Government space agencies and some of the largest space companies are talking about building a space station somewhere in cis-Lunar space, the so called "Gateway". As far as I've heard, the (still very preliminary) plans are to assemble its modules in the intended final orbit.

Wouldn't it be safer and cheaper to assemble it in Low Earth Orbit instead?

Safer given the experiences from the ISS and with much better radiation protection for the crew and the option for quick evacuation to Earth.

Cheaper given that chemical rockets would only need to launch the modules of the space station to LEO instead of to a higher delta-v orbit at the Moon. The space station, when ready to commence operations, could then be towed outwards to its intended orbit by a Solar electric propulsion spacecraft. Since the now canceled Asteroid Retrieval Mission was formulated for over half a decade ago, it has been a priority for NASA to find some kind of mission for a powerful SEP-tow to prove the abilities of this propulsion technology.

Here's a recent Boeing presentation of a Gateway station concept: http://spirit.as.utexas.edu/~fiso/telecon/Duggan_8-9-17/

  • $\begingroup$ Some advantages that you already mentioned are that it is cheaper, lower radiation contamination and it is closer to Earth. But what is our actual goal? Is it to stay in the Earth and not care about our surroundings? If so, then release it in LEO. But if our objective is to explore then assemble it in the farthest place that we can. $\endgroup$ – Matthew Aug 21 '17 at 11:30
  • $\begingroup$ @Matthew But why, really? Rocket engines are not assembled in space "because it is hard", but on the safe practical ground (which is hard enough). The assembly of entire 10+ton modules for space stations and interplanetary spacecrafts benefit from microgravity, so why not use the safest and cheapest orbit to do it at? When everything is proven to work in relative safely, then push it out to its destination. Just common sense, but I've noticed that common sense isn't always applicable in space flight, so I wonder what the Devil in the details is here. $\endgroup$ – LocalFluff Aug 21 '17 at 12:08

Making this a community wiki answer so others may add to it as they see fit:


  • LEO assembly allows less mission risk if assembly does not go completely as planned. It's much easier to schedule and launch replacements or corrective hardware.
  • LEO assembly allows for checkout and stress testing of hardware in a location where failure allows a safe return of crew in a matter of hours, rather than days.
  • "Once you're in orbit, you're halfway to anywhere." -- Keeping the upmass of individual launches small allows the use of existing, proven launch technology to get components up for assembly, lowering the program risk posture and reducing the consequences of a single launch failure. As the question implies, transport to higher orbits can then be accomplished over a longer period of time with less mechanical stress to the vehicle using a low-thrust, high-$I_{sp}$ electric thrust system.


  • Raising the apogee of a structure in the mass class of the ISS to lunar orbit levels with electric propulsion will have to happen over the course of months. During this time, the vehicle will spend an unprecedented amount of time in two particularly dangerous regions: the Van Allen belts, which could severely damage the avionics on board, and a region of LEO between roughly 800 and 1400 km altitude, in which the orbital debris threat is roughly 100 times worse than the ISS altitude. While it's impossible to get to lunar orbit without passing through these regions, all missions to date have spent as little time as possible exactly to minimize the risk from passing through them.
  • Multiple launches to LEO for assembly vs. fewer launches to lunar orbit means more wasted upmass in terms of launch vehicles. While the consequences of any one launch failure are greater, the probability of having at least one launch failure does depend on the number of launches. Though the probability per launch will not be the same for different architectures, the aggregate difference may sufficiently offset the program level risk.
  • The space environments in LEO and lunar orbit are sufficiently different that an optimal design for one area is far from optimal for the other. Among the environments that are drastically different between LEO and lunar orbit: thermal (earth albedo mellowing things out in LEO, not so much in lunar orbit), radiation (galactic cosmic rays are mitigated by the Van Allen belts in LEO, no such protection in lunar orbit), MMOD (basically no orbital debris in lunar orbit, and if we're smart, it will stay that way), atomic oxygen (LEO still has wispy tendrils of earth's atmosphere, and AO is insidious enough that it is always a design consideration, lunar orbit has basically none).
  • $\begingroup$ One thing interesting you raise (as in being adversarial to my common sense) is the change in thermal protection. But that seems relevant only to sensitive IR telescopes, not to the ISS or Apollo. Would the envisioned "Gateway" really be damaged if it were in LEO? 1) Crew wouldn't need to be aboard during the (one year or so?) passing of it through the debris and radiation fields while going from LEO to a cis-Lunar orbit. 2) LEO assembly actually requires fewer, not more, launches. A couple of SLS Skylab sized payload launches to LEO would make a pretty good space station. $\endgroup$ – LocalFluff Aug 21 '17 at 13:49
  • 1
    $\begingroup$ It's not so much about thermal protection, it's about the design and operation strategy for thermal control. The sun is just as hot in lunar orbit as it is in LEO. What is different is how much of the sky sphere is "cold". Your shadowed regions in lunar orbit are going to be much colder than they would be in LEO, as the earth, which is relatively warm, takes up much more of the sphere. $\endgroup$ – Tristan Aug 21 '17 at 14:32
  • $\begingroup$ Regarding your other comments: 1) Whether the crew is aboard doesn't really affect whether the vehicle sustains permanent damage enroute. If the pressurized part gets a hole punched in it, it's useless either way. 2) A LEO-constructed vs lunar orbit-constructed station will more than likely have wholly different designs centered around the construction concept. Yes, if you hold the design constant, it may take fewer LEO launches, but you can't assume the design would be constant. $\endgroup$ – Tristan Aug 21 '17 at 14:35

LEO is a bit dangerous in my book. If anything goes wrong you may not have enough of a margin to fix things. Put the station in a little higher orbit and make your deliveries to LEO and use tug to push building blocks to the higher orbit. That way you have advantages of both. A big problem with LEO is the space junk. When you are done the higher orbit is easier to change to lunar or whatever.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.