There is no way around it. The different resources of the Moon are in different places, and they don't seem to be close together. Mapping of resources is really patchy, low-resolution, and incomplete, so maybe there is a sweet spot that helps with this we have only to discover. Probably not, though, and even if there is, if you are talking about serious development, you are going to need to go get things from far away sooner or later.
Building first at the equator simplifies some things but runs into this as soon as the water supply comes up. For all that it is much easier to build big quickly at Lalande Crater, it has advantages in trajectories to and from Earth, and has much more iron, potassium, phosphorus, thorium, and rare earth elements than the poles, all of its water has to be delivered. A base at the poles could get its water from permanently shadowed craters once the machinery to do so had been developed and built. That could prove difficult and expensive though - we don't know yet. If you set up an efficient supply route from Earth, supplying enough water doesn't get problematic until it is time to expand the colony from the initial crew of 30 to a population of hundreds. Even then, if you skipped putting in giant swimming pools it is conceivable you could simply add a few cargo runs to your schedule loaded with nothing but water from Earth and stay within your budget. However, a place on the Moon where people are supposed to live for the rest of their lives really ought to have giant swimming pools. So, the trick is how to bring kilotons of water to the colony for those residents. And just aside from that you will want to be able to move kilotons onto and off of the Moon anyhow, so this is really just the first, most obvious case of that need.
As always, in assessing the position that follows it is important to remember that Moonwards is about showing what could be done if limitations imposed by politics were removed and serious ongoing investment was made. Actually, i should write that up as a blurb that goes in the header or sidebar of this blog... So if you see that now in that spot, this is when that occurred to me.
Our planned route to industrial-scale cargo movement around the Moon and between it and space is first, building a shuttle with nuclear thermal engines and a cargo ship with ion engines. The shuttle would handle transport between lunar orbit and the colony, and the cargo ship would do the run between lunar orbit and low Earth orbit. Then once the initial complement of scouting and construction robots and equipment had been delivered, the next mission would set up a non-rotating tether in an equatorial orbit at an altitude of 5000 km. At first that tether, or skyhook, would only descend part way to the surface. As the shuttle and further cargo missions added mass to the tether anchor and the system was perfected, the skyhook would reach to 20 km above the surface. Finally, a polar skyhook is added and further tethers reaching to low Earth orbit. Let's look at this piece by piece.
Nuclear thermal engines would require little development to create working models, their performance is superior, they are robust and can be restarted with ease, and they are safe. Their nuclear fuel elements would not be put together and start generating high radiation until in lunar orbit, so if there was a launch accident dispersion of radiation would not happen. Once operational, shielding the engines enough to block their radiation from entering the interior of the shuttle is quite easy. The shuttle would run both on straight hydrogen, and on methane. On methane it could get a specific impulse of about 500 s, on hydrogen, about 900. Using methane would give superior thrust during take-off and landing. More importantly, shipping and storing methane is far easier than doing that with hydrogen. Possibly the total payload to the colony would be maximized by not using hydrogen at all. Or maybe if hydrogen was a small portion of the mass of the propellant, and large tanks could be used that allow use of high-pressure gaseous hydrogen instead of liquid hydrogen, it would be advantageous to use a mix.
The much larger cargo vessel could run on xenon. If the HiPEP ion engines NASA has worked on were fully developed, they could have a specific impulse of
up to 9000 s. At that efficiency, and given that they have to use the slow spiral trajectory of low-thrust engines, they could deliver a ton of payload for every 115 kg of propellant.
The skyhooks take more technological development. Zylon is an existing material that is strong enough for the cables, but issues to do with maintenance on a cable of that length, and how best to make cars that travel up and down it come up. The cable can't get twisted too much, it needs to be able to take hits by micrometeorites (a big hit isn't too likely, i don't think, but it has to be engineered for), and it needs to be repairable. A loose tubular braid of a couple of dozen smaller cables sounds like a good idea. Maybe the climbing car could have sort of twin caterpillar tracks that press together from either side of the cable, meshing with it at many points to distribute the strain, something like that. At any rate, these seem like moderate hurdles to overcome, especially considering how awesome the thing you get in return is. It might take some time to get used to handling the undulations and oscillations of such a cable, but those are probably minor issues that maybe need to be compensated for a bit. I leave this for now to minds that are actually trained to study such questions. I'd just like to explain how very cool it would be.
If you have a mass in orbit around the Moon and a tether hanging down from it toward the surface, the rate the foot of the tether moves relative to the surface is determined by the orbital period of the center of mass of the whole structure. When that point is at 5000 km altitude, and the foot is at 20 km altitude, that rate is 223 m/s. In that situation, our nuclear shuttle from the surface could go meet the foot of that tether, be grappled by it, trade a payload pod on its cargo bed for one released by the foot, hang on until the orbit is back over the colony, let go, and land on a runway. The delta V it would need to do that would be a fraction of what it would need to ascend 5000 km to meet the cargo ship in orbit. (I don't know what fraction is plausible for that - i briefly took a stab at calculating it, but my math skills aren't there yet, so i have demurred for now.)
The tether anchor needs to have ion engines on it, enough to compensate for the downward drag of having the shuttle hanging on its foot, and the solar panels to power them. Unless the skyhook's anchor is many times the mass of the shuttles and payloads, the downward drag during loading at the foot would need to be compensated for quickly in order to avoid the foot dropping down so much that it smacks into the ground. At first the tether would need to stay much further from the surface in order to avoid this being an issue, and the shuttles would need to climb higher and faster to meet it. And so, each time they do, they might as well haul up a full load of rocks that can be taken by the climber back up to the anchor, where they would simply add to its mass. As the anchor gets heavier, the tether can be extended to reach closer to the surface. Once those rocks reach the anchor and the shuttle has let go, the center of mass rises to its usual spot, so the structure only loses the momentum needed to carry the shuttle while the exchange is being made. But since that takes a few days, between the payload from orbit descending 5000 km, the tradeoff being done, and the rocks ascending 5000 km, the upward rebound at the end of the process isn't relevant. Still, the amount of fuel used to do these maneuvers is reduced because the ion engines on the anchor do most of the work, and they are much more efficient than even nuclear thermal engines.
This system really takes off once a near-Earth asteroid is delivered to lunar orbit and integrated into the skyhook anchor mass. It really needs to be big though. I'm thinking the thing to model for the virtual colony would be one of about 5,000 metric tons, which is 10 times the size of what the Keck Report on asteroid retrieval was proposing. That isn't unreasonable if you have a heavy lift launcher and the next generation of ion engines at your disposal. The asteroid would arrive already equipped with ion engines and solar panels, thanks to the equipment installed to move it. A nice carbonaceous chondrite asteroid could be counted on to bear 500 tons of water bound up in its minerals. The processing facility to extract that could be installed on the asteroid surface, and it would send that down to the colony in exchange for loads of building materials. By the time that asteroid is in place, the colony would be producing those. In fact, those building materials would be used to create a space station on the asteroid. By that time, the colony would be capable of producing most of the things needed for that.
A polar skyhook would also be set up and get a near-Earth asteroid of its own. With that in place, a base at one of the poles can be done in style. The machinery and gear would be delivered from Cernan's Promise by shuttles that ride the tether, and a lot of it would have been fabricated there. So now you are talking about 1000 tons of water from the two asteroids, and a new stream of water from operations at the pole. It is going to take a while before the giant swimming pool is completely full. Once it is though, you can build another in a small fraction of the time, for a small fraction of the cost.
With reusable launchers, this architecture is definitely something a group of nations could move forward on quickly. Having a truly reusable launcher that greatly lowers the cost of the last and most difficult link in the chain, Earth to low Earth orbit, would be the key to ramping this transport system up into a genuine cargo system. Tethers between LEO and lunar orbit would amplify the capacity of the cargo run a great deal as well.
The Trans Cislunar Railroad described in Hop David's blog would be the thing to add. If some of the mass being pumped up the tethers on the Moon were directed onwards to those tethers, it could be used to build very nice space stations at the anchor masses of each of them. That system would greatly reduce the transit time for cargo between the two worlds. I was greatly helped in creating this sketch of the skyhook system that will be built in Moonwards by Hop's posts on tethers..
Another thing that would be a real boon to this system being built quickly and having lots of muscle would be successful development of the Neumann drive, which the colony could then fuel with magnesium refined on the surface and sent up to the tether anchors and the cargo ships. One of the set of calculations i look forward to doing is of how having lots of magnesium-burning Neumann drives on the cargo ships and tether anchors would speed up the whole process. However it is still not clear that there won't be development problems with that drive, promising though it seems.
Transport is the secret to everything in space. Feeling comfortable that this is an efficient, effective, realistic architecture is critical to making the virtual lunar colony feel real. We think it is. Let us know what you think.
(Note: This post was edited a month after the original posting in order to reflect improvements Sigvart made.)