[R-sig-Geo] Block spatiotemporal kriging with separable covariance model

[Edzer Pebesma]

On 01/13/2013 08:48 PM, Piero Campalani wrote:
> Dear Edzer,
> thanks for the help.
> Is block-kriging possible with this kind of covariance model, or is it just
> not implemented in the `spacetime` package?

it is just not implemented.

> 
> On 13 January 2013 18:31, Edzer Pebesma <edzer.pebesma at uni-muenster.de>wrote:
> 
>>> I saw examples of block kriging when using a 3D metric kriging, with
>>> rescaled time axis, so that a parallelepiped can be set in the krige()
>>> function.
>>
>> I'd call it a rectangle.
>>
> 
> ..why would you? Wouldn't you have a 3D xyt block ?
> 

yes, but with right angles.

I don't know what you saw, with the 3D metric model. Specifying a block
in gstat with block = c(dx, dy, dz), you get a rectangular block.

You can estimate block means for arbitrary shapes when you specify a
polygon (only 2D), or a set of points (1, 2 or 3D) that discretizes your
shape, but I've never actually seen people trying to do this for
parallelepipeds.

Bests,
-- 
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