Skip to content

spatiotemporal co-kriging?

2 messages · Santiago Beguería, Edzer Pebesma

#
Dear R-sig-geo list members,

I am working on a spatio-temporal kriging model (using gstat::krigeST) of two closely related variables. I?m getting good results on one of them, and not so good on the other. Since both dependent variables are inter-related, I expect that co-kringing would be a good option (I?ve done this in a spatial-only context with the same two variables and co-kriging proved beneficial).

So, is it possible to perform spatiotemporal CO-kriging, and if so are you aware of any examples? I have not been able to find any myself.

Cheers,

Santiago Beguer?a
CSIC
Spain
1 day later
#
As for gstat: I think you can't with krigeST, but you may succeed with
krige when you take time as the third dimension -- of course this
constrains the variogram models you can choose from to the metric model.
On 08/11/16 23:35, Santiago Beguer?a Portugu?s wrote: