This image, which is supposed to represent the current architecture of the mission to Mars by NASA, raised a wave of emotions. People find it too long and complicated. But is it really possible to make it fundamentally simpler? I think that the idea of a single-launch Starship that will land on Mars, where it will use ISRU to produce fuel and return to Earth in 2030 is far from the truth. enter image description here https://twitter.com/KenKirtland17/status/1745931455199338512

  • 2
    $\begingroup$ "I think that the idea of a single-launch Starship that will land on Mars, where it will use ISRU to produce fuel and return to Earth in 2030 is far from the truth.": ...who has suggested such a thing? SpaceX certainly hasn't. $\endgroup$ Commented Jan 13 at 15:44
  • 1
    $\begingroup$ Yeah that's right. It's more of an exaggeration. $\endgroup$
    – Saturn V
    Commented Jan 13 at 17:09
  • 1
    $\begingroup$ What exactly is your question? $\endgroup$
    – Slarty
    Commented Jan 13 at 22:46
  • 11
    $\begingroup$ @SaturnV But what's "better"? Cheaper? Faster? Safer? With fewer launches (and bigger rockets) or smaller rockets (and more launches)? Fewer orbital rendezvous? Fewer ground rendezvous? More astronaut-hours on the ground? Without knowing what parameter(s) you want to optimize, it's hard to say what a better approach would look like. $\endgroup$
    – Cadence
    Commented Jan 14 at 14:10
  • 5
    $\begingroup$ For context: The Apollo program took 8 years and 11 launches to put 3 men on the Moon ~350,000 km away, the missions lasted about a week, and the program cost a significant fraction of the US economy. Mars is 55,000,000 km away (about 150x further) and it will take over a year to get there and back. Considering NASA doesn't have anything like Moon landing money, and the problem is literally astronomically more difficult, I'd say their plan is pretty good! $\endgroup$
    – Schwern
    Commented Jan 14 at 22:15

3 Answers 3


All manned Mars proposals that do not involve a nuclear launch from earth have a LOT of launches.

Pretty much every serious plan since Von Braun's have involved multiple launches assembled in earth orbit and look much the same over the years. His scaled down 12 person/two vehicle version needed 400 launches. Looking at the plan above you can consider the impacts of deleting things:

There are mass penalties with common transfer stage and landing system, a bespoke design might allow rover or power components to cut from three to two launches but increases the cost due to additional design and testing needs.

You can scale back or delete the power system, cutting three launches to one or zero, but compromise the rest of mission (eg do Landing with batteries and surface time in hours).

You can delete the rover and probably cut the landing component to a single launch, but that means your team must walk everywhere and MUST get down near the ascent vehicle. Surface time in hours, possibly crew size of one.

You can delete the ascent vehicle and make it a one way trip, saving three or more launches.

You can delete the landing, saving 9, possibly 10, launches and replace them with rovers operated in close to real time or a visit to the Martian moons. A pure free return flyby mission might be as low as two launches, especially if the crew is cut to one small person, though this leaves the question of 'why'.

Getting to Mars is hard, and if planning to do something actually useful there to justify the trip you need to take stuff with you. In practice, a sensible mission design probably has MORE launches in it since you want to provide redundancy to avoid a failure scrubbing the entire mission (eg two ascent vehicles and two landers).


The infographic in your question references a Sept 28th, 2021 NASA document called the HEOMD Strategic Campaign Operations Plan for Exploration, where HEOMD stands for "Human Exploration and Operations Mission Directorate".

It is possible to propose a simpler plan. Mars Direct, and its revision, Mars Semi-Direct, are examples of mission plans that are arguably simpler than plan depicted in the above infographic.

enter image description here (ref) Left panels say "Year 1", "Year 3", "Year 5". Right labels alternate between "Unmanned" and "Manned".

Of course, minimizing complexity isn't always the only objective. Other objectives include reaching a satisfactory political compromise, minimizing risk, increasing crew welfare, maximizing the utility of the mission as a learning experience, achieving certain science objectives, doing the most possible within a given budget, arriving first to score points for your ideology on the world stage, executing the mission in a way that best serves an even larger long-term goal (such as establishing a colony), etc.

The Starship Approach

Missions that land on Mars and missions that land on the Moon require imparting similar amounts of delta-v. So we can review what we know about Starship's lunar HLS mission plan to get a sense of the complexity of a Starship-based Mars mission. A plan involving sending Starship to land on the moon will involve refilling. According to this Space News article...

In a presentation at a meeting of the NASA Advisory Council’s human exploration and operations committee Nov. 17, Lakiesha Hawkins, assistant deputy associate administrator in NASA’s Moon to Mars Program Office, said...

SpaceX’s concept of operations for the Starship lunar lander it is developing for the Human Landing System (HLS) program requires multiple launches of the Starship/Super Heavy system. One launch will place a propellant depot into orbit, followed by multiple other launches of tanker versions of Starship...

“It’s in the high teens in the number of launches,” Hawkins said. That’s driven, she suggested, about concerns about boiloff, or loss of cryogenic liquid propellants, at the depot.

Earlier, in August, 2021, Blue Origin made an infographic for its lawsuit with NASA over funding a second HLS program. That infographic roughly illustrates what Hawkins explained during the meeting...

enter image description here Blue Origin Lawsuit Infographic

Therefore, even if just a single Starship were sent to Mars, the overall mission complexity would still be high - likely involving on the order of 20 twenty launches from Earth.

  • 7
    $\begingroup$ I would argue that LEO refueling launches don't actually really increase the mission complexity that much. While it's yet to be demonstrated, I would actually argue that doing the same thing over and over actually reduces complexity in the system. Specifically, comparing two systems that have the same end capability: one requires many unique complex steps, while the other one requires many steps that, while complex, are not unique and have been done many times before. $\endgroup$
    – Dragongeek
    Commented Jan 14 at 14:05
  • $\begingroup$ I agree. I was thinking about getting into the topic of how one defines "complexity" but it seemed like too much of a tangent topic. $\endgroup$
    – phil1008
    Commented Jan 14 at 18:48
  • 5
    $\begingroup$ Most of the criticisms about multiple Starship launches are similar to the "nine engines is crazy" argument, and later the "33 engines is crazy" argument. People are making comparisons assuming the complexity and cost of traditional hardware. Musk's goal is to simplify the processes so that multiple launches are not complicated or expensive. We won't know for a few years if Starship can meet these goals, but based on what we are seeing with Falcon 9s constantly increasing flight rate, it is quite possible that within a few years Starships will be able to fly relatively quickly and economically $\endgroup$ Commented Jan 15 at 3:50
  • 2
    $\begingroup$ @Dragongeek: And as Musk likes to point out, SpaceX has already demonstrated fully autonomous precision docking and proximity operations with Dragon many times, and orbital refilling should actually be easier, because SpaceX is in full control of the design of both vehicles and both vehicles can maneuver. So, the docking part should be easy, that leaves in-space cryogenic liquid transfer, which has already been demonstrated at small scale on the ISS as well with the robotic refueling experiments. $\endgroup$ Commented Jan 15 at 15:59
  • 1
    $\begingroup$ @StevePemberton and people are still thinking of launches as rare events, products of heroic effort planned years in advance with months of delays being routine, with vehicles like the SLS being designed with that being considered normal...52 months lead time on a fully expendable SLS core and a production pipeline that can't even put out a core a year. Falcon 9 does multiple launches a week and Starship is designed from the start to allow an even faster launch cadence. $\endgroup$ Commented Jan 15 at 16:06

I think a lot of people are getting hung up on "complexity is bad" when the reality is that to solve complex challenges, we need complex solutions that can eventually be simplified once time and experience has been gained.

Specifically, there are plenty of "complex" systems that operate well and "seamlessly" today, to such a degree that most people don't even think about them.

An example:

Take something like the postal service. Explain to someone who has never heard of it before, that there is a system in place that enables anyone, regardless of how in-the-back-woods they live, to send letters and parcels to almost anywhere on Earth, across political and international borders, often in a matter of days. They would look at you like you've lost your mind. They could raise all sorts of questions about logistics, how it all works together, and generally just highlight how ridiculous this capability is and how monstrously complex it is.

Nevertheless, it works well. It works so well, that most people don't even think about it.

For this reason, while reducing complexity in a system is a nice thing to strive towards as an engineer, trying to jam simple solutions onto complex problems isn't always desirable.

NASA's Mars Plans

The plans, as outlined in the diagram, are knee-slapping, hilariously ludicrous. I would be mind-blow if Mars Sample Return managed to launch anything before 2030, and the idea that NASA will be launching components of a Mars base in Early 2030s is just not real. My bet is that this plan exists solely to sell the idea that SLS has plenty of work to do in the future, which is a debatable point.

Why? Multiple reasons:

NASA's focus is Luna

The Artemis program is just getting started. Currently, all the Artemis timelines are being pushed back, but, if I'm being optimistic, we will see boots-on-the-regolith before 2030. I suspect that this will result in the Moon capturing a large chunk of NASA's attention and budget over the following decade they will probably want to build out a Moon base so they can achieve the "sustainable" goal.

How exactly the ISS situation wraps up is unclear, but I could see a very realistic scenario where "moon base" becomes the type of budget line-item that the ISS is today: the premiere manned program, with widespread international cooperation, and with a dash of space tourism.

Supporting these Moon activities will eat up the SLS launches, and leave no budget for a flag-planting operation on Mars.

Travel times are too long, the stay on Mars is too short

There's just too little payoff. You've got astronauts crammed into a capsule for two years with a short, month-long surface jaunt in the middle. It's just stupid. Not only would we needlessly be psychologically damaging the astronauts, but there are all sorts of concerns associated with the short-stay mission structure.

For example, gravity and sickness. What happens if once our intrepid explorers arrive at the red planet, they are all so weak due to the sudden gravity that they can't perform mission tasks, or what happens if they fall ill (which they will) and this similarly degrades their performance? Even if the astronauts are just weakened for a week, that's already a third of the mission time.

I'm also skeptical of the scientific merit of a 30 day stay. As structured, this would essentially just be a flag-planting sample run. 30 days doesn't really leave enough time for actual "investigation". The hourly price of the astronauts there will be so extremely high in a dollars-per-minute way, that they won't have any time to do anything exploratory, and labs on Earth won't really have enough time to plan scientific activities. Beside the presumably larger volume of samples, there's little scientifically such a short stay could tell us that a completely robotic and significantly cheaper sample return mission couldn't.

Mass-is-expensive thinking paradigm

NASA and much of old-space is still stuck in a "mass-is-expensive" paradigm, meaning that one of the fundamental constraints is reducing mass down as much as possible. This is why our spacecraft have hulls thin enough that someone could poke a hole through with a moderately sharp screwdriver.

This will all (knock on wood) change when systems like Starship or New Glenn come online. The cost reduction that these will offer cannot be understated, and will require NASA to to scrap and remake basically all their plans once we cross this Rubicon where "kilotons to LEO" is no longer an unusual figure.

  • 3
    $\begingroup$ The short stay mission structure also increases the overall detrimental health effects. The astronauts are not only stuck in the interplanetary radiation environment for a longer period, but their trajectory takes them past Venus through a solar radiation environment twice as intense as that experienced at Earth and four times as intense as that experienced at Mars. It literally maximizes harm in order to minimize the benefits. It's an idiotic idea, but some people seem to be in love with it for some inexplicable reason. $\endgroup$ Commented Jan 14 at 18:58
  • $\begingroup$ 180 days to Mars is like a normal stay on ISS. Most returning astronauts are pretty functional within a week, it will likely be no worse in the 2/5 gravity on Mars, maybe even a little better. Compared to a couple of days on the Moon three functional weeks on Mars will provide a lot of useful experience as well as scientific opportunities. Do I believe this powerpoint? No. I don't think the first Mars trips will occur until we have the propulsion technology to do short stay missions with return in just over a year, however many decades that takes to happen. Multi-year missions will then follow $\endgroup$ Commented Jan 15 at 1:05
  • 5
    $\begingroup$ Mass will always be expensive. Without a space elevator or some other speculative technology, there's no escaping the tyranny of the rocket equation. New reusable launch vehicles can be more economical, but it will never be cheap climbing out of Earth's gravity well. $\endgroup$
    – bjmc
    Commented Jan 15 at 12:33
  • $\begingroup$ @bjmc the "tyranny of the rocket equation" accounts for very little of the cost of launch, it's almost all operational complexity (which might not actually be any less for an elevator) and cost of expended hardware. If Starship gets anywhere close to its operating cost targets, it will be able to put payloads into orbit for far less than a space elevator. And Starship's cost estimates are far less speculative than those for such an elevator, considering partially-reusable rockets are already in use while a space elevator requires materials we can't even fabricate yet. $\endgroup$ Commented Jan 15 at 16:22
  • 3
    $\begingroup$ @bjmc The point I'm trying to make is that most sacrifices made in NASA plans today are because mass is too expensive. This simple fact flows in from day one, and this is why we end up with ridiculously long transit times. With today's tech we could do a 100-ish-day transfer if we wanted to, but doing so is seen as a "waste" of fuel and budget: instead, the astronauts spend years aboard and we stomach the cost. With Starship-levels of lift, where we can send many, many kilotons of fuel to orbit for less than a single SLS launch, we can start being a bit "wasteful" with our mass priorities. $\endgroup$
    – Dragongeek
    Commented Jan 15 at 16:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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