Hello Dean

         The best practical method for this, is plot the variable versus
the coordinates.

It is like a cross section but with all cross section superimpose.

This works for 2D or 3D. You make Vr x Coord X or Vr x Coord Y, and so one.

By  this way you will see many aspects of your data. Perhaps a global
non-stationarity  or only a local non-stationarity.

All of the best

Armando

Dean Monroe wrote:

> Group:
>
>     How is the best to graphically show non-stationarity in spatial
> data with say one variable of interest and two dimensions?  In time
> series, one can plot realizations over time, what is the spatial analog?
>     Second, what may a person do to correct non-stationarity?  Again,
> in time series a common practice is to use first, second, etc differences.
>
> Best,
> Dean Monroe
> Oklahoma State University
> Environmental Sciences



[Non-text portions of this message have been removed]

All, I have an interesting problem that is a bit off the usual geostat
track.  The problem is the estimation of the amount of tracer deposited in a
cave system.  The cave system is composed of multiple rooms and passages
connecting the rooms.  Air flow and transport models can be used to model
the deposition of the tracer, but are difficult to set up and calibrate and
I'd like to pursue a more geostatistical approach to the problem (if
possible).

However, a straight forward application of kriging the amount of deposition
based on a number of samples will not work for at least two reasons:
1) Euclidean distance is not very meaningful as two rooms in the system may
only be separated by a 10 meter thick wall, yet the tortuous air flow path
from one room to the other may be over 500 meters.  Therefore the
connections between sample points resemble something like a connected graph
ala graph theory.  But perhaps it is possible to use this information to
remap the cave system into some sort of "connection space" and build
variograms and do kriging in that space before remapping to the actual
coordiante system (?)
2) If it were possible to develop a covariance matrix using the sample data
in some transformed coordinate system, it would not be symmetric.  Due to
the air flow patterns in the cave system, point B may be "downwind" of point
A and there is a B->A connection, but there is no A->B connectivity.  In the
parlance of graph theory, this would be a "directed graph".

I've found work across several different fields where the "best" places to
take a sample in such a system can be determined using graph theory coupled
with linear programming and/or heurisitc optimization techniques.  I have
not come across any work where estimations are made in the system based on a
finite number of existing samples, except for those that resort to physics
based models (i.e., flow and transport).

If anyone has pondered this problem before and can point me towards any
publications, I would be very appreciative.

thanks

Sean



Sean A. McKenna  Ph.D.
Geohydrology Department
Sandia National Laboratories
PO Box 5800  MS 0735
Albuquerque, NM 87185-0735
ph: 505 844-2450



--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

I'm interesting in few theorical aspects of geostatistics, today the question is
about the variogram functions:



a)      What are the properties that need some functions for be considered legal
as Theoric Variogram.

b)      If I put n legal functions in the Kriging Equation it is legal too.



I well be happy with any answer or Link.



King regards



Adrian Martínez Vargas

ISMM, Las Coloradas  s/n

www.geocities.com/adriangeologo/adrian.html





[Non-text portions of this message have been removed]

Hi Sean,

Your problem seems to bear some similarity with modeling of
river networks where meaningful distances cannot be defined
in the Euclidian space and downstream/upstream relationships
need to be fulfilled.
A few days ago, Pascal Monestiez from INRA Avignon gave a
talk on Geostatistical modelling of spatial processes on trees:
applications to drainage networks, which might be of interest
to you... and there should be other related papers in the literature.

Cheers,

Pierre
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

Dr. Pierre Goovaerts
President of PGeostat, LLC
Chief Scientist with Biomedware Inc.
710 Ridgemont Lane
Ann Arbor, Michigan, 48103-1535, U.S.A.

E-mail:  goovaert@...
Phone:   (734) 668-9900
Fax:     (734) 668-7788
http://alumni.engin.umich.edu/~goovaert/

<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

On Fri, 12 Sep 2003, McKenna, Sean A wrote:

> All, I have an interesting problem that is a bit off the usual geostat
> track.  The problem is the estimation of the amount of tracer deposited in a
> cave system.  The cave system is composed of multiple rooms and passages
> connecting the rooms.  Air flow and transport models can be used to model
> the deposition of the tracer, but are difficult to set up and calibrate and
> I'd like to pursue a more geostatistical approach to the problem (if
> possible).
>
> However, a straight forward application of kriging the amount of deposition
> based on a number of samples will not work for at least two reasons:
> 1) Euclidean distance is not very meaningful as two rooms in the system may
> only be separated by a 10 meter thick wall, yet the tortuous air flow path
> from one room to the other may be over 500 meters.  Therefore the
> connections between sample points resemble something like a connected graph
> ala graph theory.  But perhaps it is possible to use this information to
> remap the cave system into some sort of "connection space" and build
> variograms and do kriging in that space before remapping to the actual
> coordiante system (?)
> 2) If it were possible to develop a covariance matrix using the sample data
> in some transformed coordinate system, it would not be symmetric.  Due to
> the air flow patterns in the cave system, point B may be "downwind" of point
> A and there is a B->A connection, but there is no A->B connectivity.  In the
> parlance of graph theory, this would be a "directed graph".
>
> I've found work across several different fields where the "best" places to
> take a sample in such a system can be determined using graph theory coupled
> with linear programming and/or heurisitc optimization techniques.  I have
> not come across any work where estimations are made in the system based on a
> finite number of existing samples, except for those that resort to physics
> based models (i.e., flow and transport).
>
> If anyone has pondered this problem before and can point me towards any
> publications, I would be very appreciative.
>
> thanks
>
> Sean
>
>
>
> Sean A. McKenna  Ph.D.
> Geohydrology Department
> Sandia National Laboratories
> PO Box 5800  MS 0735
> Albuquerque, NM 87185-0735
> ph: 505 844-2450
>
>
>
> --
> * To post a message to the list, send it to ai-geostats@...
> * As a general service to the users, please remember to post a summary of any
useful responses to your questions.
> * To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
> * Support to the list is provided at http://www.ai-geostats.org
>


--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Interesting question...

    Anyway, despite usual bibliography regarding spatial dependecy, spatial
autocorrelation and how to measure it, If you are unaware of it, I would
suggest the usage of the "Distance" software, It is the evolution of the
TRANSECT DOS-based program, and I think you will find it interesting:

http://www.ruwpa.st-and.ac.uk/distance/distanceabout.html

(I have no affiliation with the authors)

Gaston Pezzuchi


Mensaje citado por Ramesh Krishnamurthy <tragoram@...>:

> Dear folks,
>
> my problem is to identify whether or not spatial
> autocorrelation exists in binary data. we quantified
> presence or absence of different deer species based on
> signs/tracks. Our sampling schemes was to record the
> presence/absence in a series of 250m segments (we
> called this as a transect constituted by these
> segments)and the trasects varied in length. we wanted
> to establish an index that would describe abundance
> rating for that particular transect, assuming that the
> index is linearly relatied to true abundance of the
> species. We thought of arriving at proportion value,
> by summing up the number of segments with signs and
> then divided them by the total number of segments in
> the transect. however, we wanted to determine the
> spatial autocorrelation in the data, so that we can
> obtain probability value for the individual transects.
> the data is from one time smapling, but we have about
> 250 replicate transects covering 950km.
>
> i would apprecite if any of you can provide me the
> direction (both methods if you have alreay known or
> used, and literature) to deal with this kind of binary
> data.
>
> thanks in advance.
>
> regards
>
> ramesh
>
> ________________________________________________________________________
> Yahoo! India Matrimony: Find your partner online.
> Go to http://yahoo.shaadi.com
>
> --
> * To post a message to the list, send it to ai-geostats@...
> * As a general service to the users, please remember to post a summary of any
> useful responses to your questions.
> * To unsubscribe, send an email to majordomo@... with no subject and
> "unsubscribe ai-geostats" followed by "end" on the next line in the message
> body. DO NOT SEND Subscribe/Unsubscribe requests to the list
> * Support to the list is provided at http://www.ai-geostats.org
>




-------------------------------------------------
  Internet Way - Webmail - http://www.way.com.ar

--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Thanks for your suggestion. I am aware of the distance
software and the concepts behind it. In fact, i have
used in my doctoral studies to estimate density of
pheasants in the Himalaya. But the recent work was a
landscape based survey, which means that we had to
survey large area of over 15000sq.km. within 4 months
time. so, we could not go for direct counts, which
would require both repeated measures other than
replicate transects.

i am finding it difficult to deal with this kind of
presence/absence data, as it is difficult to say, the
occurence is neighbouring segments is because of
spatial-autocorrelation or because the species density
is truly high. from the data, i picked our pair of
segments (say, first and second), and counted how many
segments have similar value..and similarly with 2 and
3, and 1 and 3. I found almost similar values in all
these pairs. This created doubt that this could be
because of high density.

one solution perhaps is to estimate density of deer
based on direct counts, and find the relationship with
the binary data.

THANKS AGIAN FOR YOUR EFFORTS.

best

ramesh

  --- gpezzuchi@... wrote: > Interesting
question...
>
>    Anyway, despite usual bibliography regarding
> spatial dependecy, spatial
> autocorrelation and how to measure it, If you are
> unaware of it, I would
> suggest the usage of the "Distance" software, It is
> the evolution of the
> TRANSECT DOS-based program, and I think you will
> find it interesting:
>
>
http://www.ruwpa.st-and.ac.uk/distance/distanceabout.html
>
> (I have no affiliation with the authors)
>
> Gaston Pezzuchi


________________________________________________________________________
Yahoo! India Matrimony: Find your partner online.
Go to http://yahoo.shaadi.com

--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

We are interested in hearing about any multivariate statistics software
that has been written for ArcView (3.x, 8.x) - both freeware and commercial.



William C. Thayer, P.E.
Environmental Science Center
Syracuse Research Corporation
thayer@...
Phone: 315-452-8424  Fax: x8440
web: http://esc.syrres.com/geosem/



--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Dear folks,

my problem is to identify whether or not spatial
autocorrelation exists in binary data. we quantified
presence or absence of different deer species based on
signs/tracks. Our sampling schemes was to record the
presence/absence in a series of 250m segments (we
called this as a transect constituted by these
segments)and the trasects varied in length. we wanted
to establish an index that would describe abundance
rating for that particular transect, assuming that the
index is linearly relatied to true abundance of the
species. We thought of arriving at proportion value,
by summing up the number of segments with signs and
then divided them by the total number of segments in
the transect. however, we wanted to determine the
spatial autocorrelation in the data, so that we can
obtain probability value for the individual transects.
the data is from one time smapling, but we have about
250 replicate transects covering 950km.

i would apprecite if any of you can provide me the
direction (both methods if you have alreay known or
used, and literature) to deal with this kind of binary
data.

thanks in advance.

regards

ramesh

________________________________________________________________________
Yahoo! India Matrimony: Find your partner online.
Go to http://yahoo.shaadi.com

--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

I recommend you to see the theory of Indicator Krigins and Simulations, with
these data you can make a long variety of studies. I recommend you to see
GSLIB manual and software, it have a long variety of possibilities that you
can use. I recommend you to use ISATIS.


----- Original Message -----
From: "Ramesh Krishnamurthy" <tragoram@...>
To: <ai-geostats@...>
Sent: Monday, September 15, 2003 12:49 PM
Subject: AI-GEOSTATS: spatial autocorrelation in binary data


> Dear folks,
>
> my problem is to identify whether or not spatial
> autocorrelation exists in binary data. we quantified
> presence or absence of different deer species based on
> signs/tracks. Our sampling schemes was to record the
> presence/absence in a series of 250m segments (we
> called this as a transect constituted by these
> segments)and the trasects varied in length. we wanted
> to establish an index that would describe abundance
> rating for that particular transect, assuming that the
> index is linearly relatied to true abundance of the
> species. We thought of arriving at proportion value,
> by summing up the number of segments with signs and
> then divided them by the total number of segments in
> the transect. however, we wanted to determine the
> spatial autocorrelation in the data, so that we can
> obtain probability value for the individual transects.
> the data is from one time smapling, but we have about
> 250 replicate transects covering 950km.
>
> i would apprecite if any of you can provide me the
> direction (both methods if you have alreay known or
> used, and literature) to deal with this kind of binary
> data.
>
> thanks in advance.
>
> regards
>
> ramesh
>
> ________________________________________________________________________
> Yahoo! India Matrimony: Find your partner online.
> Go to http://yahoo.shaadi.com
>
> --
> * To post a message to the list, send it to ai-geostats@...
> * As a general service to the users, please remember to post a summary of
any useful responses to your questions.
> * To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
> * Support to the list is provided at http://www.ai-geostats.org
>




--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Dear William,
You can find extension for ArcView 3.x (tested with version 3.2) that
allows to exchange data with ADE-4 software, to launch some multivariate
analyses (e.g. PCA, local and global PCA, CCA...) and to get results within
ArcView at the following adress:
http://www.steph280.freesurf.fr/index.html.

>We are interested in hearing about any multivariate statistics software
>that has been written for ArcView (3.x, 8.x) - both freeware and commercial.
>
>
>
>William C. Thayer, P.E.
>Environmental Science Center
>Syracuse Research Corporation
>thayer@...
>Phone: 315-452-8424  Fax: x8440
>web: http://esc.syrres.com/geosem/
>
>
>
>--
>* To post a message to the list, send it to ai-geostats@...
>* As a general service to the users, please remember to post a summary of
>any useful responses to your questions.
>* To unsubscribe, send an email to majordomo@... with no subject and
>"unsubscribe ai-geostats" followed by "end" on the next line in the
>message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
>* Support to the list is provided at http://www.ai-geostats.org

--------------------------------------------------------------------------------\
--------------------------------------------------------------------------------\
--------------------------------------------------------------------------------\
--------------------------------------
Sébastien OLLIER
Laboratoire de biométrie et biologie évolutive
UMR 5558
Bâtiment 741 - 711
43, boulevard du 11 novembre 1918
69622 VILLEURBANNE Cedex
    Tel: (33) (0)4 72 43 29 35
Fax : (33) (0)4 72 43 13 88
ollier@...
--------------------------------------------------------------------------------\
-----------------------------------------------------------



[Non-text portions of this message have been removed]

Hi,

can anyone here help me in finding a script or Plugin for ArcGIS or even
better Surfer 8, which is able to perform gridding applying classical linear
triangulation (as seen in Isaaks / Srivastava pp.251-253) ? I know the
gridding in Surfer and Arc as well, but they use Delaunay Triangulation. I
need to compare both methods for exact mapping without discontinouties.

Big thanks,

Ursula Manns...

_________________________________________________________________
MSN Messenger  -  Wer in Echtzeit kommunizieren will, lädt den MSN
Messenger. Cool, kostenlos und mit 3D Emoticons:  http://messenger.msn.ch


--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Dear  all,  is any one aware of any software package (not Surfer)
which will allow me to address the topology of a dynamical system
?

W/best wishes,

----
aldo.solari@... (fisheries biologist)

----
Friday, September 26, 2003, 9:20:26 [Canary Islands, on GMT].
----



--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Please find below the call for papers for the 1st International
Symposium on Applications of Constraint Databases (CDB'04).

One of the aims of the Symposium is to look for new research directions
and applications in constraint databases. Currently, the statistical
analysis of spatial data is mostly disregarded in the constraint
framework. With this mail we invite people from outside the database
community to submit papers to the Symposium addressing, for example,
the use of constraint databases in the statistical analysis of spatial
data.

Best regards,

Floris Geerts
Publicity Chair CDB'04

--------------------------------------------------------------------
CDB'04

First CALL FOR PAPERS

1st INTERNATIONAL SYMPOSIUM ON APPLICATIONS OF
CONSTRAINT DATABASES

in conjunction with SIGMOD-PODS 2004

Paris, France, 12-13 June 2004

http://www.luc.ac.be/cdb04


SCOPE

The last few years saw a growing interest of constraint
database theory, query evaluation, and applications in a
variety of conferences, journals, and books. This symposium
wants to bring together people from several diverse areas
that can contribute to the practice and the application of
constraint databases. It is a continuation and extension of
previous workshops held in Friedrichshafen, Germany (1995),
Cambridge, USA (1996), Delphi, Greece (1997), and Seattle,
USA (1998) as well as of the work in the comprehensive volume
"Constraint Databases" edited by G. Kuper, L. Libkin and J.
Paredaens (2000) and  the textbook "Introduction to
Constraint Databases" by P. Revesz (2002).


TOPICS OF INTEREST

We especially encourage submissions
   - opening new and future directions in constraint database
     research;
   - addressing constraints over domains other than the reals;
   - contributing to a better implementation of constraint
     database systems, in particular of query evaluation;
   - addressing efficient quantifier elimination; and
   - describing applications of constraint databases.

The following is a non-exclusive list of topics of interest.

applications:
               bioinformatics
               CAD and GIS
               computer security
               data mining
               model checking
               string databases

data and knowledge representation:
               approximation techniques
               constraint data extraction
               constraint interpolation
               incomplete information
               spatiotemporal models
               visualization

query evaluation:
               algebras
               indexing
               quantifier elimination

query languages:
               complexity
               expressive power
               new operators


INVITED SPEAKERS

Joos Heintz, Universities of Buenos Aires and of Cantabria
Leonid Libkin, University of Toronto
Andreas Podelski, Max-Planck-Institut für Informatik

PROCEEDINGS

The proceedings will be published as part of the Lecture Notes
in Computer Science (LNCS) series by Springer-Verlag. Authors
are expected to sign a copyright release form to Springer-Verlag.

SUBMISSION INSTRUCTIONS

The submission procedure will be electronic: Potential presenters
should submit e-mail copies of a paper in English and maximum 15
pages in either PostScript or PDF form to one of the program
co-chairs. Selection of papers will be mainly based on relevance,
clarity, and significance. The presented results should be original,
not published somewhere else and not under submission elsewhere.
Please follow the Springer LNCS guidelines for formatting.

IMPORTANT DATES

Abstract submission:       January 23, 2004
Submission deadline:       February 2, 2004
Notification of acceptance:  March 15, 2004
Camera ready copy deadline:  March 26, 2004
Symposium dates:           June 12-13, 2004

PROGRAM CO-CHAIRS

Bart Kuijpers, University of Limburg (LUC), bart.kuijpers@...
Peter Revesz, University of Nebraska, revesz@...

PROGRAM COMMITTEE

Saugata Basu (Georgia Tech)
Alex Brodsky (George Mason University)
Jan Chomicki (SUNY, Buffalo)
Berthe Choueiry (University of Nebraska)
Giorgio Delzanno (University of Genova)
Floris Geerts (University of Helsinki)
Marc Giusti (CNRS, Polytechnique)
Dina Goldin (University of Connecticut)
Stephane Grumbach (INRIA, Paris)
Joxan Jaffar (National University of Singapore)
Manolis Koubarakis (Technical University of Crete)
Stephan Kreutzer (Humboldt-Universität Berlin)
Bart Kuijpers (University of Limburg, LUC)
Gabriel Kuper (University of Trento)
Zoe Lacroix (Arizona State Univiveristy)
Lixin Li (Georgia Southern University)
Jan Paredaens (University of Antwerp)
Peter Revesz (University of Nebraska)
Philippe Rigaux (Universite Paris Sud)
Kai-Uwe Sattler (Technische Universität Ilmenau)
Jianwen Su (University of California, Santa Barbara)
David Toman (University of Waterloo)
Jan Van den Bussche (University of Limburg, LUC)
Dirk Van Gucht (Indiana University)
Nicolai Vorobjov (University of Bath)
Mark Wallace (Imperial College, London)

PROCEEDINGS AND PUBLICITY CHAIR

Floris Geerts (University of Helsinki)






--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Dear list members,
I would be grateful if you could suggest me any software (free domain
preferable) to apply the "scale variance analysis" developed by Moellering and
Tobler (1972).

Since, I am very new in the field of geostatistics, it does not seem to me so
easy to explore the tools offered from the software provided by ai-geostats
server. So, I would very much appreciate any indication of which of them would
be applicable in my case.

Thank you in advance!


Eleni FITOKA
Head of "Archives of Natural Areas"

The Goulandris Natural History Museum
Greek Biotope Wetland Centre
14th kilometre Thessaloniki - Mihaniona,
57001 Thermi, Thessaloniki
Greece

tel: +30 2310 473432, 475604, extension:161
fax: +30 2310 471795
email: helenf@...




[Non-text portions of this message have been removed]

Hi everyone:

I have a high-res DEM (SRTM 3-arcsec) which has local 'blackouts'/'holes',
and a low-res DEM (DTED0) of the same area, which is complete. Many of the
holes are too large for meaningful results with simple interpolation
procedures.

When I merge the two DEMs to fill in the missing data in the high-res DEM,
the generalization of the low-res DEMs elevations leads to 'troughs' (i.e.,
cliffs along the edges) instead of the desired smooth fill-ins.

My idea (taken from GIS literature) is to improve the low-res mesh prior to
merging by using the available high-res data to predict the true elevation
in the areas with holes. My question is, which would be the best available
procedure - local or global? simple (e.g., differencing) or complex (e.g.,
regression models)? I have done a few trials with global approaches but
they didn't provide satisfactory results.

Any advice or pointers are appreciated.

Cheers,

Holger Sandmann


--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

CALL FOR PAPERS

Joint meeting of TIES 2004: The International Environmetrics Society
and ACCURACY 2004: 6th International Symposium on Spatial Accuracy Assessment

For the first time, the annual TIES conference and the biennial Spatial
Accuracy Symposium will hold a joint meeting reflecting their mutual
interest in the quantification of environmental and natural resources
processes and the uncertainty associated with them. Recent
international meetings have been very successful with TIES meeting in the UK,
the USA, and South Africa and the Spatial Accuracy Symposium meeting in
Canada, the Netherlands, and Australia. The joint meeting will be held June 28
- July 1, 2004, at the Holiday Inn By The Bay in Portland, Maine, USA.

The program will consist of presentations by invited plenary speakers and
invited paper, contributed paper, and poster sessions. The website is
still under construction, but additional meeting details may be found at

http://www.ncrs2.fs.fed.us/4801/meetings/ties/

Abstracts for presentations and posters are solicited on important
quantitative aspects of environmental and natural resources problems
with emphasis on spatial data, spatial analyses, and spatial accuracy
assessments. A detailed list of conference topics appears on the
meeting website. Presentations and posters on these and any aspect of
environmental, natural resources, and spatial assessments will be
considered.

Abstracts of not more than 200 words that include title, 3 to 5
keywords, and names, affiliations, and contact information for the contact
author should be sent via regular mail, fax, or e-mail (preferably) by 21
November 2003 to:

Ronald E. McRoberts Fax: +651-649-5174
North Central Research Station
E-mail: rmcroberts@...
USDA Forest Service
1992 Folwell Avenue
Saint Paul, Minnesota 55108 USA

Notification of acceptance and instructions to authors will be sent  by 15
December 2003.

A proceedings will be published and distributed at the meeting, although
publication in the proceedings is not a requirement for presentation at the
meeting.

Rigorous adherence will be maintained to the following schedule:

01 October 2003: Initial Call for Papers
21 November 2003: Abstracts due
15 December 2003: Authors notified of acceptance
01 March 2004: Invited and Contributed manuscripts for publication due




--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Hello
I’m working with dataset discrete(countdown of foraminefera.)
I want to know if the R software works with discrete spatial variable.
Thank you very much for your help

Carlos Cuartas O.




--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Dear Carlos

I think the geoRglm package written by Ole Christensen is what you are
looking for, see

Diggle, P.J., Moyeed, R.A. and Tawn, J.A. (1998).
Model-based Geostatistics (with Discussion).
{\it Applied Statistics}, {\bf 47}, 299--350.

for applications of such spatial models.

Best regards / Venlig hilsen

Søren Lophaven
******************************************************************************
Master of Science in Engineering        |  Ph.D. student
Informatics and Mathematical Modelling  |  Building 321, Room 011
Technical University of Denmark         |  2800 kgs. Lyngby, Denmark
E-mail: snl@...                  |  http://www.imm.dtu.dk/~snl
Telephone: +45 45253419                 |
******************************************************************************

On Thu, 9 Oct 2003, cuartas oquendo carlos humberto wrote:

>
> Hello
> I’m working with dataset discrete(countdown of foraminefera)
> I want to know if the R software works with discrete spatial variable.
> Thank you very much for your help
>
> Carlos Cuartas O.
>
>
>
>
>
> --
> * To post a message to the list, send it to ai-geostats@...
> * As a general service to the users, please remember to post a summary of any
useful responses to your questions.
> * To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
> * Support to the list is provided at http://www.ai-geostats.org
>


--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Group:

     I am working on a project where we are looking at the local variance of crop
density, by using a vegetative index (NDVI).  The spot sensor readings are
approx 4m apart (grid form) and cover a 160 acre field.  Specific to my
research, I am interested in the local CV (coefficient of variation) of crop
density - but what constitutes local? My thought to answer this question is to
1.) estimate the variogram for the field, 2.) use the range as a search radius
(definition of local) and 3.) use an objective function to estimate the weighted
CV value points already on the collection grid (i.e. no prediction).  In other
words, I want to krige based on a spatially weighted CV rather than a spatially
weighted mean.

     I am relatively new to geostatistics.  May I have some feedback on this
procedure.  Is it valid? Is there a better way?  What are your thoughts?

Thanks very much and my best,
Dean Monroe
Oklahoma State University
Environmental Sciences



[Non-text portions of this message have been removed]

Dear Community,

I would like to share some thoughts with you: I studied some datasets
concerning plant distribution in arable fields:

I found first of all a large anisotropy.
Secondly I found very large nugget effects, with a tendecy to have pure
nugget effect sometimes.
The distribution of the investigated phenomenon is non-gaussiang nature.
The occurence of the phenomenon seems to be affected by several soil-
fetiliser- and weather-related factors and not by the distance of vector
(h)!

Am I -from the statistical point of view- allowed to use kriging to
interpolate these datasets or are their absoulte constraints for kriging
under these suppositions ? I think I am not allowed to krige !?

Ursula

_________________________________________________________________



--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Ursula

I think there are a few things you can try before you
abandon all hope of kriging. Certainly if you have
nugget effects which are more than (say) 2/3rds of the
total sill you will get little or no validity from
using a distance weighted method such as kriging.

Possibilities you might like to look at:

(1) normalise or otherwise transform your data. Highly
skewed data, for example, will give you spurious
anisotropies and relatively high nugget effects.

(2) study the relationship with the external factors.
Perhaps you could use kriging with external drift.

Not all patterns have a distance relationship. The
semi-variogram will tell you whether this assumption
is correct, but only when calculated properly.
Calculating a semi-variogram on data which is
extremely non-Gaussian is counter productive!

Isobel
http://geoecosse.bizland.com/whatsnew.htm

________________________________________________________________________
Want to chat instantly with your online friends?  Get the FREE Yahoo!
Messenger http://mail.messenger.yahoo.co.uk

--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

I've made an interpolation by IDW, Spline and ordinary kriging and
a cross-validation for all methods.
When I make a regression between predicted and measured values I reach an R2
of 0.0109 (spline),  0.006 (kriging), 0.0006 (idw), namely very flat curves
(lines).
What could I do? I think I'm using a small number of input values (62) for
the area that I'm considering (4000 km2).
Thanks
Duccio Rocchini

Dr. Duccio Rocchini
Dpt.of Env. Sci.
University of Siena
www.rocchini.net





--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Duccio

There are many reasons why your interpolations may not
be working. A few of these are:

# you are beyond the range of influence of any
distance relationship, that is you have too widely
spaced sampling.

# your data may have a skewed or other non-Gaussian
distribution which makes both semi-variogram and
correlation calculations invalid

# you may have discontinuities, trends or anisotropies
which have not been factored into your model

Isobel
http://drisobelclark.ontheweb.com

________________________________________________________________________
Want to chat instantly with your online friends?  Get the FREE Yahoo!
Messenger http://mail.messenger.yahoo.co.uk

--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Dear Dr. Rocchini:

Well, my first reaction  is that R squared is not an appropriate measure for
correlated data. Secondly, such a large area may well have an underlying trend
which should be removed before kriging is attempted and lastly the number of
observations seems very small if that is the only data layer available.

Sincerely

Russell Barbour Ph.D.
Research Associate in Applied Mathematics
Vector Ecology Laboratory
Yale School of Medicine
60 College St. Rm 600
New Haven CT. 06520
TEL: 203 785 2394
FAX 203  785 3604
email: russell.barbour@...


Quoting rocchini@...:

> I've made an interpolation by IDW, Spline and ordinary kriging and
> a cross-validation for all methods.
> When I make a regression between predicted and measured values I reach an R2
> of 0.0109 (spline),  0.006 (kriging), 0.0006 (idw), namely very flat curves
> (lines).
> What could I do? I think I'm using a small number of input values (62) for
> the area that I'm considering (4000 km2).
> Thanks
> Duccio Rocchini
>
> Dr. Duccio Rocchini
> Dpt.of Env. Sci.
> University of Siena
> www.rocchini.net
>
>
>
>
>
> --
> * To post a message to the list, send it to ai-geostats@...
> * As a general service to the users, please remember to post a summary of any
> useful responses to your questions.
> * To unsubscribe, send an email to majordomo@... with no subject and
> "unsubscribe ai-geostats" followed by "end" on the next line in the message
> body. DO NOT SEND Subscribe/Unsubscribe requests to the list
> * Support to the list is provided at http://www.ai-geostats.org
>

--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Hi again,

after I got so much response to my questions concerning spaital
interpolation comparison I want to present to you my conclusion:

It seems to be a very big problem to make qualitatively assumtions on the
goodness of a spaital interpolator compared to another. Nevertheless, many
ecological relevant studies use cross-validation to perform this check
(using RMSE, MAE or MRE for their pursoses). In literature we find the
advice that quantitative assumtions can be made like this e.g. (how many
locations are over or underestimated etc.) but qualitative comparision can
be misleading. Therefore I propose the follwoing:
1. Check crossvalidation results for a firsdt comparison
2. Count the numer of sites of misesimation
3. Calculate the largest differences between the estimated and true values
4. Perform a t-test for independet variables
5. Calculate the regression coefficient between estimated and true values
(output = r2)
6. Look with your expert knowledge at the maps produced and interprete the
results

Now I am very much intersted in your opinion concerning this proceeding.
Would you say this is a correct comparison of a spatial interpolator ?

Bye,

Peter

_________________________________________________________________
Schützen Sie Ihren Posteingang vor unerwünschten E-Mails.
http://www.msn.de/antispam Jetzt Hotmail-Junk-Filter aktivieren!


--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Group:

     First, thanks to all who responded to my question of CV values and kriging. 
The insights were very helpful and have caused me to revise my procedure.


     I am about to set a sampling schedule for my experiment.  At first I thought
a uniform random sampling structure could be simulated (i.e. random lat/long
coordinates) then applied to a field.  How then can I ensure that there will be
enough samples (at consecutive lags) to produce an accurate variogram?  Is there
a spatial uniform random pattern that I can use to develop my sampling schedule?

     Secondly, has anyone seen a reference as to the appropriate number of
samples to take?  Is there a reference, book, manual, or something that
delineates good sampling methods for geostatistical research?

Thanks very much..


Best,
Dean Monroe
Oklahoma State University
Environmental Sciences



[Non-text portions of this message have been removed]

Dear List members,

What would be a good interpretation for the coefficient Rho in a conditional
autoregressive (CAR) model?
I have looked through the literature (Cressie and such) but haven't found a good
verbal interpretation. Is it fair to say that the larger rho, the stronger the
spatial effect.

Also, does someone on this list know how to control rho in Splus? I would like
to force it to 0 occasionally to get a "null" model for Rsquare and AIC
calculations.

Thank you

Volker Bahn

-~^/\^~^/\~/\^~^^/\/\~^~*

Dept. of Wildlife Ecology - Rm. 210
University of Maine
5755 Nutting Hall
Orono, Maine
04469-5755, USA
Tel: (207) 581 2799
Fax: (207) 581 2858


[Non-text portions of this message have been removed]

On Thu, 16 Oct 2003, Volker Bahn wrote:

> Dear List members,
>
> What would be a good interpretation for the coefficient Rho in a
> conditional autoregressive (CAR) model? I have looked through the
> literature (Cressie and such) but haven't found a good verbal
> interpretation. Is it fair to say that the larger rho, the stronger the
> spatial effect.

Yes and no. Larger absolute values would indicate stronger spatial
effects, larger positive values stronger positive spatial autocorrelation
reducing the effective number of observations. Robert Haining's new book
on "Spatial Data Analysis" is a good source on these kinds of questions
(2003, Cambridge UP).

>
> Also, does someone on this list know how to control rho in Splus? I
> would like to force it to 0 occasionally to get a "null" model for
> Rsquare and AIC calculations.
>

Perhaps just fit lm(), AIC should be OK, as should the likelihood ratio
test. I don't have access to the software, so I'm generalising from how
similar classes work in R.


> Thank you
>
> Volker Bahn
>
> -~^/\^~^/\~/\^~^^/\/\~^~*
>
> Dept. of Wildlife Ecology - Rm. 210
> University of Maine
> 5755 Nutting Hall
> Orono, Maine
> 04469-5755, USA
> Tel: (207) 581 2799
> Fax: (207) 581 2858
>

--
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
e-mail: Roger.Bivand@...


--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Hello,

The subject of logs and back transforms has been discussed a great deal
on the list and I've seen much material concerning back transforms
following kriging of log transformed data (e.g., the approach outlined
by Cressie in his book 'Statistics for Spatial Data' and many other
texts). However, I am unsure how to proceed if the objective is
simulation.

I have applied sequential Gaussian simulation to log (base 10)
permeability data and I want to back transform the simulated
realisations. I would be grateful for any suggests from list members as
to how best to back transform the values in this case. There are too few
data to make an indicator approach feasible.

I will post a summary of answers. Many thanks in advance.

Chris Lloyd




[Non-text portions of this message have been removed]

Hi everybody,

I am looking for literature in which is explained the usage of box-
whisker plots for identifying outliers.

I have Barnett and Lewis - Outliers in Statistical data, 1994, and
Rousseeuw and Leroy - Robust Regression and Outlier Detection,
1987, but it seems i cannot find any reference to box-whisker plots.

Any help is very much appreciated. Thanks a lot,

Monica



--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any
useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org