Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Can a planet with a thin atmosphere have oceans?

  1. Sep 29, 2014 #1
    Let's say their is a Earth sized planet orbiting a sun like star in its habitable zone.

    This planet has a tilt and rotation rate similar to Earth.

    The planet however has an atmosphere of 0.1 bar.

    The planet orbits its star close enough to have an average temperature of 12 C.

    From what I understand the boiling point of water in a 0.1 bar atmosphere is 46 C.

    Would oceans of liquid water remain stable in this environment? I'd imagine air temperatures reaching above 46 C and below 0 C throughout the day.
  2. jcsd
  3. Sep 30, 2014 #2


    User Avatar
    Science Advisor
    Gold Member

    On a massive planet you would need less atmosphere to maintain liquid water, but, then again a thin atmosphere would seem less likely on a more massive planet. We still have much to learn about planetary physics.
  4. Oct 5, 2014 #3
    Yes, a planet with an atmospheric pressure of 100 millibars on its surface can have liquid oceans, if the surface temperature is between 0°C and 45.82°C.

    Water will sublimate directly from ice into gas if the atmospheric pressure is 6 millibars or less and the temperature is above 0°C.
  5. Oct 7, 2014 #4
    What if the temperatures in a location vary from 55 C in the summer days to 0 C in Summer nights? Would water boil during the summer day or would it stay liquid?
  6. Oct 9, 2014 #5


    User Avatar
    Science Advisor
    Gold Member

    The boiling point of water increases as atmospheric pressure increases.
  7. Oct 9, 2014 #6
    Assuming you are referring to the same fictitious planet mentioned by the OP with 100 millibars of atmospheric pressure on its surface: If the surface temperature exceeded 45.82°C, then the surface water would begin to boil and evaporate.

    Depending on long the temperature remains above the boiling point, the liquid surface water on the day side of the planet would become water vapor. Once the water is in the atmosphere it will eventually rise to a point where it is cooled, forms clouds, and then rains back down to the surface. Where the process will begin all over again when the temperatures rise above the boiling point the next day.

    It would not be an enjoyable place to visit, and such a place could not support complex life (as we know it on Earth) outside of the water.

    Boiling and sublimating are different in that boiling requires heat, and sublimation is the result of insufficient atmospheric pressure. Ice will go directly from solid to a gaseous state, bypassing its liquid form, if the atmospheric pressure is too low (< 7 millibars) and the temperature rises above freezing.
    Last edited: Oct 9, 2014
  8. Oct 9, 2014 #7
    If the nights are cool how could water boil in the daytime? I mean I thought it takes water a longer time to reach boiling point than air. Assume 1 day on the planet is 24 hours.
  9. Oct 9, 2014 #8


    User Avatar
    Gold Member

    Perhaps you should go back and make some assumption on how the planet is absorbing the heat from the sun. Is it atmospheric heating, surface heating, or some combination of both? This would depend upon how transparent the atmosphere is to radiation. With a non-transparent atmosphere, no direct solar radiation would reach the surface with the implication that the surface can become no hotter than the atmosphere. With a transparent atmosphere, the atmosphere is heated by conduction and convection heating from the surface, with the implication that the surface would ( or should be ) have to become hotter than the atmosphere before the atmosphere increases in temperture.

    Since transfer of heat is through either conduction, convection, or radiation, you should see how this affects the change of state of water on your planet depending upon the planets heating method. It is the temperature of the liquid or solid that determines the vapour pressure and not the temperature of the atmosphere. As an example, with solar atmospheric heating alone, on your planet of 0.1 bar atmospheric pressure, the temperature of the atmosphere can go higher than the 46C, and liquid water will not immediately boil, but will have to aquire some heat from the atmosphere to raise its temperature to 46. The time lag from beginning of daylight for this to happen necessitates some thought and calculation and not some handwaving "this is what is going to happen" when the sun shines, and you have to stack assumption upon assumption during the process. A phase chart of water will help you out in the process but it is not the whole story.

    Please re-read what chronos stated in his posts, as they are mightly applicable.
  10. Oct 9, 2014 #9
    I believe I'm asking it wrong. I'm simply wondering does water boil when water is above boiling point or when the atmosphere at the surface is above boiling point.
  11. Oct 9, 2014 #10


    User Avatar
    Gold Member

    • If the water is in a liquid state, the temperature of the liquid water will not rise above the boiling point (temperature).
    • The boiling point would vary - it would depend on - the atmospheric pressure at the surface.
    • If the atmosphere is the only source of energy, its temperature at the surface must be higher than the boiling point.
    • The liquid water will absorb energy required to change state, during which time it will remain at the boiling point temperature.
    • As the atmosphere provides the thermal energy to complete the change, molecules will move from liquid to gas.
    Because this is taking place at the surface, it isn't the type of boiling as a pot of water on a stove, where the water is rapidly bubbling as liquid moves to gas state at the bottom and sides of the pot.

    As has been mentioned, different pressures and temps may preclude a liquid state. The items above apply under proper conditions, for what I think you meant by "boil". Have you looked at a phase diagram for water? Do a Bing or Google image search on "water phase diagram" because the ranges they cover vary - you might spot the one you like easier in an image search.
  12. Oct 10, 2014 #11


    User Avatar

    Staff: Mentor

    Yes, they are different, but they both require heat and can be both thought of as happening as a result of insufficient atmospheric pressure.
  13. Oct 11, 2014 #12
    Willstaruss22, one thing to keep in mind though, without a magnetosphere of some description a favorable atmosphere is going to be temporary! (on the earth's time scale).

  14. Oct 20, 2014 #13
    Arguable. Earth is so massive that it would have to be moved to half the distance of Mercury to lose its atmosphere over 4.5 billion years. Magnetospheres seems to be essential to keep oceans so presumably plate tectonics (Venus), and to keep atmospheres on small planets (Mars). [Fig 1, "The Cosmic Shoreline", Zahnle and Catling, LPSC 2013; http://www.lpi.usra.edu/meetings/lpsc2013/pdf/2787.pdf ] A 1/10 atmosphere pressure would mean little to atmosphere lifetime at the current orbit, it is either the Sun or the Moon that will do the biosphere in. (The Sun will grow too radiant and the Moon brake the Earth too much for axis stability on the same timescale.)

    Mars is an interesting marginal case where gravity and magnetic fields seems to contribute equally to atmosphere preservation (fig 1). MAVEN will perhaps settle the question one way or another.
  15. Oct 20, 2014 #14

    D H

    User Avatar
    Staff Emeritus
    Science Advisor

    Venus falsifies this claim. Venus has but a tiny magnetic field, but it has a much thicker atmosphere than that of the Earth.

    Your reference says nothing about magnetospheres. Figure 1 shows a plot of solar intensity versus escape velocity. It assumes thermal escape is the driving mechanism, and hene that nice straight [itex]I \propto v_{esc}^4[/itex] line. It appears that the line was placed so it passes right through Pluto and Mars because they're on the border between having versus not having an atmosphere.

    From what I've read, water is essential to a having plate tectonics, which in turn is essential to having an geodynamo. Venus doesn't have a geodynamo because it doesn't have plate tectonics, and it doesn't have plate tectonics because it lost all its water through a moist or runaway greenhouse.
  16. Oct 21, 2014 #15
    Venus does nothing to falsify 'my claim' at all. The OP was talking about having a planet with liquid water on its surface, and I was suggesting that in order to have a planet with liquid water there must be a magnetosphere present. Otherwise, the solar wind will ionize the upper atmosphere, slowly blowing it away into space.
    Venus does not have liquid water, and as a result has no decerneble tectonics to remove the carbon from the atmosphere. Without a proper carbon cycle, carbon builds up in the atmosphere to the point where you have the runaway greenhouse effect we see.

    Last edited by a moderator: Oct 21, 2014
  17. Oct 21, 2014 #16
    I am beginning to wonder whether a magnetosphere is required for a planet to have an atmosphere.

    Both Venus and Mars have a much weaker magnetosphere than Earth, yet Venus has a much higher atmospheric pressure than Earth, and Mars has a much lower atmospheric pressure than Earth.

    Furthermore, it turns out that Earth, with its much stronger magnetosphere than either Venus or Mars, is losing atmosphere at a faster rate than either Venus or Mars. Earth is losing 5 x 1025 molecules, or 3 kg of hydrogen and 50 g of helium, per second. Apparently caused by thermal (Jeans) escape in the exosphere.

    See also:
    Our Planet's Leaky Atmosphere --- Kevin J. Zahnle and David C. Catling (May 11, 2009), Scientific American
    Thermally Driven Atmopsheric Escape: Transition from Hydrodynamic to Jeans Escape --- Alexey N. Volkov, Robert E. Johnson, Orenthal J. Tucker, and Justin T. Erwin (February 16, 2011), The Astrophysical Journal Letters
  18. Oct 21, 2014 #17

    D H

    User Avatar
    Staff Emeritus
    Science Advisor

    So are many atmospheric scientists. Strangeway et al. have gone so far as to ask "Does a Planetary-Scale Magnetic Field Enhance or Inhibit Ionospheric Plasma Outflows?" (AGU Fall Meeting Abstracts, Vol. 1, 2010). Based on a sample size of three (Venus, Earth, and Mars), the presence or absence of a strong magnetic field just doesn't seem to be that significant factor in determining atmospheric loss, and it certainly doesn't explain why H2O is but a trace gas in Venus' atmosphere.

    The two dominant hypotheses are those of a runaway greenhouse (Kombayashi and Ingersoll) or a moist greenhouse (Kasting). The distinction between the two is a bit subtle, and both lead to the same end result: A very hot surface that cannot sustain a drop of liquid water. Once the surface temperature reaches the critical temperature of water (374 C), a moist greenhouse transitions to a runaway greenhouse.

    In both a runaway greenhouse and a moist greenhouse, too much incoming radiation results in increasing amounts of water vapor in the atmosphere. At some critical solar intensity, a positive feedback loop sets up because water is a very powerful greenhouse gas. Increased water vapor increases surface temperatures, which increases atmospheric water vapor content. The Sun gets hotter and hotter as it ages. This critical intensity was reached on Venus long, long ago. It will happen here on the Earth a billion or so years from now.

    This doesn't quite explain how Venus lost its water. That explanation comes from photodissociation of water. The atmospheric lapse rate decreases toward the moist adiabatic rate as atmospheric moisture grows. This makes the top of the troposphere higher and warmer. The water vapor is subject to ultraviolet, which photodissociates the water. The resulting hydrogen escapes due to Jeans escape. The water eventually disappears.
  19. Oct 21, 2014 #18
    Our own planetary history would seem to invalidate the "runaway greenhouse effect." During the Ordovician/Silurian ice-age atmospheric CO2 levels were between 7,000 ppm and 11,000 ppm. Furthermore, at the Permian/Triassic boundary mean surface temperatures soared to between 35°C and 40°C, when atmospheric CO2 levels were less than 370 ppm. Just the opposite should have happened under the "runaway greenhouse effect" theory.

    A far more likely scenario is the "moist greenhouse effect." It is at least consistent with Earth's past climate conditions. Although, the Permian/Triassic boundary high mean surface temperature anomaly still remains unexplained. A "moist greenhouse effect" would also be self-regulating. In other words, there would be no "runaway" heating. As radiative forcing increases, so does the water vapor in the atmosphere due to increased evaporation. Once the water vapor reaches a certain altitude it begins to cool and form clouds, thus increasing albedo. The clouds produce rain, which in turn washes CO2 from the troposphere in the form of carbonic acid, lowering the radiative forcing due to less atmospheric CO2.

    A simpler explanation for Venus' high atmospheric pressure, and only trace amounts of atmospheric H2O, may be linked to its volcanic state. Venus has ~1,600 major volcanic features with anywhere between 100,000 to 1,000,000, or more, volcanoes. Compared to Earth's meager ~1,500 volcanoes. The curious thing is that Venus shows no evidence of plate tectonics, which should imply less volcanic activity than is found on Earth.
  20. Oct 21, 2014 #19


    User Avatar
    2017 Award

    Staff: Mentor

    That is true, but not the important point: venus can keep its atmosphere without a magnetic field. It is true that hydrogen is much more volatile than carbon and oxygen, but venus is also hotter. Some small loss rate would be acceptable.
  21. Oct 21, 2014 #20

    D H

    User Avatar
    Staff Emeritus
    Science Advisor

    The solar intensity at the Earth was not then and has never been high enough to trigger a runaway greenhouse on Earth. That won't happen for another billion years or so. The same goes with a moist greenhouse.

    You don't understand what a moist greenhouse is. Both a runaway greenhouse and moist greenhouse spell absolute doom for life. Surface temperatures of well in excess of 1000 Kelvin are not very good for water-based life. The only way out of a moist greenhouse is a runaway greenhouse. A moist greenhouse results in a positive feedback that doesn't stop until the surface temperature exceeds the critical temperature of water, and then you have a runaway greenhouse. A runaway greenhouse doesn't stop until all the water disappears via photodissociation and atmospheric escape. There is no self regulation. It's the disappearance of a self-regulation mechanism that marks the onset of a moist greenhouse. The difference between a moist greenhouse versus a runaway greenhouse is rather subtle. The distinction lies in conditions at the top of the troposphere versus conditions at the top of the stratosphere.

    There are three mechanisms via which a terrestrial planet can release heat from the core to space: Magma ocean, stagnant lid, and plate tectonics convection. A very, very young planet is at least partially covered with a magma ocean. That phase is extremely efficient at transferring heat, but ends as the magma ocean freezes. A planet with plate tectonics is not as efficient as a magma ocean, but it is still does a good job at transferring heat to outer space. A stagnant lid is markedly inefficient compared to the other two. Venus doesn't have a magma ocean and it doesn't have plate tectonics. It has a stagnant lid. This, BTW, is part of why Venus doesn't have a magnetic field.

    Venus doesn't have plate tectonics because it doesn't have any water on the surface. Water strongly modifies the liquidus and solidus points of rock. It acts as a lubricant that enables plate tectonics to occur. Non-hydrated crustal rock becomes is stiff and brittle. If Venus did have plate tectonics when it was young, it stopped shortly after Venus lost its surface water. The only mechanism Venus now has for transferring heat from the core and mantle to outer space is stagnant lid convection. One conjecture is that this is so inefficient that Venus's interior gets hotter and hotter -- that is, until it blows it's stagnant lid wide open with a planet-wide volcanic event.
  22. Oct 23, 2014 #21

    D H

    User Avatar
    Staff Emeritus
    Science Advisor

    Here's a paper that has direct bearing on the original question:

    Vladilo, et al., "The habitable zone of Earth-like planets with different levels of atmospheric pressure." The Astrophysical Journal 767.1 (2013): 65.
    arxiv preprint: http://arxiv.org/abs/1302.4566

    A big problem with a thin atmosphere is that the habitability zone shrinks drastically as pressure decreases. There's little tolerance for variation in solar intensity with a low pressure atmosphere. The problem with this is that solar intensity is not constant over long periods of time. The Earth became habitable 3.5 to 4 billion years ago and will cease to be so a billion or so into the future. A planet with a much reduced atmosphere will have a much shorter period during which it is habitable.
  23. Oct 23, 2014 #22
    The mass of water is definitely a consideration here. Puddles probably would boil off, but I'm not so confident about lakes or oceans. It's late fall here where I live and temperatures drop below freezing at night and are above freezing during the day. There is a concept called "degree days" that is applicable. If the average temperature between the high and the low is above freezing, you add that number of degrees to the total from whenever it was that you started counting. If it's below freezing, you subtract. The last three days here, for example, the high was 5C, the low was -8C. That's -3 degrees for three days, so -9 degree days. That means that freezing conditions are in place and more ice should form on a pond at night, than could melt during the day. In other words, even though temperatures get above freezing during the day, bodies of water are icing over and the ice is getting thicker.

    For what you're talking about here, we have a boiling point of ~46C, a high of 55C and a low of 0C, so we'll set our benchmark at 46C. The average temperature over that 24 hour period (length of day you gave somewhere) is 27.5C which gives us -18.5 degree days with respect to boiling. Lakes and oceans should be stable. There may be some minor localized boiling at the shoreline if the shallow water/surface temperature reaches 46C+ and there are no waves mixing the water.

    Another interesting effect would be that at night, as the upper layers of the lake / ocean cooled below the more constant temperatures of deeper water, the water would begin to 'turn over.' Density differences would allow convective mixing of the waters. To what depth this would occur would be dependent on the temperature gradients in the water. This has positive implications for oxygen (assuming oxygen content in the atmosphere) transport into deeper waters.
  24. Oct 29, 2014 #23
    Exactly, the generic escape mechanism doesn't depend on magnetic fields. The observations favor thermal escape.

    Yes, it is a complicated feedback process.

    People have argued that the hydrogen escape, which is much helped by solar wind proton sputtering at max efficiency against hydrogen atoms (after UV hydrolysis), was the main difference between Venus and Earth that led to today's situation. Loss of water meant thermal runaway, and what little plate tectonics Venus had was lost then. As the crust thickens under much the same temperature above as below, mantle convection stops and eventually what little outer core geodynamo Venus had shut down.

    The solar wind and even CMEs are much deflected by the geomagnetic field.

    But you can easily argue some other way, and modeling remains difficult. I think Zahnle and Catling comes down in support of hydrogen loss because of too weak initial geodynamo.

    [I wrote this, then saw that you have a reference arguing the other way. So it goes.

    But I would be glad if we can put the magnetic field away. Maybe MAVEN will help decide this, as per above - Mars being on the sensitive cusp of thermal escape.]

    Yes, but that is because we still have hydrogen left (because we have oceans left). Venus (and to some degree Mars) has little hydrogen left to lose.
  25. Oct 29, 2014 #24
    Earth was habitable ("cold early Earth") about 4.4 billion years ago. That is what a painstaking 6 year work on our oldest Jack Hill zircon told us this spring. And with the Nice/Nice 2 models the impact rate was no worse than the (apparently, according to models) survivable late bombardment at about that time. Since tentative molecular clocks of the genome places the first splits right before 4 billion years, the UCA lineage ought to derive sometime after 4.4 but before 4.0 billion years ago.

    [I would favor 4.4 billion years, because emergence seems simple while the DNA LUCA was about as complex as the average bacteria today. It took several 100s of millions of years from the mitochondria event to the first body plans of Metazoa, so the generic "body plan" of bacteria could also take much the same time to evolve.]

    For the latter upper end habitability date, I don't find any support in the supplied reference. That is the usual date for latest habitability re complex life (consensus 0.5 - 1.5 billion years more), which goes when the partial pressure of atmospheric CO2 becomes too small to sustain plants. (It seems to be a 1D model sometimes used as pseudo-2D, which isn't good - it is only 3D models that solves the faint early Sun problem of the Archean.)

    Typically, refugias for prokaryotes last 1-2 billion years longer, which ought to define the tail end of the radiative HZ lifetime.
    Last edited: Oct 29, 2014
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook