On Mar 22, 12:30 am, wfz...@[EMAIL PROTECTED]
wrote:
> On 3=D4=C221=C8=D5, =CF=C2=CE=E710=CA=B136=B7=D6, wfz...@[EMAIL PROTECTED]
>
>
>
> > Hi, all
> > I meet a strange problem,when I use the contour procedure to generate
> > precipitation contour plot,the total AWS stations is about 24, just
> > one has rainfall at 0.6mm,others are zero.
> > so I use this data and a contour plot can be created(around this
> > station with 0.6mm rainfall, there is few station)
>
> > but if the total stations is about 150, and the same situation that
> > ony this station has 0.6mm rainfall, there will be no contour plot,
> > which means the whole map will be the same colors presenting zero. the
> > distribution of stations' location is not even, some places are many
> > stations ,some are not. but around this station with 0.6mm rainfall
> > there are many stations.
>
> > by the way, my color table scale is like '0,0.2,1,2,...'
> > so anyone can tell me what happen?
>
> Hi,all
>
> I know what happen, when I use the procedure'griddata', the keyword
> 'smoothing' is 0.05 (I specify it),so the contour plot is nothing,when
> I change it to 0.01,the contour plot is correct.
> then I check the help information about smoothing, just few
> information
>
> so I want to know , in the griddata procedure, for the Inverse
> Distance method, how to specify the smoothing keyword correctly?
>
> thanks
> wenfang
Hi Wenfang,
I think you are not getting a response from the group (which is very
responsive) because of the way you have framed your question. There
isn't anything easy to framing a question well - it takes a lot of
time - and sometimes the solution appears even as you compose the
question! YeeHa!
I happened to have questions about GridData a short while ago (see
http://tinyurl.com/2w5d2h)
, and I got many productive responses. I
think I got good responses because I made it as easy as I could on the
newsgroup readers to help - an short reproducible example, a
description of how I knew the results were "wrong", etc.
Anyway, the short of it is I would love to help, but I can't penetrate
the issue because you have unveiled the problem in a way I can
understand.
The one thing I can say is that I agree, the do***entation for the
SMOOTH keyword to GridData is not helpful. There doesn't seem to be
any reference cited for this method, but this might be a good start...
http://en.wikipedia.org/wiki/Inverse_distance_weighting
Cheers,
Ben
|
