Can We Engineer The Sun?

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
(ambient music) - Well, we used to look up in the sky and wonder at our place in the stars. Now we just look down and worry about our place in the dirt. - In antiquity, our ancestors looked up to the stars and saw a perpetual, eternal canvas, one inhabited by gods, deities, and mythical legends. The heavens were immortal, untouched by the ravages of time that would cause our buildings to crumble, our art to fade, and our faculties to decline. On earth, everything dies, but those sparkling stars seem to be the exception, a source of solace in a brutal cosmos. When we first learn that stars die, it troubles us. It feels wrong. Children often panic at the thought, but truly, the light in our sky, the engine of our planet's biosphere, is gradually stepping towards a dramatic end, albeit one billions of years from now. Before it dies, the sun will boil off our oceans, vaporize our planetary surface, and eventually smother the earth into itself in a final embrace. And it's not just the sun. Whilst the pathways might be different, a similar fate awaits all stars. And one day in the far, far future, there will be no lights in anyone's sky, a dark veil that swallows the cosmos whole. But there are some who resist. After all, it's human nature. Even with our own lives, we're compelled to try and extend our lifespans through medicine, improved health, and diet. So could we somehow extend the lifespan of our sun? Could we buy ourselves more time through our ingenuity and technology? Thinking of our planet's demise, usually folks assume that it's the engulfment by the sun that defines the end, for in around seven and a half billion years, the sun will appear unrecognizable as a swollen red giant star, one that has devoured the inner solar system. But actually, earth's engulfment by the sun is not really the problem. Well, don't get me wrong, that is definitely a problem, but actually, the end for complex life is far nearer, because as the sun ages, its luminosity rises, and it's that increased power output that really spells the doom for life. You see, this leads to greater insulation on the earth and hence greater weathering. I mean, essentially, more energy means more evaporation, which leads to more precipitation. Rain dissolves carbon dioxide out of the air, forming carbonic acid, and it's that acid which eventually leads to the formation of solid carbonates and limestones. The net effect of this is that making the earth warmer causes carbon dioxide levels to drop. Yes, I said drop. And no, before you start, this won't save us from climate change, because this process takes millions of years to complete, although there are some ideas to artificially speed up, known as enhanced weathering. Now, whilst lower carbon dioxide levels might sound great in our current era of a climate crisis, it's actually the death now for complex life. In just under a billion years from now, which is still some six and a half billion years before the sun engulfs the earth, CO2 levels will drop close to zero, terminating photosynthetic life and collapsing the global food chain. This is the end of complex life on earth. I've skipped through the gory details of our demise, but if you have the stomach for the more detailed account, then be sure to check out our previous video, "Watching the End of the World." What all of this means is that if we want to maintain a habitable earth for longer, then we need to somehow keep the irradiation striking the earth stable over that time, and the sun doesn't wanna cooperate with that. So that begs the question, is there something that we could do, or perhaps something that our more technologically advanced descendants might be able to do? There are basically three options. Option one is just to pack up and move the entire population to another solar system, a journey of daunting scale and complexity. In that choice, we sacrifice our home, watch it burn to the ground as we sail away into the sky, a final goodbye to the parent that gave us life. But options two and three refuse to let the earth die so easily and seek to preserve it. Starting with option two, the idea here, somewhat audaciously, is to physically move the earth further back from the sun. And it's this idea which has garnered the greatest scientific attention in recent years. In 2001, Don Korycansky authored a provocative paper that suggested that our distant descendants could indeed pull this off using asteroids. The idea starts by first finding a very large asteroid, something like the size of 324 Bamberga, but one in a highly elliptical and distant orbit, such that it periodically swings into the inner solar system every few millennial or so. The asteroid's orbit is then manipulated, perhaps using reflective paint, or even just smashing into it, like the DART Mission recently did. Either way, this manipulation causes the asteroid to make a near pass of the earth. In doing so, the asteroid will whip around the earth, performing a gravitational sling shot. In such exchanges, orbital energy is transferred between the asteroid and the earth, with the direction depending upon the angle of attack. Korycansky cleverly arranges the slingshot such that the asteroid's orbital energy is lowered, but the earth gains. Now, because the earth is so much more massive than the asteroid, this has a huge effect on the asteroid's orbit, but just marginally increases the earth's orbital distance around the sun. Congratulations, you have just moved the earth back a tiny bit, and thus slightly reduced its temperature. Now, to really make a difference, we're gonna have to do this many times, so rinse, wash, repeat over and over. And the real elegance of this scheme is that that asteroid can be reused to this end. The first step is that the asteroid's orbit is nudged with thrusters towards Jupiter, where it performs another slingshot to regain its lost energy. Additional adjustments are then made to the asteroid's orbit to ensure that it returns back to the earth in its next roundtrip. And in fact, the authors propose that the outer planets, such as Saturn, could help here. For an asteroid with a period of, say, 6,000 years, it would perform 1 million of these close encounters with the earth over the next 6 billion years. Each one pushes the Earth's orbit back a little until we reach an orbital distance comparable to Mars's current location. That distance is important, because in 6 billion years' time, the sun will be about 2.2 times more luminous. So pushing the earth square root 2.2 times further out, which is where Mars lives now, means our irradiation stays the same as the modern value. The scheme is brilliant, and I remember being quite inspired by this paper when I first read it many years ago, just the idea that we can use the components of our solar system to modify our environment. Astroengineering. In fact, many have even suggested that we could use this to solve climate change in the short term. However, it has to be said that this idea comes with some considerable challenges. Perhaps most alarmingly, we're deliberately aiming one of the largest asteroids in the solar system to close proximity of our fragile home planet. Indeed, that asteroid would need to be so large that if the unthinkable happened and we miscalculated, leading to a strike on the earth, it won't just trigger a mass extinction. It would completely exterminate life on earth. Game over. The authors are certainly aware of this, and acknowledge it in a rather haunting final sentence of their paper, saying, "This danger cannot be overemphasized." In this scheme, it's also deeply unclear what happens to Mars, because we're pushing earth further back into a Martian-like orbit, and two planets cannot stably occupy the same orbit at the same time. The authors also briefly comment on this, somewhat enigmatically, by stating, "The fate of Mars in this scenario remains unresolved." This orbital overlap is important, because we know that the inner solar system can easily become chaotic. For example, Batygin and Laughlin showed in 2008 that instabilities are plausible on a time scale of just 1 billion years from now. And now, in this asteroid deflection scheme, we're really stirring the pot by pushing planets around into unnatural positions. Okay, so you might think that we are finally done with our list of side effects here, but there is just one other small thing. I mean, it's just a minor hiccup, really. We lose the moon. Now, maybe you're okay with losing the moon, but I'm not. - I got nukes out the yin-yang. Just let me launch one, for Christ's sake! - Sir! Are you suggesting that we blow up the moon? - Would you miss it? Would you miss it? - Look, the moon isn't just a pretty light in our night sky. Its existence may be central to our survival, because, as shown decades ago by Jacques Laskar, the moon is thought to stabilize Earth's spin obliquity. That's the tilt which gives us our seasons. Take the moon away, and climate stability could run wild, transforming our planet into a nightmarish environment for civilization. With options one and two covered, is there any other alternative? Is there some other way that we could save our home? Well, that's where my research group stepped in, the Cool Worlds Lab at Columbia University. The asteroid deflection business made a lot of headlines when it was first published, but buried in the literature is an alternative, an idea that we felt was underdeveloped, and that is starlifting. Could we reduce the mass of our sun? The first published discussion of starlifting was in a fairly obscure conference proceeding in 1985 by David Criswell with, then, Greg Matloff reintroducing the idea in 2017. Stars naturally lose small amounts of mass all the time, in particular through their stellar winds. So Matloff proposed enhancing that mass loss rate through the use of lasers aimed at the star. One could also imagine siphoning mass off the surface with a compact massive object in close orbit, causing so-called mass overflow. Friend of the channel Isaac Arthur has a great video introducing this work on starlifting that I'll link to down below in the description. But despite this work, it's really not clear how much mass we have to take off the sun, nor exactly how that affects the solar output. Stellar evolution is an notoriously challenging field, and I knew this wasn't gonna be an easy problem to solve. Each year at Columbia, we welcome new graduate students to our Astronomy Program, and professors such as myself line up in a shark tank setting to pitch new research ideas to them. I remember last year how I spoke to the students about all the different exoplanet projects we were working on, but I mentioned that, "Hey, on the side, we also like working on "these wacky little side projects." Afterwards, one of the students, the brilliant Matt Scoggins, came up to me and said, "Tell me more." So for the last year, Matt has been chipping away at this problem with me, and we're thrilled to be able to share our new work with you today, research that finally reveals what it takes to save not just ourselves but the sun itself. Perhaps one day, then, just as the sun has provided life unto us, we can use our ingenuity to return the favor and give back to the sun the gift of life. But how exactly do we propose to do this? - The key to making a star live forever, or at least a lot longer, is balance. You have to find just the right mass loss rate such that the temperature of your planet remains the same at all times. In simple terms, if you remove mass from the star, it reduces the gravitational pressure at the center, sort of like this. (balloon hissing) Reducing the pressure at the center reduces the rate of fusion, and reducing the rate of fusion reduces the luminosity output. This should intuitively make sense. We know that small red dwarfs like Proxima Centauri produce far less output than our sun, but the other effect in play here is time. As stars age, their cores accumulate more and more helium ash in their centers. Since helium is denser than hydrogen, this raises the internal pressure, and hence increases the rate of fusion, luminosity output. So our challenge is this balancing act where we need to find the rate of mass loss such that these two effects cancel each other out, and we produce a stable luminosity. You can immediately see that we're creating a kind of strange star over time, because after doing this for a sustained period, you have a much lower mass star, but one whose luminosity equals that of our modern sun. And the only reason that that statement is true is because this engineered star is what we call fairly evolved. It's aged to the point where its output is enhanced and it's teetering on the edge of becoming a red giant, but we keep spooning mass off at just the right rate to prevent that from happening. You can see this behavior on this plot. Let's put a dot down for the sun and evolve it forward in time. Without any starlifting, it would naturally do this, barely change in mass, but increase in luminosity, represented by the changing color shown here. Eventually, the track stops, because the star transitions into a giant phase. We can also put on tracks for stars of different masses, which are the other horizontal lines shown here. However, in starlifting, we play this game where we pull just enough mass to keep the luminosity the same. All of these dots here have the same luminosity as our modern sun, so we just need to cause the sun to follow this new black track, an engineered one. What's interesting is that even with starlifting, the game eventually ends, because we're gradually getting closer and closer to that red giant termination point. So the game eventually ends, but you can see that we've actually extended the sun's main sequence lifetime to 20 billion years, whereas it would naturally only go to 10 billion. That's a 15 billion-year future of stable luminosity for us. There's no need to move earth, no need to do anything else, and the best part of this is the mass loss rate required for this is pretty timid. It's only 2.5% the mass of Ceres each year, which is the largest asteroid in the solar system. This is way below the rate that Matloff predicts should be possible with lasers, yet more, the energy cost to do this is easily supplied by the star itself. Just harvest the sun's output and feed a fraction of it back to create the necessary mass loss. So our paper shows, using the latest stellar evolutionary models at our disposal, that starlifting can sustain a stable luminosity for tens of billions of years. But much like the asteroid deflection scheme, this is something with considerable challenges. First, it isn't clear how we would actually implement starlifting. If we wanna keep the material near the star, we need some sort of stellar excavator, or if we want to eject the material, as Matloff suggested, would earth need to be protected from this increased stellar wind? Of course, there is also the small issue of currently being nowhere near technologically advanced enough to starlift, and it's unlikely to happen in the next few millennia, but it gives us something to hope for. (ambient music) - Starlifting could be the key to sustaining the sun for far longer than its natural life allows. It's a staggering thought that human ingenuity could one day greatly extend the lifespan of an object that massively dwarfs our tiny pale blue dot. Although our paper is largely satisfied with the result that we could theoretically extend the sun's lifespan by 10 giga-years, one can't help but ask, "Could we go further?" Could we somehow tune this method such that we could reach hundreds of billions, perhaps even trillions, of years? One option is to pick on a different host star. The sun is rather old and massive, but what if we picked a smaller red dwarf star instead? Certainly, one could apply our method to smaller stars, but there's a limit to how far one can go. The candle that burns half as bright burns twice as long, so the stars that will last the longest naturally are the smallest red dwarfs, especially those barely massive enough for hydrogen fusion in the first place. Sadly, these skinny little or late end dwarfs don't really work with starlifting, because these stars don't have a lot of fat to spare. Remember, these stars are barely massive enough for hydrogen fusion in the first place. When it comes to maximum life extension, we found that the best stars were orange dwarfs, stars in between the mass of our sun and the smallest stars. With initial masses of about half that of our sun, there's plenty of excess fat to skin off here over time. In fact, it's possible to get these stars to live for trillions of years, and crucially, that's trillions of years of stable luminosity output. So civilizations living around these orange dwarves may have struck the cosmic jackpot, because their home star could be engineered to live more than a trillion years. Who knows? Perhaps civilizations even migrate to these stars for that reason, but do we really have to abandon the sun to last a trillion years? Maybe not. One final trick we conceived combines the asteroid deflection scheme from earlier with starlifting, a cosmic one-two knockout punch for longevity. So how does this work? One way to make the sun last far longer is to take off more mass than we previously conceived. Remember, originally, we take off mass at just the right rate such that the sun's luminosity remains stable. Instead, let's now take off more mass, which will let it live for longer. But the consequence of that is that the sun's luminosity will now fall. This would ordinarily be very bad news for life on earth, but using the asteroid deflection scheme from earlier, we could move the earth, but now not outwards, but rather inwards, preventing freeze-over. By combining the two, there's no clear barrier to having your cake and eating it, a sun that could last a trillion years or more. Do other civilizations do this? Will we ever do this? Will our technology ever develop to the point where this is even possible? Those are all great questions, and ones that we do not have the answers to. Research progresses in step-by-step increments. Previous work showed that we can imagine ways of starlifting mass off the sun, and our work shows how it could greatly extend our star's lifespan, but we've only really scratched the surface of what's possible in astroengineering. And it makes you wonder, could we even tell whether a distant star was engineered or not? Sci-fi author Karl Schroeder once paraphrased Arthur C. Clarke by saying that, "Any sufficiently advanced civilization "is indistinguishable from nature." The original version said "magic." In this research, we touched on some ideas for how to detect engineered stars, but frankly, it remains a major challenge. Ultimately, that means for now that more research is needed down the road. This isn't the kind of research which is easy to fund, but to some of us, questions about life and the universe are foundational. They're endlessly fascinating, and if you believe that, you can support our work by bumping up this video, subscribing, and, if you have the resources, you can even support our research program by using the link in the description. If you have any thoughts or questions about starlifting, please do put them down below in the comment section. I'll do my best to get back to you. So until next time, stay thoughtful and stay curious. (light electronic music) Thank you so much for watching this video, everybody. If you enjoyed it, be sure to like, share, subscribe, all the YouTubey stuff. And as I said, this research was supported by our donors. Let me thank our latest two, that is, Richard Williams and Joseph Gillmer. Thank you so much for your support, guys. So have a cosmic awesome day out there, and I'll see you 'round the galaxy. (music continues)
Info
Channel: Cool Worlds
Views: 216,842
Rating: undefined out of 5
Keywords: Astronomy, Astrophysics, Exoplanets, Cool Worlds, Kipping
Id: IY0KWLanlLM
Channel Id: undefined
Length: 21min 58sec (1318 seconds)
Published: Sat Oct 08 2022
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.