Why do we live around a yellow (*technically
white but appears to us yellow!) star, and not a red one? It’s the kind of question you’ve probably
not given much thought to before, but in fact it’s a challenging one to explain. Count up the stars, and one finds that red
dwarfs greatly outnumber Sun-like stars, but then why don’t we live around one? Join me today as we explore a new research
paper from the Cool Worlds Lab in which we outline the Red Sky Paradox. Paradoxes hold a special fascination to us,
because they represent riddles to be solved, brain teasers that tickle the intellect. The term paradox is rather broad, with one
example being self-contradictory statements, for example saying “This statement is false”,
the famous Liar’s Paradox. But more generally paradoxes simply mean statements
that run contrary to expectation. For example, the Tea Leaf Paradox describes
the phenomenon where the tea leaves in a stirred cup of tea migrate to the center of the bottom
of the cup, but physical intuition tells us that the centrifugal force should push them
out to the edges. In 1857, physicist James Thomson suggested
the solution may be related to secondary flows caused by friction with the container, which
was rigorously proved by Albert Einstein in 1926. Was there anything Einstein couldn’t solve? One of the most famous paradoxes in science
is the so-called Fermi paradox. Whilst working at Los Alamos National Laboratory
in 1950, physicist Enrico Fermi was discussing the probability of alien life in the universe
when he famously asked, “so where is everybody?”. Fermi pointed out a seemingly logical contradiction,
if there are billions of stars and life occasionally spawns, surely we should see evidence for
any aliens? The Fermi paradox is the theme of dozens of
books, hundreds of lectures and countless after dinner conversations, trust me. Of course, it’s only really a paradox if
one accepts the premise, that alien life is common and should be evident to us. And so efforts to resolve the paradox focus
on trying to refute some aspect of that premise, for example that life is perhaps rare, or
that aliens are shy. There is literally a hundred different speculative
resolutions to the Fermi paradox, each of which in some way challenge that premise in
different ways. This really isn’t any different from the
Tea Leaf paradox where the premise that centrifugal forces dominates was ultimately wrong, and
perhaps one day someone will prove one of the resolutions to the Fermi paradox. So now we have this new Red Sky Paradox that
I’m introducing in my paper, which I’m happy to share was just published in the prestigious
Proceedings of the National Academy of Sciences. To understand this paradox, let’s begin
by considering three simple facts established by modern astronomy. First, M-dwarf stars, that is stars with masses
between one twelfth and one half that of the Sun, are the most abundant type of star in
the Universe. Indeed, they make up about three-quarters
of all stars. Second, these M-dwarfs live far longer lives
than Sun-like stars, in fact they’re so long lived that some will last ten trillion
years, whereas the Universe is only 13.8 billion years old. And third, M-dwarfs apparently have no shortage
of planets, especially rocky planets. Indeed they seem to have just as many habitable-zone
Earth-sized as Sun-like stars do, and possibly even more. So as we sit here, thinking about these three
facts under the light of our yellow star (*technically white but appears to us yellow!), one is faced
with an apparent logical contradiction. If these red M-dwarf stars are far more common,
far longed lived and have just as many rocky planets around them, surely we would expect
to be born around one of them, under a red sky? For years, I’ve laid awake at night wondering
this, why is why recently I felt compelled to finally write it up. Now immediately, you might wonder, as I did,
perhaps the surprisingness here isn’t so extreme. If the odds of being born around a yellow
star (*technically white but appears to us yellow!) versus a red star are say 1:5, we
could perhaps write this off as a mere coincidence, an outcome which is somewhat uncommon but
not highly unlikely. So what are the odds of this being mere coincidence? We know that M-dwarfs stars are, altogether,
about 5 times more common than Sun-like stars. For the astronomers out there, I’m being
generous here by defining Sun-like to include F, G and K type dwarfs. Ok so 5:1, not so extreme. Now let’s factor in the time aspect. M-dwarfs live for, on average, about 20 times
longer than Sun-like stars. Yes some live for a thousand times longer,
but when we average over the entire population the factor drops down to 20. If we simply combine this factor of 20 with
the earlier factor of 5, that gives us a factor of 100. So now things start to become rather uncomfortable. In fact, we are really violating the famous
cosmological principle here, which assumes that our astronomical view of the universe
is typical. Now before we start thinking about resolutions
to this, and certainly we can think of some, let’s just step back to that factor of 100
for a second. In that calculation, I’m really assuming
that a star lasting 20 times longer gives you a 20 times greater chance of spawning
an intelligent civilisation, all things beings equal. This makes sense if life and intelligence
evolution is a slow, gradual process that infrequently occurs. It’s a lottery and M-dwarfs are simply buying
more lottery tickets. But what if planets typically spawn civilisations
within a billion years or so, a fast and inevitable process. This clearly erodes the longevity advantage
that M-dwarfs enjoy. If so, it doesn’t matter as much whether
you live for a trillion years or a billion, either is generally sufficient time for a
civilisation to pop up. This discussion of timescales might remind
some subscribers of earlier work we published which calculated the statistical probability
distribution of this timescale using our evolutionary record as a guide. Using that approach, we can ask, how likely
is it we would live around a Sun-like star versus a red dwarf, depending upon this evolutionary
timescale? As shown here, if the timescale is fast, a
hundred million years for example, the odds are about 10%. Unlikely but not incredibly so. Following the curve to longer timescales though,
the temporal advantage of red dwarf stars kicks in and we end up with the 1% odds as
discussed earlier. So which one is right? 10% or 1%? Well with current data it’s impossible to
solve this completely, but what we can say is that everything in this red region is actually
inconsistent with Earth’s evolutionary history. That’s because it would entail complex multicellular
life emerging on the Earth far earlier than the geological record shows it. Incidentally, life everywhere also exacerbates
the Fermi paradox. The long timescales, shown in blue, are actually
all consistent with Earth’s history. This might surprise you because if the typical
timescale is a trillion years, surely that’s inconsistent with our emergence within a few
billion years? The answer to this is selection bias, we’re
only here because success occurred and so we can’t actually say that a trillion-year
emergence timescale is any less likely than billion-year timescale. If that sounds confusing, don’t worry, this
is highly unintuitive. We did a video about this a while back using
prisoners picking locks as an analogy, so check that out for a deeper explanation. Now coming back to our graph, there is a much
larger region of this plot consistent with us being a 1% fluke than a 10% fluke, in fact
you really have to fine tune the evolutionary timescale quite a bit to end with the 10%
number. Yet more, if we consider that planets can
spawn multiple civilizations, not just one, then the longevity factor of M-dwarfs gives
an unequivocal advantage. So the 1% fluke value really looks to be the
correct value. OK, 1%. Well, maybe that’s the answer. We really are just a fluke. It’s certainly possible and indeed my paper
formally states this as a Resolution 1 to the Red Sky Paradox. But frankly it’s a not a very satisfying
resolution. We’re basically just saying it’s luck,
no explanation whatsoever, and a statement that flies in the face of the cosmological
principle. So I’m leaving this on the table, but it’s
not my preferred resolution. Perhaps like the Fermi paradox, and indeed
the Tea Leaf paradox, one of the assumptions built into the Red Sky Paradox might be wrong,
in such a way that it resolves the paradox altogether. Aside from the fluke possibility, my paper
suggests three additional possibilities, resolutions 2, 3 and 4. Before we dive into those, I want to quickly
thank the sponsor of this video, that’s Brilliant.org. Whether it’s research like this or working
as a data scientist, excellence requires mastery of core skills. brilliant.org is a great way learn in a fun
interactive way at home, because let me tell you as an educator that lectures shouldn’t
be the only way you learn. The act of doing cements skills and brilliant.org
have developed a terrific platform to help you do just that. Courses span high school math to software
development, from engineering to science, with a new astrophysics course including sections
on cosmology and our favourite exoplanets. You can subscribe, and help support us out,
by using the link brilliant.org/CoolWorlds and the first 200 of you to use that link
will get 20% off the annual Premium subscription. Now back to the Red Sky Paradox, in the second
half here I want to talk about these 3 other resolutions. They’re pretty straight-forward so let’s
go through them swiftly. Resolution 2 is that life or by extension
intelligent life just rarely emerges on M-dwarf planets. In fact, to resolve the paradox, it occurs
about 100 times less often, or even less. There is some theoretical ideas out there
motivating this. For example, M-dwarfs spit out high energy
flares, perhaps that impinges life. As another example, they take far longer than
Sun-like stars to settle down into a mode of stable output, this could actually desiccate
nearby planets of their precious water leading to Dune-like desert worlds. As a final example, M-dwarfs rarely have Jupiter-sized
planets, perhaps those are crucial for life, by protecting us from asteroid impacts. We don’t know which or indeed if any of
these are true, but we have good reasons to be suspicious that M-dwarfs are just as good
for life as yellow dwarfs. Resolution 3 is somewhat similar to 2, except
that it looks at the temporal aspect more closely. Perhaps M-dwarfs and Sun-like stars are equally
likely to start life, it’s just that Sun-like stars are the ones who enjoy far longer windows
of opportunity. This seems counter-intuitive because M-dwarfs
live so long, but maybe the actual temporal window when life is possible is very short. In fact, if it were 5 times shorter than Sun-like
stars, that would be enough to resolve things. One possibility is that Mars-like planets
on the outer edge of the habitable-zone become habitable during that early, violent M-dwarf
phase, where the extra energy could actually be useful. Indeed, if we studied young M-dwarf planets,
we could potentially test that hypothesis directly. Now finally resolution 4 is that habitable-zone
rocky planets are in fact rare around M-dwarfs. A paucity of pale-red dots. Now you’re probably thinking, hold on, you
said earlier that they seem to have just a many habitable rocky planets as Sun-like stars. Yes, *seem* to. But appearances can be deceptive. You see, M-dwarfs come in a diverse range
of masses, and frankly we’ve only really surveyed the heaviest examples for planets. That’s because they’re the brightest and
easiest to look at. But what if the smallest red dwarfs are different
beasts and rarely have habitable-zone planets? If planets here are a 100 times less common
we’d have our solution. This works out because the smallest red dwarfs
are in fact the most numerous. So if they’re different it’s going to
be dominant. Again, the good news here is that this is
testable, we are now looking at these smaller stars, and already have detected some systems
like Proxima Centauri b and the TRAPPIST-1 system. It’s still early days, but I’m little
skeptical of this resolution given these early examples. So there you have it, four resolutions, luck,
inhibited evolution, truncated habitable windows, or a paucity a pale-red dots. Right now, we really don’t know which one
is right. Just like how we don’t know which resolution
dissolves the Fermi paradox. Now, you might be thinking this Red Sky Paradox
is purely academic but it actually has major practical consequences. Figuring this out is crucial because it would
reveal where we should be pointing our future life hunting telescopes. In resolution 2, M-dwarfs are perhaps best
avoided altogether. In resolution 3, they could be worthwhile
but only if we target young systems. And in resolution 4, we should focus on the
heaviest M-dwarfs, skipping systems like Proxima Centauri. Who knows, one day, we might plan a interstellar
mission to a distant star. If life is off the cards for 3/4 of all stars
in the universe, that seems like something we should probably figure out before we leave. So let me know down below, what is your favourite
resolution? Do you have others in mind not discussed here? I want to finish by just saying a special
thank-you to our supporters. This research was supported by many of you,
donors to our research team at Columbia University. The idea is that we conduct original research,
make videos about them, which then lead to many of your supporting us, which then leads
to more research and more videos. A cycle of research and outreach. It’s an pretty unique way to help fund research
and I’m so grateful to all who have believed in this dream. So if you want to support a research team,
not just a YouTube channel, then please do consider helping us out. So until next time, stay thoughtful and stay
curious.
