The setting I'm interested in would involve a cold planet (cold enough for most bodies of water to be frozen - I'm thinking of an average global temperature of -20C or below). The simplest way would have to have the planet far away from its star. However I'd rather not deal with the low-light aspect, so I was wondering if there's a way to avoid that. The luminosity level I aim for is above 'Sun as seen from Mars'.

Having no atmosphere would be easy, but not exactly what I'm aiming for. I'd prefer an atmosphere that allows human life. (I mean in the breathing sense. Separate shielding for cosmic rays can be provided, if necessary)I'd prefer weather (eg. ground covered in snow, whether water-based or something else, which would also increase the albedo). I realize that Earth had several Snowball eras, but I'm wondering if there could be a planet with the above characteristics whose main state is 'snowball'.

Also - could this be possible with a blue star?

  • 3
    $\begingroup$ You do of course know that our own Earth has spent multiple long spells frozen all over, for a total of about one hundred million years of being a "very cold planet that still received plenty of light from the Sun". $\endgroup$
    – AlexP
    Commented Aug 3, 2019 at 14:32
  • $\begingroup$ Higher albedo, for example by being covered in ice, would suffice. You want plenty of light, but not much heat retained, hence, you need heat being reflected. Lack of greenhouse gasses (as noted in one answer) trapping thermal radiation and high albedo reflecting other wavebands should do the trick. Even for Earth's otherwise identical twin. $\endgroup$
    – M i ech
    Commented Aug 6, 2019 at 11:38

5 Answers 5


Just remove greenhouse gases from the atmosphere: CO2, H2O, CH4, for example.

This would lower the amount of infrared radiation trapped by atmosphere, lowering as a consequence the temperature.

Keep in mind that since the climate is a complex system with several feed-back and feed-forward, a snowball planet would reinforce its conditions: more water trapped as ice ==> less water in the atmosphere ==> lower temperatures ==> more water trapped as ice.

  • 2
    $\begingroup$ I was typing a much more convoluted answer, involving captured rogue planets with frozen cores. This one is much simpler and feasible. $\endgroup$ Commented Aug 3, 2019 at 14:35
  • $\begingroup$ Would humans be able to breathe in such an atmosphere? I know things can get pretty messy when it comes to changing atmospheric composition and the affects on life $\endgroup$
    – Foosic17
    Commented Aug 4, 2019 at 16:00
  • 2
    $\begingroup$ @Foosic17 that depends largely on how cold it gets, and whether there is any life cranking out oxygen. Earth went through some severe cold periods and the atmosphere would have been breathable, if very dry. You would lose water fast, similar rate to a hot desert. oxygen levels were lower than now comparable ot high altitude today. $\endgroup$
    – John
    Commented Aug 5, 2019 at 5:08

Our planet is just a few thousand years out of an ice age. To get a habitable planet covered in ice (or mostly covered in ice):

  • Make the sun slightly dimmer (less radiated heat)
  • Make the planet slightly further from the sun (less intercepted heat)
  • Make the planet have slightly less green house gas (less retained heat)
  • Be at an extreme cold point in a planet's natural heat/cold cycle

The difference you'd need to have a planet in your required range would be surprisingly slight. Any of these listed tweaks could result in a planet colder than Earth for thousands or maybe millions of years.

As a note: to have native terrestrial life, the planet would have had to have open ocean and warm land at some point to encourage flora and fauna onto the land. Also keep in mind seasonal variation. With seasons due to axial tilt, that would mean open ocean near the equator and further toward one pole for half the year. For seasons caused by elliptical orbits, that would mean open ocean for part of the year and pack ice for the other part.

  • 1
    $\begingroup$ Welcome to the site Helion, when you have a few minutes, please take the tour and read up in our help center about how we work: How to Answer. We expect answers to be "clear, authoritative and explain why they are correct". If you can edit your answer to address the question as written, then great. IE. How is the main state of your "snowball" maintained? $\endgroup$ Commented Aug 3, 2019 at 16:26
  • $\begingroup$ Great answer Helion! For my own curiosity (maybe this is helpful for the OP as well): what role would an inactive/less-productive iron core have on the planet at large? I've always heard how Mars's lack of a molten core caused it to not have a good magnetic field, which would cause all kinds of radiation issues for future explorers, but what about it's effects on the climate? Would a molten-core planet be guaranteed to be 'warm' like Earth? Please let us know if you have info on that. $\endgroup$
    – arpanet101
    Commented Aug 3, 2019 at 19:58
  • 2
    $\begingroup$ @cyber101, a molten core planet like earth would not be guaranteed to be warm on the surface. The sun provides almost all the heat to the Earth's surface: physics.stackexchange.com/questions/183235/… $\endgroup$
    – Helion
    Commented Aug 3, 2019 at 23:35
  • 1
    $\begingroup$ @cyber101 The lack of a magnetic field on Mars has resulted in most of its atmosphere being stripped away. If it were Earth-sized and had been able to hold onto a thick atmosphere, it may have been able to retain enough heat to support liquid water at the surface. In fact, it is hypothesised that early Mars had oceans. $\endgroup$ Commented Aug 4, 2019 at 22:19
  • $\begingroup$ Thanks for the answers guys! $\endgroup$
    – arpanet101
    Commented Aug 6, 2019 at 5:39

The answers you've received are great! I'd like to piggy-back off of them - specifically L.Dutch's:

We can add to this answer by incorporating volcanic activity. A decent supervolcano will start a nuclear winter. An overactive, planet-wide 'ring of fire' with many small volcanoes could potentially prevent a forever-cold feedback loop. I could have mentioned something like ozone-creation-supporting environments to prevent the needless build-up of ice + providing a weak greenhouse gas, but volcanoes are cooler haha.


If you are talking about low light problems in the sense of people not being able to see well, you should be fine. The human eye can adapt to massive changes in light level. For example, even though Pluto receives about 1/1000 of the light as does the Earth, if you were standing on Pluto at noon you could read a book by sunlight.

To get a -20c average temperature, the orbit of Mars or just a bit beyond would be fine for a sunlike star with the right planetary conditions. But even at Jupiter's distance (which would be VERY cold), you'd get about 3% of the light you get on Earth. That may not sound like a lot but your eyes can easily handle it, as the human eye can handle about 4-stops of brightness change - a range of about 1,000,000:1. For example, a bright day outside will be about 100,000 lumens, but even a bright room like a classroom or an office may be only about 2,000-3,000 lumens. And once your eyes adapt, you can see just as well inside.

So on a planet out where Jupiter is, it would seem pretty much as bright as it does on Earth, but people's eyes would be dilated more. The amount of light would be about what you get in a well lit room indoors.

But you don't really need to go out as far as Jupiter. The equilibrium temperature at Mars' orbit is -63 degrees C. So a planet at Mars' distance with a modest greenhouse could produce the temps you need. And Mars gets plenty of sunlight - about half of what the Earth gets. The human eye would scarcely notice the difference. People would have slightly more dilated pupils and then still perceive the same amount of light.

For a blue star, just scale out the distance to correct for the higher luminosity of the star. The habitable zone of a Blue star is larger than for a yellow star like our sun, so you'd have some more room to play with distance.

Solar power, on the other hand, would have to be scaled up to match the available light.

  • $\begingroup$ Very helpful! I'm not sure if I can accept multiple answers, but it helps a lot too. $\endgroup$
    – Laura
    Commented Aug 7, 2019 at 13:45

Some nice answers already. If you don't mind doing some math, look up effective temperature of a planet. You can design the star size, the planet's albedo and calculate the effective temperature.

Earth's effective temperature isn't too far off from -20 degrees C, in fact it's -21 C.


Some problems you'd run into is that with no CO2, your planet would have a hard time undergoing photosynthesis, but you could make the planet a light color so a lot of the heat is reflected off, not absorbed, and that would significantly increase the cooling. Covering a planet with ice is one way to make it reflective.

Somebody mentioned volcanoes, but volcanoes are tricky. Volcanic soot tends to be dark and it's thought one way Earth emerged from it's snowball phase is the dark volcanic soot darkening the ice. (hypothesis not certainty).

Cover your planet with white sand and it can be nice and cold. White sand could increase it's apparent brightness as well. Create sufficient sea ice on the North and South pole or if you prefer, a permanent glacier on either pole.

YOu mentioned no atmosphere, but no atmosphere actually makes planets hot under direct sunlight. The moon for example gets very hot under direct sunlight. The atmosphere spreads the heat around, so you probably want an atmosphere.

YOu can also play around with axial tilt and seasons and cold locally vs cold globally if you like. Lots of ways to go with this.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .