How we could get to Mars

NASA rendering of person on spacesuit on Mars


I became a space fan when in the summer before my 10th birthday Voyager 2 flew by Neptune. I was rather saddened by the limited coverage I got from the press in my native Greece and was quite happy when we moved about a year later to California –temporarily for two years- which allowed me to devour all the space and astronomy books in the public libraries. I read books published at different decades but one thing was in common with the more recent books: an unshakable belief in the Space Shuttle to deliver cheap and routine space access. However I did remember the Challenger disaster, so even to my prepubescent mind it was obvious that this was not going to happen. Nowadays I see too many articles in published books, the traditional press and on the web treating SpaceX’s aspirations and possibilities as a given with the certainty of the Shuttle aspirations of the 1970s and 1980s. This a rather cool headed analysis of what is actually taking place in regards to human deep space exploration today and what we are likely to see in terms of developments.

The failed promise of the Space Shuttle

Modern rocketry is a byproduct the two World Wars. Rockets were one of the few military technologies that Germany could freely develop and use under the Treaty of Versailles. Following the Second World War both superpowers sought unstoppable means of delivering nuclear weapons, exploiting Nazi knowhow. Von Braun and the German rocket pioneers did have idealistic aspirations for the technology they produced, though they were very willing to turn a blind eye to the horrors of the Third Reich: V-2 production took place in underground facilities manned by Concentration Camp slave labor. Nuclear tipped ballistic missiles are intended to bring similar horrors to enemy population centers. Superpower competition during the Cold War took many forms; most relevant in our case was the International Geophysical Year when both superpowers promised to launch an artificial satellite. Sputnik was treated with panic in the US because it showed that the USSR possessed technological superiority in rocketry and could thus strike the US with impunity. Much as the fears had been somewhat allayed after the US also launched artificial satellites, Vostok 1 demonstrated that at that point it was the Soviet Union that held the technological advantage. Hence President Kennedy declared and Congress willfully funded a crash program to send a man on the moon, telling NASA that they could waste anything except time. NASA did rise to the challenge, President Kennedy’s deadline was met, but as soon as American superiority had been demonstrated the political will to fund further exploration disappeared.

The Shuttle was born of the desire to provide employment for the armies of engineers rendered redundant by the end of Apollo. A reusable space launcher was to be everything for everyone; in the end it proved a jack of all trades but master of none. It has been hard to create a unified narrative of the Space Transportation System in the way that Tom Wolfe’s The Right Stuff narrates the Mercury program and Andrew Chaikin’s A Man on the Moon the Apollo program. We are after all talking about a 30 year program of 135 flights parts of which are still classified. However over time parts of its story have been told.

KH-9 imagery satellite during integration. Courtesy NRO

The biggest launch customer in the United States is the Department of Defense. When NASA went forward with the Shuttle they reached out asking them for the specifications of its preferred payload so as to select payload size and mass. The preferred payload in the 1970s seems to have been optical reconnaissance film satellites. Film reconnaissance meant that they could operate in space only as long as their film lasted, a maximum of 275 days in the last KH-9 satellite and often 3 months or less. As a result they required continuous replenishment, especially since after the 4 canisters of film had reached the earth, the satellite could no longer send any intelligence back. A shuttle would not only be able to send them up cheaply, it might allow to simply replace the film rather than the entire satellite. In the late 1970s though a new type of reconnaissance satellite appeared that used electro optical means rather than film to capture intelligence. Much as that meant lower resolution or scene width, since CCDs surpassed film for resolution only in the last decade or so, the image could now reach the ground far faster through the use of ground receiving stations and images could be taken as long as the satellite lived. There were many still payloads the Shuttle could launch to space, but even before Columbia reached the launch pad for STS-1 its design payload had evaporated. Still, the Shuttle did launch 15 payloads for DoD, mostly in the 1980s before most payloads were removed from the Shuttle post-Challenger. NASA did reach out from early on to commercial satellite users and did launch several such payloads, but in the end the shuttle never quite reached the one flight per week or so rate that would have been required to be economical.

The causes for this high required rate were both programmatic and economic. Simply put from the start the Space Shuttle was undersold, Nixon run with the cheapest estimate having insufficient margin. When, as typical in aerospace engineering, there were overruns, Congress never quite appropriated all the funding necessary to cover them. NASA sacrificed the last two moon missions, a large part of the space science program and ever larger parts of reusability to get the Shuttle to the launchpad. The most expensive part of the rocket, the engine, was reusable. Unfortunately no one expected that so much refit would be required, especially for the thermal tiles. The Centaur-G which would have allowed heavy payloads to geostationary orbit was never popular with the astronauts since it was to be located in the payload bay, its cancellation following Challenger was met with relief but denied several missions for the shuttle. The Hubble was refit several times, but in the end considering the cost of a single flight it would have been cheaper to launch a new telescope every time. In the end in 1986 the pre-Challenger plan was to fly the Shuttle 15 times in that year so as to help lower the unit cost. The push to meet the schedule was what caused the disaster with contractor safety concerns being overruled by NASA management. But even if all missions had taken place NASA would likely have run out of astronauts sufficiently trained in the mission by the end of 1986. The policy changes after Challenger disaster seriously reduced the number of potential payloads available. What Columbia showed was that humans were not really necessary to a launch vehicle and keeping them abroad was at times rather detrimental for the cost of the payload which had to become human rated at great cost. There was one type of payload that already had to be man rated, the International Space Station (ISS) modules. Yet launching with expendable rockets would have been cheaper than on the Shuttle Bay, assuming assembly would not require human labor. The shuttle was a great idea in the 1960s, a great technological achievement of the 1970s, but its operations proved problematic. A hypothetical Shuttle 2 post Columbia, made from a clean sheet would have been lighter, more capable, safer but in the end still a jack of all trades, master of none. This is what led to cancellation.

Martian Mission Architecture

Apollo launched in a single stack to the Moon. Three astronauts spent two weeks in a capsule the size of a car’s cabin. Orbital mechanics dictates that it is very difficult to have a mission that lasts less than 15 months. You simply cannot keep people in a space the size of the Apollo Command Module for over a year and expect them to emerge psychologically sane. In any case the food alone for such a long trip will require more space than what was inside the Apollo CM. Mars requires a different architecture than the Moon.

The first NASA plans for a mission to Mars date to the mid-1960s when Von Braun realized that post Apollo Marshall Space Flight Center would either need a new mission or would have to do layoffs. Over the years many organizations have produced many plans, best outlined in David Portree’s Humans to Mars: Fifty years of mission planning with a consensus architecture eventually emerging:

1. Rockets launch a human Transit Habitat with a propulsion module to Earth orbit

2. Propulsion module launches habitat from Earth orbit to Trans Mars Injection

3. Transit Habitat enters Mars orbit. Astronauts use descent vehicle to land on Mars. The descent vehicle could travel with the habitat or be already waiting for them in Mars orbit.

4. Landing takes place near prepositioned supply vehicle-surface habitat which through In Situ Resource Utilization (ISRU) will have produced rocket fuel and oxidizer for return trip from Mars surface to orbit. Astronauts spend between 30 days in conjunction and 1 (earth) year in opposition missions on the surface of Mars exploring

5. Astronauts launch to Mars orbit and back to the transit habitat. Habitat leaves for Earth orbit

6. Astronauts use return capsule to go from the habitat to the Earth surface

Technology improvements necessary for Mars

The emergence of consensus on architecture in the mid-1990s meant the identification of the knowledge gaps necessary to launch the mission. Below is a list of known issues so far to the best of my judgment.

The transfer habitat and propulsion module will most likely weigh far more than any rocket either in operation or under development is capable of launching. The only exception would be Elon Musk’s Interplanetary Transfer System. The lightest version so far of a transfer stack would be an inflatable habitat, such as an advanced version of Bigelow’s B330 module along with an engine and fuel that might marginally fit in the heaviest of rockets. Much as this configuration on a Falcon Heavy has been proposed several times by enthusiasts, it is not being actively developed by anyone. Bigelow Aerospace has signed a contract with the United Launch Alliance to develop B330 fairings for launch on an Atlas 5 and does have a contract for Deep Space Habitat development, but not for the Falcon Heavy. No other inflatable module is under development by any other company or space agency, at least as far as I know.

Chemical rockets have high thrust but low efficiency. They are necessary for pulling out of earth’s gravity well but if used for a Mars mission the architecture becomes too heavy. This has been known since very early which is why already in the 1960’s there was a NASA program for a Nuclear Thermal Rocket. More recently there has been funding for development of high power Solar Electric Propulsion. While there have been several science missions and geostationary satellites with ion engines of various types, no high power non chemical propulsion system has been demonstrated yet, either on the ground or in space.

Moving outside the Earth’s magnetosphere means exposure to galactic cosmic rays. Much as research on the ISS has shown that radiation is more of an issue of Informed Consent rather than a showstopper, mitigation is necessary. Another human factors issue is the prolonged effect of microgravity on human physiology. The Russians did have several missions on Mir to understand the factors involved and more recently there has been the American and Russian Year in Space mission on the ISS, which will be followed by similar missions. Less visible that they physical factors but equally important are the psychological factors associated with locking people for months at a time in a small enclosed space.

The ISS has demonstrated the ability to maintain life support and other equipment functioning for years, capable of recycling gases and water and treating waste. On the other hand the ISS receives 8 or so resupply missions on a typical year, Mars bound spacecrafts will not be able to resupply until they reach prepositioned material no earlier than Mars orbit. The ISS is to the Transfer Habitat what Gemini was to Apollo, proving ground for operations, containing several subsystems but less than the prototype necessary for the trip.

Much as everyone has cheered when robotic spacecraft have landed on Mars, Entry Descent and Landing (EDL) for a human sized 20 t habitat needs to be demonstrated. Curiosity weighs a little under 800 kgs, Mars 2020 is a little over 1 ton but the Skycrane system is not scalable for a two story apartment-sized human habitat. The Red Dragon intends to demonstrate supersonic retropropulsion under Martian conditions whenever it gets to Mars. NASA’s Low Density Supersonic Decelerator (LDSD) failed in its most recent test due to issues with the parachute which ripped apart despite ground modeling showing that it should not. LDSD in any case is capable of landing payloads heavier than Mars 2020 but lighter than a human sized 20 tons

ISRU has become critical in any architecture due to the great mass penalty that carrying fuel to and from earth would cause. It has yet to be demonstrated under Martian conditions, though there is an experiment, MOXIE, on the Mars 2020 rover. Another thing that would need demonstration would be how long a descent vehicle can remain functional in Mars orbit. The Soyuz-TMAM is capable of lasting 6 months in space without use in the mean time between missions and Federatsia is intended to last for a year. Orbital mechanic might require that the descent vehicle be launched in the previous synod, 26 months in advance, of the landing mission, if the astronauts do not take it with them.

In the end there are also the unknown unknowns, factors we are unaware at this time that might prove a showstopper in the future. We cannot know them before we actually start building and testing spacecraft. Alas, we are nowhere near this point so we need to guess with our current knowledge issues of the future.

Comparison of the status of active Mars mission actors

There are several players, in cooperation and competition, which propose missions to Mars. Below is an overview of proposed actions and how likely they are to take us there.


Per its detractors, NASA’s plan is not more concrete than this graphic

Despite the criticism that NASA regularly receives, especially from libertarians, the agency has the most complete plan to go to Mars. Unfortunately it is the most complete plan possible under the current budget realities as dictated by American politics, not the best possible plan under current technology. Still NASA has moved further along than any other player in the business, though considering that the Trump administration seems to refocus towards the Moon it remains to be seen if and when the plan will reach fruition. After all the ISS had been in serious planning before even the launch of Skylab, as its follow up, yet systematic assembly begun only in 2000.

The current plan hinges on the successful use of the Space Launch System (SLS). This Congress mandated mega rocket, made mostly from Shuttle derived parts sourced from politically important districts, will launch the transfer habitat in four pieces to be assembled in orbit. The habitat will reach Mars and the first time around will explore Phobos and Deimos but will not land on the Martian surface. The second time a surface mission will follow lasting 30 days, and the third time closer to an earth year. Up to the first surface mission the cost is about as much as 22 Apollo missions. At the end of the project not much infrastructure will be left on the Martian surface and the architecture does not lead to a sustained colonization of Mars. For that matter the Transfer Habitat is not currently intended to be reusable. More importantly is that simply there isn’t much of a political consensus to fund and go through with the project. On the other hand traditional NASA contractors like Lockheed Martin and Boeing have proposed more detailed plans base on the NASA SLS/Orion framework.

The US share of the ISS currently costs $5 billion/year to operate, including developing Commercial Crew Transportation. After the ISS is abandoned it will not lead to complete freeing up of funding for Mars, it will likely be replaced by another commercial Low Earth Orbit space station, cheaper to operate but not free to NASA. For that matter abandoning the Space Shuttle did not free funding for SLS/Orion, the money lost in that transition was fully restored only in 2016, 5 years after the Shuttle was abandoned. The next step in the current plan is an international plan for a Lunar Orbital space station. For NASA it will be proving ground for the Transfer Habitat, the other partners though would like to use it for lunar exploration. It is very possible that between the LEO Station, Lunar Orbital station and operating the SLS more than once a year for resupply NASA will simply not have any more money to send people to Mars. In any case no plan expects any serious Mars specific human space flight action to be undertaken for the next 10 years, the current plan talks about test flights after the lunar station is assembled in the late 2020s. Even if the Lunar Orbital station is built in time and on money, something seriously debatable, we can guess even less what NASA’s financial condition will be in 10 years. Still NASA’s plans, ever more detailed with every formulation, will prove what other plans will be judged against.


SpaceX’s Mars Architecture. Copyright SpaceX

Elon Musk founded Space Exploration Technology so as to colonize Mars. So far they have repeatedly proven to have the ambition and have delivered a series of ever more impressive feats, but getting to Mars is still mostly a plan on paper. As Elon Musk unveiled in Guadalajara in 2016 his plan to colonize Mars depends on the Interplanetary Transport System (ITS). The ITS launch vehicle will be a monster rocket capable of launching 550 tons in expendable mode to Low Earth Orbit. Its payload will be the Interplanetary Spaceship, capable of sending some 100 people to Mars in one go. It will be launched by the ITS launch vehicle, refueled in LEO by the ITS tanker, land on Mars with the colonists, get refueled from ISRU produced propellants and return to Earth for more colonists. In terms of ambition it is a return to the 1960s, the first missions carrying mostly cargo will hold 12 people to set up the colony as NASA’s plans of the time did, when NASA’s current mission plan is for 4 people per mission. Somehow the whole plan was intended to bear flight prototypes in 2022 and each Interplanetary Spacecraft will cost less than an A380.

SpaceX has most definitely proven the ability to perform what was though impractical to impossible in terms of launch technology. However it is one thing to be the 20th or so company to build a launcher, albeit a very innovative launcher capable of reuse, and another thing to be the first to send people to Mars. The ITS booster is to be powered by 42 Raptor Engines. Raptor is under development and has an Air Force contract. Its funding purpose is to be an upper stage for Falcon 9/Falcon Heavy which currently suffers from an insufficiently energetic upper stage. However the last rocket to have that many engines was the failed Soviet launcher N-1, which exploded spectacularly all 4 times it was tried. The most people in space at any time has been 13, when the Space Shuttle Discovery visited the ISS during STS-119. No one has demonstrated what is necessary to have 100 people in space at any one time. Curiosity, as mentioned earlier, is the heaviest object landed on Mars; if NASA’s 20 tons are hard to land, ITS’ 300 ton surface landing is an even bigger challenge. ISRU is yet to be demonstrated. To reduce risk SpaceX does have precursor missions on the books. The first is the Red Dragon, originally planned for 2018 but now delayed for 2020, which will be a modified Dragon launched to Mars to demonstrate supersonic retropropulsion and deliver yet to be announced experiments –perhaps related to ISRU- on the surface of Mars. A second Red Dragon will follow on the next synod and then the plan called for unmanned demonstration of the whole ITS. Since Red Dragon 1 slipped, the demonstration has also slipped to 2024.

Another potential architecture that has been proposed by enthusiasts and has proven to have a long life on the internet is a proposal to replace the SLS with the Falcon Heavy. This is not as far fetched as it may sound because the payloads that the SLS will carry are still in the conceptual stage. Instead of flying 4 SLS block Is to assemble the transfer habitat, fly 6 Falcon Heavies. For that matter instead of having a megarocket flying large pieces, have smaller pieces flown and assembled on smaller more economical rocket with lower unit cost. Our main experience with space assembly comes from space stations, especially Mir and ISS. Both took several years to assemble with a pause in-between, caused in Mir’s case by the collapse of the USSR and in ISS’ case by the Columbia disaster. The current blueprint for the Mars Transfer Habitat is based on the heavy SLS block IIb, capable of launching twice what a Falcon Heavy can send. The lowest mass SLS requirement for the Phobos mission requires 8 block II launches for Transfer habitat plus two for Orion and fuel would mean 16 Falcon Heavies. Assuming 97% reliability, which has been demonstrated by Soyuz and Ariane V but not by Falcon 9 (or the SLS), at 16 there is a 1-(0.97)¹⁶ = 38.6% chance one of the flights will fail. Mathematics states that for a 50% failure chance at 16 flights, assuming the chance of failure in one flight is independent of the chance in another flight, required reliability is 95,7%. Actual reliability of the Falcon 9 so far is 30/33 = 90.9% full success plus 2 complete failures and 1 partial failure. Fans of a Falcon Heavy replacing SLS approach propose that we could limit the number of launches through in space propellant depots and asteroid mining to refill the depots. That requires significant development and has a very low Technological Readiness Level. In any case SpaceX itself is not developing a Falcon 9/Falcon Heavy base architecture, they prefer the ITS.If a Falcon Heavy architecture was to be pursued its sponsors would need to completely retool the current plan that passes through a Lunar Orbital station optimized for SLS and has reached a high level of maturity towards smaller pieces.

Even if the technical challenges were not daunting, there are also major financial challenges. Simply put even if SpaceX captures 60% of the $5.5 billion commercial launch market, has a profit margin of 10% which it reinvests to the Mars plan, $330 million a year is simply insufficient. Thus the SpaceX plan is to create a communication satellite constellation which by 2020 will give it a profit of over $1 billion a year despite competing several other communication constellations. For comparison since 2005 NASA has spent some $24 billion on Ares, SLS and Orion which are far less ambitious projects and still needs to spend a few billion dollars more in development until EM-2s debut of the full SLS Orion stack. There is good reason to believe that SpaceX can spend less to develop the same capabilities as NASA, but not a small fraction of what NASA has spent to dar. The whole Journey to Mars until the first landing is estimated to cost hundreds of billions of dollars. Let’s assume that by a minor miracle SpaceX only needs $50 billion for its very ambitious infrastructure. There is no funding source available for that kind of money. Even if NASA gives it all the SLS/Orion money, as several screamers on the internet demand, something unlikely to happen since SpaceX made the mistake of building its rocket factory in California as opposed to Alabama, on top of the whatever it gains from the commercial arm, it would take almost a decade to pile up the money. If SpaceX had a less ambitious project, something like what its internet cheerleaders are proposing it could have been realistic. In its current form ITS is a fantasy.


Artist rendition of the Soviet TMK-MAVR spacecraft in Venus orbit. Origin: Wikipedia

The Russian Federation would like to send cosmonauts to Mars. As far as I can tell it’s not happening anytime soon. Sending someone to Mars is a megaproject and Roskosmos has several megaprojects already running. By that I refer to the Vostochny spaceport, intended to replace the otherwise fully functional Baikonur that is now located in Kazakhstan, the Angara rocket that will replace the half Ukrainian Proton Rocket and the Federatsia crew capsule to replace the Soyuz. With the exception of Federatsia which is intended to add new capacity, the other megaprojects are belated responses to the collapse of the USSR. So far, like the other space powers, Russia is more interested in a return to the moon. Still Russia is capable of eventually producing a multicore version of Angara of similar lift capacity with the SLS or resurrecting the Energia rocket, they have the best know how on space station modules which can be adapted into transfer habitats and Federatsia is intended to survive Mars reentry, or at least can be uprated to do so. Russia has the capacity to send cosmonauts to Mars, if it had the will and the funding they could do it. Their current plan is to participate in the Lunar Orbital station, and for that matter Anatoly Zak’s site, one the authoritative source on Russian space, is the best public source for the Lunar Orbital station concept.


Artist’s concept of the Shenzhou 11 (left) and Tiangong 2 (right) spacecraft docking in orbit. Credit: CCTV

China is only the third county to achieve autonomous human spaceflight. Its ambitions so far have been quite limited, a space station of the same class as early Salyut stations, robotic lunar missions and perhaps a lunar landing in 15 years. The Mars plan so far has been limited to robotic probes. On the human side China talks of a better space station with a Hubble like space telescope nearby, no taikonauts on Mars plan has been released or even talked about. Their Mars plans are likely where NASA and the USSR was in the 1960s, there is likely a plan in the formulation somewhere, but its implementation is a very low priority.


Artist impressions of ESA’s Aurora Exploration Program. Copyright ESA

The European Space Agency does not have an autonomous Mars program per se. Rather they are willing to share part of the burden of Mars spaceflight in exchange for an astronaut or cosmonaut seat. So far they are funding the service module of Orion and have promised a module for the Lunar Orbital station. They could in theory fund the environmental system of ITS, or contribute to a Russian or Chinese architecture in exchange for seats, if a realistic proposal was presented. So far it does not seem anyone has reached out to them for that, or presented a realistic plan.


It is hard to calculate what has been spent since the start of the Space Age to explore Mars, often due to classification issues. If the Lunar Orbital station is used as a waypoint for lunar surface exploration, would its cost count towards Mars or Lunar exploration? On the robotic side all countries have spent in the order of US$35 billion on Mars probes. What is certain is that several times that figure will be required to send people to Mars. I am quite enthusiastic about a return to the Moon, though I realize that it might just lead to a decades long freeze on Martian exploration on the NASA side or even complete abandonment of humans to Mars plan. Private efforts into funding space plans have proven theoretical so far; beyond telecommunication satellites and remote sensing satellites there simply isn’t a commercial case to explore space, and if anything the Google Lunar X Prize proves it is very hard to fund private space exploration. We have an actor with the knowhow to get to Mars that lacks funding and another with the ambition that also lacks funding. SpaceX might win this race by default, but its current plans should not be taken as infallible or particularly realistic.

I am a space enthusiast, not someone working in the sector. I have not received any money from anyone for my arguments. My opinions are my own, they do not reflect those of my employer the City of Fresno.