Kym:
What you want to do is not easy. I would use a low tension (like 4)
with only the offshore points to establish the "floor" of the "bowl".
Then I would make the resolution somewhat coarser, and use something
like r.stats -xlq to "resample" this floor and create a new site list.
These sites will serve to fasten down the floor of the bowl.
Then I would combine all 3 sets of points into a single sites file and
respline with higher tension (say 150). The result should be a bowl where
everything connects smoothly.
I'd be interested in any feedback on this idea.
--------------------------------------------------------------
____ Oak Ridge National Laboratory
/ ) / / / / Environmental Sciences Division
/--< o / / /--/ P.O. Box 2008, M.S. 6038
/___/__<__/__/_ / / _ Oak Ridge, TN 37830-6038
William W. Hargrove (615) 574-1902
hnw@mtqgrass.esd.ornl.gov (615) 576-8646 (fax)
--------------------------------------------------------------