Kermack's reasoning is rock-solid, and there is essentially no possibility of snorkelling sauropods. As Kermack himself noted in the closing words of his paper, "If the sauropods were, in fact, aquatic, they probably lived much the same sort of life as the present-day Hippopotamus, swimming and diving in the water, and walking along the bottom. To breathe, however, they would have needed to raise their body nearly, if not quite, to the surface."
Did some sauropods live that way? It seems enormously unlikely that, among such a huge clade, widespread in time and space, there were not at least some species that were as habitually aquatic as hippos. In determining which those might be, the osteological correlates to look for would include stout limbs, barrel-shaped (as opposed to slab-sided) torso, and reduced skeletal pneumaticity. In the first two of these, Opisthocoelicaudia looks like a strong candidate; but it fails the last.
-- Mike.