[R-sig-Geo] another krigeST question

[Benedikt Gräler]

Dear Erin,

as this warning indicates, krigeST could not automatically check whether the temporal distances used in the variogram model and your data set coinside (eg. being both reported in days). KrigeST would in case of a mismatch correct for it.

In the later package versions, variogramST stores the temporal unit as an attribute. The attribute is then used by the fitting routine an passed to the variogram model, allowing krigeST to compare it with temporal distances used in the prediction. In the case when no attribute is found, the below warning is issued and the model is evaluted as is. 

HTH, 

 Ben

On February 18, 2014 7:46:13 AM CET, "Hodgess, Erin" <HodgessE at uhd.edu> wrote:
>Hi again:
>
>Now I'm getting the following:
>
>
>Warning message:
>In krigeST(index.index ~ 1, w, newdata = pred.grd, modelList = stv) :
>The spatio-temporal variogram model does not carry a time unit
>attribute: krisgeST cannot check whether the temporal distance metrics
>coincide.
>>
>
>
>Here is the code to generate the model:
>    stv <- vgmST("sumMetric",
>    space=vgm(spsil1,smod1,sran1),
>    time=vgm(epsil1,emod1,eran1),
>    joint=vgm(60,"Exp",100),
>    stAni=20)
>
>Any help much appreciated.
>
>Thanks,
>Erin
>
>
>	[[alternative HTML version deleted]]
>
>_______________________________________________
>R-sig-Geo mailing list
>R-sig-Geo at r-project.org
>https://stat.ethz.ch/mailman/listinfo/r-sig-geo

-- 
Benedikt Gräler

Institute for Geoinformatics
University of Münster

http://www.ifgi.de/graeler

+49 251 83 33082