[R-sig-Geo] Question on lags in h-Scattergrams

[Edzer Pebesma]

On 04/01/2011 10:22 PM, Rich Shepard wrote:
>   My question is not R specific, but about h-scattergrams as described in
> Pierre Goovaerts's "Geostatistics for Natural Resource Evaluation." No
> matter how many times I read these sections I still cannot grok the
> relationship between delta-h (which he sets at 100m in the discussions
> beginning in Section 2.3.2 on page 25) and the examples in following
> sections.
> 
>   For example, on page 27, Figure 2.8 shows Cd concentrations at 6 classes
> of distances, 'h' with distance tolerance 100m (the delta-h). Where does
> the
> 100m distance show up in the plots with h=43m through h=1024m? This is what
> confuses me.

It seems to me that in figure 2.8 the average distance of all point
pairs in a particular lag class is shown, but the exact class boundaries
are not given (delta h = 100, but the h values in h +/- delta h are
not). Looking at the average values, I would assume lag intervals 0-100,
100-300, 300-500 etc. If I try to reproduce this, as in

library(gstat)
data(jura)
example(jura)
hscat(Cd~1, jura.pred, c(0,100,300,500,700,900,1100)/1000,
variogram.alpha=90, tol.hor=22.5, asp="iso")

I'm getting quite similar, but not identical results, so some of my
assumptions must be false. Maybe Pierre can help us out.

Note that the intervals in this plot are in km.

With best regards,

>   Learning spatial statistics on my own (all my stat classes in grad school
> were biometry focused) leaves me lacking a local resource to ask.
> 
> TIA,
> 
> Rich
> 
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo

-- 
Edzer Pebesma
Institute for Geoinformatics (ifgi), University of Münster
Weseler Straße 253, 48151 Münster, Germany. Phone: +49 251
8333081, Fax: +49 251 8339763  http://ifgi.uni-muenster.de
http://www.52north.org/geostatistics      e.pebesma at wwu.de