GEOSTATS: Variograms with multiple sills
- Lisa Dent's questions on variograms with multiple sills refer:
1) Variograms stabilizing and then fluctuating up again could mean
one of several things, most notably:
(a) a trend in the data (or drift)
(b) not enough samples at the larger lags (rule of thumb is to cut
off at half the maximum lag distance).
2) Assuming it is because of a trend, try to figure out why. I assume
that for a stream, nutrient concentrations would not be abrupt
isolated patches, but rather something more gradual, as the stream
cuts more of the mainland to reach maturity over time. There is also
a big contrast between upsteam area where there is no flood plain and
further downstream where it starts to meander. Intuitively then, there
should be a trend here somewhere although I am no stream expert.
3) Re (2), the moral here is there is no substitute for a subjective,
intuitive understanding onf the problem domain before using geostatistics
to complement that understanding. The tools shouldn't be used as a
black box. I would read up on 'stream science' if there is such a thing
before trying to infer 'patches' of nutrient concentrations. Good luck.
*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!