[Gstat-info] Oceanographic data interpolation

[Edzer J. Pebesma]

Nelson Luna Silvestre wrote:
>> > The ellipsoid axis where defined by extracting the models parameters,
>> > sorting the range values for each variable and choosing the max/min
>> > pair of values. This gives me the ellipsoid axis directions and their
>> > ranges, if the pair of directions is orthogonal - deep trouble,
>> > otherwise.
>> >
>> I see; so your fitted major/minor direction are one of the four
>> directions chosen, and hopefully a perpendicular pair. Well, this is
>> rather course fitting, and subject to which four (or however many)
>> directions you start off with.
>
> I am not sure if I fully understand your answer - what do you mean by
> 'this is rather course fitting'?
> Do you suggest any other alternative on finding major/minor axis 
> directions?
I meant coarse, not course --- my impression was that you end up with 
one of the directions 0, 45, 90 or 135 as fitted value, but perhaps I 
got this wrong?

> It seems to me that this an iterative process of trial-and-error,
> anyway - but, then again, I am just a beginner in the geostat field...
>
> Regards,
> Nelson