Colonizing Red Dwarfs
This video is sponsored by CuriosityStream. Get access to my streaming video service, Nebula, when you sign up for CuriosityStream using the link in the description. The greatest and mightiest of civilizations will be around the larger stars, but the longest lasting ones will be around the red dwarfs. There are billions of galaxies, each containing many billions of stars, with billions and billions of worlds around them all, and around 3 in 4 of those alien suns are red dwarfs, so we should assume an awful lot of folks will live under a red sun in the future and I thought we should explore some of the difficulties and special challenges of colonizing such red dwarf systems, with a particular focus on tidally locked planets around such suns, where one side exists in eternal daytime and the other in perpetual darkness. First we should start by recognizing that regardless if it is a red dwarf or blue giant, all stars are white, we just call them red and yellow and blue as a means of categorizing them simply. White light sources that are generated by heat, whether it’s a sun or incandescent light bulb, have a temperature that controls what the peak wavelength is for the light emitted, but the peak is not the entirety of the light.
For a glowing hot red metal rod, that peak isn’t even red, it's somewhere in the infrared and there’s just a little bit of red being given off that we can see. Most stars give off the majority of their light in the infrared range of the spectrum, but still give off a wide spectrum of visible light too and the peaks and spectra can be compared to a light bulb’s color temperature. Our sun, called a yellow dwarf, has a surface temperature of 5800 Kelvin, which makes for a peak wavelength of green not yellow, and on the blue end of green too, but it is a white light source.
So too is an incandescent light bulb, like the kind many of us grew up with, and those had a color temperature of about 2400 Kelvin, whereas ‘warm white’ bulbs are about 3000 Kelvin, and red dwarfs can be anywhere from 2300 to 3800 Kelvin, it is a pretty broad range from that old cold incandescent light up to the natural bright white LEDs we often have now, but it’s all white light. Probably one of the most common concerns I hear about colonizing red dwarfs is that folks wouldn’t like the lighting, but its lighting you’re already very used to, both indoor and out, not some deep red as folks tend to imagine. Red dwarfs not only have a wide spectrum of color, they are a wide spectrum of star, too. The vast majority of stars are smaller than our yellow dwarf Sun, the term yellow dwarf is a relic of when we could only see the biggest and brightest stars, and so is the term red dwarf.
We had no idea how common they were when we started cataloguing them, and what a broad range of characteristics they had. For instance it is common to assume planets around these stars would be tidally locked, always showing the same side to their star, so that one side was dark and the other perpetually lit, and indeed these might be common and we’ll spend a lot of time on them today, but with the Luminosity of Red Dwarfs ranging from 0.015% to 7.2% of the Sun’s light, a factor of 500, that means their habitable zones range in distance from their sun by a factor of 22, making their habitable zones, in terms of light for heating worlds, between about 1% to 40% of what Earth’s distance from our Sun is. Alternatively K-Type Orange Dwarfs, the next biggest star type, vary much less in brightness, and G-Type yellow dwarfs like our sun, or the slightly brighter and bigger F-Type yellow-white dwarfs, also occupy a narrow band compared to M-type red dwarfs. The classes above that in brightness or mass do vary more, but these represent around 1% of stars and the shortest lived ones, either being short lived giants or stars in their end of life red giant phases and thus not usually considered ideal for colonization. That’s a point I’d dispute, but a topic for another day.
The key notion though is that M-Type Red Dwarfs are a wide class of stars that contain more stars than all the other types combined, and maybe should be broken into more types as a result, but being the most common by far they should make up the bulk of future colonized systems. They also live far longer than other stars and are the most efficient with their fuel, some able to live trillions of years, indeed no red dwarf, even the oldest, has yet left the main sequence let alone died. This longevity might make them preferred colonies, and also gives them far more time for life to evolve naturally on the planets orbiting them. However this doesn’t mean they’d make for the places where the majority of people lived. As an example, we often picture the final state of colonization for most stars as a Dyson Swarm, where all that star’s light is being absorbed by various artificial constructs and space habitats rather than just natural planets, and one around our Sun would receive 14 to 7000 times more light than a red dwarf’s Dyson swarms would have, again a very large range, but it’s a good reminder that while red dwarfs make up the majority of stars, they do not make up the majority of starlight, indeed they make up less than a percent of it, so if that translated directly into population, so many watts of sunlight equals so many people, single-sun red dwarf systems would make up the majority of systems but less than a percent of the galactic population.
At least at any given moment, considering they last longer and are more efficient with their fuel, a red dwarf might be home to as many people as a larger, brighter, but shorter lived star over the entirety of their own lifespans. Now on the planet-colonization end of things, our main interest for today, they are also a little less likely to give us the worlds we want for direct colonizing or terraforming. The bigger a star is, the more planetary mass it is likely to have, because it would have had a bigger disc around it for those stars to form from. Furthermore, the larger the star, the larger the habitable zone. Habitable Zone size is debatable, Venus and Mars might be considered in our Sun’s Goldilocks Zone in spite of being a literal molten hellscape and frozen desert respectively, so suns our size might have room for two planets in their habitable zone, but smaller ones do not.
Indeed those very smallest M-type red dwarfs have habitable zones more on par with what we’d think of as a moon’s distance rather than interplanetary. So while red dwarfs will have planets, they will not likely be where the majority of Earth-analogues for size and temperature can be found, and many of these might be tidally locked, eternal day on one side and night on the other. Though let me add a big caveat on that, we are less than 30 years out from the discovery of the first exoplanet, and most of our basic notions of planet formation and setup pre-date that. As our catalogue of worlds has grown, our expectations and theories have been regularly shattered. Alan Boss, an exoplanet theorist, once quipped back in 2007 in regard to our discoveries not matching our theories “The score is something like observers: 230, theorists: 0.”
And we’ve found dozens of times more planets since that comment. And sadly virtually none of those are planets of roughly Earth Mass in a Habitable Zone, and even fewer around red dwarfs. This is not because we think they are rare, but because planets are easier to see the more light they are reflecting, and big gas giants right next to their Suns, what we call a hot Jupiter, are the easiest to see. But it means we don’t have hard data yet on what red dwarf habitable worlds would be like, and as mentioned, our theories have a less than ideal track record where exoplanets are concerned.
So let's begin with this notion that planets around red dwarfs are tidally locked. There’s a few reasons why this was assumed to be the norm for red dwarfs. First, almost every moon is tidally locked to its planet, and second, we had incorrectly assumed that both Mercury and Venus were tidal locked or near to it for quite a while. But Mercury is in a 3:2 orbital resonance, 3 days for every 2 years exactly, not a 1:1 day equals year tidal locking.
Venus has a day and year of about the same length, indeed the day is longer than the year, but we are not inclined to think this is strictly from the classic model of tidal locking. Objects orbiting something are not rigid, even when rocky, and stretch a bit in the direction of the gravity, as opposed to staying a perfect sphere. Since they spin, the bit stretched will begin to fall back down once it spins past that “Noon” position, but it takes a bit of time to sag back so you get a moon that is stretched a bit off-center. As a result you have a little extra gravity pulling on the side that just rotated past the primary body and this acts as a slow brake. Over a long enough time, any object orbiting another body will slow in its rotation till its day equals the same length as its year, and eventually the main body will do the same. Pluto for instance is locked to its largest Moon Charon, and vice-versa, and we expect Earth to eventually have a day as long as its month too, but not for many billions of years.
We have a formula for loosely calculating this time for lockdown, and while it includes the mass and size of the primary and the satellite, the biggest factor by far is the radius the object orbits at. For instance we figure a star half our mass would take 4 times longer to tidally lock Earth, at the same distance, and one a tenth our mass 100 times longer, as it goes with the inverse square of mass. However we estimate locking to go with the sixth power of distance, so a planet half the distance from a star would take a 64th the time to lock, and a tenth the distance would lock a million times faster. Hence why moons are almost all locked, the Earth may be 333,000 times less massive than the Sun, to act as a locking mechanism for our moon, but is 400 times closer to Earth than Earth to the Sun.
So while the Sun’s mass exerts over a hundred billion times the locking power that Earth’s Mass does, 400^6 is over 4 quadrillion, making the distance factor 40,000 times bigger than that mass effect. Red dwarfs have less mass than our Sun, so a planet as far away as Earth would take longer to lock, but it couldn’t be as far away as Earth and be habitable. A Red Dwarf a quarter of our mass would have only about a hundredth of our sun’s luminosity, so a planet would need to be ten times closer to get the same lighting, as sunlight falls off in strength with the inverse square of distance.
All things being equal, a satellite would tidally lock 10^6 or a million times faster by being closer, though also 16 times slower from the lower mass, or 62,500 times faster accounting for both. Such a star is fairly average for a red dwarf, and indeed is on the edge of one of the most important boundaries for red dwarfs, of being fully convective. Red dwarfs below a quarter our Sun’s Mass are expected to use almost all their hydrogen fuel before burning out, it just mixes together, but larger stars are much less fuel efficient and often burn only a fraction of the hydrogen before dying because the core is where most of the burning goes on and in larger stars it doesn’t mix with the upper layers very well. Stars below this quarter-solar-mass level are not expected to ever turn into red giants, and are doubly long-lived as stars go, as they burn their fuel slowest and burn it all. I’d also tend to guess that above this mass you would not see tidally locked habitable zone planets. We still don’t know where the cutoff would be in terms of distance and mass for tidal locking, and again it is a function of time, but newer modeling with better computers has led us to believe that atmospheres act a bit like a lubricant against tidal locking.
There’s a major concern that the more volatile solar activity of smaller stars might blast the atmospheres off of their planets, but it seems likely with the larger red dwarfs that any planet that has retained a reasonable atmosphere would not be tidally locked. Where that cutoff is, time will tell as our modeling and observations improve, but as I said earlier, M-type red dwarfs are such a broad category of stars they really deserve sub-classification, and I would bet two of the common ones in the future will be which side of fully convective they are, and if they can have non-tidally locked habitable zone planets with atmospheres. That is age dependent too, as the older a red dwarf is the more likely planets around it will have lost their atmosphere, and the more likely they are too have tidally locked, and even more so if they have lost their atmospheres. That’s fairly important considering how long they can live. By which I mean natural atmospheres, though for colonization purposes we would include worlds we could add an atmosphere onto and expect it to last geological periods, not mere centuries.
There may be naturally occurring tidally locked worlds with atmospheres in habitable zones of red dwarfs, but this staple of fiction now seems a lot less likely, with the caveat that much of this is still guesswork, and probably would have to be on either younger systems or the least massive of red-dwarfs. Or on terraformed planets. A planet might be tidally locked with no atmosphere, but that is not a strong bar to adding air and sea. What are these worlds like? Well we often call them Twilight Worlds, which can seem like a bit of a misnomer given that on such planets most areas never experience twilight. The Sun is either never in sight, on roughly half of it, or always in sight and in about the same place. There is a band of twilight between these two hemispheres where it is assumed most folks would choose to live, and indeed on most of them there would be spots where the sun might rise and set occasionally, just not rise much or set much.
For all that we say the Moon always shows us the same face, 59% of the moon’s surface is visible from Earth, not 50%. And this is due to Libration, or wagging and wobbling, which comes in a few types. First we have Diurnal Libration, which means as Earth spins you can see the Moon from slightly different angles, and is probably the easiest to visualize but only accounts for a tiny portion of that extra visible surface of the Moon and doesn’t really apply to lighting, both sides of a star are equally bright, and thus light the tidally locked planet, but from that planet there might be sunspots near the edge of the Sun visible from one part of the planet but not others. The effect is minimal but would vary by how big the Sun was in terms of angular size, and it is worth noting that red dwarfs appear far larger in the sky of the planets they warm than our Sun would, which in turn would be larger than a giant star, for the same planetary temperatures, you need to be closer to smaller stars to stay warm and they will look bigger in the sky over all.
This effect though, that the star and planet both have real width and are not point-like, contributes to this band of twilight. Libration from Latitude is our next one, and is the same concept as seasonal variation. Earth is tilted relative to its orbit around the Sun, and a tidally locked world can be too. In this context it means the polar regions will alternate in level of lighting over the course of the year.
Now it is worth noting that years can be short around red dwarfs, anywhere from a few days for hotter worlds around the smallest red dwarfs to a few months for worlds at the cold edge of the habitable zone of the largest of red dwarfs, though I’d doubt these last kind would be tidally locked. That is earth days incidentally, again the day on a tidally locked world is identical to its year. Of course a planet could have had some event push that tidal locking ahead quicker than schedule, again Venus’s own day and year are nearly the same and these days we guess it was from a large collision event, much like what we suspect created our own Moon or flipped Uranus on its side. Some planets far from its sun could be tidally locked, simply because planets don’t seem to circle their stars for eons without major events happening to them, and it would not be that improbable one got slammed by another planet in a way that accelerated that tidal locking. Of course planets rarely circle stars, not in perfect circles anyway, but rather in ellipses, and this results in our third type of Libration, wobbling in Longitude.
An object orbiting a star in an ellipse will have two interesting effects. First and easiest to understand, the planet is sometimes closer to its star, and sometimes further, and so will be brighter or dimmer. Indeed, the Earth gets about 7% more light when it’s at perihelion, closest to the sun, then when it’s at aphelion, furthest from the sun. Amusingly, at least for those of us living in the Northern Hemisphere, perihelion occurs in early January. Our seasonal variation from axial tilting is much more potent than our planet’s orbital eccentricity, but Earth has a fairly low eccentricity, or nearly circular orbit, compared to most planets. Pluto and Mercury have the highest eccentricity, and planets on such very elliptical orbits would get larger seasonal effects from that eccentricity.
Given that years on these planets might last days to weeks, such seasonal variation from eccentricity might more closely approximate our day than our season. The other aspect of Libration from elliptical orbits is a bit trickier to picture, but what has to be remembered is that tidally locked planets still rotate, they just do it once a year. However they spin at the same rate all day long, they do not orbit at the same speed all year long though. Satellites orbit fastest near a star – they are falling and picking up speed as they do it – and slowest when furthest from the star.
As a result the actual angle the star is relative to a fixed place on that planet is changing over the course of the year, and the more the more eccentric the orbit is. For this reason the sun can rise and set over the course of that year, though even for the most eccentric orbits it is probably only going to seem to bounce up and down on the horizon a bit. One other thing about red dwarfs and eccentric orbits. Because the orbital period is so short, far more elliptical orbits can exist and permit habitability. Planets take a while to warm and cool, there is a reason why the longest and shortest days of the year, the solstices, are not considered the middle of summer and winter, but more like their beginning. If a planet orbits its sun every 30 earth days, it probably can get away with getting much closer and further from its star without boiling or freezing those on it, and this eccentricity might cause quite a weather cycle, as the intensity would vary a lot and the regions over which the Sun actually rose and set would be pretty large.
We’ve got a few extremes for each of these variables incidentally. On the extreme side of Mass, the least massive red dwarf offers day and year lengths to planets of only a few days while the most massive could have years as long as 90 days, and can have regular days too, large red dwarfs probably don’t have tidally locked habitable planets but still aren’t likely to have Earth-like seasons because even though they are as likely to have axial tilting as Earth, they just don’t have the year length to allow a large rise or dip in temperature. On the other end of that spectrum, the smaller red dwarfs can have seasonal variations simply from highly elliptical orbits, if the planet is eccentric, simply because the orbit is so short in duration, and would see strong weather.
Tidally locked worlds with highly circular orbits though might be quite stagnant weather-wise, one eternally shining sun with little variation in position or strength in the sky would just blow all the moisture slowly to the dark side, so that you might get a dead barren sun side, a glacial dark side, and a twilight band full of meltwater and icebergs. This is one example of what we call an Eyeball Earth, and they can also come in the hot eyeball and icy eyeball varieties. It depends on how deep in the habitable zone a planet is, as for instance one barely inside the outer edge might be a big frozen iceball whose center region under the eternal noon is melted out ocean and islands, or the reverse where the sun side is totally unlivable but the dark side is cool enough for liquid water, and it just doesn’t have photosynthesis running its ecology. You could have separate ecologies too, same as we have many biomes, including those which are in utter darkness like the ocean depths. You could have some effect like marine snow too, where photosynthetic plants and organisms growing in the twilight band are pushed by winds from the sunny band feeding a weaker detritivore ecology on the night side. It might not always be dark there too, even ignoring the twilight bands, such planets can have their own moons, though we would expect them to be less likely to tidally lock if they had a large moon.
They also could be pretty tightly packed planets, much like the Galilean Moons around Jupiter, nearly planets in their own right, and they’d make for pretty bright objects in the night sky of each other if Jupiter were a Second Sun, something we discussed more in our Summer on Jupiter Episode. You can also have binary star systems, where the planets might orbit a red dwarf which was a smaller and distant companion of another larger star, or planets in orbit around a pair of close binary red dwarfs. For that matter though, speaking of binaries and moons of gas giants, you could have habitable moons of a gas giant or brown dwarf that was orbiting a red dwarf. Jupiter mass compared to the smallest of red dwarfs is comparable to the Moon’s Mass to Earth, and there’s a broad range for satellite options between a near binary star of a brown dwarf and a smaller red dwarf to a larger red dwarf and something more in the Neptune Mass or Super-Earth category, but all permit plausible scenarios for Earth-sized moons stably orbiting a giant planet itself orbiting that star. Such a moon is likely to be tidally locked to its gas giant planet but still gets regular sunlight, and thus would have a day in the sense of the sun rising and setting, anywhere from a few days to a few months, while it orbits its planet, which it would not necessarily need to do more often. We have some weird options for distant moons and Lagrange point objects.
We usually hold that a moon needs to be no more than a ninth as far from its planet as that planet is from its sun to be stable. Our Moon orbits every month, at about a 400th the Earth-Sun distance, but should theoretically be stable at that distance even if Earth moved to say 0.025 AU, 10 moon orbital distances from the Sun, where the Sunlight would be 1600 times brighter.
A star a 1600th as bright would only have about 9% of our Sun’s mass, and Earth, at that distance from that mass of star, and would orbit it about every five days, while our own moon would still orbit us every 29 days. So you could have habitable moons orbiting their planet much faster than the planet orbited that star, as we’d expect, tidally locked to the planet but getting a day and night as they rotated relative to the star, and you could get some rather peculiar examples where that lunar orbital period was longer than the planetary one. But a fairly critical concept to settling red dwarf systems in the first place is that things will be tighter packed than our own solar system.
Their habitable sphere will be between 3000 to 300 billion times smaller in volume than our solar system’s, space is still pretty empty, but not as empty as it is here. That matters a lot for a Dyson swarm for instance. One thing we often comment on is that it’s easy to find enough raw materials for a simple solar collector Dyson Swarm around our Sun but resources get pretty scarce as you try to replace all of that with artificial rotating habitats for people to live in.
We also point out that even though a Dyson swarm englobes an entire star, it’s still composed of individual components distanced from each other by enormous spaces that make collisions unlikely and the whole thing less dense than air, even pretty thin air. This is less true of a Red Dwarf. For instance, while we’d expect red dwarf systems to have less building material orbiting around them in the forms of planets, moons, and asteroids, they also need way less of that material for a complete Dyson, between 0.01% and 7% of our Sun’s, given that is the range of red dwarf luminosity. It does need to be a great deal denser though, the smallest M9 Red Dwarfs might need ten-thousand times less building material, but they have 300 billion times less volume to stick them in, making them 30 million times denser than a Dyson around our own Sun. That happens to be ideal for connectivity, if you’re building a civilization mostly focused on short light speed internal communication, like a giant digital brain or a vast uploaded civilization.
They are very easy to Dyson Swarm up, whether it's for such digital architecture or for more classic space habitats. Indeed, given that hybrid Dyson approaches are likely to be common, you might get around the redder light spectra, if that bothered you, by filtering it into your habitats, letting mirrors with selective reflectivity to certain wavelengths send in light so more of the blue got into the habitats and more of the red and infrared went into power generation. You can do that same trick with orbital mirrors to heat up the light spectrum hitting the planet, and to get light onto the dark side of tidally locked worlds too.
This is the other reason the twilight-banded tidally locked worlds might be less common too. We might find a lot of them where they were initially airless, and we could comet bombard them to get that atmosphere, red dwarfs should have plenty of cometary bodies too, but it takes a lot less effort to setup orbital mirrors and shades to patch the eternal day and night side issue than it takes to bring in even a very modest amount of air and water to permit shallow lakes. Planets already having air and water are not likely to be tidally locked, as we explained earlier, and the airless ones are just so easy to fix the lighting on with mirrors and shades compared to adding in that air and water that you’d assume if they terraformed them to have that air and water they would have fixed the lighting issue too.
So what is the lure to colonize red dwarfs in the end? Well, they are likely to be more the equivalent of a tight and fertile archipelago in comparison to the more desolate but large continents of bigger stars. Everything is tight packed and easily gotten. There’s less of it, but it will be around a lot longer, and more to the point, the people who colonized lands across our own seas didn’t skip islands in favor of larger landmasses, indeed in terms of land value, area for area islands tend to be more valuable than continents by a wide margin. Someone will want a claim on the red dwarfs because they are valuable, less valuable than a yellow dwarf, but still immensely valuable, and far more numerous, and far longer lived.
The greatest and mightiest of civilizations will be around the larger stars, but the longest lasting ones will endure around the red dwarfs, long after those mighty civilizations around the larger stars are nothing but distant memories. As a heads up, our sponsor for this episode, curiositystream, has extended their holiday promotional deal till the end of the month of January, and they’re currently offering 41% off subscriptions to their streaming service. Curiosity Stream has thousands of fun and educational videos, including awesome science content like their Exoplanets Documentary discussing some of the worlds we might find around stars like those we looked at today.
However, they’ve also partnered up with us at Nebula, our Streamy-Award Nominated streaming service, to offer Nebula’s content along with their own, if you sign up at the link in the episode description. That means you will not only get Curiosity Stream, and at a 41% discount, but also lets you catch SFIA episodes early and without ads, and help support our show while you’re doing it, as well as see our Exclusive Coexistence with Alien Series and other great content from our sibling shows. Again you can get a year of both Curiositystream and Nebula for less than $15, get to support the show and see our episodes early, and get all that for less than $15 by using the link in the episode’s description. Some of you might have noted we had previously had today’s episode listed as airing at the start of January, and we moved it forward a week, swapping it with our episode Zero Gravity Civilizations which will now be coming out next Thursday, and we will be following our look at colonizing dwarfs stars up by looking at colonizing Giant Stars at the end of February. Before we finish January though, we do have our Monthly Livestream Q&A coming out this Sunday, January 31st, at 4 pm Eastern Time, and as usual we will be answering your questions as you submit them to the livestream chat. If you want alerts when those and other episodes come out, make sure to subscribe to the channel, and if you’d like to help support future episodes, you can donate to us on Patreon, or our website, IsaacArthur.net, which are linked in the episode description below, along
with all of our various social media forums where you can get updates and chat with others about the concepts in the episodes and many other futuristic ideas. You can also follow us iTunes, Soundcloud, or Spotify to get our audio-only versions of the show. Until next time, thanks for watching, and have a great week!