Note that the depth wherever shallows are not drawn on map_C.tga is treated as being either 1,000m or 400m. Compare this to the English Channel, North Sea and Baltic Sea, where much of those areas are no deeper than a few tens of meters!
This leads me to become more desirous of a better treatment of water depth determination. My new map_W.tga idea, which is the consolidated map_C.tga, would be treated as an entirely separate entity from map_C itself. Where map_C might have no shallows drawn at all, map_W could have vast areas of shallows, they being used only to set the depth to which a ship would sink no farther. Map_C draws the shallows texturing, whereas map_W would supply a number for water depth; they can be quite discordant, if that's what's intended.
This new scheme would NOT require that all maps with water areas have a new map_W. If such a map is absent, the stock scheme by itself to determine water depth would still be in effect.
And while I mull this matter over yet more, it occurs to me that in the existing scheme where a distance limit of just 355m from a ship within which to do the 'spiral search' for land seems rather small. The search is done over what seems to be a rather small step size of just 10m. Why, I wonder, would be not extend the land search to, say 1km, with steps in distance of about 30m being in effect?