Hi, sorry about redirecting you here, but GitHub issues are for feature and bug tracking. This forum is a better place for discussions.
As I mentioned on GitHub, rendering undersea terrain is possible out-of-the-box, with a few caveats. With a little bit of work it should work nicely.
The first caveat is that the CesiumTerrainProvider does not currently support heights less than -100 meters relative to the WGS84 ellipsoid. It’s just a limitation of the current format. We’ll be moving to a new format in the next few months that will not have that limitation. Also, it is not as big a limitation as it may sound even now, because TerrainProviders are a plugin point, so writing a new one that interfaces with your data, expressed in a different format without this limitation, is not hard at all.
The second caveat is that you may run into some problems with horizon culling. I describe the problem at the end of my second horizon culling blog post (http://cesium.agi.com/2013/05/09/Computing-the-horizon-occlusion-point/):
Also, please keep in mind an important caveat when using this approach. In the real world, objects occluded by the WGS84 ellipsoid are not necessarily occluded by the real surface of the Earth. This is because the Earth’s surface is actually slightly below the ellipsoid in parts of the world. Depending on your application, using WGS84 as the occluding volume may be acceptable, or you may need to use a more conservative ellipsoid.
With undersea terrain, it is clearly the case that terrain is substantially below the WGS84 ellipsoid. Thus, the horizon culling is likely to be too aggressive and cause parts of the surface to not render. As a quick work around, we can just hack out the horizon culling. A better solution is to use a more conservative (smaller) ellipsoid for horizon culling; specifically, one that doesn’t stick up past the bathymetric surface anywhere.