Because in
> > > "Set Region" we have North, South, East and West Boundaries. And my
> > > regions are not perfect squares.
>
> > A region don't need to be square. And you can use an analysis mask.
> > Raster > Mask, or r.mask.
>
> Or "r.mapcalc". I remember at least one occasion where "r.mask" did not
> work
> as expected. Don't know if this has been fixed.
>
> Note that there might be some modules that ignore the MASK!
>
> Cheers, Nikos
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/grass-user/attachments/20100518/afd736a2/attachment.html