I have the below snippet describing a planet in a (hard) sci-fi setting. This harsh planet has some human settlers, who can survive the climate in light weight environmental suits.

The planet is a large, dry, desert world without any notable resources and there is barely any surface water, although underground water exists. Its surface entirely covered by sand and dry craggy rock formations. The days are blistering hot and the nights icy cold. It has an atmosphere which is frequently ravaged by violent, electrically charged, sandstorms, often lasting several days.

The planet is accompanied by two moons, one, a ball of frozen rock and ice, the other just a small bare rock, encircled by a debris field.

Within the constraints of the description above, I'm trying to find a way to satisfy the above described 'The days are blistering hot and the nights icy cold' for the entire planet, if possible.

The obvious problem is of course that if the planet is very hot near the equator, it won't be nearly as hot at its polar regions.

So, my question is: is there a method to basically create a single biome (hot) desert planet?

I thought of things like:

  • erratic orbit?
  • extreme axial tilt (maybe like Uranus)?
  • wobbling axial tilt?
  • wind transporting the heat?

But I don't know how those options would/could affect the climate on the planet and what other unforeseen consequences those changes would bring about.

  • $\begingroup$ Somewhat related. $\endgroup$
    – HDE 226868
    Commented Dec 9, 2016 at 19:30
  • 5
    $\begingroup$ Do you need it to be uniformly hot? If you allow your equator to be like the Sahara and your poles to be like the Gobi, then you still have un-hospitable desert and it is much easier to explain. $\endgroup$
    – kingledion
    Commented Dec 9, 2016 at 19:39
  • 2
    $\begingroup$ There is no need for the planet to uniformly hot. I'm fine with the equator region being even hotter than other regions (but it should not be a 'you die within minutes' region) but the coldest places should still be way hotter than comfortable (I'm not looking for a desert in the sense of 'no water'). $\endgroup$
    – Jacco
    Commented Dec 9, 2016 at 19:49
  • 1
    $\begingroup$ As a side note: if the planet is close enough to the sun to be habitable, let alone a blistering desert, it won't have an icy moon. $\endgroup$
    – ckersch
    Commented Dec 9, 2016 at 21:53
  • 1
    $\begingroup$ Question: what about something like Venus? A strong greenhouse effect can raise the temperature a lot, even on the darker side of the planet... Beware that this can cause some bad effects on life on the planet.. $\endgroup$
    – frarugi87
    Commented Feb 28, 2017 at 14:00

5 Answers 5


So in general terms, if the planet was like Earth in size and atmosphere, but had an orbit of .8 AU's instead of 1 AU, then the planet would have an average surface temperature of Kelvin: 322, Celsius: 49, Fahrenheit: 120

At the orbit of Venus (0.723 AU) the average temperature would be Kelvin: 338, Celsius: 65, Fahrenheit: 149

For reference Earth has an average temperature of Kelvin: 288, Celsius: 15, Fahrenheit: 59

The poles would be pretty warm.

Throw in no large oceans and you won't have much water vapor in the air. That means few clouds, and very rare rain. Because of the lack of clouds all the heat will radiate off into space at night. The Sahara Desert gets cold at night.

If there is no axial tilt then there would be no seasons, so no winter.

Edit: Alternately for more creativity, the star could be part of a binary/trinary star system and so that's a lot of extra heat being pumped at it. Though nights might be iffy at that point.

A reason why there might not be much water is if there was something that is splitting the water. It could be photobiological, with a super algae that lies dormant in the sand until it gets wet. Then it reproduces as fast as it can, and in the process splits the water in to oxygen and hydrogen. This would solve the oxygen problem too. Eventually some of the oxygen and hydrogen join back up again into water, and starts the process over.

  • $\begingroup$ I was writing something along these lines, but you've done it better than I could. +1. $\endgroup$ Commented Dec 9, 2016 at 20:25
  • $\begingroup$ I was hoping for a more imaginative method :-) $\endgroup$
    – Jacco
    Commented Dec 9, 2016 at 20:43
  • $\begingroup$ @Jacco It's your story, you can literally do anything you want with it. :) Limited/no water means you get a desert, and a closer orbit will get you the heat. The reason Arrakis is a desert in Dune is a really silly one, and it's a huge hit. And you don't even have to explain it! Just say that's the way it is. Or maybe have it be just a little closer and some natural process that keeps the water away from the surface. Play with this until you get a result you like: astro.indiana.edu/ala/PlanetTemp/index.html $\endgroup$
    – AndyD273
    Commented Dec 9, 2016 at 21:02
  • $\begingroup$ @Jacco Added a few other ideas. Hopefully they'll help some. with the science based tag I usually try not to let my imagination get to far away if I can help it. $\endgroup$
    – AndyD273
    Commented Dec 9, 2016 at 21:15
  • 1
    $\begingroup$ Nice answer but the picky terminology part of me wants to point out that "a lot of extra heat being pumped" isn't correct - heat cannot travel through a vacuum. There is however more radiation :) $\endgroup$
    – Zxyrra
    Commented Dec 10, 2016 at 3:52

Not actually sure if this will fit what you had in mind, but, what about having the planet just turn slowly?

I mean, days are warmer than nights, and sunny days warmer than cloudy ones, nice and simple - so if the sun is in the sky for longer days, the days should get a lot hotter with more sunlight and radiation, and the nights a lot cooler, since it's longer until the sun warms them up again. The more rapidly the planet turns, the more the temperature would be equalized, I think, and the slower it turns, the hotter the day and colder the night (like Mercury, though your planet will have an atmosphere and so will not be so extreme)

So, the idea that popped into my head as a starting point was a day about three times as long as our day. The middle third of that day would probably be hotter, and the middle third of that night probably colder, than probably even the hottest day and coldest night of the equivalent area on earth - since the area won't have cooled down or heated up in the meantime. Diurnal temperature variances can range from a few degrees (4*C), to more than a hundred degrees (102*C, granted that is a world record), depending on landscape and season, but someplace like a desert would have the highest variations. One example given of the diurnal variation of a low lying plain was 30*C - which would increase a lot with longer days, since a day three times as long (with no other differences) could mean a gain and loss of 90*C, and could also mean more as the rise in temperature during the extra hours of sunlight is heating an area still warm from the hours equivalent to our day, with no chance to cool off first. Likewise the night would have more time to loose heat, and would likely get colder the longer it was out of sunlight.

A couple other thoughts - if the temps get higher every day and cool a lot more every night, then any plant that survives would have to be adapted to those temperature variations - it might favor vegetation that is more desert-like (since those adaptions work on our world, in similar conditions). And areas that are more desert-like experience more diurnal temperature variations, since the vegetation doesn't hold the temperatures as well, so it could be a cycle tending towards desertifying any flat areas. If the world is a little warmer than ours (perhaps by being closer to the sun, as AndyD273 suggested), or a little drier, or even a little smoother that might reduce the frequency other ecosystems (like rainforests, the other adaption to hotter temperatures) to a much less observable level. So where we have deserts might be mostly uninhabitable, where we have plains would be deserts, forests would likely be plains (and eventually adapt towards deserts), rainforests would be more like temperate forests (after adapting to colder nighttime temps), and so on.

Second point, the violent storms make sense in a world with high temperature variations - the atmosphere would try to equalize the extreme temperatures from the day side to the night side, meaning violent winds that can easily play into your giant electrical sandstorms.

I'm not entirely sure about the details of this, but the size of your planet might play a role as well - a larger planet might have more thermal mass to equalize temperatures, and a smaller might not be able to hold onto extra heat, and so have more extreme temperatures. Also, a planet with a warmer core might be warmer overall (as you wanted a hot planet), but it might also be less prone to temperature extremes, since that inner heat could keep the temperature from varying as much between day and night. I know less about this, though, take it as a suggestion rather than a fact.


The icy moon could be relatively large, have a highly reflective surface and be relatively close so as to reflect significant amounts of solar radiation onto the planet’s surface. If it was in an eccentric polar orbit it would spend a disproportionate amount of its time over the poles. So the poles should get extra heat and light. If the moon did not spin on its own axis and was tidally locked to the parent star its orbit would always be from pole to pole over the dawn / dusk solar terminator so would only ever appear directly overhead over the Polar Regions and for shorter periods at dawn and dusk in other regions.

The side effects of this would be some extra heat and light at dawn and dusk in some areas and at some times (depending on the orbital period) as the moon rapidly passed across the equatorial plane. It would also illuminate the night side of the planet to some extent, but would be low on the horizon as viewed from the equator at night and would also be a crescent moon so would provide much less light and heat. Ignoring the planets axial tilt, the moon would only ever appear in its half-moon phase from the poles, a crescent moon from the equatorial night side and a gibbous moon from the equatorial day side.

As seen from the poles, moonrise and moonset would always be 180 degrees apart and the points would slowly progress around the horizon over the course of a year. As seen from the equator the moon would appear in its gibbous phase low on the horizon at midday and in its crescent phase low on the horizon at midnight. The moons transition would depend on the time of the year. At one point call it “spring equinox” the moon would always appear low on the horizon and would move from one side of the horizon to the other over the course of the day changing phase from gibbous at midday to crescent at midnight. Over the next 3 months the moon would rise higher in the sky until at “mid-summer” it would pass directly overhead. Over the next 3 months it would sink again until by the “autumnal equinox” it would be back near the horizon after which it would start to dip below the horizon. At “mid-winter” the moon would only be visible at dawn and dusk.

The exact size of the planet and moon, the shape of the moons orbit and any axial tilt or orbital eccentricity of the planet could significantly affect the situation.


Is it possible? Well, I wrote "no" but in defending that position, I thought of a possible scenario where it was. First, how do you illuminate a sphere uniformly? Well, I can't think of any way even with 3 sources (stars). So, could you spin it fast enough so that the entire surface was over a local 'day' exposed to the same heating? Maybe, but I can't wrap my mind around that. I'm guessing you can't. My original model is the primary sun always over the equator, and two fainter suns directly above the N and S rotational axes. The amount of illumination averages to nearly the same, and with winds the weather is the same (why constant illumination (at the poles) would result in the same biome as a (fast?) day-night cycle is a different problem). So, maybe you can do something with that. A couple of easy calculations would determine the amount of insolation per day per square meter at the poles, equator, and at 45°. They'd need to be close. OK, that was a digression. If you had clouds, you could modify the insolation, but I don't think that works. Water clouds = rain, and its a dry world, sand clouds = contribution to different weather/climate. The thicker the atmosphere, the more the illumination can be averaged. So, here's my idea: why not a very thick atmosphere around a planet which is far from any star which relies on its internal heat - the star contributing virtually none of its energy. These type planets are theorized to exist, and aren't necessarily attached to a star. (wandering planet).


AndyD273's answer has a lot of merit but I think there's a simpler answer; the thinner and drier the atmosphere the higher the daylight temperature will be, and the greater the day-night temperature difference is going to be as the atmosphere bleeds heat faster too. Also the thinner the atmosphere the stronger the winds can be, so an atmosphere of only barely breathable density is going to make the world hotter in daylight and colder at night as well as much winder than it would otherwise be. What I can't help you with is polar temperature distribution, the polar regions are always going to be colder than the tropics even if you funnel air directly north-south from the equator to heat them. A thinner atmosphere is automatically drier than a thick one because the water carrying capacity drops as the pressure does. So a world with a lot of water but a thin atmosphere is still going to have less cloud cover, hot days and cold nights. You can dig into the regolith and recover water for drinking, farming etc... allowing you to have a relatively high population if you want one.

You can also change other aspects of the atmosphere to suit your purposes; less Ozone would make the sunlight stronger and more blue and UV wavelengths would make planet fall. More Carbon Dioxide and/or Methane would let you a have higher heat retention but the days would still feel much hotter than the nights due to solar radiation exposure, actually that might let you have something approaching global temperature equality in nightly minimums, the day temperatures are still going to vary depending on latitude.


You must log in to answer this question.

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