GEOSTATS: Tom Mueller's comments
- I think that you could identify the problem a bit more explicitly. The
difference between the different choices of the parameters in the
exponential and gaussian variogram models is the difference between a
parameter and an "effective range". This distinction has long been known in
the geostatistical literature, the problem arises because neither the
exponential nor the gaussian has a true range, i.e., a distance at which
the variogram is constant (as occurs with a spherical model). Instead the
"effective range" is the distance at which the value of the variogram is
95% of the sill.
It is likely that the reason that Surfer and SAS do not recognize this
distinction is that neither is really a geostatistical package, early
versions of Surfer did not allow the user to specify a variogram model for
kriging and a linear model was assumed. In the case of a linear model, the
slope does not affect the kriging weights (it does affect the kriging
variances). It is likely that the programmers were not really familiar with
geostatistics and the geostatistical literature.
It is nice to see other software packages incorporate geostatistical tools
but sometimes the implementations leave a lot to be desired.
Department of Mathematics
University of Arizona
Tucson, AZ 85721
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!