Peter:
 > It is 200 nm from the baseline, not from the coast. This probably 
> won't make much difference, but I wonder whether it makes any all the 
> same. 
 
Much less impact, of course, at the 200nm level than at 12nm, since the 
baseline has to be quite long before it begins to show a serious effect 
at the 200nm limit - most river estuary baselines become irrelevant, 
for instance. Big bays are, as you say, much more of a worry. And I 
haven't looked at somewhere like the Maldives yet, but I think diffuse 
archipelagic countries with long interisland baselines just won't 
render well at all in my model.
As I say, I look on this exercise as experimental geography - I render 
the 200nm limits from GTOPO30, and then muse on the difference between 
my result and the real world of maritime treaties and disputes. I would 
never, for instance, have known about Isla del Coco (another 
potentially habitable island a few km on a side that's missing from 
GTOP30) if I hadn't noticed its EEZ linking the Galapagos to Central 
America.
 >(The enclave of high seas in the Barentsz Sea has this nickname loop 
>hole; the one in the Sea of Okhotsk is the peanut hole, the one in 
>the Bering Sea the donut hole).
 
Lovely. (Maybe we should call the north polar high seas enclave 
the "ozone hole".) After while I hope to be able to generate a global 
view of 200nm EEZs which more or less matches the real world at the 
(low) resolution I'm using. I think a north polar view would be 
particularly interesting, to help judge the true extent of the polar 
high seas enclaves, which are always seriously distorted on 
conventional world maps.
Grant