[R-sig-Geo] ArcGIS Geostatistical Analyst -- how does it display /fit variograms?

[ONKELINX, Thierry]

Hi,

A lot of people in our institute use ArcGIS. But I always advise them
not to use ArcGIS for kriging etc. Mainly because kriging in ArcGIS is a
black box tool. You only know the input and the output, but not how
things are calculated. More over people tend to try the different
options without really knowing what (and why) they are doing. The just
stick with the interpolated map that "looks" the best.

For kriging I promote R and gstat.

HTH,

Thierry


------------------------------------------------------------------------
----
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature
and Forest
Cel biometrie, methodologie en kwaliteitszorg / Section biometrics,
methodology and quality assurance
Gaverstraat 4
9500 Geraardsbergen
Belgium 
tel. + 32 54/436 185
Thierry.Onkelinx at inbo.be 
www.inbo.be 

To call in the statistician after the experiment is done may be no more
than asking him to perform a post-mortem examination: he may be able to
say what the experiment died of.
~ Sir Ronald Aylmer Fisher

The plural of anecdote is not data.
~ Roger Brinner

The combination of some data and an aching desire for an answer does not
ensure that a reasonable answer can be extracted from a given body of
data.
~ John Tukey

-----Oorspronkelijk bericht-----
Van: r-sig-geo-bounces at stat.math.ethz.ch
[mailto:r-sig-geo-bounces at stat.math.ethz.ch] Namens D G Rossiter
Verzonden: zaterdag 6 september 2008 16:07
Aan: r-sig-geo at stat.math.ethz.ch
Onderwerp: [R-sig-Geo] ArcGIS Geostatistical Analyst -- how does it
display /fit variograms?

Hi, I know this is mostly an R-spatial list but this is where the most  
computational geostats experts hang out, so please forgive me for  
asking an ArcGIS question.

I use R almost exclusively for my own work, but have been asked to  
supervise the development of an introductory geostats course for our  
partner at the University of Rwanda. They have standardized on ArcGIS  
for all of their GIS work (and SPSS for non-spatial stats), and the  
prospective students (mostly centre workers and collaborating  
researchers) are familiar with it. The decision was taken by their  
administration not to use my R/gstat material from the ITC distance  
education course, rather to develop the course with ArcGIS.

My counterpart is now with me developing the course. The deficiencies  
of ESRI documentation are well-known. I have dug around quite a bit  
both within the ESRI docs (on-line and with the program) and through  
various mailing lists and the web and can not find out some basic  
information. I hope you can shed some light,

1. What exactly is the display of the empirical variogram? The doc.  
implies there is one average semivariance per bin (as is usual) but  
the display often has several at the same bin. The variogram can be  
exported as a table, where it shows multiple (2 - 6 or so)  
semivariances for each bin; the table also shows a "weight" for each  
of these, but they do not add to 1 or 100 or anything I can  
recognize!  The close-range bins usually have one, then the number  
increases. So I guess each dot represents some number of point-pairs.

2. How is the variogram being fit? What weighting, what solver?  If  
the user changes the cutoff/bin width, the solution changes (as it  
should); but I can't see how it's solving, and I can't find any option  
to change the weighting (as in e.g. gstat).

3. When fitting direct and cross-variograms for co-kriging, it seems  
that a linear model of co-regionalization is being enforced (i.e. same  
range). Again, how is the fit being done? Like fit.lmc in gstat?

Naturally we want the students to understand what the program is doing  
for them!  Although ESRI promotes "press the button and look at the  
cross-validation". I do like their disclaimer in the ArcGIS Desktop  
9.3 help: "Kriging is a complex procedure that requires greater  
knowledge about spatial statistics than can be conveyed in this  
command reference". They then ref. Burrough (1986! not even the  
revised book), Heine (1986), McBratney & Webster Journal of Soil Sci.  
37:317 (1986), Oliver IJGIS 4 (1990), Press etc. Numerical Recipes,  
and Royle et al. Geoprocessing 1 (1981). Not exactly the most up to  
date or accessible reference list (no offrence to the fine authors  
mentioned).

Thanks for your help.

D. G. Rossiter
Senior University Lecturer
Department of Earth Systems Analysis
International Institute for Geo-Information Science and Earth  
Observation (ITC)
PO Box 6, 7500 AA Enschede, The Netherlands
Internet: http://www.itc.nl/personal/rossiter/pubs/list.html#pubs_m_R



International Institute for Geo-Information Science and Earth
Observation (ITC)
Chamber of Commerce: 410 27 560

E-mail disclaimer
The information in this e-mail, including any attachments, is intended
for the addressee only. If you are not the intended recipient, you are
hereby notified that any disclosure, copying, distribution or action in
relation to the content of this information is strictly prohibited. If
you have received this e-mail by mistake, please delete the message and
any attachment and inform the sender by return e-mail. ITC accepts no
liability for any error or omission in the message content or for damage
of any kind that may arise as a result of e-mail transmission.

	[[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-geo

Dit bericht en eventuele bijlagen geven enkel de visie van de schrijver weer en binden het INBO onder geen enkel beding, zolang dit bericht niet bevestigd is door een geldig ondertekend document.%CRLF%The views expressed in  this message and any annex are purely those of the writer and may not be regarded as stating an official position of INBO, as long as the message is not confirmed by a duly signed document%CRLF%