... the weight distribution would have them bobbing around tail-down...
Finally, a physics-based reason for the Scout NOT to have a flat belly: it's to compensate for being tail-heavy, so it floats "flat" with half the tail submerged... After all, who else has greater need of rough-landing capability than a Scout?

I think that "50 Starbases" may have suggested 5 or 6 thousand tons as an upper limit on hard stands. Can't remember if the more recent "Starports" says anything.
At Tavonni (Beowulf Down), I went with 0.5m of "permacrete", a kind of engineering "unobtanium" even stronger than concrete (but I used real-world concrete prices) and multiple landing pads of different sizes, up to a maximum of 6000 tons displacement. Anything larger and I assumed it would have to land in water, and I (IMTU, there's no canon) put the ceiling on that at 20,000t displacement. You're starting to get into the size of large real-world (wet) merchants and super-tankers, and they're pretty fragile for their size. I figure if you tried lifting a super-tanker on grav you'd break it's back - it needs the water all around to support & cushion it.
Never really bothered to calculate ground pressure, tho'. Just made sure we could fit that ol' triangular 1000t merchant from FASA's _Merchant Ships_.
We did attach a runway, though (one of my few disagreements with John M. Ford's magnificent tome: it should say 'ports "mostly" do not have runways, rather than they "never" do. Hard & fast rules don't work across 11,000 worlds). 'Course the main reason we use it is not for "airframe"-style starships, but just-for-fun old-fashioned aircraft we rent out to conference attendees and other tourists. Either they fly themselves or charter a plane - kinda the equivalent of going on an old stagecoach or steam train at any number of theme parks worldwide...
