Dear all
This discussion about kriging interpolators and so on brings me to
stress the fact that geostatistics methodology is not only kriging.
1) The explorative part (including in this one also the study, not
inference, of spatial continuity) of geostatistics is fundamental.
It is in this phase of the study, that the user can develop and
analyze the relationships between data and expert knowledge (i.e.
informations about the physical and chemical processes about the
studied phenomena). This kind of study, in which data are studied
spatially, has its own autonomy: it is useful whenever you have
spatial (or spatio-temporal) data, independently by the fact that
then you need to try the inference of a variogram in order to perform
a more or less complex interpolation or geostatistical simulation.
2) Secondly, if we think about geostatistical simulations and the
quantification of spatial uncertainty, kriging algorithm couldn't be
necessary: this happens for example in simulated annealing as
developed in gslib library and, if I'm not wrong, in some way also in
transition probability simulation algorithm.
This just to say that also in the unfortunate case that kriging is
wrong, geostatistics (or if you want spatial statistics...) will be
still alive.......
and really useful.
Sincerely
Sebastiano Trevisani
At 19.13 01/06/2006, JW wrote:
Hello Tom,
Thanks for making my case against geostatistics even more
compelling. Just assume, krige, model, smooth and be happy!
Kind regards,
Jan W Merks
+
+ To post a message to the list, send it to [email protected]
+ To unsubscribe, send email to majordomo@ jrc.it with no subject and "unsubscribe
ai-geostats" in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the
list
+ As a general service to list users, please remember to post a summary of any
useful responses to your questions.
+ Support to the forum can be found at http://www.ai-geostats.org/
