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

GEOSTATS: Coordinate conversions (off-topic)

Expand Messages
  • Simon Brewer
    Hi First, my apologies for a question which is a little off-topic for this group, but I hope that someone might have an answer! I have a set of sites
    Message 1 of 2 , Jul 18, 2000
    • 0 Attachment
      Hi

      First, my apologies for a question which is a little off-topic for this
      group, but I hope that someone might have an answer!

      I have a set of sites distributed across Europe, and I wish (ultimately)
      to use kriging to interpolate between them. The problem is that the
      coordinates that I have, are in Latitude/Longditude. I want to convert
      these into a cartesian system, in order to work with them. However, as
      the study area is quite large (approx 15°W-35°E 35°N-72°N), I am worried
      that reprojecting them (e.g. using the Albers Equal Area proj.) will
      distort the spatial relation of the sites.

      My questions therefore, are:
      1) Will the distortion be too great? Does anyone have experience of
      working at this scale?

      2) Are there alternative conversion which will better preserve the
      spatial relation? Presumably this will take into account the curvature
      of the earth.

      3) There is a piece of software listed under the Soft-FAQS of
      ai-geostats called spherekit
      (http://curie.ei.jrc.it/software/spherekit.htm), which appears to be
      designed to work at this scale. Has anyone managed to get it to compile
      under Linux?

      Thanks in advance
      Simon
      --
      ------------------------------------------------------
      Simon Brewer
      European Pollen Database
      Place de la Republique tel: 00 33 (0)4 90 96 18 18
      13200 Arles fax: 00 33 (0)4 90 93 98 03
      France email: simon.brewer@...
      ------------------------------------------------------
      --
      *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!
    • Noemi Barabas
      Hi Simon, This may not be the best solution but a possible alternative: I have just recently started working on a problem with similar issues of coordinate
      Message 2 of 2 , Jul 18, 2000
      • 0 Attachment
        Hi Simon,

        This may not be the best solution but a possible alternative:
        I have just recently started working on a problem with similar issues of
        coordinate conversion and distortion, although the spatial nature of my
        problem is somewhat different from yours. I am unbending a river so that
        kriging is done in a domain that follows the course of the river instead
        of going over land. I have converted the coordinates (which were already
        in meters and on a small scale geographically) by applying a grid
        generation scheme. (Gridgen is one software that can be used).

        Grid generation is widely used in the field of flow simulation in channels
        or around bodies where boundary conditions apply along irregular surfaces
        (hydrodynamics, aeronautics etc).

        The physical boundary is defined into four segments that correspond to the
        four sides of a rectangle. An equal number of grid points are laid out
        along each of the pairs of facing boundaries. Various algorithms (Laplace
        transforms to linear interpolation, depending on the problem) calculate
        grid lines between the extreme grid point pairs. The grid is generated in
        the original coordinate system, so each sample has to be assigned to the
        nearest grid point within limits of resolution and coordinate precision.

        In flow field simulations, the differential equations and boundary
        conditions are transformed based on the mathematical relationship between
        the two coordinate systems. This is not necessary in this application, as
        all grid points to be estimated are calculated before kriging.
        Semivarogram calculation, modeling and kriging takes place in the
        transformed space and the results need merely be mapped to the original
        coordinate points corresponding to the initial grid generation output.

        An excellent book on the theory is:
        Numerical grid generation : foundations and applications
        by Thompson, Joe F. (1985)
        Published New York : North-Holland : Elsevier Science Pub. Co.

        Gridgen is available from Pointwise for a free 90 day trial period:
        www.pointwise.com


        --
        *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.