[GRASS-user] interpolating with a covariate - v.vol.rst

Hi,

For some crazy reason I was under the impression that it is possible to do
interpolation with a covariate with v.vol.rst. Are there any examples on how
to parameterize this module, when a 2D surface is requested, rather than a 3D
volume. I noticed the 'cellinp' argument for a cross-section, but this is not
quite what I am after. I am looking to do something very similar to
interpolation of rainfall data, taking into account the orographic effect of
terrain.

I would be happy to add an example or two to the man pages.

Cheers,

Dylan

--
Dylan Beaudette
Soil Resource Laboratory
http://casoilresource.lawr.ucdavis.edu/
University of California at Davis
530.754.7341

Dylan,

On Tue, Jan 6, 2009 at 10:01 PM, Dylan Beaudette
<debeaudette@ucdavis.edu> wrote:

Hi,

For some crazy reason I was under the impression that it is possible to do
interpolation with a covariate with v.vol.rst. Are there any examples on how
to parameterize this module, when a 2D surface is requested, rather than a 3D
volume. I noticed the 'cellinp' argument for a cross-section, but this is not
quite what I am after. I am looking to do something very similar to
interpolation of rainfall data, taking into account the orographic effect of
terrain.

This was my main business (say, of our cluster) over the last months :slight_smile:
You can do that. I am using the elevation model as auxiliary variable:

# something like this:
v.vol.rst in=vectpoints cellinp=dem wcolumn=pointval cellout=rst2d

cellout delivers the 2D map, extracted from the volume along the
dem map.

Hope this helps
Markus

PS: If you are interested, I have a patch of v.vol.rst which reduces the
computational time to roughly 30%. Done by a friend of mine
(mathematician) who removed unneeded function calls from the code.
Needs more testing...

On Tue, Jan 6, 2009 at 2:33 PM, Markus Neteler <neteler@osgeo.org> wrote:

Dylan,

On Tue, Jan 6, 2009 at 10:01 PM, Dylan Beaudette
<debeaudette@ucdavis.edu> wrote:

Hi,

For some crazy reason I was under the impression that it is possible to do
interpolation with a covariate with v.vol.rst. Are there any examples on how
to parameterize this module, when a 2D surface is requested, rather than a 3D
volume. I noticed the 'cellinp' argument for a cross-section, but this is not
quite what I am after. I am looking to do something very similar to
interpolation of rainfall data, taking into account the orographic effect of
terrain.

This was my main business (say, of our cluster) over the last months :slight_smile:
You can do that. I am using the elevation model as auxiliary variable:

# something like this:
v.vol.rst in=vectpoints cellinp=dem wcolumn=pointval cellout=rst2d

cellout delivers the 2D map, extracted from the volume along the
dem map.

Hope this helps
Markus

Thanks Markus. One more question: have you found a good compromise in
the 3D region settings- i.e. some ratio of horizontal:vertical
resolution that gives good results and doesn't take too long to
compute?

Cheers,

Dylan

On Wednesday 07 January 2009, Helena Mitasova wrote:

On Jan 7, 2009, at 11:19 AM, Dylan Beaudette wrote:
> On Tue, Jan 6, 2009 at 2:33 PM, Markus Neteler <neteler@osgeo.org>
>
> wrote:
>> Dylan,
>>
>> On Tue, Jan 6, 2009 at 10:01 PM, Dylan Beaudette
>>
>> <debeaudette@ucdavis.edu> wrote:
>>> Hi,
>>>
>>> For some crazy reason I was under the impression that it is
>>> possible to do
>>> interpolation with a covariate with v.vol.rst. Are there any
>>> examples on how
>>> to parameterize this module, when a 2D surface is requested,
>>> rather than a 3D
>>> volume. I noticed the 'cellinp' argument for a cross-section, but
>>> this is not
>>> quite what I am after. I am looking to do something very similar to
>>> interpolation of rainfall data, taking into account the
>>> orographic effect of
>>> terrain.
>>
>> This was my main business (say, of our cluster) over the last
>> months :slight_smile:
>> You can do that. I am using the elevation model as auxiliary
>> variable:
>>
>> # something like this:
>> v.vol.rst in=vectpoints cellinp=dem wcolumn=pointval cellout=rst2d
>>
>> cellout delivers the 2D map, extracted from the volume along the
>> dem map.
>>
>> Hope this helps
>> Markus
>
> Thanks Markus. One more question: have you found a good compromise in
> the 3D region settings- i.e. some ratio of horizontal:vertical
> resolution that gives good results and doesn't take too long to
> compute?

you need to set your vertical resolution based on the spatial
variability you want to capture
and then don't forget to use appropriate zmult that will stretch the
vertical distances so that
they are about the same magnitude as horizontal distances (e.g. if
your horizontal res. is 10m
and vertical is 0.1m you need zmult around 100, if you want
anisotropic effects it may be less -
see more on how to find the parameters here:
http://skagit.meas.ncsu.edu/~helena/gmslab/papers/
TGIS2002_Hofierka_et_al.pdf

I hope this helps, Helena

> Cheers,
>
> Dylan

Thanks for the tips Helena. I'll check on that paper, and report back with the
results.

Dylan

--
Dylan Beaudette
Soil Resource Laboratory
http://casoilresource.lawr.ucdavis.edu/
University of California at Davis
530.754.7341

Dear grass users..
I'm working with grass6 devel.
Trying to use v.vol.rst with option "cellout" I have obtained a strange
result:

The "elev" 3D map is correctly created (I have seen it by means of nviz)
but the cellout map is created as a small map (like a miniature of the
map that I would expect) that is placed at the up-left corner of the
region.. (I'm using monitors and I haven't MASK in my mapset). The
remaining part of the cellout map has only values = 0.

I think I can obtain the same thing using r3.cross.rast starting from
the "elev" map created from v.vol.rst, but: could the problem I have
found be a little bug? or it is a problem o f zmult or wmult (but the
resolution is the same along the x,y,z directions...)

thanks...

Ivan

Il giorno ven, 09/01/2009 alle 12.41 -0800, Dylan Beaudette ha scritto:

On Wednesday 07 January 2009, Helena Mitasova wrote:
> On Jan 7, 2009, at 11:19 AM, Dylan Beaudette wrote:
> > On Tue, Jan 6, 2009 at 2:33 PM, Markus Neteler <neteler@osgeo.org>
> >
> > wrote:
> >> Dylan,
> >>
> >> On Tue, Jan 6, 2009 at 10:01 PM, Dylan Beaudette
> >>
> >> <debeaudette@ucdavis.edu> wrote:
> >>> Hi,
> >>>
> >>> For some crazy reason I was under the impression that it is
> >>> possible to do
> >>> interpolation with a covariate with v.vol.rst. Are there any
> >>> examples on how
> >>> to parameterize this module, when a 2D surface is requested,
> >>> rather than a 3D
> >>> volume. I noticed the 'cellinp' argument for a cross-section, but
> >>> this is not
> >>> quite what I am after. I am looking to do something very similar to
> >>> interpolation of rainfall data, taking into account the
> >>> orographic effect of
> >>> terrain.
> >>
> >> This was my main business (say, of our cluster) over the last
> >> months :slight_smile:
> >> You can do that. I am using the elevation model as auxiliary
> >> variable:
> >>
> >> # something like this:
> >> v.vol.rst in=vectpoints cellinp=dem wcolumn=pointval cellout=rst2d
> >>
> >> cellout delivers the 2D map, extracted from the volume along the
> >> dem map.
> >>
> >> Hope this helps
> >> Markus
> >
> > Thanks Markus. One more question: have you found a good compromise in
> > the 3D region settings- i.e. some ratio of horizontal:vertical
> > resolution that gives good results and doesn't take too long to
> > compute?
>
> you need to set your vertical resolution based on the spatial
> variability you want to capture
> and then don't forget to use appropriate zmult that will stretch the
> vertical distances so that
> they are about the same magnitude as horizontal distances (e.g. if
> your horizontal res. is 10m
> and vertical is 0.1m you need zmult around 100, if you want
> anisotropic effects it may be less -
> see more on how to find the parameters here:
> http://skagit.meas.ncsu.edu/~helena/gmslab/papers/
> TGIS2002_Hofierka_et_al.pdf
>
> I hope this helps, Helena
>
> > Cheers,
> >
> > Dylan

Thanks for the tips Helena. I'll check on that paper, and report back with the
results.

Dylan

--
Ti prego di cercare di non inviarmi files .dwg, .doc, .xls, .ppt.
Preferisco formati liberi.
Please try to avoid to send me .dwg, .doc, .xls, .ppt files.
I prefer free formats.
http://it.wikipedia.org/wiki/Formato_aperto
http://en.wikipedia.org/wiki/Open_format

Ivan Marchesini
Department of Civil and Environmental Engineering
University of Perugia
Via G. Duranti 93/a
06125
Perugia (Italy)
Socio fondatore GFOSS "Geospatial Free and Open Source Software" http://www.gfoss.it
e-mail: marchesini@unipg.it
        ivan.marchesini@gmail.com
tel: +39(0)755853760
fax (university): +39(0)755853756
fax (home): +39(0)5782830887
jabber: geoivan73@jabber.org

Hi Ivan,

On Tue, Jan 27, 2009 at 1:43 PM, ivan marchesini <marchesini@unipg.it> wrote:

Dear grass users..
I'm working with grass6 devel.
Trying to use v.vol.rst with option "cellout" I have obtained a strange
result:

The "elev" 3D map is correctly created (I have seen it by means of nviz)
but the cellout map is created as a small map (like a miniature of the
map that I would expect) that is placed at the up-left corner of the
region.. (I'm using monitors and I haven't MASK in my mapset). The
remaining part of the cellout map has only values = 0.

this is strange - I am using v.vol.rst regularly and do not observe
such a problem.
What was the command line?

I think I can obtain the same thing using r3.cross.rast starting from
the "elev" map created from v.vol.rst, but: could the problem I have
found be a little bug? or it is a problem o f zmult or wmult (but the
resolution is the same along the x,y,z directions...)

please tell us how you launched it (preferably a North Carolina
example to be able to replicate it)

Markus