Using VisAD with oceanographic data

Hi Bill et al,

This must be something that occurs often but I can't find the answer in the
FAQ, examples or tutorials.

I want to visualise data from a coastal ocean model. For starters I am looking
at the depth map. So I have depth as a function of (latitude,longitude). First
I tried a Linear2DSet with the depth ScalarMap set as Display.RGB. This worked
fine giving (with the right ColorTable) a grading of colour from blue in the
deep water through green to red in the shallow. However the land (represented
in the data as zero depth) just appears red i.e. with no distinction from the
shallow coastal water. I would like the land to be something distinct, like
black. I also tried setting the depths over the land to a large negative value
but this just biassed the autoscaling of the colour map so I got a land/sea
mask with no detail in the sea part.

Next I tried Gridded2DSet and Irregular2DSet, by compressing the depth data to
remove land points and passing in a latlong[2][nactive] array of latitudes and
longitudes for the active (sea) points. This gave some black areas where the
land should be BUT there appears to be some interpolation going on so that land
areas between to sea areas get filled in with some mess^H^H^H^H, sorry,
interpolated data points. I also noticed this went very slowly, maybe due to
this interpolation.

What is the best way of doing this ?
Can I turn off the interpolation for the Gridded2DSet/Irregular2DSet ?
Can I apply a land/sea mask to the Linear2DSet ?
Is the vistutor.jar file OK ? (I have an old copy but I got an error when I
tried to unpack the latest version)
And why is there no sound file for the guy playing the bagpipes ?

Cheers and TIA,

Mike.

-- 
____________________________________________
Dr. Mike Ashworth
Head of Advanced Research Computing Group
Computational Science & Engineering Dept.
CLRC Daresbury Laboratory, Warrington WA4 4AD, UK

Tel: +44 1925 603663
Fax: +44 1925 603634
Email: M.Ashworth@xxxxxxxx
http://www.dl.ac.uk/TCSC/CompEng/Ashworth_M/
____________________________________________