| Re: Interpolation on a sphere [message #37671 is a reply to message #2948] |
Tue, 20 January 2004 05:08   |
Haje Korth
Messages: 651
Registered: May 1997
|
Senior Member |
|
|
Ken,
thank you for your input. I (temporarily?) solved the problem by going back
to a combination of TRIANGULATE/TRIGRID, which is essentially SPH_SCAT. From
the functionality of GRIDDATA I thought that this routine will obsolete the
other ones in the future, especially since QHULL is much more stable in
calculating the Delauney triangulation than TRIANGULATE is. As a result, all
artifacts I suffered from before are gone. My problem is that I do not know
whether there is a bug in GRIDDATA or if I used a poor choice of keywords.
Therefore, I cannot report my issues to RSI. But something is definitely
screwy here.....
Greetings,
Haje
--
"Kenneth P. Bowman" <kpb@null.com> wrote in message
news:kpb-E20603.14132819012004@news.tamu.edu...
> In article <bugttk$qo8$1@houston.jhuapl.edu>,
> "Haje Korth" <haje.korth@jhuapl.edu> wrote:
>
> Are you sure it's not either a case of too few points or of real
> features of the data?
>
> You could try a least-squares fit to spherical harmonics (truncating
> appropriately) and then reconstruct a gridded field from the spherical
> harmonics.
>
> Ken Bowman
>
>> Now, my problem is that no matter what interpolation method I use, I
obtain
>> ARTIFACTS (e.g., saw teeth, see attached picture) in the gridded output.
>> Does anyone know how get a decent interpolated data set? Am I using the
>> right key words? Or should I attempt a completely different approach?
Any
>> help is appreciated.
>>
>> Thanks,
>> Haje
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
