[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: log-normal transformation
Gstat does not give you the Lagrange parameter, and it is not
trivial to report it because it doesn't actually solve the
kriging systems in terms of Lagrange parameters (but rather in
terms of Eqs. A.2 and A.4 in the manual). It can give you the
estimation variance related with estimating the trend, when
however I cannot see directly how you derive the Lagrange
parameters from there (that is, without hacking the code).
Another issue here is that universal kriging (or kr. with
external drift if you like) would give you two lagrange
parameters; Cressie's section on lognormal kriging only deals
with one, the case of ordinary kriging.
For your question regarding other people with experience on
kriging rainfall using altitude please write to the ai-geostats
list, try http://www.ai-geostats.org/