What else can you do with a Big Dumb Booster?

So, this Tuesday SpaceX pulled back the curtain to announce their Interplanetary Transport System—a monstrously large rocket, fully reusable and about two and a half times the size of a Saturn V moon rocket—capable of transporting a hundred people to Mars, and with a goal of initial flight testing within a decade.

It's not total vaporware: in the past couple of weeks they also tested the first full-up Raptor engine that will power the ITS (a cryogenic methalox engine with a closed-cycle gas generator, which gives it a specific impulse head and shoulders higher than Apollo-era kit and the capability to operate on fuel generated from the Martian atmosphere for return flights). They've also unveiled the biggest carbon fiber tank ever assembled (the fully-reusable ITS will use carbon composites extensively), and have unveiled a bunch of targets for what the ITS stack will be able to achieve: in non-reusable form it will be able to deliver a 500 tonne payload to LEO, and with reusability in mind a 320 tonne interplanetary craft capable of landing vertically on Mars (and, when refuelled, of returning to Mars orbit without staging).

So, here's my question:

What are the other possible commercial applications of the ITS, besides sending a million optimists to Mars?

Here's what I can see:

1-2 order of magnitude cost reduction in cost/ton of payload to orbit: this is axiomatic. ITS won't be commercially viable for Musk's proposed Mars colonization bid if the per-launch cost of this big-ass fully reusable rocket significantly exceeds that of the big-ass but not fully reusable (the second stage is disposable) Falcon Heavy that flies later this year. So let's posit a cap of $100M on flight costs, or maybe $400M for a disposable shot (which would only really be necessary for a single monolithic payload that can't be broken down into sub-elements massing less than 300 tons—candidates for which, see below). (Here are SpaceX's cost estimates.)

Big, dumb, comsats: Currently the mass of a geosynchronous comsat is constrained by the payload of the available boosters, which are tailored to fit the perceived requirements of the comsat market. About half the mass of a comsat in GEO is fuel, used for positioning (satellites in geosynchronous orbit drift, very gradually, away from their parking longitude). Their power output is constrained by the solar panels they can carry and the size of their emitters. So a big GEO comsat today is on the order of 5-8 tons. A current advanced geosynchronous comsat such as Inmarsat-4A F4 has a 12 kW electrical system; this obviously puts a ceiling on its broadcast power; but ITS raises the bar so high that it effectively disappears. The first post-ITS generation of comsats could have power outputs in the megawatt range if necessary. So I'm going to guess that 1-2 decades after ITS flies, we're going to see satellite phones converge with regular cellphones in terms of size, convenience, and bandwidth capacity (although they're going to cost more). Upshot: terrestrial 5G and hypothetical 6G high bandwidth service will look more like municipal-area gigabit wifi, and your phone will cut over to satellite bandwidth if you roam into rural areas (or even suburban areas, by the US definition). But you won't notice anything except a slight increase in latency. It's as if your cell tower just moved into orbit.

No more Kessler syndrome nightmares: the launch stack is fully reusable. Anyone not aiming to operate a reusable launch stack by 2030 at this point is a buggy whip manufacturer. So that's one source of debris gone. And another source of the problem is the number of objects in space. A few giant satellites are less likely to shed debris or risk a collision hazard than a large number of small satellites. And we'll have so much spare lift capacity that cleanup becomes a practical possibility, paid for by the insurance underwriting industry: sending up a fleet of cubesats to hunt down, grapple with, and de-orbit 1960s paint chips is cheap compared to the payout if said paint chip holes your orbital Hilton.

Space tourism, for realz: the Bigelow BA-2100 spacehab only needs a 70-90 ton LEO launch capacity and has half the volume of the entire ISS. We can conservatively estimate that a space hotel with a ~300 ton mass fabricated using Bigelow's expandable tech and flown on the ITS would have 3-4 times the habitable space of the ISS, so room for 20-40 tourists and staff. (The inflatable hab tech isn't vapourware either: there's one docked to the ISS right now.) A week in space won't be a cheap vacation, but Virgin Galactic think people will pay $25K for 10 minutes in free fall; I reckon $250,000 for a honeymoon in orbit will find some takers among the 1%. (Passengers would travel as a sub-cargo aboard an ITS which would be mostly carrying other types of paying cargo.)

Return to the Moon, this time for good: a huge problem with proposals to build a permanent base on the Moon is that the Moon is short on volatiles that you can turn into fuel, and has no atmosphere worth mentioning for aerobraking purposes. (Lithobraking is not recommended. Or should I say lithobreaking.) One serious proposal for a long-term Lunar presence requires the construction of a Lunar space elevator. This would not run from surface to geosynchronous orbit—the moon, being tidally locked, has no GEO—but instead to the L1 (near-side) or L2 (far-side) Earth-Moon libration points, 56,000 and 67,000 kilometers from the surface (points where the effect of the Moon's gravity and the effect of the centrifugal force resulting from the elevator system's synchronous, rigid body rotation cancel each other out and an elevator could be stable). Unlike a terrestrial space elevator sufficiently high tensile strength materials for such a tether already exist. There is, however, the slight problem of fabricating and shipping a 120,000 kilometer long cable out to near-Lunar orbit (and capturing a near-Earth asteroid to act as a counterweight). This is just a wild-ass Charlie guess, but I suspect shipping up 500 tonne cable drums will work out cheaper in the end than trying to build a carbon fiber factory in space (at least, until space industries are sufficiently developed to go the whole eat-your-own-dogfood distance). (Upshot: ITS probably makes the folks at LiftPort Group very, very happy.

Stupidly enormous space telescopes: Because there is a budget and a booster that can lift primary mirrors 17 meters in diameter is going to make the astronomical community need a change of underwear when the implications sink in. (Put it this way: one part of the value proposition is "maps of continent-sized features on terrestrial exoplanets" by 2040.)

(Speculative) Wake shield molecular beam epitaxy fab lines: with a wake shield you can produce an ultra-hard vacuum suitable for growing rystalline semiconductor thin films. I don't know wht the commercial implications are other than really pure GaAs and AlGaAs semiconductor substrates, but with rock-bottom launch costs and the ever spiralling cost of semiconductor fab lines (part of which is down to the requirement for clean room air flow on a large scale) we might see some semiconductor manufacturing activities planned for deployment in orbit after 2030. (After all, high-end microprocessors—at least before they're sliced, diced, and packaged in pin grid arrays—are some of the few objects that cost so much per unit weight that they'd be worth retrieving from orbit even with current generation flight costs.)

Anyway: these are the first non-Mars short term applications of ITS that I can come up with off the top of my head. Stuff I don't think is plausible: ITS upper stage derivatives used as ballistic point to point passenger transports on Earth (because reasons), pick-axe wielding asteroid miners going out to the belt to hew mineral ore and bring it back to Earth orbit (yeah, we'll get asteroid mining, but probably by using the smallest feasible robotic gravity tractor—you don't need the ITS for that job), microgravity crystallography factories for pharmaceuticals (oh come on), Lunar 3He mining for aneutronic fusion reactors (because if you can do aneutronic fusion at all Boron is much cheaper). Anything else? Chip in!