Re: AI-GEOSTATS: gstat in R doubts

[Edzer J. Pebesma]

(I added gstat-info to the addressees, as I believe this message and
possibly further discussion better belong there)
--
Edzer
Marta Rufino wrote:

Dear Collegues,

I have some particular doubts about gstat for R (most probabily very 
basic for what I apoligise):

1. Does it compute indicator kriging? How to define it? It should be 
I=.5, but where? I could not find the argument in 'variogram' 
If your variable is called z, and cutoff is .5, define the formula for 
the data as:
I(z < .5) ~ 1

2. I could not manage to make it do the backtransformation of the 
kriging predictions when using log data. Is it possible and how? 
You can take the exponent to get a median-like estimator, for
the expected value you will need to use the kriging variance as
well. See any book on lognormal kriging. The lagrange parameter
is not available.


3. How exactly we specify universal kriging? I have found different 
ways in different places:

kk=krige(abu~x+y, ~x+y, dat, bl, model=wls) # I think this one 
indicates a quadractic internal trend :-)
or
kk=krige(abu~(covariate), ~x+y, dat, bl, model=wls)
or
kk=krige(abu~x+y+(covariate), ~x+y, dat, bl, model=wls) 
These are indeed three forms of universal kriging (although some people
call the second "external drift" kriging -- but how then would they call
the third? Universal external drift kriging?).
BTW abu~x+y indicates a first order linear trend in x and y, use 
abu~I(x^2)+I(y^2)+I(x*y)+x+y
for a second order linear trend (is that what you meant by quadratic trend?)

And the variogram of the residuals for the universal kriging... how do 
we do it? the same way? 
Yes: variogram(abu~x+y, ~x+y, dat)

4. The function plot.variogram as the same name as in geoR. This 
causes bugs and problems runing it, and does not allow us to run both 
packages which would be important. Could this be solved by any of the 
authors, simply by renaming the function... probabily there are more 
like this, although I did not proceed because I got stuck.
(To solve this the only way I found is to attach and detach the 
packages, ...) 
Paulo and I are aware of this; we'll solve it, at some stage.

5. The variogram produced by both packages (gstat and geoR) are 
different. Why? 
Because Paulo Ribeiro and I are different persons. gstat is much older 
than geoR,
but gstat in R is much younger. I copied the "old" gstat variogram 
properties to the
gstat in R variogram object. I was in a hurry and didn't look too much 
around. I can
regret this, but it's too late. I agree that they should have different 
names. Give me
a good suggestion!

It was excelent and extremely usefull for me the implementation of 
gstat in R!!! Thank you so much!
Thanks,
--
Edzer