How I’d Go To Mars

Over on Arcturus, I rashly promised to post something about how I’d do the whole Moon-Mars thing. And so I shall, but with no pretense of technological or future-historical accuracy, though I’ll mention some technologies and dates; instead, I’ll be building a strawman proposal, with attention to its project-management aspects (in what follows, all definitions are taken from A Guide to the Project Management Body of Knowledge [PMBOK® Guide], 2000 Edition; Project Management Institute).
The first of those being assumptions, whose formal definition is “factors that, for planning purposes, are considered to be true, real, or certain.” In particular, I’ll assume that stakeholder (“individuals and organizations that are actively involved in the project, or whose interests may be positively or negatively affected as a result of project execution or project completion; they may also exert influence over the project and its results”) interests have already been balanced.
They haven’t, of course, and the game-theoretic aspects of a program slated to stretch through as many as seven future Administrations, twice that number of Congresses, and nearly thirty Federal budgets, render the prospects for the new space policy rather bleak, I’m afraid. So let’s pretend that isn’t true. The American public is united in its support (or perhaps cowed into submission, or merely indifferent enough to raise no objection): we’re going to Mars! How do we get there?

Another key project-management concept is constraints — “factors that will limit the project management team’s options.” More specifically, “the work [of the project team] typically involves competing demands for scope, time, cost, risk, and quality.” The most common framework for evaluating constraints is the triple constraint, colloquially expressed as “good, fast, cheap — pick two.” I’ve decided I want lots of scope, which means either lots of time or plenty of budget.
I’m vicariously choosing to spend more money and finish up sooner, because I assume that we’ll have strong nanotechnology around 2020, after which space exploration requires several orders of magnitude less expense and headcount — see this section from Engines of Creation for how we might be mass-producing rocket engines superior to the SSME by growing them in vats. Since I’m amusing myself here by envisioning a multi-billion-dollar program with projects all over the world, I’ll have none of this cheap, ubiquitous stuff cutting into my bureaucratic empire-building.
Also, because “projects manage to requirements, which emerge from needs, wants, and expectations,” I should begin by listing some:


  1. I want the mission to be easily repeatable. That means the creation of processes, vehicles, standards, and infrastructure for follow-on use.

  2. I want to manage the risks associated with long periods of isolation and inaccessibility. That means working our way up to a Mars landing by stages, and it certainly means in situ resource utilization (ISRU) wherever possible.

  3. I want to promote international cooperation, while retaining the lead role for the US and avoiding the execution of critical-path tasks in unstable countries.

  4. I want to engage the public, especially in the US but elsewhere as well.

With all that said, my program would include, but not be limited to, the following projects:


  • Develop an equatorial launch site (ELS), to be widely but not universally available; first preference goes to the space agencies of free countries with a track record of orderly transfers of power, and to private firms on a case-by-case basis.
    Unfortunately, few nation-states on or near the equator are politically stable, prosperous, or entirely friendly to the US; in longitudinal order, they are: São Tomé and Príncipe; Gabon; Congo/Brazzaville; Congo/Kinshasa; Uganda; Kenya; Somalia; Maldives; Indonesia; Nauru; Kiribati; Ecuador; Colombia; Brazil. The US does administer Baker and Jarvis Islands, which are quite near the equator in the mid-Pacific, and are uninhabited. But both are very small — too small to have, for example, sources of fresh water — and have been constituted as National Wildlife Refuges. The largest and most economically diversified equatorial nation is, of course, Brazil, but Brazil-US relations are not at their best at the moment.
    So something like Sea Launch, but larger and more permanent, might be necessary. Platform technology is well in hand; this source notes that construction of oil-production platforms in 500 meters of water is becoming common, and this news release describes an oil-drilling platform that “has a deckload capacity of around 4,000 tons and is rated to drill to 30,000 ft in just over 3,000 ft of water” — I note that the Saturn V moon rocket weighed 3,100 tons. I also found a good overview of deepwater development systems, which explains various types of platforms capable of operation in as much as 7,500′ (2,300 m) of water. An oceanic launch site might even permit the use of nuclear thermal propulsion in a booster rocket. Speaking of which …

  • Develop or scale up high-efficiency propulsion systems, including both nuclear thermal and nuclear electric. A nuclear-thermal booster launched from Earth’s surface could theoretically place three-eighths of its mass in low Earth orbit; compare the aforementioned Saturn V, which put 240,000 pounds out of 6.2 million at launch, about one twenty-sixth of its mass, into orbit — a nuclear Saturn V would have had a 2-million-pound payload!
    Nuclear-electric propulsion efficiencies are even greater, but thrust (and therefore acceleration) is low, so their best use is for transporting cargo from Earth orbit to lunar orbit, Mars orbit, and other destinations.
    Concomitantly, develop long-term cryogenic storage technology, perhaps with nanomaterials, so that the liquid hydrogen used by both nuclear-thermal and the most efficient chemical rockets can be carried through interplanetary distances.

  • Evaluate moving the ISS to a low-inclination orbit, possibly even an equatorial one. In any case, this would limit its accessibility from Russian launch sites. Continue to use Russian, Ukrainian, etc, production facilities where possible, and pay to relocate personnel from Baikonur (Tyuratam) Cosmodrome to the ELS (actually to a relatively nearby port city). The ISS then becomes (as slated in the Bush proposal) a medical testbed for gaining understanding of the effects of lengthy spaceflight on the human body, but one with far more economical access from Earth.

  • Conduct a lunar program largely as a testbed for the Mars program, using the Earth orbit rendezvous (533 kB *.pdf; see page 4) technique to assemble large manned spacecraft and cargo.

  • Nonetheless, strategically select the lunar landing sites. Some places on the Moon are much more valuable than others. The permanently-shadowed regions at the poles which contain trillions of kilograms of lunar ice are the best known, but there are others as well; turning to T.A. Heppenheimer’s Colonies In Space (pp 115-116), we find:
    By computing orbits to be followed by the payloads, it has been found that if they are launched from a suitable point on the lunar surface and aimed at a target, they will hit that target even if the launch velocity is slightly off. That is, the gravity of the earth and moon acts to focus the trajectories so they arrive at the target despite slight errors in velocity. The best such launch site is at 33.1° east longitude, near the craters Censorinus and Maskelyne. Then, the target can be chosen as a point in space 40,000 miles behind the moon, known as the L2 point. A catcher or target located there will stay on station, since there the gravity of the earth and moon is cancelled by the centrifugal force due to the orbital motion of the catcher.
    This idea was used by James P. Hogan in The Two Faces of Tomorrow. Wikipedia has a good explanation of mass drivers. Coincidentally, Maskelyne crater is near the Apollo 11 landing site (see bottom of page).

  • Carry out a “dress rehearsal” at the Sun-Earth L2 point, 1.5 million km from Earth. This would allow the crew module’s various subsystems to be thoroughly tested without committing the crew to an actual trip to Mars (Wikipedia also has a good entry on Lagrangian points). I would expect that unless the trip to Mars could be made relatively short (like the 1½-month trajectory I computed in this incredibly geeky post), part or all of said crew module would rotate to provide artificial gravity and prevent bone loss, etc. A rotating module with a radius of 50 meters would provide 0.5 g at 3 RPM.

  • Build a base on Phobos or Deimos first, only then descend to the planet — and send robots first. This is not an original idea. Besides providing cheap radiation shielding in the form of regolith, the moons are most likely captured asteroids whose bulk composition, in spite of highly desiccated surfaces (134 kB *.pdf), resembles that of carbonaceous chondrites. If so, then Deimos alone could contain over 100 trillion kilograms of water, and therefore over 10 trillion kg of hydrogen, making it an enormous, conveniently located propellant cache.

  • Deploy Martian comsats and GPS. Mars-synchronous orbital altitude is 17,400 km (20,800 km from center of planet), about 11,400 km above Phobos and 2,700 km below Deimos. Three satellites spaced 120° apart would provide continuous communication with Earth and everywhere on Mars below ~78° N/S latitude. Assuming both the transmitting antennae on the satellites and the receiving antennae on Earth are accurately pointed 10-meter-diameter paraboloids, the transmitter power is 1 kW, and the system is using X-band wavelengths (8 GHz) with a bandwidth of 10 MHz, then at a typical Mars-Earth distance of 160 million kilometers, the channel capacity would be over 40 megabits per second (I built a spreadsheet a few years ago, based on a chapter in this book, to do such messy calculations). The actual maximum data rate from (for example) Mars Odyssey is 110 kbits/sec, because its transmitter power is only a few watts and its high-gain antenna is only about a meter in diameter.

  • Land someplace cool. Several places, actually, starting with a high-latitude site for access to subsurface ice for refueling. Charitum Montes would do. Then on to the spectacular Valles Marineris, and be sure to visit Olympus Mons. If the energy requirements of rocketing from one site to another are prohibitive, drive instead. Deploy lots of drone aircraft while you’re at it, and send back plenty of video!

And now for my larger point. Besides the imminence of nanotech making a high-level-of-effort program obsolete within two decades, the above scenario is clearly strewn with arbitrary elements. Even within the constraints of bulk technology, there are a vast number of possible approaches to exploring Mars. But the political process will necessarily close most of them off, and there is simply no reason to believe that even if a manned Mars program somehow survives (at least) four or five Administrations, eight or ten Congresses, and fifteen or twenty Federal budgets, it will have selected and executed the most scientifically productive of the alternatives. Any directly publicly-financed effort will make plenty of idiosyncratic choices, to say nothing of political rewards to powerful legislators. So as silly as my ideas may be, they almost certainly are no worse than what the NASA-Congress-Administration system will (attempt to) produce.
I don’t always agree with Bob Zubrin, but his Mars Prize idea, which simply dangles a $35 billion carrot to be munched by whoever gets there first, is far more attuned to the decentralized nature of a free society than the way we (try to) do things now.


33 thoughts on “How I’d Go To Mars”

  1. Jay, whoa, great post.

    Some queries. Pricetag on this project? Way, way more than W has proposed spending?

    Any speculation on likelihood of useful spinoff technology for military or commercial use? (I suppose you would learn things developing a really good mass driver that would apply to constructing a good rail gun to zap boost phase ICBMs.)

    I’m not very tecchie, so bear with me. Why does “the imminence of nanotech mak[e] a high-level-of-effort program obsolete within two decades”?

  2. Nice start… I’d add Zubrin’s idea to be dropping redundant materials onto Mars long before the mission, including oxygen generators and storage units, and habitat modules.

    When I first saw “Hogan” for a second I thought you were referencing Hoagland. ;-)

  3. Any budget numbers I quote would be wild guesses, and in any case W’s outlays — if I understand the proposal correctly — cover only the first few years of developing a modular manned spacecraft suitable for operations between Earth’s surface and cislunar space, plus some robotic exploration.

    Having said that, I suspect that the use of an equatorial launch platform, Russian/Ukrainian/Kazakh contractors, and (possibly) nuclear propulsion would result in substantial cost savings over a Cape Canaveral-launched, all-American, chemical propulsion combination. The largest cost in most space operations today is the launch vehicle. The Shuttle costs $25,000 per kilogram of payload. At the other extreme, the Dnepr (a converted SS-18 ICBM) could cost as little as $1,000 per kilogram. The Saturn V cost (corrected for inflation) about $19,000 per kg of payload, even though only 13 were ever launched, vs over 100 Shuttles; there is some evidence that a suitably designed large chemical rocket could reduce this to only a few hundred dollars. At $1000/kg, we could put 1000 tonnes in low Earth orbit, and about half that on the lunar surface, for $1 billion.

    So my ROM (rough order of magnitude) estimate for the whole program would be “closer to $10 billion than $100 billion.”

    As “strong nanotechnology” gets closer and closer, the answer to what to do about more and more big projects — like going to Mars, or processing waste from nuclear power plants, or building a bridge across the Strait of Gibraltar — becomes “wait.” With strong nanotech, we will be able to mass-produce booster rockets for essentially the price of the feedstock materials, typically well under $1/kg. A Saturn V-class launch vehicle, but noticeably more efficient and far safer than the original, would cost $~1 million. Payload costs to orbit would be expressed in cents rather than (thousands of) dollars per kilogram. Mars exploration could come within the range of a well-funded university geology department, with lead times measured in months rather than decades.

  4. Definitely, staging consumables, structures, and mechanisms via unmanned flights; I should have admitted up front that I’m nowhere near as familiar with “Mars Direct” as I should be.
    Hoagland? Gawd — I’d almost managed to forget about him. ;)

  5. After reading up on both the PhD and Mars Direct plans, my feeling is the the PhD plan adds cost and complexity while having very little effect on mission safety or success. Radiation shielding can easily be found on Mars as well in the form of “dirt”, and we now know that there is plenty of water, of which much is near the surface. Hopefully we’ll be able to find some really concentrated ice deposits within the next few years. Robots should obviously be deployed to the landing sites before humans land, but it would be much better to have them at the landing site before the humans left earth, rather waiting until the astronauts are already on Phobos. I’m all for a Phobos mission at some point in the exploration program, but if the goal is to explore Mars, go to Mars!

  6. I suspect that the most important period for the planning of this project will always be the period between now and the next election.

  7. The idea for sea-platform-based launch makes me wonder if there are any equatorial offshore oil fields. Would be nice for the platform to have an additional (and lucrative) source of revenue.

  8. I dont understand why should we be waiting for nanotech .. mass production of rockets would be possible right now, if the rockets were designed for mass production methods we have.
    Neither dont i understand the fascination with nuclear thermal rockets … IMO its a well published fact that the _mass or fuel efficiency_ of the rocket has little to do with its launch costs.
    I dont think nuclear propulsion would change that, on the contrary actually. Just for a second, lets imagine a nuclear shuttle .. say with a payload of whopping 100 tons. Given that current STS launch has cost us approximately half a billion per shot, nuclear one could easily be around billion.

  9. “A nuclear-thermal booster launched from Earth’s surface”: my word, the Greens would scream themselves sick over that one.
    (Giggle)
    Go for it!

    Though as an alternative, whatever happened to that laser launch concept I remember reading about some years ago?

    Personally, I’d say don’t wait for nanotech. I’m optimistic enough to hope and think MM nano will arrive, but pessimistic enough to want to avoid relying on it.

    How much extra would it cost to add in some asteroid survey/RV operations? Maybe, if you mainly go the NASA route, that could be contracted out.

    As for launch sites, what about Nauru? Also, how close to the Equator do you need to be? Because if close but not on it will do, how about the US/UK island base at Diego Garcia?

  10. Commenters are proving adept at pointing out the arbitrary and idiosyncratic elements in my plan. In order:
    Going to Phobos/Deimos first allows optimization of separate vehicles for truly interplanetary transport and Mars surface-to-orbit shuttle. Adds to schedule, leaves cost relatively unchanged, improves quality/safety.
    Yes, there are equatorial offshore oil fields. Prograde launch would be over inhabited areas, however, and all nearby nations, Gabon possibly excepted, are impoverished and unstable. Brazil also has an offshore field near Fortaleza, but Brazil claims its entire continental shelf, so a facility there would not be in truly international waters.
    Waiting for nanotech doesn’t make sense yet. I think it will in 10-15 years. American political culture and processes do not make me optimistic about our prospects for getting anyone to Mars, perhaps even back to the Moon, in that timeframe. A propulsion alternative I neglected to mention is simple, pressure-fed boosters; Truax has claimed that these could cost as little as 1% as much as current launchers. I don’t think anything should be inferred from Shuttle launch costs other than that the economics of government space programs are horrendous.
    Laser launch is good for very small payloads only — which is not to say that it has no utility; but large finished assemblies or manned spacecraft probably can’t be orbited this way. Nauru is inhabited, and I don’t know how they’d feel about leaving. Diego Garcia is reasonably close to the equator, but has, uh, other uses at the moment. The nearby Chagos Archipelago was forcibly evacuated by the British and has become the subject of a human-rights dispute.
    Keep those cards and letters coming in, folks!

  11. “I don’t think anything should be inferred from Shuttle launch costs other than that the economics of government space programs are horrendous.”

    Yeah, and nuclear propulsion has a little chance of being anything else but government program. Very politically loaded too.

  12. Going to Phobos/Demios has an added advantage not really mentioned. Assume for a moment that the Mars missions are Mars Semi-Direct (that is the return vehicle is not sent ahead and only a hab and fuel creation is sent ahead and the return vehicle stays in orbit) and the return vehicle lands on Phobos where it the regolith provides radiation protection while the Mars teams do their duty on the planet bellow.

    Followup trips would end up leaving a lot of gear on Phobos that could be used and bring the price down. You create a beachhead that has got to be valuable. Leaving infrastructure behind means we’re less likely to abandon Mars for thirty years as we did the moon.

  13. re: Nauru– wasn’t it just recently hit hard by a hurricane ? So hard that rebuilding will be so costly that the government is considering disbanding and returning to being a dependency on New Zealand again? If so, somebody rich might be able to buy it cheap, and turn it into a spaceport.

  14. Waiting for strong nanotech and real AI would be like continually waiting until NEXT year to buy that new computer… whenever we start getting serious about it will certainly be more expensive than the next year (unless we’re talking about 1972.)

    Plus, as tinfoil-hat as it sounds, it wouldn’t be a bad idea to have self-sufficient outposts before we spread serious nontech capability around this turbulent globe.

  15. “A nuclear-thermal booster launched from Earth’s surface”: my word, the Greens would scream themselves sick over that one.
    (Giggle)
    Go for it!”

    Is there any extraterrestrial source of nuclear fuel nearby? Does the moon have a supply we can take advantage of for interplanetary travel. I’m not expecting any significant amount of nuclear fuel to leave Earth in the foreseeable future…

    “Going to Phobos/Demios has an added advantage not really mentioned. Assume for a moment that the Mars missions are Mars Semi-Direct (that is the return vehicle is not sent ahead and only a hab and fuel creation is sent ahead and the return vehicle stays in orbit) and the return vehicle lands on Phobos where it the regolith provides radiation protection while the Mars teams do their duty on the planet bellow. ”

    Better yet, no return vehicle. Plan on the human population being there to stay. Make sure half of them are female. If we abandon Mars the way we did the moon, they’re effectively independent, and after they’ve built up their infrastructure for a while, they can start running ships to Earth and back. If they think it’s worth the trouble…

  16. Wow, you’re talking about a hell of a lot of people if you want to get through the genetic bottleneck without inbreeding problems. Believe me, I know – I live in Kentucky.

  17. Wanting to be somewhere else when everyone on Earth has easy access to nanotech isn’t tinfoil-hat at all. We can all think of people we’d just as soon not be living within thousands of miles of if they could grow missiles like so many cornstalks. I know I’ll sure want to be able to scram if the occasion arises.
    The Moon probably isn’t a great place to find uranium and thorium (though there are areas of higher concentration, including Mare Fecunditatis (40 kB *.pdf) — the same source suggests that “deposits of uranium cover most of the nearside of the Moon”). A colorful map in this paper (2.2 MB *.pdf) indicates that thorium concentrations vary on a logarithmic scale by about 0.15, that is, around 40%, with the highest beingin Mare Insularum, southeast of Copernicus crater.
    Presumably, type M asteroids are elevated in all heavy metals, including uranium and thorium, and at least two near-Earth metallic asteroids (3554 Amun and 6178 [1986 DA]) are known.
    Paul Davies, writing in the NYTimes, suggested a one-way Mars mission.
    It’s my understanding that the minimum viable population size to avoid inbreeding problems is ~10,000, but with genetic engineering techniques, I would expect this limitation to be more or less removed.

  18. Have you thought about the praccticality of using either an electric rail gun or “super gun” technology for boosting cargo to low earth orbit? Both systems are relatively low tech and essentially provide a first stage where all the hardware remains on the ground. Solid rocket engines would provide the remainder of boost to orbit. Because of extreme acceleration (1000s of g), these systems are useless for manned flight or transport of fragile components, but might be ideal for low cost shipping of fuel, building materials, water, food and other bulky low-tech stuff to low earth orbit.

    To be practical, either system would probably have to be built at high altitude (moutain or alto-plano) to reduce air resistance for the projectile leaving the gun. For equatorial launch that means the Andes. Within CONUS, the closest mountain sites to the equator would be in south New Mexico.

  19. Mars Express has found better evidence for water ice on Mars. That changes things a bit.

    And Spirit has sent two streams of good data back – so today is a good day for space activists.

  20. Regarding the floating launch platform concept Jay, what PM strategies would you recommend to counter the risk of deliberate, malicious, and perhaps suicidal interference ranging from small boats or aircraft(or animals released) in the launch fan, to interdiction its supporting sea lanes and/or blockade its supporting seaports, to terrorist acts employing a wide variety of technologies, to seaborne or air attack by legitimately constituted national military forces? The difficulties NASA has had in the past with securing its launch envelopes, and the Navy’s debacle surrounding their practice range on Viaques island don’t bode well for a distant platform in international waters.

    Perhaps the Boyz can lease NASA their fleet for a suitable charge?

  21. I’m surprised that in all the talk of sealaunch platforms and nanotech, nobody’s mentioned the platform-based space elevator, a possibly nearer-term “weak-nanotech” solution to high launch costs.

    http://www.isr.us/SEHome.asp

    Nanotube composite ribbon with the necessary strength is still in the realm of unobtanium, but the announcement of suitable stuff could come any day. As long as we’re talking multi-decade billion$ programs, makes sense to include it in the idea matrix.

  22. Interestingly enough, much of what you have called for is already in the NASA plans, dispersed in various places. These include: propulsion, lunar base as testbed for Mars (a long-standing idea in the life support community), Comsat/GPS for Mars – a comsat is in the pipeline now and GPS is actively under study, ISRU is baselined in most Mars mission models (particularly for propellant production).

  23. David Rothman correctly suggests that an Earth-based mass-driver could work. Back in the ’80s, I corresponded, met, and spoke with (but of course can’t now remember the name of) a physicist who had worked out quite a few of the details, partly by sneaking into supposedly classified sessions of SDI-related conferences. It’s my understanding that the only good location is Cayambe, and Ecuador is politically problematic. I remain irrationally fond of this idea, however.
    Michael Sargent also correctly notes that there are (my term) topological issues with security for an oceanic platform — this is what, er, sank the Oceania project. I’d manage the risk by either locating it as far from land as possible or including defensive measures in the project — perhaps both. A search at this nifty database finds, for example (warning: big *.jpg’s), Ulu Seamount, which tops out at 1500 meters below the surface, and Winslow Reef, which reaches to -1650 m. Both are just east of the aforementioned Baker Island, about 2,500 km from Hawaii, and are nearly 4,000 km from large land masses (Australia, New Guinea, New Zealand).
    As Patrick notes, I’m a bit behind the curve in ignoring the space elevator. I’ve noticed that mention of it has a catnip-like effect on engineers of my acquaintance. Well, they don’t actually roll around on the floor and make funny noises, but they certainly seem to find it … stimulating. And as “ech” implies, few if any of my ideas are original. You want original thinking, start paying me to blog. ;)

  24. Christmas Island is a likely spot.
    See http://www.apsc2orbit.com/thespaceport/theSpaceport.html

    The problem with Mass Drivers on Earth is that pesky atmosphere. Assuming an evacuated tube to accelerate in, you have the problems of frictional heating, and sheer g-force deceleration when you hit the atmosphere. Even Everest-level atomspheric density is problematic.

    Now on the Moon, or even Mars, they become much more attractive. The Moon especially, lots of free soolar power…

  25. Jay, suggest to your engineer friends that they actually look at the numbers. They might stop laughing. I expect that elevator to be in operation long before a manned mission to Mars could be launched. And the nice thing is, it not only gets you into orbit, it can launch you to many of the planets, as well.

    Speaking of skyhooks, though, have you looked at the possibility of forgetting about landing craft at the Mars end of things? A geosynch skyhook is just barely feasible for earth, but Mars has a substantially shallower gravity well, and a natural satalite quite near Mars synchronous orbit. You don’t even need nanotubes, standard graphite composite, or one of the higher strength polymers would be sufficient. They could be manufactured on site.

    While a skyhool lowered from Demos wouldn’t be perfectly stationary relative to Mars surface, it would be traveling slow enough to use very low performance craft to handle the last few miles of the trip.

  26. If you don’t want this to be another checklist space program (gone to the moon, we don’t need to be back) it needs to generate enough income to be profitable. You can get some government funds but it’s all impractical nonsense if you don’t find a way to make money reasonably soon.

  27. I should have used a different metaphor; what I meant by “catnip-like effect,” etc, is that just about every engineer I know is irresistibly drawn to the space elevator project and wants to work on it! Skyhooks could definitely figure into the project, as well.

    And I have no doubt that a permanent presence in space will be predicated on profitability. The commercial space industry is just beginning to branch out beyond remote sensing and telecommunications; I wish them all the best.

Comments are closed.