EDIT: I am not asking if you think finding water is necessary for a return flight. I'm asking for the scientific evidence that there is mineable water on Mars that could be used for this purpose. Please direct your answers to the question as asked.

If one is inclined to write an answer about the necessity of finding mineable water on Mars for this purpose, one can always ask it as a new and separate question, and answer there.


  1. Nature Geoscience Granular flows at recurring slope lineae on Mars indicate a limited role for liquid water
  2. ibid. Supplementary data
  3. NASA JPL Recurring Martian Streaks: Flowing Sand, Not Water?
  4. Space.com Weird Dark Streaks on Mars May Not Be Flowing Water After All (a relatively concise summary)
  5. Wikipedia Seasonal flows on warm Martian slopes

It seems that one significant body of evidence for accessible subsurface water on Mars that could be potentially used along with atmospheric CO2 to produce methalox propellants (CH4 and LOX) is substantially called into question.

THE QUESTION: What is the (remaining) body of scientific evidence of potential, mineable sources of water for return trip propellant production on Mars?

As just one example of a plan to use water on Mars to produce methane, see Elon Musk's article Making Humans a Multi-Planetary Species or the video of the 29-Sep-2017 talk at the International Astronautical Congress (IAC) in Adelaide, Australia, starting at 33:20.

From Spaceflight Now's SpaceX’s Elon Musk announces vision for colonizing Mars:

The moon is not an good option because it lacks an atmosphere, has weaker gravity than Mars, and lacks resources like large reservoirs of frozen ice and carbon dioxide that could be converted into water, air and rocket propellant on the red planet.

below: Warm Season Flows on Slope in Newton Crater (animated), from here.

enter image description here

  • $\begingroup$ Out of curiosity, where did you see people planning to gather water from Mars to create fuel? $\endgroup$
    – Phiteros
    Commented Nov 21, 2017 at 7:22
  • $\begingroup$ @Phiteros I've added links to Spaceflight Now and Elon Musk for you. I didn't originally because I wanted to keep the focus of the discussion on the nature of any scientific evidence of available/mineable water, rather than start a discussion of Musk's current plan. $\endgroup$
    – uhoh
    Commented Nov 21, 2017 at 9:08
  • $\begingroup$ See also the new Phys.org article Martian landslides caused by underground salts and melting ice? which links to the Science Advances open access article Martian subsurface cryosalt expansion and collapse as trigger for landslides View ORCID Profile $\endgroup$
    – uhoh
    Commented Feb 4, 2021 at 0:36
  • $\begingroup$ Whats the evidence for water on Mars? I do not understand the question. Surely the presence of a several-meter thick water ice sheet on the north pole counts for something? From radar data the total volume of water ice in the north pole is estimated at 820 000 cubic kilometers. There's also a lot of water ice on the south pole, but that's harder to estimate as it is mixed up with much CO2 ice also. $\endgroup$ Commented Mar 3, 2021 at 1:04
  • $\begingroup$ @PcMan clearly you do understand the question, as have several other users. Have you looked through the answers yet? I've written two of them myself. The goal of Stack Exchange is the generation of good answers to on-topic questions so that future readers can benefit from those answers. There's some good ones here, so mission accomplished! In Stack Exchange you don't have to not know the answer to post a question. $\endgroup$
    – uhoh
    Commented Mar 3, 2021 at 1:19

6 Answers 6


There is an interesting podcast that I listen to - We Martians. Last Nov they had an episode that touches heavily on this. The episode is here: http://www.wemartians.com/home/015 and it goes into far more detail than I can, but here's a brief summary:

The SHARAD (SHAllow RADar) instrument on the Mars Reconnaissance Orbiter used ground penetrating radar on the area of Mars known as Utopia Planitia. The radar results, while not definitive, strongly suggest "a mixture of ice, air, and dust" in the area. "SHARAD detection and characterization of subsurface water ice deposits in Utopia Planitia, Mars".

This information was presented in context with info from the Phoenix lander. The lander was equipped with a robot arm which dug in the soil. This article describes it further: "The Dirt on Mars Lander Soil Findings"

Using its robotic arm, Phoenix dug into the Martian surface to see if it could reach the ice below.

In one trench, dubbed "Dodo-Goldilocks," the lander exposed what mission scientists described as "bright material" about 4 to 5 centimeters below the surface (a similar patch was seen below the lander itself, likely exposed by the spacecraft's landing thrusters). Over the next two months, the team watched as Phoenix's cameras showed that the material was sublimating away, which would be expected of water ice exposed to the Martian atmosphere.

The tendency of soil samples scooped up by the lander to clump together made it difficult to get the samples into Phoenix's onboard instruments, but after several attempts, a sample was coaxed in and the lander's detectors confirmed that there was indeed water ice hiding under the regolith. The confirmation was originally announced on July 31.

Interestingly, the ice seems to occur at different depths under the surface depending on the terrain. The plains where Phoenix landed feature polygonal mounds surrounded by troughs that result from the seasonal expansion and contraction of the ice underneath the surface, which creates cracks and crevices.

And from this related article: "Water Ice on Mars Confirmed"

NASA's Phoenix Mars Lander has confirmed the existence of water ice on Mars.

Mission scientists celebrated the news after a sample of the ice was finally delivered to one of the lander's instruments.

later in the article:

"I'm very happy to announce that we've gotten an ice sample," said the University of Arizona's William Boynton, co-investigator for Phoenix's Thermal and Evolved-Gas Analyzer (TEGA), which heats up samples and analyzes the vapors they give off to determine their composition.

"We have water," Boynton added. "We've seen evidence for this water ice before in observations by the Mars Odyssey orbiter and in disappearing chunks observed by Phoenix last month, but this is the first time Martian water has been touched and tasted."

The news that ice had fallen into TEGA came on Thursday morning, surprising scientists who had run into problems delivering a sample of the icy dirt because of its unexpected stickiness.

Opinion: I'm not sure any of this is 100% definitive proof of there being sufficient mineable water for producing Methalox, but it does strongly suggest that at least in this part of the planet there is significant water just below the surface.

Future missions will study this issue as well.

The Trace Gas Orbiter may provide more information once it's fully online - it has the ability to detect neutrons4 from sub-surface hydrogen (which may or may not be water) and trace H2O in the atmosphere5.

The ExoMars rover and NASA's Rover 2020 will both carry instruments that can contribute to this.

ExoMars Rover, specifically, will be carrying the WISDOM ground-penetrating radar which should be able to see subsurface ice6, and it's core drill is equipped with Ma-MISS, an infrared spectrometer that should be able to determine the presence of water or ice in the drill site7.

NASA's Rover 2020 will also carry the RIMFAX ground penetrating radar8, as well as other instruments for chemical analysis - I am not sure if any of them are specifically aimed at ice detection. Of related interest is the MOXIE experiment, which will attempt to extract O2 from the martian atmosphere.9

  • $\begingroup$ Thanks for your answer, a quick look shows that you've taken the question seriously and taken a lot of time to explain the scientific evidence for mineable water useful for methalox production. I'll check out your links later today. $\endgroup$
    – uhoh
    Commented Nov 23, 2017 at 0:15
  • $\begingroup$ While it's been closer to a month than "later today", I finally got to it. I've added a bit more Phoenix and SHARAD information in this supplemental answer. Thanks again! $\endgroup$
    – uhoh
    Commented Dec 20, 2017 at 4:16

Some of the SHARAD evidence discussed in this answer is also described in Phys.org's Mars ice deposit holds as much water as Lake Superior. This image shows evidence of subsurface ice at Utopia Planitia at about 47° N latitude.

enter image description here

For comparison, here is some SHARAD data for Mars' northern polar ice cap, from here. Note that CO2 ice and H2O ice have very different dielectric properties and can be distinguished using SHARAD's chirped radar.

enter image description here

This image shows a cross-section of a portion of the north polar ice cap of Mars, derived from data acquired by the Mars Reconnaissance Orbiter's Shallow Radar (SHARAD), one of six instruments on the spacecraft.

The data depict the region's internal ice structure, with annotations describing different layers. The ice depicted in this graphic is approximately 2 kilometers (1.2 miles) thick and 250 kilometers (155 miles) across. White lines show reflection of the radar signal back to the spacecraft. Each line represents a place where a layer sits on top of another. Scientists study how thick the pancake-like layers are, where they bulge and how they tilt up or down to understand what the surface of the ice sheet was like in the past as each new layer was deposited.

As far as the direct evidence of water (as ice) from Phoenix is concerned, I've found non-paywalled copies of the following papers:

  1. Arvidson, R. E. et al. (2009): Results from the Mars Phoenix Lander Robotic Arm experiment
  2. Smith, P. T. et al. (2009): H2O at the Phoenix Landing Site
  3. Supporting Online Material for Smith et al.

The TEGA mass spectroscopy data for mass = 18 AMU shown in Figure S6 of the Supporting Online Material of Smith at al. is a little complicated to interpret/explain, but the melting of ice shown in TEGA's differential scanning calorimetry data in Figure S9 of the data is pretty easy to grasp. Once the oven's ramping temperature reached -2°C the sample became "endothermic" and absorbed more heating power for a given change in temperature.

The area under the curve is 0.35 Joules, (or 0.35/4.2 = 0.08 calories). If it were pure ice that value would correspond to only about one milligram of water based on a heat of fusion of 334 J/g.

Fig. S9. TEGA finds ice and aqueous minerals. (A) Thermal analysis for “Wicked Witch” subsurface sample, sublimation lag in the Snow White trench. The endothermic peak with an onset of -2ºC indicates the melting of water ice.

enter image description here


I don't think liquid water is needed. My understanding of both NASA's and SpaceX's plan for ISRU is to use the Sabatier process to create the fuel from the Martian atmosphere. They'd bring H2 feed-stock, and combine it with CO2 from the atmosphere in a catalytic reactor to create methane and water. The water would then be electrolyzed into O2 (for oxidized) and H2 (to feed back into the reactor).

More here: https://en.m.wikipedia.org/wiki/Sabatier_reaction#Manufacturing_propellant_on_Mars

Edit to add:

The "bring H2 with you" approach was an element of the "Mars Direct" proposal from Zubrin, laid out in his 1996 book "The Case for Mars". This is summarized in the Wikipedia page here: https://en.wikipedia.org/wiki/Mars_Direct#First_launch

The first flight of the Ares rocket (not to be confused with the similarly named rocket of the now defunct Constellation program) would take an unmanned Earth Return Vehicle to Mars after a 6-month cruise phase, with a supply of hydrogen, a chemical plant and a small nuclear reactor. Once there, a series of chemical reactions (the Sabatier reaction coupled with electrolysis) would be used to combine a small amount of hydrogen (8 tons) carried by the Earth Return Vehicle with the carbon dioxide of the Martian atmosphere to create up to 112 tonnes of methane and oxygen. This relatively simple chemical-engineering procedure was used regularly in the 19th and 20th centuries,[8] and would ensure that only 7% of the return propellant would need to be carried to the surface of Mars.

Here's a paper form 2001 discussing it in the context of a planned-but-cancelled NASA mission: "Sizing of a Combined Sabatier Reaction and Water Electrolysis Plant for Use in In-Situ Resource Utilization on Mars"

The carbon dioxide used in the reaction will be obtained from the Martian atmosphere, but the hydrogen used will be imported from Earth.

However, it seems that this approach is now outdated, as it was formulated before we knew much about the presence of water on Mars.

More recent approaches seem to be to use Martian water. For example, there's this paper from 2017 specifically addressing the SpaceX program: "Proposed ITS Pressurized Cargo Modules To Initiate a Chemical Industry on Mars" p. 13. This approach assumes a small amount of H2 brought from Earth, followed by more H2 extracted from Martian water, as well as that reclaimed from the Sabatier process itself.

The H2 input may initially be brought from Earth to start up the process, but after start-up the H2 will be derived from water electrolysis...

The input H2O will derive from a water mining operation. Water is known to be abundant on Mars, but unlike Earth, it occurs in solid state rather than liquid. Water mining for underground ice would proceed by employing soil-moving machines analogous to strip mining equipment. Alternatively, wells may be drilled to reach buried glacial ice, heat and pressurize the water to liquid state, and pump the liquid into a storage module.

Another interesting approach would skip water altogether and generate O2 and H2 from hematite, which is abundant on the surface: "On the in situ production of oxygen and hydrogen from martian hematite deposits via a two-step thermochemical CO2/H2O splitting process."

Now, opinion time: I suspect that the very earliest missions will still use the "BYOH2" approach. It requires only 7% of the material by mass, and is way simpler, mechanically and technologically, than mining ice or collecting hematite. But once we have a longer-term presence (manned or unmanned), there will be a shift to other approaches.

  • 1
    $\begingroup$ Can you add a link supporting both NASA's and SpaceX's plan to bring "H2 feedstock" all the way from Earth? Large quantities of hydrogen is particularly difficult to transport and store for long periods of time as a liquid, or as a highly pressurized gas. At least in the case of SpaceX I had thought the idea was to round up hydrogen from local water. Thanks! $\endgroup$
    – uhoh
    Commented Nov 21, 2017 at 4:44
  • 1
    $\begingroup$ The water output of the reaction is not enough to provide all the necessary hydrogen. Additional water or hydrogen is needed to continue the production of methane. An initial feedstock of 100 hydrogen molecules is enough for 50 molecules of methane, but not more without an additional source of hydrogen. $\endgroup$
    – Uwe
    Commented Nov 21, 2017 at 11:17
  • $\begingroup$ The plan to bring feedstock was from a podcast, so hard to link. I'll dig around for it. $\endgroup$
    – jgalak
    Commented Nov 21, 2017 at 13:24
  • $\begingroup$ Bringing H2 feedstock is something Zubrin's mentioned as part of Mars Direct, but SpaceX's plans haven't mentioned it at all. $\endgroup$
    – DylanSp
    Commented Nov 21, 2017 at 14:23
  • $\begingroup$ Why was this answer downvoted? $\endgroup$
    – mike
    Commented Nov 21, 2017 at 15:08

The strongest evidence would probably be that the Phoenix lander actually dug some up, which suggests that there is enough water present to be useful for colonization.

However, jgalak’s answer is also correct. Mining local ice is important for long term plans to build an independent colony, but for early missions, a pump and a hydrogen tank is much lighter and less complex than mining equipment.

Note that hydrogen is only a small part of the fuel / oxidizer mix by mass. The reaction CH4 + 2O2 => 2H2O + CO2 uses 4 parts hydrogen to 14 carbon and 64 oxygen. Even if you bring the hydrogen from Earth, using local carbon and oxygen reduces imported fuel mass by over 95 percent.

If storage is a serious problem, you can also bring water from earth for a more modest 56 percent reduction, though you can accept quite a lot of hydrogen leakage before that becomes the better option.

  • $\begingroup$ Please support your conjecture with quantitative, reliable, factual sources. Your proposed BYOH2 (Bring Your Own H2) or worse, BYOH2O is currently indistinguishable from an opinion. Also please add a link showing to what extent Phoenix actually identified H2O. Thanks! $\endgroup$
    – uhoh
    Commented Nov 21, 2017 at 11:32
  • 1
    $\begingroup$ @uhoh being a little harsh today aren’t you? Actual ice dug up isn’t good enough because it is small scale in one location, but it seems you also don’t accept orbital evidence of distribution and quantity because it’s “indirect” $\endgroup$ Commented Nov 21, 2017 at 19:36
  • $\begingroup$ I'm asking for the evidence. Is it a spectrum, or an image, or a time series, or a combination of all of them? What is it about the data that says that it's water? I'm not asking if it's there or not, I'm asking for the nature of the scientific evidence that it's there. I'm not sure and likely many future readers are not sure what "orbital evidence" really means. Enlighten us! $\endgroup$
    – uhoh
    Commented Nov 21, 2017 at 19:41
  • 1
    $\begingroup$ @uhoh You seem to have some misconceptions about the meaning of “evidence”. All the answers take it as meaning “why do we think there is minable water on mars”, but you seem to mean something more like “why does sublimation rate identify water” or “why does hydrogen imply the presence of water”, which is both not what you actually asked and too broad to be answerable. $\endgroup$ Commented Nov 21, 2017 at 19:56
  • $\begingroup$ Title: "What's the scientific evidence of...", body of question; "What is the (remaining) body of scientific evidence of..." I think what happened is that the first answer didn't address the question, and the following two users took their interpretation from previous answers. I'm asking for the scientific evidence itself, not for an interpretation, evaluation, or opinion based on it. An answer addressing my question as asked would describe the scientific evidence of mineable water for methane production, and support the descriptions with verifiable scientific references. $\endgroup$
    – uhoh
    Commented Nov 22, 2017 at 3:02

There are two possibilities. The first, proposed by Robert Zubrin in the Case for Mars, is as has been suggested, bring a feedstock of Hydrogen, and use it to make the fuel. Hydrogen is by far the lightest part of a methalox system, and could easily be included.

The second is using liquid water to make the hydrogen. There's a large number of sources that support that there is liquid water near the surface of Mars. The following image shows a rough distribution of the water on Mars near the surface, based on neutron flux data. The poles have a much higher portion that is water ice. These have been partially confirmed by the landers on the surface. One of the key objective of any lander is to provide ground truth to help verify orbital observations.

Bottom line, we could either bring hydrogen (Likely for the first missions), or look for water ice to use, but either will likely get us what we need to make rocket fuel to get home.

enter image description here

  • $\begingroup$ The neutron data is interpreted as evidence of protons - hydrogen atoms. It might be water, it might not be. The hydrogen might be easily extracted for methane production, or it might be very difficult to extract. I don't believe that the neutron data can address if the hydrogen is in the form of H2O i.e. water, or not. What landers? What partial confirmation? Without supporting factual evidence, this is not a good stackexchange answer. If the moderators start leaving unsupported statements like this, the site could quickly become a free-for-all of unsupported assertions. $\endgroup$
    – uhoh
    Commented Nov 21, 2017 at 18:54
  • $\begingroup$ Also note that I've included several links and a short discussion of "likely evidence" for water on Mars that is currently being "re-evaluated". I didn't ay discounted, but it's now somewhat suspect. I put that there for a reason; to illustrate the point that inferring the presence of water from indirect evidence is suspect. $\endgroup$
    – uhoh
    Commented Nov 21, 2017 at 19:05
  • $\begingroup$ There is a large number of sources. Phoenix and Curiosity have both seen evidence. There is evidence seen of recent liquid water on the surface by MRO in the form of new gullies and salt deposits seen. Spectroscopy has also determined that water is on Mars. I linked one Wikipedia article with some sources, I'll also add en.wikipedia.org/wiki/… $\endgroup$
    – PearsonArtPhoto
    Commented Nov 21, 2017 at 19:09
  • $\begingroup$ First words of the title: "What's the scientific evidence..." A link-only answer is not acceptable. I've looked at the Wikipedia article, and it does not name the evidence. This is not a trivial question. $\endgroup$
    – uhoh
    Commented Nov 21, 2017 at 19:16

What's the scientific evidence of water for return trip methalox on Mars?

The Business Insider article NASA is helping SpaceX scout these 9 places on Mars for landing the first Starship rocket missions is an interesting read, and mostly documents the electronic "paper" trail of scouting potential landing sites using targeted, stereoscopic HiRise image pairs.

elevation map of Mars showing the nine candidate landing sites SpaceX is considering for its first Martian voyages of Starship Click for full size

An elevation map of Mars showing the nine candidate landing sites SpaceX is considering for its first Martian voyages of Starship. NASA/USGS/ESA/DLR/FU Berlin (G. Neukum) via Google Earth Pro; Business Insider

The article points out that the candidate sites all cary a significant possibility of having substantial subsurface frozen water near the surface.

The article mentions one piece of evidence specifically:

Some evidence for this is in the shape of nearby craters, which appear to sink after a meteorite impact because they expose ice to Martian air, which is about 1% as thick as Earth's.

Functionally that is a vacuum, causing the now-exposed ice to sublimate away into the air in the same way a block of dry ice does when it warms up.

Ice exposed to the thin martian air sublimates into a gas and collapses soil around the original crater

One line of evidence for ice on Mars are impact sites. Ice exposed to the thin martian air sublimates into a gas and collapses soil around the original crater. NASA/JPL/University of Arizona; Business Insider

  • 1
    $\begingroup$ Informative answer, but what's the actual evidence for water at the impact sites ? Is it shown in images at one site over time that the soil really collapses ? $\endgroup$
    – Cornelis
    Commented May 25, 2020 at 14:39
  • 1
    $\begingroup$ @Cornelisinspace to the other bits of evidence "of water for return trip methalox on Mars" discussed in other answers, I've just added this as an additional piece that I'd read about. To answer the question in your comment would require some investigation, starting with identifying which crater this is, how big it is and if there is a deeper story here. Therefore I think that would be an excellent new question! $\endgroup$
    – uhoh
    Commented May 25, 2020 at 17:28

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.