Loading ...
Sorry, an error occurred while loading the content.
 

GEOSTATS: variogram models with multiple sills

Expand Messages
  • Lisa.Dent@asu.edu
    Hi, I m fairly new to spatial statistics and have come across a problem. I m using semi-variograms as a descriptive tool to quantify spatial heterogeneity in
    Message 1 of 1 , Apr 23, 1997
      Hi,

      I'm fairly new to spatial statistics and have come across a problem. I'm
      using semi-variograms as a descriptive tool to quantify spatial
      heterogeneity in nutrient concentrations in stream water. I'm primarily
      interested in fitting a theoretical variogram to my experimental data to
      obtain a range, which I interpret as patch size, and can then compare to
      patch sizes for other nutrients or for the same nutrient at different
      times of the year. My initial data are all one-dimensional transects
      (since streams flow in one direction), with samples every 25 m over a
      total of 10 km.
      My problem is this: in some of the experimental variograms, it looks to
      me as if the variance increases to a sill, remains at that sill for
      several lags, and then increases again to a second sill. Since I am
      interested in finding the range, I would like to fit a model with two
      ranges. I have been using GSplus, which does not have any models with
      multiple ranges. I also tried using a double spherical model, which I
      found in the literature, but it does not really level off at the first
      sill the way my experimental variograms do. The only solution I can
      think of is to first fit a spherical model to an empirical variogram
      calculated using one maximum lag (the lag where the variogram begins to
      increase for the second time), getting one range, and then fit an
      empirical variogram calculated using a larger maximum lag, to get the
      second range. Is this legitimate? Is there any objective way to
      determine what the maximum lags should be?

      Any advice is appreciated. As well as GSplus, I have also been using
      Splus spatial stats.

      Thanks,
      Lisa

      ***************************************************************************
      * Lisa Dent lisa.dent@... *
      * Department of Zoology (602) 965-6478 *
      * Arizona State University *
      * Tempe, AZ 85287-1501 *
      * *
      ***************************************************************************


      --
      *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!
    Your message has been successfully submitted and would be delivered to recipients shortly.