Quantcast

Searching Docs about 3D geological modelisation

classic Classic list List threaded Threaded
21 messages Options
12
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Searching Docs about 3D geological modelisation

Jean-Roc Morreale
Hello Grasslist,

I'm searching for papers, books or usecases covering the use of GRASS in
geology, mainly on the stratigraphic representations (well log,
schematic block diagram) and geomorphological (erosion, georelief
reconstruction) aspects.

This is an area I've not practiced so far so I would be interested to
know how people here deal with it  in Grass (especially their rules for
building a 3d model from logs and exploiting it correctly but also any
kind of spicy stuff you can obtain from it).

Thre is some related papers I've already find on these subjects :

HydroGIS 96: Application of Geographic Information Systems in Hydrology
and Water Resources Management (Proceedings of the Vienna Conference,
April 1996). IAHS Publ. no. 235, 1996.

Spatial Geologic Hazard Analysis in Practice, J. David Rogers, 2004

Three-dimensional Geological Modeling by FOSS GRASS GIS, Atsushi
Kajiyama et al., 2004

Volume modeling of soils using GRASS GIS 3D-Tools, Markus Neteler, 2001

Regards,
MORREALE Jean Roc
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Benjamin Ducke
Well, you are in luck and trouble at the same time.
Out of all GIS I know, GRASS is the only one that's
competent at handling 3D, volumetric rasters (aka
voxels, which is really what you want for
stratigraphic models). On the other hand, efficient
voxel modelling from sparse 3D data, such as well logs
and transects remains really challenging.

The problem with well logs is also that often the data
is of a "qualitative" nature, indicating the presence
and absence of types of material. There is no interpolation
algorithm in GRASS currently which can handle that sort
of data well.

For visualization, you can export the voxel data to VTK format
and use ParaView or VisIt, which are both superbly
powerful.

I have attached two screenshots of models that I once
produced from log data. One has been interpolated with
GRASS' 3D splines algorithm (which was a bit cheating,
really), the other is some sort of 3D IDW interpolation
done directly in VisIt.
I don't think they turned out too great.

But, hey: it's still more realistic than drawing on a piece
of paper and playing the old "connect the logs with
lines" game...

Ben


MORREALE Jean Roc wrote:

> Hello Grasslist,
>
> I'm searching for papers, books or usecases covering the use of GRASS in
> geology, mainly on the stratigraphic representations (well log,
> schematic block diagram) and geomorphological (erosion, georelief
> reconstruction) aspects.
>
> This is an area I've not practiced so far so I would be interested to
> know how people here deal with it  in Grass (especially their rules for
> building a 3d model from logs and exploiting it correctly but also any
> kind of spicy stuff you can obtain from it).
>
> Thre is some related papers I've already find on these subjects :
>
> HydroGIS 96: Application of Geographic Information Systems in Hydrology
> and Water Resources Management (Proceedings of the Vienna Conference,
> April 1996). IAHS Publ. no. 235, 1996.
>
> Spatial Geologic Hazard Analysis in Practice, J. David Rogers, 2004
>
> Three-dimensional Geological Modeling by FOSS GRASS GIS, Atsushi
> Kajiyama et al., 2004
>
> Volume modeling of soils using GRASS GIS 3D-Tools, Markus Neteler, 2001
>
> Regards,
> MORREALE Jean Roc
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
>

--
Benjamin Ducke
Geospatial Consultant

Oxford Archaeology Digital
Janus House
Osney Mead
OX2 0ES
Oxford, U.K.

Tel: +44 (0)1865 263 800 (switchboard)
Tel: +44 (0)1865 980 758 (direct)
Fax :+44 (0)1865 793 496
[hidden email]
http://oadigital.net






------
Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user

idw-3d-model.png (116K) Download Attachment
fence-diagram-rst.png (80K) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Rich Shepard
On Thu, 7 Jan 2010, Benjamin Ducke wrote:

> The problem with well logs is also that often the data is of a
> "qualitative" nature, indicating the presence and absence of types of
> material. There is no interpolation algorithm in GRASS currently which can
> handle that sort of data well.

   So what is needed is a political algorithm. :-)

   Kidding aside, I suspect that a fuzzy interpolation algorithm would solve
the problem.

Rich
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Benjamin Ducke
In reply to this post by Jean-Roc Morreale
Rich Shepard wrote:
>> material. There is no interpolation algorithm in GRASS currently which
>> can
>> handle that sort of data well.
>
>   So what is needed is a political algorithm. :-)

That's actually right: given the presence of n different
layer types in the vicinity of an empty voxel, the algorithm
would need to decide by some sort of "majority vote"
which type to assign to that voxel.

>
>   Kidding aside, I suspect that a fuzzy interpolation algorithm would solve
> the problem.

How? You could make the interpolated value depend on a
fuzzy set member function, I suppose, but the situation
here is actually so well defined that I think a probabilistic
approach would be preferable. Since each voxel can only
store one value, a second output map could store the
classification probability. That may be very useful
for visualization (you could show voxels with little
probability hazier).

Ben

>
> Rich
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
>


--
Benjamin Ducke
Geospatial Consultant

Oxford Archaeology Digital
Janus House
Osney Mead
OX2 0ES
Oxford, U.K.

Tel: +44 (0)1865 263 800 (switchboard)
Tel: +44 (0)1865 980 758 (direct)
Fax :+44 (0)1865 793 496
[hidden email]
http://oadigital.net





------
Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Rich Shepard
On Fri, 8 Jan 2010, Benjamin Ducke wrote:

> That's actually right: given the presence of n different layer types in
> the vicinity of an empty voxel, the algorithm would need to decide by some
> sort of "majority vote" which type to assign to that voxel.

Ben,

   I'm not a geologist so I don't know how best to extrapolate from well
logs. I've used water well logs to determine the presence of impervious
layers and aquifer surfaces, but not tried to imcorporate those data into a
GRASS analysis. But your comment above makes me wonder if calculating
Veronoi diagram of these various layers would provide that majority vote.

Rich
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Christian Kaiser-2
In reply to this post by Benjamin Ducke
It seems to me that this is a 3D interpolation problem with categorical variables.

Maybe the Bayesian Maximum Entropy approach could help. There are some interesting publications around also for geology and soil sciences, and they can deal with categorical data as well. Look for example here: http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.html#Soil%20Science

Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a tool for 3D geostatistics.

None of them is available through GRASS, but the algorithms are freely available (I think open-source, but not verified).

I am not a geologist, so please forgive if it is not adequate...

Christian Kaiser





On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:

> Rich Shepard wrote:
>>> material. There is no interpolation algorithm in GRASS currently which
>>> can
>>> handle that sort of data well.
>>
>>  So what is needed is a political algorithm. :-)
>
> That's actually right: given the presence of n different
> layer types in the vicinity of an empty voxel, the algorithm
> would need to decide by some sort of "majority vote"
> which type to assign to that voxel.
>
>>
>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would solve
>> the problem.
>
> How? You could make the interpolated value depend on a
> fuzzy set member function, I suppose, but the situation
> here is actually so well defined that I think a probabilistic
> approach would be preferable. Since each voxel can only
> store one value, a second output map could store the
> classification probability. That may be very useful
> for visualization (you could show voxels with little
> probability hazier).
>
> Ben
>
>>
>> Rich
>> _______________________________________________
>> grass-user mailing list
>> [hidden email]
>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>>
>
>
> --
> Benjamin Ducke
> Geospatial Consultant
>
> Oxford Archaeology Digital
> Janus House
> Osney Mead
> OX2 0ES
> Oxford, U.K.
>
> Tel: +44 (0)1865 263 800 (switchboard)
> Tel: +44 (0)1865 980 758 (direct)
> Fax :+44 (0)1865 793 496
> [hidden email]
> http://oadigital.net
>
>
>
>
>
> ------
> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user


_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Jean-Roc Morreale
In reply to this post by Benjamin Ducke
Thanks for the answers,

Having used it a little with archaeological structures it was no wonder
that the first software I thought about was grass, but the main
difference is, as you pointed out, that in this case I do not have a
well defined volume.

I was thinking about using r3.in.ascii and then r.to.rast3 but it
doesn't seems to be the right thing to do on a large scale (150km²) due
to potential multiple top/bottom crossing of the generated layers.

Your screenshots are exactly the kind of data visualization I'm seeking
to obtain.

Jean Roc

Le 08/01/2010 00:34, Benjamin Ducke a écrit :
> Well, you are in luck and trouble at the same time. Out of all GIS I
> know, GRASS is the only one that's competent at handling 3D,
> volumetric rasters (aka voxels, which is really what you want for
> stratigraphic models). On the other hand, efficient voxel modelling
> from sparse 3D data, such as well logs and transects remains really
> challenging.
 >
 > The problem with well logs is also that often the data is of a
 > "qualitative" nature, indicating the presence and absence of types of
 > material. There is no interpolation algorithm in GRASS currently
 > which can handle that sort of data well.
 >
> I have attached two screenshots of models that I once produced from
> log data. One has been interpolated with GRASS' 3D splines algorithm
> (which was a bit cheating, really), the other is some sort of 3D IDW
> interpolation done directly in VisIt. I don't think they turned out
> too great.
 >
> But, hey: it's still more realistic than drawing on a piece of paper
> and playing the old "connect the logs with lines" game...

You missed the use of Illustrator

> Ben
>
>
> MORREALE Jean Roc wrote:
>> Hello Grasslist,
>>
>> I'm searching for papers, books or usecases covering the use of
>> GRASS in geology, mainly on the stratigraphic representations (well
>> log, schematic block diagram) and geomorphological (erosion,
>> georelief reconstruction) aspects.
>>
>> This is an area I've not practiced so far so I would be interested
>> to know how people here deal with it  in Grass (especially their
>> rules for building a 3d model from logs and exploiting it correctly
>> but also any kind of spicy stuff you can obtain from it).
>>
>> Thre is some related papers I've already find on these subjects :
>>
>> HydroGIS 96: Application of Geographic Information Systems in
>> Hydrology and Water Resources Management (Proceedings of the Vienna
>> Conference, April 1996). IAHS Publ. no. 235, 1996.
>>
>> Spatial Geologic Hazard Analysis in Practice, J. David Rogers,
>> 2004
>>
>> Three-dimensional Geological Modeling by FOSS GRASS GIS, Atsushi
>> Kajiyama et al., 2004
>>
>> Volume modeling of soils using GRASS GIS 3D-Tools, Markus Neteler,
>> 2001
>>
>> Regards, MORREALE Jean Roc
>> _______________________________________________ grass-user mailing
>> list [hidden email]
>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>>
>
>
>
>
> _______________________________________________ grass-user mailing
> list [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Benjamin Ducke
In reply to this post by Jean-Roc Morreale
Woohoo, this forum is always a treasure trove
of good advice. I had not idea SGemS existed!
The Voronoi idea is also good, I am just not sure
that the 3D Voronoi diagram is quite what one
would instinctively think it is.

http://en.wikipedia.org/wiki/Voronoi_diagram

says: "In general a cross section of a 3D Voronoi
tessellation is not a 2D Voronoi tessellation itself."

Need to look into that.

I don't have much practical experience
with Bayes models, so can't really comment on
that.

Cheers,

Ben


Christian Kaiser wrote:

> It seems to me that this is a 3D interpolation problem with categorical variables.
>
> Maybe the Bayesian Maximum Entropy approach could help. There are some interesting publications around also for geology and soil sciences, and they can deal with categorical data as well. Look for example here: http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.html#Soil%20Science
>
> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a tool for 3D geostatistics.
>
> None of them is available through GRASS, but the algorithms are freely available (I think open-source, but not verified).
>
> I am not a geologist, so please forgive if it is not adequate...
>
> Christian Kaiser
>
>
>
>
>
> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
>
>> Rich Shepard wrote:
>>>> material. There is no interpolation algorithm in GRASS currently which
>>>> can
>>>> handle that sort of data well.
>>>  So what is needed is a political algorithm. :-)
>> That's actually right: given the presence of n different
>> layer types in the vicinity of an empty voxel, the algorithm
>> would need to decide by some sort of "majority vote"
>> which type to assign to that voxel.
>>
>>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would solve
>>> the problem.
>> How? You could make the interpolated value depend on a
>> fuzzy set member function, I suppose, but the situation
>> here is actually so well defined that I think a probabilistic
>> approach would be preferable. Since each voxel can only
>> store one value, a second output map could store the
>> classification probability. That may be very useful
>> for visualization (you could show voxels with little
>> probability hazier).
>>
>> Ben
>>
>>> Rich
>>> _______________________________________________
>>> grass-user mailing list
>>> [hidden email]
>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>>
>>
>> --
>> Benjamin Ducke
>> Geospatial Consultant
>>
>> Oxford Archaeology Digital
>> Janus House
>> Osney Mead
>> OX2 0ES
>> Oxford, U.K.
>>
>> Tel: +44 (0)1865 263 800 (switchboard)
>> Tel: +44 (0)1865 980 758 (direct)
>> Fax :+44 (0)1865 793 496
>> [hidden email]
>> http://oadigital.net
>>
>>
>>
>>
>>
>> ------
>> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>>
>> _______________________________________________
>> grass-user mailing list
>> [hidden email]
>> http://lists.osgeo.org/mailman/listinfo/grass-user
>
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
>


--
Benjamin Ducke
Geospatial Consultant

Oxford Archaeology Digital
Janus House
Osney Mead
OX2 0ES
Oxford, U.K.

Tel: +44 (0)1865 263 800 (switchboard)
Tel: +44 (0)1865 980 758 (direct)
Fax :+44 (0)1865 793 496
[hidden email]
http://oadigital.net





------
Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

hamish-2
In reply to this post by Jean-Roc Morreale
Jean Roc wrote:
> I was thinking about using r3.in.ascii and then r.to.rast3

eg  http://grass.osgeo.org/wiki/Help_with_3D


> but it doesn't seems to be the right thing to do on a large
> scale (150km²) due to potential multiple top/bottom crossing
> of the generated layers.
...
> Your screenshots are exactly the kind of data visualization
> I'm seeking  to obtain.

see Helena's GRASS 4 screenshots near the end of this page:
  http://grass.osgeo.org/screenshots/viz.php
  http://skagit.meas.ncsu.edu/~helena/gmslab/viz/ches.html


GRASS's own NVIZ has support for custom isosurfaces and multiple
cutting planes.


this thread might interest you:
 http://thread.gmane.org/gmane.comp.gis.grass.user/12869



Hamish




_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

hamish-2
In reply to this post by Jean-Roc Morreale
Jean Roc wrote:
> I was thinking about using r3.in.ascii and then r.to.rast3

eg  http://grass.osgeo.org/wiki/Help_with_3D


> but it doesn't seems to be the right thing to do on a large
> scale (150km²) due to potential multiple top/bottom crossing
> of the generated layers.
...
> Your screenshots are exactly the kind of data visualization
> I'm seeking  to obtain.

see Helena's GRASS 4 screenshots near the end of this page:
  http://grass.osgeo.org/screenshots/viz.php
  http://skagit.meas.ncsu.edu/~helena/gmslab/viz/ches.html


GRASS's own NVIZ has support for custom isosurfaces and multiple
cutting planes.


this thread might interest you:
http://thread.gmane.org/gmane.comp.gis.grass.user/12869



Hamish






_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Dylan Beaudette-2
In reply to this post by Benjamin Ducke
Two more ideas:

1. conditional simulation, based on a 3D variogram model
2. transition probability-based interpolation of categories

Check out gstat for the conditional simulation, and TPROGS for the
transition probability. If anything is interested, I have done some
programming to connect GRASS and TPROGS.

Cheers!

Dylan

On Fri, Jan 8, 2010 at 1:24 PM, Benjamin Ducke
<[hidden email]> wrote:

> Woohoo, this forum is always a treasure trove
> of good advice. I had not idea SGemS existed!
> The Voronoi idea is also good, I am just not sure
> that the 3D Voronoi diagram is quite what one
> would instinctively think it is.
>
> http://en.wikipedia.org/wiki/Voronoi_diagram
>
> says: "In general a cross section of a 3D Voronoi
> tessellation is not a 2D Voronoi tessellation itself."
>
> Need to look into that.
>
> I don't have much practical experience
> with Bayes models, so can't really comment on
> that.
>
> Cheers,
>
> Ben
>
>
> Christian Kaiser wrote:
>> It seems to me that this is a 3D interpolation problem with categorical variables.
>>
>> Maybe the Bayesian Maximum Entropy approach could help. There are some interesting publications around also for geology and soil sciences, and they can deal with categorical data as well. Look for example here: http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.html#Soil%20Science
>>
>> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a tool for 3D geostatistics.
>>
>> None of them is available through GRASS, but the algorithms are freely available (I think open-source, but not verified).
>>
>> I am not a geologist, so please forgive if it is not adequate...
>>
>> Christian Kaiser
>>
>>
>>
>>
>>
>> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
>>
>>> Rich Shepard wrote:
>>>>> material. There is no interpolation algorithm in GRASS currently which
>>>>> can
>>>>> handle that sort of data well.
>>>>  So what is needed is a political algorithm. :-)
>>> That's actually right: given the presence of n different
>>> layer types in the vicinity of an empty voxel, the algorithm
>>> would need to decide by some sort of "majority vote"
>>> which type to assign to that voxel.
>>>
>>>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would solve
>>>> the problem.
>>> How? You could make the interpolated value depend on a
>>> fuzzy set member function, I suppose, but the situation
>>> here is actually so well defined that I think a probabilistic
>>> approach would be preferable. Since each voxel can only
>>> store one value, a second output map could store the
>>> classification probability. That may be very useful
>>> for visualization (you could show voxels with little
>>> probability hazier).
>>>
>>> Ben
>>>
>>>> Rich
>>>> _______________________________________________
>>>> grass-user mailing list
>>>> [hidden email]
>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>>
>>>>
>>>
>>> --
>>> Benjamin Ducke
>>> Geospatial Consultant
>>>
>>> Oxford Archaeology Digital
>>> Janus House
>>> Osney Mead
>>> OX2 0ES
>>> Oxford, U.K.
>>>
>>> Tel: +44 (0)1865 263 800 (switchboard)
>>> Tel: +44 (0)1865 980 758 (direct)
>>> Fax :+44 (0)1865 793 496
>>> [hidden email]
>>> http://oadigital.net
>>>
>>>
>>>
>>>
>>>
>>> ------
>>> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>>>
>>> _______________________________________________
>>> grass-user mailing list
>>> [hidden email]
>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>>
>> _______________________________________________
>> grass-user mailing list
>> [hidden email]
>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>>
>
>
> --
> Benjamin Ducke
> Geospatial Consultant
>
> Oxford Archaeology Digital
> Janus House
> Osney Mead
> OX2 0ES
> Oxford, U.K.
>
> Tel: +44 (0)1865 263 800 (switchboard)
> Tel: +44 (0)1865 980 758 (direct)
> Fax :+44 (0)1865 793 496
> [hidden email]
> http://oadigital.net
>
>
>
>
>
> ------
> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Benjamin Ducke
Cheers for these. They are certainly all highly interesting.
Do you have an actual link for the T-PROGS software itself?
All I can seem to come up with are interfaces from other
software and publications mentioning it.

I would certainly be interested in taking a look at your
GRASS interface. Is T-PROGS open source?

My gut feeling is that the T-PROGS approach would give better
results than 3D kriging, as it seems better able to to
follow 3D shape trends:

http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;37&g=50

... but that certainly would need testing.

Having said that, I also like this approach for a more
heuristic model:

http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;41&g=50

It's very simple and could easily be implemented directly
in GRASS GIS. In fact, I coded something very similar to this
for archaeological stratigraphy reconstruction a while back.

Cheers,

Ben


----- Original Message -----
From: "Dylan Beaudette" <[hidden email]>
To: "Benjamin Ducke" <[hidden email]>
Cc: "GRASS user list" <[hidden email]>
Sent: Saturday, January 9, 2010 4:30:40 AM GMT +01:00 Amsterdam / Berlin / Bern / Rome / Stockholm / Vienna
Subject: Re: [GRASS-user] Searching Docs about 3D geological modelisation

Two more ideas:

1. conditional simulation, based on a 3D variogram model
2. transition probability-based interpolation of categories

Check out gstat for the conditional simulation, and TPROGS for the
transition probability. If anything is interested, I have done some
programming to connect GRASS and TPROGS.

Cheers!

Dylan

On Fri, Jan 8, 2010 at 1:24 PM, Benjamin Ducke
<[hidden email]> wrote:

> Woohoo, this forum is always a treasure trove
> of good advice. I had not idea SGemS existed!
> The Voronoi idea is also good, I am just not sure
> that the 3D Voronoi diagram is quite what one
> would instinctively think it is.
>
> http://en.wikipedia.org/wiki/Voronoi_diagram
>
> says: "In general a cross section of a 3D Voronoi
> tessellation is not a 2D Voronoi tessellation itself."
>
> Need to look into that.
>
> I don't have much practical experience
> with Bayes models, so can't really comment on
> that.
>
> Cheers,
>
> Ben
>
>
> Christian Kaiser wrote:
>> It seems to me that this is a 3D interpolation problem with categorical variables.
>>
>> Maybe the Bayesian Maximum Entropy approach could help. There are some interesting publications around also for geology and soil sciences, and they can deal with categorical data as well. Look for example here: http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.html#Soil%20Science
>>
>> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a tool for 3D geostatistics.
>>
>> None of them is available through GRASS, but the algorithms are freely available (I think open-source, but not verified).
>>
>> I am not a geologist, so please forgive if it is not adequate...
>>
>> Christian Kaiser
>>
>>
>>
>>
>>
>> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
>>
>>> Rich Shepard wrote:
>>>>> material. There is no interpolation algorithm in GRASS currently which
>>>>> can
>>>>> handle that sort of data well.
>>>>  So what is needed is a political algorithm. :-)
>>> That's actually right: given the presence of n different
>>> layer types in the vicinity of an empty voxel, the algorithm
>>> would need to decide by some sort of "majority vote"
>>> which type to assign to that voxel.
>>>
>>>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would solve
>>>> the problem.
>>> How? You could make the interpolated value depend on a
>>> fuzzy set member function, I suppose, but the situation
>>> here is actually so well defined that I think a probabilistic
>>> approach would be preferable. Since each voxel can only
>>> store one value, a second output map could store the
>>> classification probability. That may be very useful
>>> for visualization (you could show voxels with little
>>> probability hazier).
>>>
>>> Ben
>>>
>>>> Rich
>>>> _______________________________________________
>>>> grass-user mailing list
>>>> [hidden email]
>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>>
>>>>
>>>
>>> --
>>> Benjamin Ducke
>>> Geospatial Consultant
>>>
>>> Oxford Archaeology Digital
>>> Janus House
>>> Osney Mead
>>> OX2 0ES
>>> Oxford, U.K.
>>>
>>> Tel: +44 (0)1865 263 800 (switchboard)
>>> Tel: +44 (0)1865 980 758 (direct)
>>> Fax :+44 (0)1865 793 496
>>> [hidden email]
>>> http://oadigital.net
>>>
>>>
>>>
>>>
>>>
>>> ------
>>> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>>>
>>> _______________________________________________
>>> grass-user mailing list
>>> [hidden email]
>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>>
>> _______________________________________________
>> grass-user mailing list
>> [hidden email]
>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>>
>
>
> --
> Benjamin Ducke
> Geospatial Consultant
>
> Oxford Archaeology Digital
> Janus House
> Osney Mead
> OX2 0ES
> Oxford, U.K.
>
> Tel: +44 (0)1865 263 800 (switchboard)
> Tel: +44 (0)1865 980 758 (direct)
> Fax :+44 (0)1865 793 496
> [hidden email]
> http://oadigital.net
>
>
>
>
>
> ------
> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>



------
Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Jean-Roc Morreale
it seems to me that the TPROGS word only refers tp a method and not a
software, here is slide I found about if by Gary Weissmann (from
google's cache as the pdf's link is dead) :

http://209.85.229.132/search?q=cache:5JU5t4eQ4HYJ:www.isgs.illinois.edu/research/3DWorkshop/2005/pdf-files/Weissmann-GSA-2005_ppt.pdf+TPROGS&cd=24&hl=fr&ct=clnk&gl=fr&client=firefox-a

 From forum's post he seems to have a software package available for
transition probability geostatistics which he then import into GSM (the
software linked in the previous mails)

Le 09/01/2010 11:51, Benjamin Ducke a écrit :

> Cheers for these. They are certainly all highly interesting.
> Do you have an actual link for the T-PROGS software itself?
> All I can seem to come up with are interfaces from other
> software and publications mentioning it.
>
> I would certainly be interested in taking a look at your
> GRASS interface. Is T-PROGS open source?
>
> My gut feeling is that the T-PROGS approach would give better
> results than 3D kriging, as it seems better able to to
> follow 3D shape trends:
>
> http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;37&g=50
>
> ... but that certainly would need testing.
>
> Having said that, I also like this approach for a more
> heuristic model:
>
> http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;41&g=50
>
> It's very simple and could easily be implemented directly
> in GRASS GIS. In fact, I coded something very similar to this
> for archaeological stratigraphy reconstruction a while back.
>
> Cheers,
>
> Ben
>
>
> ----- Original Message -----
> From: "Dylan Beaudette"<[hidden email]>
> To: "Benjamin Ducke"<[hidden email]>
> Cc: "GRASS user list"<[hidden email]>
> Sent: Saturday, January 9, 2010 4:30:40 AM GMT +01:00 Amsterdam / Berlin / Bern / Rome / Stockholm / Vienna
> Subject: Re: [GRASS-user] Searching Docs about 3D geological modelisation
>
> Two more ideas:
>
> 1. conditional simulation, based on a 3D variogram model
> 2. transition probability-based interpolation of categories
>
> Check out gstat for the conditional simulation, and TPROGS for the
> transition probability. If anything is interested, I have done some
> programming to connect GRASS and TPROGS.
>
> Cheers!
>
> Dylan
>
> On Fri, Jan 8, 2010 at 1:24 PM, Benjamin Ducke
> <[hidden email]>  wrote:
>> Woohoo, this forum is always a treasure trove
>> of good advice. I had not idea SGemS existed!
>> The Voronoi idea is also good, I am just not sure
>> that the 3D Voronoi diagram is quite what one
>> would instinctively think it is.
>>
>> http://en.wikipedia.org/wiki/Voronoi_diagram
>>
>> says: "In general a cross section of a 3D Voronoi
>> tessellation is not a 2D Voronoi tessellation itself."
>>
>> Need to look into that.
>>
>> I don't have much practical experience
>> with Bayes models, so can't really comment on
>> that.
>>
>> Cheers,
>>
>> Ben
>>
>>
>> Christian Kaiser wrote:
>>> It seems to me that this is a 3D interpolation problem with categorical variables.
>>>
>>> Maybe the Bayesian Maximum Entropy approach could help. There are some interesting publications around also for geology and soil sciences, and they can deal with categorical data as well. Look for example here: http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.html#Soil%20Science
>>>
>>> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a tool for 3D geostatistics.
>>>
>>> None of them is available through GRASS, but the algorithms are freely available (I think open-source, but not verified).
>>>
>>> I am not a geologist, so please forgive if it is not adequate...
>>>
>>> Christian Kaiser
>>>
>>>
>>>
>>>
>>>
>>> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
>>>
>>>> Rich Shepard wrote:
>>>>>> material. There is no interpolation algorithm in GRASS currently which
>>>>>> can
>>>>>> handle that sort of data well.
>>>>>   So what is needed is a political algorithm. :-)
>>>> That's actually right: given the presence of n different
>>>> layer types in the vicinity of an empty voxel, the algorithm
>>>> would need to decide by some sort of "majority vote"
>>>> which type to assign to that voxel.
>>>>
>>>>>   Kidding aside, I suspect that a fuzzy interpolation algorithm would solve
>>>>> the problem.
>>>> How? You could make the interpolated value depend on a
>>>> fuzzy set member function, I suppose, but the situation
>>>> here is actually so well defined that I think a probabilistic
>>>> approach would be preferable. Since each voxel can only
>>>> store one value, a second output map could store the
>>>> classification probability. That may be very useful
>>>> for visualization (you could show voxels with little
>>>> probability hazier).
>>>>
>>>> Ben
>>>>
>>>>> Rich
>>>>> _______________________________________________
>>>>> grass-user mailing list
>>>>> [hidden email]
>>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>>>
>>>>>
>>>>
>>>> --
>>>> Benjamin Ducke
>>>> Geospatial Consultant
>>>>
>>>> Oxford Archaeology Digital
>>>> Janus House
>>>> Osney Mead
>>>> OX2 0ES
>>>> Oxford, U.K.
>>>>
>>>> Tel: +44 (0)1865 263 800 (switchboard)
>>>> Tel: +44 (0)1865 980 758 (direct)
>>>> Fax :+44 (0)1865 793 496
>>>> [hidden email]
>>>> http://oadigital.net
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> ------
>>>> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>>>>
>>>> _______________________________________________
>>>> grass-user mailing list
>>>> [hidden email]
>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>>
>>> _______________________________________________
>>> grass-user mailing list
>>> [hidden email]
>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>>
>>
>>
>> --
>> Benjamin Ducke
>> Geospatial Consultant
>>
>> Oxford Archaeology Digital
>> Janus House
>> Osney Mead
>> OX2 0ES
>> Oxford, U.K.
>>
>> Tel: +44 (0)1865 263 800 (switchboard)
>> Tel: +44 (0)1865 980 758 (direct)
>> Fax :+44 (0)1865 793 496
>> [hidden email]
>> http://oadigital.net
>>
>>
>>
>>
>>
>> ------
>> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>>
>> _______________________________________________
>> grass-user mailing list
>> [hidden email]
>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>
>
>
> ------
> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Dylan Beaudette-2
In reply to this post by Benjamin Ducke
On Sat, Jan 9, 2010 at 2:51 AM, Benjamin Ducke
<[hidden email]> wrote:

> Cheers for these. They are certainly all highly interesting.
> Do you have an actual link for the T-PROGS software itself?
> All I can seem to come up with are interfaces from other
> software and publications mentioning it.
>
> I would certainly be interested in taking a look at your
> GRASS interface. Is T-PROGS open source?
>
> My gut feeling is that the T-PROGS approach would give better
> results than 3D kriging, as it seems better able to to
> follow 3D shape trends:
>
> http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;37&g=50
>
> ... but that certainly would need testing.
>
> Having said that, I also like this approach for a more
> heuristic model:
>
> http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;41&g=50
>
> It's very simple and could easily be implemented directly
> in GRASS GIS. In fact, I coded something very similar to this
> for archaeological stratigraphy reconstruction a while back.
>
> Cheers,
>
> Ben

Hi Ben,

Yes. It would be very interesting to have these functions within GRASS
libraries, as opposed to the kludgy interfacing that I did via shell
scripting + awk. Here are some of the details, from *several* years
ago (GRASS 5.x):

http://169.237.35.250/~dylan/grass_and_tp/

... note that this is rather old work, and somethings may have changed
since then.

Here is the reference for the software:

Carle, Steven F. T-PROGS Transition Probability Geostatistical
Software Version 2.1 manual. University of California, Davis. 1999.

I can get in touch with Graham Fogg here at UC Davis, whom I believe
is in charge of maintaining the current implementation of T-PROGS--
basically fortran source + a tcl/tk interface. Having this
functionality in GRASS would greatly add to the capabilities of the
voxel framework.

Also, by 'conditional simulation' I wasn't referring to kriging per
se, rather the conditional simulation of an indicator (categorical)
variable, based on random fields + variogram model. the gstat library
can perform both unconditional simulation (randomness only tied to a
variogram model), and conditional simulation (randomness tied to real
point data + variogram model).

I'll report back here with my findings.

Cheers,
Dylan



>
>
> ----- Original Message -----
> From: "Dylan Beaudette" <[hidden email]>
> To: "Benjamin Ducke" <[hidden email]>
> Cc: "GRASS user list" <[hidden email]>
> Sent: Saturday, January 9, 2010 4:30:40 AM GMT +01:00 Amsterdam / Berlin / Bern / Rome / Stockholm / Vienna
> Subject: Re: [GRASS-user] Searching Docs about 3D geological modelisation
>
> Two more ideas:
>
> 1. conditional simulation, based on a 3D variogram model
> 2. transition probability-based interpolation of categories
>
> Check out gstat for the conditional simulation, and TPROGS for the
> transition probability. If anything is interested, I have done some
> programming to connect GRASS and TPROGS.
>
> Cheers!
>
> Dylan
>
> On Fri, Jan 8, 2010 at 1:24 PM, Benjamin Ducke
> <[hidden email]> wrote:
>> Woohoo, this forum is always a treasure trove
>> of good advice. I had not idea SGemS existed!
>> The Voronoi idea is also good, I am just not sure
>> that the 3D Voronoi diagram is quite what one
>> would instinctively think it is.
>>
>> http://en.wikipedia.org/wiki/Voronoi_diagram
>>
>> says: "In general a cross section of a 3D Voronoi
>> tessellation is not a 2D Voronoi tessellation itself."
>>
>> Need to look into that.
>>
>> I don't have much practical experience
>> with Bayes models, so can't really comment on
>> that.
>>
>> Cheers,
>>
>> Ben
>>
>>
>> Christian Kaiser wrote:
>>> It seems to me that this is a 3D interpolation problem with categorical variables.
>>>
>>> Maybe the Bayesian Maximum Entropy approach could help. There are some interesting publications around also for geology and soil sciences, and they can deal with categorical data as well. Look for example here: http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.html#Soil%20Science
>>>
>>> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a tool for 3D geostatistics.
>>>
>>> None of them is available through GRASS, but the algorithms are freely available (I think open-source, but not verified).
>>>
>>> I am not a geologist, so please forgive if it is not adequate...
>>>
>>> Christian Kaiser
>>>
>>>
>>>
>>>
>>>
>>> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
>>>
>>>> Rich Shepard wrote:
>>>>>> material. There is no interpolation algorithm in GRASS currently which
>>>>>> can
>>>>>> handle that sort of data well.
>>>>>  So what is needed is a political algorithm. :-)
>>>> That's actually right: given the presence of n different
>>>> layer types in the vicinity of an empty voxel, the algorithm
>>>> would need to decide by some sort of "majority vote"
>>>> which type to assign to that voxel.
>>>>
>>>>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would solve
>>>>> the problem.
>>>> How? You could make the interpolated value depend on a
>>>> fuzzy set member function, I suppose, but the situation
>>>> here is actually so well defined that I think a probabilistic
>>>> approach would be preferable. Since each voxel can only
>>>> store one value, a second output map could store the
>>>> classification probability. That may be very useful
>>>> for visualization (you could show voxels with little
>>>> probability hazier).
>>>>
>>>> Ben
>>>>
>>>>> Rich
>>>>> _______________________________________________
>>>>> grass-user mailing list
>>>>> [hidden email]
>>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>>>
>>>>>
>>>>
>>>> --
>>>> Benjamin Ducke
>>>> Geospatial Consultant
>>>>
>>>> Oxford Archaeology Digital
>>>> Janus House
>>>> Osney Mead
>>>> OX2 0ES
>>>> Oxford, U.K.
>>>>
>>>> Tel: +44 (0)1865 263 800 (switchboard)
>>>> Tel: +44 (0)1865 980 758 (direct)
>>>> Fax :+44 (0)1865 793 496
>>>> [hidden email]
>>>> http://oadigital.net
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> ------
>>>> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>>>>
>>>> _______________________________________________
>>>> grass-user mailing list
>>>> [hidden email]
>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>>
>>> _______________________________________________
>>> grass-user mailing list
>>> [hidden email]
>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>>
>>
>>
>> --
>> Benjamin Ducke
>> Geospatial Consultant
>>
>> Oxford Archaeology Digital
>> Janus House
>> Osney Mead
>> OX2 0ES
>> Oxford, U.K.
>>
>> Tel: +44 (0)1865 263 800 (switchboard)
>> Tel: +44 (0)1865 980 758 (direct)
>> Fax :+44 (0)1865 793 496
>> [hidden email]
>> http://oadigital.net
>>
>>
>>
>>
>>
>> ------
>> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>>
>> _______________________________________________
>> grass-user mailing list
>> [hidden email]
>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>
>
>
> ------
> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

kapo coulibaly-2
Another way would be to use grass to generate surfaces or horizon based on individual layer picks from the logs and use a script to convert them to vtk format. For one thing vtk is written in python and grass can import vtk. I'm attaching an example of what i get. In all fairness i used R and paraview (based on vtk libraries) but the whole procedure can easily be ported to GRASS.

On Sat, Jan 9, 2010 at 12:51 PM, Dylan Beaudette <[hidden email]> wrote:
On Sat, Jan 9, 2010 at 2:51 AM, Benjamin Ducke
<[hidden email]> wrote:
> Cheers for these. They are certainly all highly interesting.
> Do you have an actual link for the T-PROGS software itself?
> All I can seem to come up with are interfaces from other
> software and publications mentioning it.
>
> I would certainly be interested in taking a look at your
> GRASS interface. Is T-PROGS open source?
>
> My gut feeling is that the T-PROGS approach would give better
> results than 3D kriging, as it seems better able to to
> follow 3D shape trends:
>
> http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;37&g=50
>
> ... but that certainly would need testing.
>
> Having said that, I also like this approach for a more
> heuristic model:
>
> http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;41&g=50
>
> It's very simple and could easily be implemented directly
> in GRASS GIS. In fact, I coded something very similar to this
> for archaeological stratigraphy reconstruction a while back.
>
> Cheers,
>
> Ben

Hi Ben,

Yes. It would be very interesting to have these functions within GRASS
libraries, as opposed to the kludgy interfacing that I did via shell
scripting + awk. Here are some of the details, from *several* years
ago (GRASS 5.x):

http://169.237.35.250/~dylan/grass_and_tp/

... note that this is rather old work, and somethings may have changed
since then.

Here is the reference for the software:

Carle, Steven F. T-PROGS Transition Probability Geostatistical
Software Version 2.1 manual. University of California, Davis. 1999.

I can get in touch with Graham Fogg here at UC Davis, whom I believe
is in charge of maintaining the current implementation of T-PROGS--
basically fortran source + a tcl/tk interface. Having this
functionality in GRASS would greatly add to the capabilities of the
voxel framework.

Also, by 'conditional simulation' I wasn't referring to kriging per
se, rather the conditional simulation of an indicator (categorical)
variable, based on random fields + variogram model. the gstat library
can perform both unconditional simulation (randomness only tied to a
variogram model), and conditional simulation (randomness tied to real
point data + variogram model) .

I'll report back here with my findings.

Cheers,
Dylan



>
>
> ----- Original Message -----
> From: "Dylan Beaudette" <[hidden email]>
> To: "Benjamin Ducke" <[hidden email]>
> Cc: "GRASS user list" <[hidden email]>
> Sent: Saturday, January 9, 2010 4:30:40 AM GMT +01:00 Amsterdam / Berlin / Bern / Rome / Stockholm / Vienna
> Subject: Re: [GRASS-user] Searching Docs about 3D geological modelisation
>
> Two more ideas:
>
> 1. conditional simulation, based on a 3D variogram model
> 2. transition probability-based interpolation of categories
>
> Check out gstat for the conditional simulation, and TPROGS for the
> transition probability. If anything is interested, I have done some
> programming to connect GRASS and TPROGS.
>
> Cheers!
>
> Dylan
>
> On Fri, Jan 8, 2010 at 1:24 PM, Benjamin Ducke
> <[hidden email]> wrote:
>> Woohoo, this forum is always a treasure trove
>> of good advice. I had not idea SGemS existed!
>> The Voronoi idea is also good, I am just not sure
>> that the 3D Voronoi diagram is quite what one
>> would instinctively think it is.
>>
>> http://en.wikipedia.org/wiki/Voronoi_diagram
>>
>> says: "In general a cross section of a 3D Voronoi
>> tessellation is not a 2D Voronoi tessellation itself."
>>
>> Need to look into that.
>>
>> I don't have much practical experience
>> with Bayes models, so can't really comment on
>> that.
>>
>> Cheers,
>>
>> Ben
>>
>>
>> Christian Kaiser wrote:
>>> It seems to me that this is a 3D interpolation problem with categorical variables.
>>>
>>> Maybe the Bayesian Maximum Entropy approach could help. There are some interesting publications around also for geology and soil sciences, and they can deal with categorical data as well. Look for example here: http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.html#Soil%20Science
>>>
>>> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a tool for 3D geostatistics.
>>>
>>> None of them is available through GRASS, but the algorithms are freely available (I think open-source, but not verified).
>>>
>>> I am not a geologist, so please forgive if it is not adequate...
>>>
>>> Christian Kaiser
>>>
>>>
>>>
>>>
>>>
>>> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
>>>
>>>> Rich Shepard wrote:
>>>>>> material. There is no interpolation algorithm in GRASS currently which
>>>>>> can
>>>>>> handle that sort of data well.
>>>>>  So what is needed is a political algorithm. :-)
>>>> That's actually right: given the presence of n different
>>>> layer types in the vicinity of an empty voxel, the algorithm
>>>> would need to decide by some sort of "majority vote"
>>>> which type to assign to that voxel.
>>>>
>>>>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would solve
>>>>> the problem.
>>>> How? You could make the interpolated value depend on a
>>>> fuzzy set member function, I suppose, but the situation
>>>> here is actually so well defined that I think a probabilistic
>>>> approach would be preferable. Since each voxel can only
>>>> store one value, a second output map could store the
>>>> classification probability. That may be very useful
>>>> for visualization (you could show voxels with little
>>>> probability hazier).
>>>>
>>>> Ben
>>>>
>>>>> Rich
>>>>> _______________________________________________
>>>>> grass-user mailing list
>>>>> [hidden email]
>>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>>>
>>>>>
>>>>
>>>> --
>>>> Benjamin Ducke
>>>> Geospatial Consultant
>>>>
>>>> Oxford Archaeology Digital
>>>> Janus House
>>>> Osney Mead
>>>> OX2 0ES
>>>> Oxford, U.K.
>>>>
>>>> Tel: +44 (0)1865 263 800 (switchboard)
>>>> Tel: +44 (0)1865 980 758 (direct)
>>>> Fax :+44 (0)1865 793 496
>>>> [hidden email]
>>>> http://oadigital.net
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> ------
>>>> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>>>>
>>>> _______________________________________________
>>>> grass-user mailing list
>>>> [hidden email]
>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>>
>>> _______________________________________________
>>> grass-user mailing list
>>> [hidden email]
>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>>
>>
>>
>> --
>> Benjamin Ducke
>> Geospatial Consultant
>>
>> Oxford Archaeology Digital
>> Janus House
>> Osney Mead
>> OX2 0ES
>> Oxford, U.K.
>>
>> Tel: +44 (0)1865 263 800 (switchboard)
>> Tel: +44 (0)1865 980 758 (direct)
>> Fax :+44 (0)1865 793 496
>> [hidden email]
>> http://oadigital.net
>>
>>
>>
>>
>>
>> ------
>> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>>
>> _______________________________________________
>> grass-user mailing list
>> [hidden email]
>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>
>
>
> ------
> Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user


_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user

geolModel.png (60K) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Searching Docs about 3D geological modelisation

Dylan Beaudette-2
In reply to this post by Dylan Beaudette-2
Hi Thomas,

I am pretty sure that the program (maybe not the source code) are in
the public domain.

I'll contact the author, and post back.

Cheers,
Dylan

On Mon, Jan 11, 2010 at 6:56 AM, Thomas Adams <[hidden email]> wrote:

> Dylan,
>
> Can you tell me how to obtain TPROGS? Is it only available commercially?
>
> Thanks,
> Tom
>
>
> Dylan Beaudette wrote:
>>
>> Two more ideas:
>>
>> 1. conditional simulation, based on a 3D variogram model
>> 2. transition probability-based interpolation of categories
>>
>> Check out gstat for the conditional simulation, and TPROGS for the
>> transition probability. If anything is interested, I have done some
>> programming to connect GRASS and TPROGS.
>>
>> Cheers!
>>
>> Dylan
>>
>> On Fri, Jan 8, 2010 at 1:24 PM, Benjamin Ducke
>> <[hidden email]> wrote:
>>
>>>
>>> Woohoo, this forum is always a treasure trove
>>> of good advice. I had not idea SGemS existed!
>>> The Voronoi idea is also good, I am just not sure
>>> that the 3D Voronoi diagram is quite what one
>>> would instinctively think it is.
>>>
>>> http://en.wikipedia.org/wiki/Voronoi_diagram
>>>
>>> says: "In general a cross section of a 3D Voronoi
>>> tessellation is not a 2D Voronoi tessellation itself."
>>>
>>> Need to look into that.
>>>
>>> I don't have much practical experience
>>> with Bayes models, so can't really comment on
>>> that.
>>>
>>> Cheers,
>>>
>>> Ben
>>>
>>>
>>> Christian Kaiser wrote:
>>>
>>>>
>>>> It seems to me that this is a 3D interpolation problem with categorical
>>>> variables.
>>>>
>>>> Maybe the Bayesian Maximum Entropy approach could help. There are some
>>>> interesting publications around also for geology and soil sciences, and they
>>>> can deal with categorical data as well. Look for example here:
>>>> http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.html#Soil%20Science
>>>>
>>>> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a
>>>> tool for 3D geostatistics.
>>>>
>>>> None of them is available through GRASS, but the algorithms are freely
>>>> available (I think open-source, but not verified).
>>>>
>>>> I am not a geologist, so please forgive if it is not adequate...
>>>>
>>>> Christian Kaiser
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
>>>>
>>>>
>>>>>
>>>>> Rich Shepard wrote:
>>>>>
>>>>>>>
>>>>>>> material. There is no interpolation algorithm in GRASS currently
>>>>>>> which
>>>>>>> can
>>>>>>> handle that sort of data well.
>>>>>>>
>>>>>>
>>>>>>  So what is needed is a political algorithm. :-)
>>>>>>
>>>>>
>>>>> That's actually right: given the presence of n different
>>>>> layer types in the vicinity of an empty voxel, the algorithm
>>>>> would need to decide by some sort of "majority vote"
>>>>> which type to assign to that voxel.
>>>>>
>>>>>
>>>>>>
>>>>>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would
>>>>>> solve
>>>>>> the problem.
>>>>>>
>>>>>
>>>>> How? You could make the interpolated value depend on a
>>>>> fuzzy set member function, I suppose, but the situation
>>>>> here is actually so well defined that I think a probabilistic
>>>>> approach would be preferable. Since each voxel can only
>>>>> store one value, a second output map could store the
>>>>> classification probability. That may be very useful
>>>>> for visualization (you could show voxels with little
>>>>> probability hazier).
>>>>>
>>>>> Ben
>>>>>
>>>>>
>>>>>>
>>>>>> Rich
>>>>>> _______________________________________________
>>>>>> grass-user mailing list
>>>>>> [hidden email]
>>>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>> --
>>>>> Benjamin Ducke
>>>>> Geospatial Consultant
>>>>>
>>>>> Oxford Archaeology Digital
>>>>> Janus House
>>>>> Osney Mead
>>>>> OX2 0ES
>>>>> Oxford, U.K.
>>>>>
>>>>> Tel: +44 (0)1865 263 800 (switchboard)
>>>>> Tel: +44 (0)1865 980 758 (direct)
>>>>> Fax :+44 (0)1865 793 496
>>>>> [hidden email]
>>>>> http://oadigital.net
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> ------
>>>>> Files attached to this email may be in ISO 26300 format (OASIS Open
>>>>> Document Format). If you have difficulty opening them, please visit
>>>>> http://iso26300.info for more information.
>>>>>
>>>>> _______________________________________________
>>>>> grass-user mailing list
>>>>> [hidden email]
>>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>>>
>>>>
>>>> _______________________________________________
>>>> grass-user mailing list
>>>> [hidden email]
>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>>
>>>>
>>>>
>>>
>>> --
>>> Benjamin Ducke
>>> Geospatial Consultant
>>>
>>> Oxford Archaeology Digital
>>> Janus House
>>> Osney Mead
>>> OX2 0ES
>>> Oxford, U.K.
>>>
>>> Tel: +44 (0)1865 263 800 (switchboard)
>>> Tel: +44 (0)1865 980 758 (direct)
>>> Fax :+44 (0)1865 793 496
>>> [hidden email]
>>> http://oadigital.net
>>>
>>>
>>>
>>>
>>>
>>> ------
>>> Files attached to this email may be in ISO 26300 format (OASIS Open
>>> Document Format). If you have difficulty opening them, please visit
>>> http://iso26300.info for more information.
>>>
>>> _______________________________________________
>>> grass-user mailing list
>>> [hidden email]
>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>>
>>
>> _______________________________________________
>> grass-user mailing list
>> [hidden email]
>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>
>
> --
> Thomas E Adams
> National Weather Service
> Ohio River Forecast Center
> 1901 South State Route 134
> Wilmington, OH 45177
>
> EMAIL:  [hidden email]
>
> VOICE:  937-383-0528
> FAX:    937-383-0033
>
>
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: [GRASS-user] Searching Docs about 3D geological modelisation

Dylan Beaudette
In reply to this post by Dylan Beaudette-2
Quick update:

I recently heard back from Graham Fogg here on campus, and he is in favor of
allowing T-PROGS to be used within GRASS. However, he is still waiting for
the final go-ahead from the original author.

Dylan

On Monday 11 January 2010, Thomas Adams wrote:

> Dylan,
>
> Can you tell me how to obtain TPROGS? Is it only available commercially?
>
> Thanks,
> Tom
>
> Dylan Beaudette wrote:
> > Two more ideas:
> >
> > 1. conditional simulation, based on a 3D variogram model
> > 2. transition probability-based interpolation of categories
> >
> > Check out gstat for the conditional simulation, and TPROGS for the
> > transition probability. If anything is interested, I have done some
> > programming to connect GRASS and TPROGS.
> >
> > Cheers!
> >
> > Dylan
> >
> > On Fri, Jan 8, 2010 at 1:24 PM, Benjamin Ducke
> >
> > <[hidden email]> wrote:
> >> Woohoo, this forum is always a treasure trove
> >> of good advice. I had not idea SGemS existed!
> >> The Voronoi idea is also good, I am just not sure
> >> that the 3D Voronoi diagram is quite what one
> >> would instinctively think it is.
> >>
> >> http://en.wikipedia.org/wiki/Voronoi_diagram
> >>
> >> says: "In general a cross section of a 3D Voronoi
> >> tessellation is not a 2D Voronoi tessellation itself."
> >>
> >> Need to look into that.
> >>
> >> I don't have much practical experience
> >> with Bayes models, so can't really comment on
> >> that.
> >>
> >> Cheers,
> >>
> >> Ben
> >>
> >> Christian Kaiser wrote:
> >>> It seems to me that this is a 3D interpolation problem with categorical
> >>> variables.
> >>>
> >>> Maybe the Bayesian Maximum Entropy approach could help. There are some
> >>> interesting publications around also for geology and soil sciences, and
> >>> they can deal with categorical data as well. Look for example here:
> >>> http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.ht
> >>>ml#Soil%20Science
> >>>
> >>> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a
> >>> tool for 3D geostatistics.
> >>>
> >>> None of them is available through GRASS, but the algorithms are freely
> >>> available (I think open-source, but not verified).
> >>>
> >>> I am not a geologist, so please forgive if it is not adequate...
> >>>
> >>> Christian Kaiser
> >>>
> >>> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
> >>>> Rich Shepard wrote:
> >>>>>> material. There is no interpolation algorithm in GRASS currently
> >>>>>> which can
> >>>>>> handle that sort of data well.
> >>>>>
> >>>>>  So what is needed is a political algorithm. :-)
> >>>>
> >>>> That's actually right: given the presence of n different
> >>>> layer types in the vicinity of an empty voxel, the algorithm
> >>>> would need to decide by some sort of "majority vote"
> >>>> which type to assign to that voxel.
> >>>>
> >>>>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would
> >>>>> solve the problem.
> >>>>
> >>>> How? You could make the interpolated value depend on a
> >>>> fuzzy set member function, I suppose, but the situation
> >>>> here is actually so well defined that I think a probabilistic
> >>>> approach would be preferable. Since each voxel can only
> >>>> store one value, a second output map could store the
> >>>> classification probability. That may be very useful
> >>>> for visualization (you could show voxels with little
> >>>> probability hazier).
> >>>>
> >>>> Ben
> >>>>
> >>>>> Rich
> >>>>> _______________________________________________
> >>>>> grass-user mailing list
> >>>>> [hidden email]
> >>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
> >>>>
> >>>> --
> >>>> Benjamin Ducke
> >>>> Geospatial Consultant
> >>>>
> >>>> Oxford Archaeology Digital
> >>>> Janus House
> >>>> Osney Mead
> >>>> OX2 0ES
> >>>> Oxford, U.K.
> >>>>
> >>>> Tel: +44 (0)1865 263 800 (switchboard)
> >>>> Tel: +44 (0)1865 980 758 (direct)
> >>>> Fax :+44 (0)1865 793 496
> >>>> [hidden email]
> >>>> http://oadigital.net
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>> ------
> >>>> Files attached to this email may be in ISO 26300 format (OASIS Open
> >>>> Document Format). If you have difficulty opening them, please visit
> >>>> http://iso26300.info for more information.
> >>>>
> >>>> _______________________________________________
> >>>> grass-user mailing list
> >>>> [hidden email]
> >>>> http://lists.osgeo.org/mailman/listinfo/grass-user
> >>>
> >>> _______________________________________________
> >>> grass-user mailing list
> >>> [hidden email]
> >>> http://lists.osgeo.org/mailman/listinfo/grass-user
> >>
> >> --
> >> Benjamin Ducke
> >> Geospatial Consultant
> >>
> >> Oxford Archaeology Digital
> >> Janus House
> >> Osney Mead
> >> OX2 0ES
> >> Oxford, U.K.
> >>
> >> Tel: +44 (0)1865 263 800 (switchboard)
> >> Tel: +44 (0)1865 980 758 (direct)
> >> Fax :+44 (0)1865 793 496
> >> [hidden email]
> >> http://oadigital.net
> >>
> >>
> >>
> >>
> >>
> >> ------
> >> Files attached to this email may be in ISO 26300 format (OASIS Open
> >> Document Format). If you have difficulty opening them, please visit
> >> http://iso26300.info for more information.
> >>
> >> _______________________________________________
> >> grass-user mailing list
> >> [hidden email]
> >> http://lists.osgeo.org/mailman/listinfo/grass-user
> >
> > _______________________________________________
> > grass-user mailing list
> > [hidden email]
> > http://lists.osgeo.org/mailman/listinfo/grass-user



--
Dylan Beaudette
Soil Resource Laboratory
http://casoilresource.lawr.ucdavis.edu/
University of California at Davis
530.754.7341
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: [GRASS-user] Searching Docs about 3D geological modelisation

Benjamin Ducke
Hey, good news. Please keep us updated!

Ben

----- Original Message -----
From: "Dylan Beaudette" <[hidden email]>
To: "Thomas Adams" <[hidden email]>
Cc: "grass list" <[hidden email]>
Sent: Wednesday, January 20, 2010 8:10:45 PM GMT +01:00 Amsterdam / Berlin / Bern / Rome / Stockholm / Vienna
Subject: Re: [GRASS-user] Searching Docs about 3D geological modelisation

Quick update:

I recently heard back from Graham Fogg here on campus, and he is in favor of
allowing T-PROGS to be used within GRASS. However, he is still waiting for
the final go-ahead from the original author.

Dylan

On Monday 11 January 2010, Thomas Adams wrote:

> Dylan,
>
> Can you tell me how to obtain TPROGS? Is it only available commercially?
>
> Thanks,
> Tom
>
> Dylan Beaudette wrote:
> > Two more ideas:
> >
> > 1. conditional simulation, based on a 3D variogram model
> > 2. transition probability-based interpolation of categories
> >
> > Check out gstat for the conditional simulation, and TPROGS for the
> > transition probability. If anything is interested, I have done some
> > programming to connect GRASS and TPROGS.
> >
> > Cheers!
> >
> > Dylan
> >
> > On Fri, Jan 8, 2010 at 1:24 PM, Benjamin Ducke
> >
> > <[hidden email]> wrote:
> >> Woohoo, this forum is always a treasure trove
> >> of good advice. I had not idea SGemS existed!
> >> The Voronoi idea is also good, I am just not sure
> >> that the 3D Voronoi diagram is quite what one
> >> would instinctively think it is.
> >>
> >> http://en.wikipedia.org/wiki/Voronoi_diagram
> >>
> >> says: "In general a cross section of a 3D Voronoi
> >> tessellation is not a 2D Voronoi tessellation itself."
> >>
> >> Need to look into that.
> >>
> >> I don't have much practical experience
> >> with Bayes models, so can't really comment on
> >> that.
> >>
> >> Cheers,
> >>
> >> Ben
> >>
> >> Christian Kaiser wrote:
> >>> It seems to me that this is a 3D interpolation problem with categorical
> >>> variables.
> >>>
> >>> Maybe the Bayesian Maximum Entropy approach could help. There are some
> >>> interesting publications around also for geology and soil sciences, and
> >>> they can deal with categorical data as well. Look for example here:
> >>> http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.ht
> >>>ml#Soil%20Science
> >>>
> >>> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a
> >>> tool for 3D geostatistics.
> >>>
> >>> None of them is available through GRASS, but the algorithms are freely
> >>> available (I think open-source, but not verified).
> >>>
> >>> I am not a geologist, so please forgive if it is not adequate...
> >>>
> >>> Christian Kaiser
> >>>
> >>> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
> >>>> Rich Shepard wrote:
> >>>>>> material. There is no interpolation algorithm in GRASS currently
> >>>>>> which can
> >>>>>> handle that sort of data well.
> >>>>>
> >>>>>  So what is needed is a political algorithm. :-)
> >>>>
> >>>> That's actually right: given the presence of n different
> >>>> layer types in the vicinity of an empty voxel, the algorithm
> >>>> would need to decide by some sort of "majority vote"
> >>>> which type to assign to that voxel.
> >>>>
> >>>>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would
> >>>>> solve the problem.
> >>>>
> >>>> How? You could make the interpolated value depend on a
> >>>> fuzzy set member function, I suppose, but the situation
> >>>> here is actually so well defined that I think a probabilistic
> >>>> approach would be preferable. Since each voxel can only
> >>>> store one value, a second output map could store the
> >>>> classification probability. That may be very useful
> >>>> for visualization (you could show voxels with little
> >>>> probability hazier).
> >>>>
> >>>> Ben
> >>>>
> >>>>> Rich
> >>>>> _______________________________________________
> >>>>> grass-user mailing list
> >>>>> [hidden email]
> >>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
> >>>>
> >>>> --
> >>>> Benjamin Ducke
> >>>> Geospatial Consultant
> >>>>
> >>>> Oxford Archaeology Digital
> >>>> Janus House
> >>>> Osney Mead
> >>>> OX2 0ES
> >>>> Oxford, U.K.
> >>>>
> >>>> Tel: +44 (0)1865 263 800 (switchboard)
> >>>> Tel: +44 (0)1865 980 758 (direct)
> >>>> Fax :+44 (0)1865 793 496
> >>>> [hidden email]
> >>>> http://oadigital.net
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>> ------
> >>>> Files attached to this email may be in ISO 26300 format (OASIS Open
> >>>> Document Format). If you have difficulty opening them, please visit
> >>>> http://iso26300.info for more information.
> >>>>
> >>>> _______________________________________________
> >>>> grass-user mailing list
> >>>> [hidden email]
> >>>> http://lists.osgeo.org/mailman/listinfo/grass-user
> >>>
> >>> _______________________________________________
> >>> grass-user mailing list
> >>> [hidden email]
> >>> http://lists.osgeo.org/mailman/listinfo/grass-user
> >>
> >> --
> >> Benjamin Ducke
> >> Geospatial Consultant
> >>
> >> Oxford Archaeology Digital
> >> Janus House
> >> Osney Mead
> >> OX2 0ES
> >> Oxford, U.K.
> >>
> >> Tel: +44 (0)1865 263 800 (switchboard)
> >> Tel: +44 (0)1865 980 758 (direct)
> >> Fax :+44 (0)1865 793 496
> >> [hidden email]
> >> http://oadigital.net
> >>
> >>
> >>
> >>
> >>
> >> ------
> >> Files attached to this email may be in ISO 26300 format (OASIS Open
> >> Document Format). If you have difficulty opening them, please visit
> >> http://iso26300.info for more information.
> >>
> >> _______________________________________________
> >> grass-user mailing list
> >> [hidden email]
> >> http://lists.osgeo.org/mailman/listinfo/grass-user
> >
> > _______________________________________________
> > grass-user mailing list
> > [hidden email]
> > http://lists.osgeo.org/mailman/listinfo/grass-user



--
Dylan Beaudette
Soil Resource Laboratory
http://casoilresource.lawr.ucdavis.edu/
University of California at Davis
530.754.7341
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user



------
Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: [GRASS-user] Searching Docs about 3D geological modelisation

Dylan Beaudette
In reply to this post by Benjamin Ducke
On Saturday 09 January 2010, Benjamin Ducke wrote:

> Cheers for these. They are certainly all highly interesting.
> Do you have an actual link for the T-PROGS software itself?
> All I can seem to come up with are interfaces from other
> software and publications mentioning it.
>
> I would certainly be interested in taking a look at your
> GRASS interface. Is T-PROGS open source?
>
> My gut feeling is that the T-PROGS approach would give better
> results than 3D kriging, as it seems better able to to
> follow 3D shape trends:
>
> http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;37&g=50
>
> ... but that certainly would need testing.
>
> Having said that, I also like this approach for a more
> heuristic model:
>
> http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;41&g=50
>
> It's very simple and could easily be implemented directly
> in GRASS GIS. In fact, I coded something very similar to this
> for archaeological stratigraphy reconstruction a while back.
>
> Cheers,
>
> Ben

Hi Ben and others,

Here are some concerns from the author of the TPROGS software:

---------------------------------------------------------------------------------------------------------
Steve is hesitant because he's not sure what the finished product would be. I
think he's probably concerned about misapplication or perhaps some kind of
ripoff. Can you provide a bit more background on where you see this going?
---------------------------------------------------------------------------------------------------------

I think that it would be helpful to put together a small proposal, regarding
how the TPROGS source code / ideas would be integrated into GRASS. It seems
like the author is worried about use without citation, and once he
understands what GRASS developers have in mind, should be for it.

To start the discussion, I propose that the methods used within the TPROGS
software be integrated (with proper citations) into a GRASS library, so that
a series of modules can perform the multi-step process associated with
modeling transition probabilities. Furthermore, the GRASS rast3 (voxel)
datatype should be used to store the resulting structures-- this will make
visualization with NVIZ / Paraview a snap.

Alternatively, we may be able to link GRASS with TPROGS with a little bit of
python glue. While this may work if there are limitations regarding the use
of the TPROGS source, I think that having these algorithms present in the
GRASS libraries would be a real benefit.

I have CC-ed Graham, so that  we can keep him in the conversation.

Cheers,
Dylan


>
> ----- Original Message -----
> From: "Dylan Beaudette" <[hidden email]>
> To: "Benjamin Ducke" <[hidden email]>
> Cc: "GRASS user list" <[hidden email]>
> Sent: Saturday, January 9, 2010 4:30:40 AM GMT +01:00 Amsterdam / Berlin /
> Bern / Rome / Stockholm / Vienna Subject: Re: [GRASS-user] Searching Docs
> about 3D geological modelisation
>
> Two more ideas:
>
> 1. conditional simulation, based on a 3D variogram model
> 2. transition probability-based interpolation of categories
>
> Check out gstat for the conditional simulation, and TPROGS for the
> transition probability. If anything is interested, I have done some
> programming to connect GRASS and TPROGS.
>
> Cheers!
>
> Dylan
>
> On Fri, Jan 8, 2010 at 1:24 PM, Benjamin Ducke
>
> <[hidden email]> wrote:
> > Woohoo, this forum is always a treasure trove
> > of good advice. I had not idea SGemS existed!
> > The Voronoi idea is also good, I am just not sure
> > that the 3D Voronoi diagram is quite what one
> > would instinctively think it is.
> >
> > http://en.wikipedia.org/wiki/Voronoi_diagram
> >
> > says: "In general a cross section of a 3D Voronoi
> > tessellation is not a 2D Voronoi tessellation itself."
> >
> > Need to look into that.
> >
> > I don't have much practical experience
> > with Bayes models, so can't really comment on
> > that.
> >
> > Cheers,
> >
> > Ben
> >
> > Christian Kaiser wrote:
> >> It seems to me that this is a 3D interpolation problem with categorical
> >> variables.
> >>
> >> Maybe the Bayesian Maximum Entropy approach could help. There are some
> >> interesting publications around also for geology and soil sciences, and
> >> they can deal with categorical data as well. Look for example here:
> >> http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.htm
> >>l#Soil%20Science
> >>
> >> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a
> >> tool for 3D geostatistics.
> >>
> >> None of them is available through GRASS, but the algorithms are freely
> >> available (I think open-source, but not verified).
> >>
> >> I am not a geologist, so please forgive if it is not adequate...
> >>
> >> Christian Kaiser
> >>
> >> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
> >>> Rich Shepard wrote:
> >>>>> material. There is no interpolation algorithm in GRASS currently
> >>>>> which can
> >>>>> handle that sort of data well.
> >>>>
> >>>>  So what is needed is a political algorithm. :-)
> >>>
> >>> That's actually right: given the presence of n different
> >>> layer types in the vicinity of an empty voxel, the algorithm
> >>> would need to decide by some sort of "majority vote"
> >>> which type to assign to that voxel.
> >>>
> >>>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would
> >>>> solve the problem.
> >>>
> >>> How? You could make the interpolated value depend on a
> >>> fuzzy set member function, I suppose, but the situation
> >>> here is actually so well defined that I think a probabilistic
> >>> approach would be preferable. Since each voxel can only
> >>> store one value, a second output map could store the
> >>> classification probability. That may be very useful
> >>> for visualization (you could show voxels with little
> >>> probability hazier).
> >>>
> >>> Ben
> >>>
> >>>> Rich
> >>>> _______________________________________________
> >>>> grass-user mailing list
> >>>> [hidden email]
> >>>> http://lists.osgeo.org/mailman/listinfo/grass-user
> >>>
> >>> --
> >>> Benjamin Ducke
> >>> Geospatial Consultant
> >>>
> >>> Oxford Archaeology Digital
> >>> Janus House
> >>> Osney Mead
> >>> OX2 0ES
> >>> Oxford, U.K.
> >>>
> >>> Tel: +44 (0)1865 263 800 (switchboard)
> >>> Tel: +44 (0)1865 980 758 (direct)
> >>> Fax :+44 (0)1865 793 496
> >>> [hidden email]
> >>> http://oadigital.net
> >>>
> >>>
> >>>
> >>>
> >>>
> >>> ------
> >>> Files attached to this email may be in ISO 26300 format (OASIS Open
> >>> Document Format). If you have difficulty opening them, please visit
> >>> http://iso26300.info for more information.
> >>>
> >>> _______________________________________________
> >>> grass-user mailing list
> >>> [hidden email]
> >>> http://lists.osgeo.org/mailman/listinfo/grass-user
> >>
> >> _______________________________________________
> >> grass-user mailing list
> >> [hidden email]
> >> http://lists.osgeo.org/mailman/listinfo/grass-user
> >
> > --
> > Benjamin Ducke
> > Geospatial Consultant
> >
> > Oxford Archaeology Digital
> > Janus House
> > Osney Mead
> > OX2 0ES
> > Oxford, U.K.
> >
> > Tel: +44 (0)1865 263 800 (switchboard)
> > Tel: +44 (0)1865 980 758 (direct)
> > Fax :+44 (0)1865 793 496
> > [hidden email]
> > http://oadigital.net
> >
> >
> >
> >
> >
> > ------
> > Files attached to this email may be in ISO 26300 format (OASIS Open
> > Document Format). If you have difficulty opening them, please visit
> > http://iso26300.info for more information.
> >
> > _______________________________________________
> > grass-user mailing list
> > [hidden email]
> > http://lists.osgeo.org/mailman/listinfo/grass-user
>
> ------
> Files attached to this email may be in ISO 26300 format (OASIS Open
> Document Format). If you have difficulty opening them, please visit
> http://iso26300.info for more information.
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user



--
Dylan Beaudette
Soil Resource Laboratory
http://casoilresource.lawr.ucdavis.edu/
University of California at Davis
530.754.7341
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: [GRASS-user] Searching Docs about 3D geological modelisation

Benjamin Ducke
Why not just ask Steve what he is concerned about and
what he would like us to do so that he can shed his
concerns?
And then try to find a way to accommodate him?
If he got more directly involved into this process,
it might make him feel less uneasy about it.

Ben

----- Original Message -----
From: "Dylan Beaudette" <[hidden email]>
To: [hidden email]
Cc: "Benjamin Ducke" <[hidden email]>, "Graham Fogg" <[hidden email]>
Sent: Friday, January 22, 2010 9:38:18 PM GMT +01:00 Amsterdam / Berlin / Bern / Rome / Stockholm / Vienna
Subject: Re: [GRASS-user] Searching Docs about 3D geological modelisation

On Saturday 09 January 2010, Benjamin Ducke wrote:

> Cheers for these. They are certainly all highly interesting.
> Do you have an actual link for the T-PROGS software itself?
> All I can seem to come up with are interfaces from other
> software and publications mentioning it.
>
> I would certainly be interested in taking a look at your
> GRASS interface. Is T-PROGS open source?
>
> My gut feeling is that the T-PROGS approach would give better
> results than 3D kriging, as it seems better able to to
> follow 3D shape trends:
>
> http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;37&g=50
>
> ... but that certainly would need testing.
>
> Having said that, I also like this approach for a more
> heuristic model:
>
> http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=ARTICLES;41&g=50
>
> It's very simple and could easily be implemented directly
> in GRASS GIS. In fact, I coded something very similar to this
> for archaeological stratigraphy reconstruction a while back.
>
> Cheers,
>
> Ben

Hi Ben and others,

Here are some concerns from the author of the TPROGS software:

---------------------------------------------------------------------------------------------------------
Steve is hesitant because he's not sure what the finished product would be. I
think he's probably concerned about misapplication or perhaps some kind of
ripoff. Can you provide a bit more background on where you see this going?
---------------------------------------------------------------------------------------------------------

I think that it would be helpful to put together a small proposal, regarding
how the TPROGS source code / ideas would be integrated into GRASS. It seems
like the author is worried about use without citation, and once he
understands what GRASS developers have in mind, should be for it.

To start the discussion, I propose that the methods used within the TPROGS
software be integrated (with proper citations) into a GRASS library, so that
a series of modules can perform the multi-step process associated with
modeling transition probabilities. Furthermore, the GRASS rast3 (voxel)
datatype should be used to store the resulting structures-- this will make
visualization with NVIZ / Paraview a snap.

Alternatively, we may be able to link GRASS with TPROGS with a little bit of
python glue. While this may work if there are limitations regarding the use
of the TPROGS source, I think that having these algorithms present in the
GRASS libraries would be a real benefit.

I have CC-ed Graham, so that  we can keep him in the conversation.

Cheers,
Dylan


>
> ----- Original Message -----
> From: "Dylan Beaudette" <[hidden email]>
> To: "Benjamin Ducke" <[hidden email]>
> Cc: "GRASS user list" <[hidden email]>
> Sent: Saturday, January 9, 2010 4:30:40 AM GMT +01:00 Amsterdam / Berlin /
> Bern / Rome / Stockholm / Vienna Subject: Re: [GRASS-user] Searching Docs
> about 3D geological modelisation
>
> Two more ideas:
>
> 1. conditional simulation, based on a 3D variogram model
> 2. transition probability-based interpolation of categories
>
> Check out gstat for the conditional simulation, and TPROGS for the
> transition probability. If anything is interested, I have done some
> programming to connect GRASS and TPROGS.
>
> Cheers!
>
> Dylan
>
> On Fri, Jan 8, 2010 at 1:24 PM, Benjamin Ducke
>
> <[hidden email]> wrote:
> > Woohoo, this forum is always a treasure trove
> > of good advice. I had not idea SGemS existed!
> > The Voronoi idea is also good, I am just not sure
> > that the 3D Voronoi diagram is quite what one
> > would instinctively think it is.
> >
> > http://en.wikipedia.org/wiki/Voronoi_diagram
> >
> > says: "In general a cross section of a 3D Voronoi
> > tessellation is not a 2D Voronoi tessellation itself."
> >
> > Need to look into that.
> >
> > I don't have much practical experience
> > with Bayes models, so can't really comment on
> > that.
> >
> > Cheers,
> >
> > Ben
> >
> > Christian Kaiser wrote:
> >> It seems to me that this is a 3D interpolation problem with categorical
> >> variables.
> >>
> >> Maybe the Bayesian Maximum Entropy approach could help. There are some
> >> interesting publications around also for geology and soil sciences, and
> >> they can deal with categorical data as well. Look for example here:
> >> http://www.enge.ucl.ac.be/staff/curr/Bogaert/biblioBME/BMEbibsubject.htm
> >>l#Soil%20Science
> >>
> >> Or maybe you can have a look at SGeMS (http://sgems.sourceforge.net), a
> >> tool for 3D geostatistics.
> >>
> >> None of them is available through GRASS, but the algorithms are freely
> >> available (I think open-source, but not verified).
> >>
> >> I am not a geologist, so please forgive if it is not adequate...
> >>
> >> Christian Kaiser
> >>
> >> On 8 janv. 2010, at 11:04, Benjamin Ducke wrote:
> >>> Rich Shepard wrote:
> >>>>> material. There is no interpolation algorithm in GRASS currently
> >>>>> which can
> >>>>> handle that sort of data well.
> >>>>
> >>>>  So what is needed is a political algorithm. :-)
> >>>
> >>> That's actually right: given the presence of n different
> >>> layer types in the vicinity of an empty voxel, the algorithm
> >>> would need to decide by some sort of "majority vote"
> >>> which type to assign to that voxel.
> >>>
> >>>>  Kidding aside, I suspect that a fuzzy interpolation algorithm would
> >>>> solve the problem.
> >>>
> >>> How? You could make the interpolated value depend on a
> >>> fuzzy set member function, I suppose, but the situation
> >>> here is actually so well defined that I think a probabilistic
> >>> approach would be preferable. Since each voxel can only
> >>> store one value, a second output map could store the
> >>> classification probability. That may be very useful
> >>> for visualization (you could show voxels with little
> >>> probability hazier).
> >>>
> >>> Ben
> >>>
> >>>> Rich
> >>>> _______________________________________________
> >>>> grass-user mailing list
> >>>> [hidden email]
> >>>> http://lists.osgeo.org/mailman/listinfo/grass-user
> >>>
> >>> --
> >>> Benjamin Ducke
> >>> Geospatial Consultant
> >>>
> >>> Oxford Archaeology Digital
> >>> Janus House
> >>> Osney Mead
> >>> OX2 0ES
> >>> Oxford, U.K.
> >>>
> >>> Tel: +44 (0)1865 263 800 (switchboard)
> >>> Tel: +44 (0)1865 980 758 (direct)
> >>> Fax :+44 (0)1865 793 496
> >>> [hidden email]
> >>> http://oadigital.net
> >>>
> >>>
> >>>
> >>>
> >>>
> >>> ------
> >>> Files attached to this email may be in ISO 26300 format (OASIS Open
> >>> Document Format). If you have difficulty opening them, please visit
> >>> http://iso26300.info for more information.
> >>>
> >>> _______________________________________________
> >>> grass-user mailing list
> >>> [hidden email]
> >>> http://lists.osgeo.org/mailman/listinfo/grass-user
> >>
> >> _______________________________________________
> >> grass-user mailing list
> >> [hidden email]
> >> http://lists.osgeo.org/mailman/listinfo/grass-user
> >
> > --
> > Benjamin Ducke
> > Geospatial Consultant
> >
> > Oxford Archaeology Digital
> > Janus House
> > Osney Mead
> > OX2 0ES
> > Oxford, U.K.
> >
> > Tel: +44 (0)1865 263 800 (switchboard)
> > Tel: +44 (0)1865 980 758 (direct)
> > Fax :+44 (0)1865 793 496
> > [hidden email]
> > http://oadigital.net
> >
> >
> >
> >
> >
> > ------
> > Files attached to this email may be in ISO 26300 format (OASIS Open
> > Document Format). If you have difficulty opening them, please visit
> > http://iso26300.info for more information.
> >
> > _______________________________________________
> > grass-user mailing list
> > [hidden email]
> > http://lists.osgeo.org/mailman/listinfo/grass-user
>
> ------
> Files attached to this email may be in ISO 26300 format (OASIS Open
> Document Format). If you have difficulty opening them, please visit
> http://iso26300.info for more information.
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user



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



------
Files attached to this email may be in ISO 26300 format (OASIS Open Document Format). If you have difficulty opening them, please visit http://iso26300.info for more information.

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
12
Loading...