[R-sig-Geo] more about semivariogram

[Tomislav Hengl]

Dear Marcia,

Why would you mask "0"? This is hard data and throwing it out would lead to complete different
results. This reminds of Alex McBratney's quote "You should aim at developing tools that can fit
your data, and not the other way around."

If the purpose of your analysis is spatial prediction of counts, then you should use some link
function (the best candidate is probably Zero Inflated Poisson model) to fit the regression model,
then model variogram for residuals (transformed scale) and interpolate them using ordinary kriging.

This is all well explained in:

E.J. Pebesma, R.N.M. Duin, P.A. Burrough, 2005. Mapping Sea Bird Densities over the North Sea:
Spatially Aggregated Estimates and Temporal Changes. Environmetrics 16, (6), p 573-587 
http://dx.doi.org/10.1002/env.723 

See also the demo script: 

> library(gstat)
> demo(fulmar) # Fulmaris glacialis

T. Hengl
http://spatial-analyst.net/wiki/index.php?title=Species_Distribution_Modelling 



> -----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 Marcia Muñoz
> Sent: Saturday, July 18, 2009 7:08 AM
> To: r-sig-geo at stat.math.ethz.ch
> Subject: [R-sig-Geo] more about semivariogram
> 
> Dear all,
> 
> I want to do a variogram with 520 points, in each point I have abundance of
> a species. My problem or doubt is this, many of these points are empty, they
> have zero (0). My question is, can I do my variogram with all the 520 points
> or just with points that have abundance greater than zero?
> 
> Thanks a lot!
> 
> Mar
> 
> 	[[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