Well, it's probably a little bit more complicated than that. If there's a cavity of known size, then one could predict with a high degree of precision how the gravitational pull should be affected. Then you see a different pull and you calculate that there must be a mass of size X causing the change. Then you say, ok, we have a mass X and a cavity of size Y; making a density calculation is trivial at that point. I'm guessing they did this calculation, and found that it is 1000kg/m^3 within error, and there probably aren't any other candidate liquids that fit that density. (For example, liquid methane, which has been observed on other moons, is about 450kg/m^3, or something like that. Not really even close.) Couple the density calculation with the fact that they know from observation that there are ice and ice geysers on the surface, and I think it's an if-it-looks-like-a-duck-and-quacks-like-a-duck type situation. These are just my speculations, however :)