Extrapolating WW3 data near the shore

[Patrick Reuter]

Dear Bill,

This partially solves my problem, thank you! I was waiting for an  
answer and solution for a while.

However : the const function supposes 0 value at missing grid points.  
With the GraDS bilinear interpolation, coastlines that are close to  
missing grid points will receive a smaller value than coastlines that  
are farther away from missing grid points -- and this is not wat we  
want.

I was having a look at your website (nice one!), and I think that  
yourself experienced the same problem with the sea surface temperature.

So is there no "const" function that would extrapolate the values over  
the ocean to the mainland?

Do you have a solution for this? I would again appreciate your help.

I attach you three images : 1 without 'const', and 2 with 'const'  
(north atlantic and france).

Thanks in advance and cheers!
   Patrick


Thanks in a


Bill Reilly <bill_reilly at COMPUSERVE.COM> a écrit :

> Try this:
>
> d const(htsgw_surface, 0, -u)
>
> Bill Reilly
> www.PassageWeather.com
> bill at passageweather.com
>


-------------- next part --------------
A non-text attachment was scrubbed...
Name: nww3.png
Type: image/png
Size: 26311 bytes
Desc: not available
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20080627/4f43c906/attachment.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nww3-const.gif
Type: image/gif
Size: 65301 bytes
Desc: not available
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20080627/4f43c906/attachment.gif 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nww3-const-france.gif
Type: image/gif
Size: 42126 bytes
Desc: not available
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20080627/4f43c906/attachment-0001.gif