Re: contouring within a domain

From: David Reusch (dbr AT XXXXXX)
Date: Fri Aug 09 2002 - 08:24:59 MDT

  • Next message: Erik Kluzek: "Masking out line plots with markers of _FillValue..."

    On Thu, Aug 08, 2002 at 06:34:16PM -0600, Dave Brown wrote:
    >I'm not familiar with the EASE grid but I just did a Google search for
    >it. I'm guessing you would be using a South azimuthal equal-area grid
    >which I believe should work with the new implementation.

    Yes, that's the same grid I'm using. Mine is just tailored to my
    specific application (naturally).

    >I'm not sure I totally understand your situation, but if the invalid
    >data locations are clustered into identifiable regions (i.e. not
    >scattered throughout the whole data domain) then I think you could
    >simply stuff missing values into these locations in the data grid,
    >and the contouring code would be able to handle it.

    We have two types of "invalid": where we really have no data and
    where the data are outside our domain of interest. The domain itself
    is just a nominally convex interior subset of the larger grid that
    is also nowhere near rectangular (hence resorting to lists of data
    and assoc. lat/long). However, if this new contouring code is as
    good as it sounds, I'll find a way to produce a rectangle if I have
    to and populate with UNDEFs as needed.

    The possibility also exists, depending on grid resolution, that
    there could be interior missing data. I'll assume that will also
    work until shown otherwise.

    So is this functionality available yet? If so, how do I use it?
    If not, how soon can I expect to see it? I'd be happy to test it
    for you (on linux) if that would help.

    thx,
    dave reusch
    dbr AT geosc.psu.edu
    _______________________________________________
    ncl-talk mailing list
    ncl-talk AT ucar.edu
    http://mailman.ucar.edu/mailman/listinfo/ncl-talk



    This archive was generated by hypermail 2b29 : Fri Aug 09 2002 - 08:39:24 MDT