I'm trying to use GSLIB's 'kt3d' program to krig a dataset with an
external drift described by a second variable. My problem is how to
estimate the semivariogram (SV) of the residuals. My understanding of kt3d
is that it scales the second variable internally during each kriging
estimate and so it is not possible to know beforehand what the
residual will be, thus it is not possible to plot the semivariogram
of the residuals.
The papers that I have read indicate that one should look
for directions or areas where the trend can be ignored, and here
the SV of the data values will be equivalent to the SV of the
residuals. In my case doing this means reducing the number of points
Am I right in thinking this, and if so does anyone have any advice about
balancing the need to limit the trend in the areas used for SV modelling to give
an estimate of the SV of the residuals while still having sufficient points to
model? Alternatively, is there a more robust way to produce the required SV ?
Dr Tom Charnock
National Radiological Protection Board
Didcot, Chilton, Oxon, OX11 0RQ, UK.
*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!