I have so many problems with the whole 'Ark Swarm' concept that it's hard to put them into words.
Any space-based plan is significantly bottle-necked by launch capacity. Even with all the resources thrown at it, we see that pretty clearly. It's not just that most of the 'casting of lots' chosen couldn't catch a ride, but within the Gen-pop we see important roles critically under-filled. The prioritization of that launch capacity was really poorly thought out too. Whats-his-face is apparently the only guy who thought that the Ark-Swarm's fundamentally constrained resources might be a problem to be solved sooner rather than later. Other characters recognize the problem when it's pointed out and even consider that the oversight might be indicative that the Ark-Swarm isn't meant to be more than a put-on to distract and placate the masses.
The Ark Swarm makes sense as part of a 'many baskets approach, and even as the public face of a many baskets plan (if for no other reason than that it's much harder for desperate people to mob Low Earth Orbit than a random salt mine) But when we see how other plans were implemented, they didn't have nearly the same amount of resources thrown at them, despite the bottle-neck to orbit indicating that there should have been plenty to go around even if the Ark Swarm was the main priority
And yet, it's clear from the Crater Lake accords and that the Ark is where Julia chose to make her escape to (in flagrant and pointless violation of the accords) that the ark is pretty much the basket.
Digging in makes a lot more sense. Making an underground colony self-sufficient and sustainable is basically the same problem as making one in space, but you have gravity, and air and water, and much less radiation, and it takes a lot fewer resources just to get there, so instead of (or even in addition to) focusing the entire planet's effort on one space-colony, everybody can pretty much build their own under-ground shelter and fill it however they want.
The book mentions that it would be difficult to expand the volume of the under-ground shelter due to the tailing taking up as much or more volume as is created, but that's not a real problem. You can process the tailing for volatiles and useful materials, and the slag can then be reformed into higher-density materials which can be used to reinforce the structure. The volume gains from such processing would be minimal, but by no means non-existent. However, while the surface may be inhospitable, there's no reason for it to be inaccessible. There's no reason you couldn't just dig a protected outflow, with an airlock if necessary, and dump the slag outside. As long as the living areas aren't too close to the surface, if an impact does bust the outflow, oh well, dig another one. Outflows can be tiny and cheap to dig, just a small bore-hole to the surface where slag material is melted and forced up via pressure. Once you can dig around and gain habitable some volume, the underground shelter would have basically unlimited resources and space to expand.