On Friday, the British Library finally sent me the paper on lava tubes  I'd been referred to by Tom Billings (getting hold of it was a non-trivial exercise). It was definitely worth the effort, however: it's very interesting indeed.
As I suggested on 2nd March, the problem with designing a lava tube-based outpost is that, at the moment, we've no way of knowing whether there actually is a lava tube where we think there is one. I was pleasantly suprised to find my speculation confirmed by Coombes & Hawke; they say, "There is one major problem to consider... the difficulty in confirming, absolutely, that a tube does in fact exist... and determining what its exact proportions are."
The best thing about this paper, however, is that it provides a table of very strong lava tube candidates, some more than a kilometre in width. Some of these lava tubes could be ideal sites, except for the unfortunate fact that none of the tubes assessed as 'prime candidates' are close to the equator. That's a shame, because an equatorial position would be the most fuel-efficient to get to.
Even so, it's possible that strategists might decide it better to go with a lava tube we know rather than spend a lot of money on techniques such as SPR in the hope of finding a more conveniently located site.
 Coombs, C.R., and B.R. Hawke, 'A Search for Intact Lava Tubes on the Moon: Possible Lunar Base Habitats', in The Second Conference on Lunar Bases and Space Activities of the 21st Century (W.W.Mendell, Ed.), NASA CP-3166, Vol. I, p. 219, 1992