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

GEOSTATS: kriging in arc/info

Expand Messages
  • Altaf Mubaraki
    Dear GIS gurus: My problem: I have a global dataset of DMS (all Oceanic) emissions. To have a better resolution I split the dataset into four geochemical
    Message 1 of 2 , Nov 4, 1997
    • 0 Attachment
      Dear GIS gurus:

      My problem:

      I have a global dataset of DMS (all Oceanic) emissions. To have a
      better resolution I split the dataset into four geochemical
      regions -atlantic, pacific, indian and southern. Each region is then
      again split into a number of smaller regions. My objective
      is to start at the regional level (atlantic/pacific etc) and krig the
      emissions using the different methods arc/info offers. And
      then join the krigged lattices into one.

      Now when I run the kriging function without specifying a barrier cover,
      the kriging proceeds quickly and I get the output lattice
      and the kriging variance. But since I do not want the emissions to get
      interpolated on to land (these, as I said earlier, are
      strictly oceanic emissions), I created a line coverage consisting of
      the world outline map with just the continental boundaries; and used
      that as the barrier cover. The online help does indicate that barriers
      option will significantly increase the processing time required to
      interpolate the out_lattice but surely it shoudn't take more than few
      hours at most, no?

      I tried to krig the atlantic region first. I decreased the number of
      arcs in teh barrier line coverage by deleting the western shoreline of
      US, S. America and the arcs beyond the western shoreline of Asia/africa
      (in the hope to expedite the processing). My problem is that despite
      allowing the kriging procedure to run for more than 48 hours, it still
      didn't finish!. Surely something is amiss.

      I think the choice of the barrier is somehow the reason for the
      inordinate time its taking to process. Is there anything wrong
      in my using the line coverage as an embankment? Its a regular arc
      coverage(line) I created from a data file using the generate command.

      I have approx. about 4000 data points in each of the four regions.

      Any inputs/ideas would be greatly appreciated...

      I will summarize ...

      Thanks,

      Altaf

      --
      *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!
    • szz@zycor.lgc.com
      ... Consider generalizing the world outline map-derived boundary! Certainly adding boundary files will cost you CPU time, but I bet that you could get away
      Message 2 of 2 , Nov 4, 1997
      • 0 Attachment
        >
        > From: Altaf Mubaraki <altaf@...>
        > Mime-Version: 1.0
        > To: pgis <ai-geostats@...>, esri-l@...
        > Subject: GEOSTATS: kriging in arc/info

        >
        > Dear GIS gurus:
        >
        > My problem:
        >
        > But since I do not want the emissions to get
        > interpolated on to land (these, as I said earlier, are
        > strictly oceanic emissions), I created a line coverage consisting of
        > the world outline map with just the continental boundaries; and used
        > that as the barrier cover. The online help does indicate that barriers
        > option will significantly increase the processing time required to
        > interpolate the out_lattice but surely it shoudn't take more than few
        > hours at most, no?
        >


        Consider generalizing the world outline map-derived boundary! Certainly
        adding boundary files will cost you CPU time, but I bet that you could
        get away with a lot less segments in the boundary file you are using.
        And CPU time should be roughly proportional to the number of boundary
        segments.

        SZoraster
        --
        *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.