[R-sig-Geo] stationary test

[Tomislav Hengl]

Dear Alessandro,

If I understand correctly, you have densely collected LiDAR data and would like to see if the
population/variogram parameters of your target variable (elevation) differ drastically in different
part of the area. In the case of elevation data and larger areas - I am sure they do.

In order to do something like this, you would need a tool that can estimate these parameters locally
i.e. in a moving window (Haas, 1990). This would then allow you to map the parameters over the whole
area of interest and then do statistical comparison for the purpose of stratification (Lloyd and
Atkinson, 1998) or localized prediction (Walter et al., 2001).

As far as I know, such analysis is not available in R, but you might try to try playing with a small
tool Budiman Minasny developed few years ago (Bishop et al. 2006):
http://www.usyd.edu.au/su/agric/acpa/software/digeman.htm

I imagine that something like this could be easily either run from R (it is distributed as windows
*.exe) or even implemented as a package (Budi works mainly in Matlab). I find Digeman extremely
interesting (it actually produces maps of variogram parameters!), but have not much time to test it.

See also: http://www.usyd.edu.au/su/agric/acpa/vesper/vesper.html 


all the best,

T. Hengl
http://spatial-analyst.net 


References:

Bishop, T.F.A., Minasny, B., McBratney, A.B., 2006. Uncertainty analysis for soil-terrain models.
International Journal of Geographical Information Science 20, 117-134.
Lloyd, C.D., Atkinson, P.M., 1998. Scale and the spatial structure of landform: optimizing sampling
strategies with geostatistics. In: Proceedings of the Third International Conference on
GeoComputation, University of Bristol, UK, 17–19 September 1998. University of Bristol, Bristol, UK,
16pp.
Haas, T. C., 1990. Kriging and automated semivariogram modelling within a moving window. Atmospheric
Environment 24A: 1759–1769.
Walter, C., McBratney, A. B., Donuaoui, A., Minasny, B., 2001. Spatial prediction of topsoil
salinity in the Chelif valley, Algeria, using local ordinary kriging with local variograms versus
whole-area variogram. Australian Journal of Soil Research 39: 259–272.




-----Original Message-----
From: r-sig-geo-bounces at stat.math.ethz.ch [mailto:r-sig-geo-bounces at stat.math.ethz.ch] On Behalf Of
Alessandro
Sent: donderdag 7 augustus 2008 19:51
To: r-sig-geo at stat.math.ethz.ch
Subject: [R-sig-Geo] stationary test

Hi All,

 

is there un code to do a stationary test in R? because I have a data-base
and I wish to know there is or not a stationary in my data-set.

 

Alessandro 


	[[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
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: UserGuide.txt
URL: <https://stat.ethz.ch/pipermail/r-sig-geo/attachments/20080808/419758eb/attachment.txt>