Hi,
I want to perform spatial interpolation of soil moisture (here named 'sm')
using regression kriging (RK) following the example given by Tom Hengl
http://spatial-analyst.net/wiki/index.php?title=Spatial_prediction_of_soil_moisture)
and compare results with OK.
The covariate is the soil el
I would take a very good look at the regression kriging literature (or
that of universal kriging, or external drift kriging -- terms that were
there before regression kriging came up). It looks like the method
cannot be applied in your case.
--
Edzer
On 02/23/2011 08:51 PM, giuseppe calamita wrote
Thank you vary much Edzer,
actually I experimented the same problem (before RK) trying to apply external
drift kriging.
Your comment makes me feel better (after months spent thinking how to solve the
problem).
The only pseudo-solution I've found has been to apply a 'idw' on the covariate
variab
So you have values of 'sm' (target) and 'res' (covariate) at all
sampling locations. Now you would like to predict at some other
locations? You obviously have to have values of 'res' at those locations
(This is why you get an error in gstat)
"krige" function needs either a SpatialGrifDataFra