[R-sig-Geo] Error-Mapping

[ulrike.grimm at uni-leipzig.de]

Dear all,

I am a R- starter and not really experienced, but I try my best. For  
my PhD thesis I want to create an error map of a palaeo-relief and I  
found the instruction from Tomislav Hengl 2009 in "A practical Guide  
to Geostatistical mapping" (page 221 ff) really interesting. He also  
recommended this mailing list.

Actually I have two questions.

1. Question: Is it possible to create an empty grid with a individual  
shape of the research area? I found several instruction at the  
Internet but nothing really useful. Useful in this case means, that I  
can further follow the script from Hengl.

2. Question: The gstat package developed further and the function  
"overlay", which he used, is not available anymore. Instead, I think,  
the function "over" should be used, but then I don´t get the result  
which I need for further steps. Because the message reveals: Error in  
plot(cross.ov at coords[, 1], cross.ov at data[[1]], type = "l", xlab = "X",  
  : trying to get slot "coords" from an object of a basic class  
("numeric") with no slots. But of course, the initial objects had  
slots. If I got it right the function "over" just produces numeric  
values. What is the equivalent function for "overlay" from 2009?

I would be really happy about any hint or maybe newer/better solutions  
for error mapping of DEM´s. In my case paloae-DEM´s.

Best regards,
Ulrike Grimm


-- 
Dipl. Geographin

Universität Leipzig
Institut für Geographie
Johannisallee 19a
D-04103 Leipzig

http://geographie.physgeo.uni-leipzig.de/landschaft/mitarbeiter/ulrike-grimm/