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

Fw: GEOSTATS: philosophical question

Expand Messages
  • Gedas Vaitkus
    ... From: Gedas Vaitkus To: Sent: Monday, June 05, 2000 9:40 AM Subject: Re: GEOSTATS: philosophical question ...
    Message 1 of 1 , Jun 5, 2000
    • 0 Attachment
      ----- Original Message -----
      From: Gedas Vaitkus <seabird@...>
      To: <Eeqmc11@...>
      Sent: Monday, June 05, 2000 9:40 AM
      Subject: Re: GEOSTATS: philosophical question


      > Hi there,
      >
      > A good question, really. I was always taught, that GIS is kind of a
      digital
      > model of the reality, but the more I work in GIS, the more I understand,
      > that the overall quality and "performance" of a GIS-based model first of
      all
      > depends on it's "resolution" (=scale of data capture), the quality of
      > information source (there are many maps of insufficient quality, from
      which
      > GIS data is produced), as well as topological quality of GIS data
      ("spagetti
      > data" has almost no use in GIS analysis). Therefore, in most cases one
      must
      > admit, that the map he produces in a GIS is just a nice picture.
      >
      > When we talk about "a model", we have in mind "modelling". Modelling on a
      > typical GIS data is rather limited (one can cut, overlay certain types of
      > objects, measure distances, create buffers, query certain types of
      > data/objects, etc.). To be able to perform real modelling on spatial data,
      > one needs to have a regularly spaced attribute data set related to
      identical
      > type of spatially referenced "geodetic markers", which usually are points
      or
      > polygons. In fact, here we are talking about a collection of grids, made
      > from geostatistically processed raw data. Therefore, we can conclude here,
      > that geostatistical approach is the best for data analysis and modelling,
      > whereas GIS approach provides strong basis for cartographic presentation
      of
      > the data/modelling product combined with base maps of relevant type. And
      > this is what most of the professionals do every day. This means a lot of
      > work, in fact - a double work.
      >
      > For the last 2-3 years I am working on a new concept - a high-resolution
      > pseudo-raster in GIS, which includes topologically perfect high-resolution
      > coverage in GIS, which allows to link and perform any kind of digital
      > manipulation on external datasets, produced not only from geostatistically
      > processed raw data, but also from imagery and even "standard" GIS data.
      This
      > works as a large collection of "normal" geostatistical grids under a
      > standard "cover" of GIS coverages. Working this way, if resolution of the
      > pseudo-raster coverage is high enough (here you need lots of computing
      time
      > on powerful hardware), I get rid at least of problems, related with GIS
      > topology, spatial resolution and do not bother about limitations of
      > mathematical modelling on my data. The only thind to do is get high
      quality
      > data (both field sampling and GIS). The best GIS environment (for me) to
      do
      > this is MapInfo Pro.
      >
      > You can see some earlier examples of this approach on the BaltSAS web page
      > (http://www.5ci.lt/users/seabird/sb_atlas/se_baltic).
      >
      > Regards,
      >
      > Gedas Vaitkus
      >

      --
      *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.