Hacker News new | past | comments | ask | show | jobs | submit login
NASA proposes a magnetic shield to protect Mars' atmosphere (phys.org)
475 points by cyanbane on Mar 5, 2017 | hide | past | web | favorite | 173 comments

This is incredibly exciting but a few important details are missing.

The first is how big does the structure need to be? I can buy a 1 Tesla magnet online right now but that's probably not what they're thinking of. Would we need a city-sized coil or something like that?

The second is the time scale. They say that the temperature could rise by 4 Celsius and trigger a greenhouse effect, but is that an immediate effect (10 years or so) or century-scale effect? I'm hoping the scientists put out a paper because I'd love to learn more about the specifics of their proposal.

For people asking about the amount of energy we're talking about, the energy density of a magnetic field is |B|^2 /(2mu_0). It appears that the described structure has something on the order of the cross section of Mars, and I'll assume that it's depth is in that ballpark as well. For a fictitious uniform field that would require on the order of 1.6e19 J, which is roughly 1.2 times the total electrical energy output of the US in 2001:



Assuming you had no losses, so say ideal superconducting coils, that's how much energy you have to dump into magnetic field. You can do this as slow as you like, so with 1/10 the US electrical energy production it would take 10 years to build up that magnetic field. Hypothetically, if you had thin-film plastic PV cells (like cellophane thin to be reasonable to build and get into space) with 100% efficiency covering the footprint of this system (Mars) you could generate enough power to charge up this field in 471 seconds:


Of course the big story here is that efficient thin film solar in space could generate obscene amounts of power.

I'm not sure on the depth of the field volume required, so that might make it easier, too.

I wonder if a very large array of solar panels would not also act like a solar sail that would would catch enough solar wind to escape the pull of the L1 Lagrange point.

Not a problem. The L1 point is where the gravitational pull of the Sun is balanced by the gravitational pull of the planet plus the centrifugal force; if you move slightly closer to the sun, its gravity will balance planetary gravity, centrifugal force, and the pressure on the sail.

I think this would happen if L1 was a stable lagrange point, like L4 and L5. That's not the case though, L1, L2 and L3 are unstable. To put it differently, ignoring the centrifugal force, at L1 the gravitational pull of Mars is equal to the gravitational pull of the Sun; when you move the sail towards Mars, the pull towards Mars increases, the one towards the Sun decreases, and the resultant points in the same direction as the pressure on the sail (i.e. towards Mars). Nothing is balanced, in the end the sail falls towards Mars.

No, the fact that it's not stable isn't what's being talked about. It's merely the fact that an additional force from the solar wind would act to shift the equilibrium point of L1 slightly towards the sun.

Makes sense. Thanks for clarifying.

Absolutely this would have an effect: https://en.wikipedia.org/wiki/Solar_sail#Trajectory_correcti...

Presumably, though, some of the power generated could be used to accelerate propellant to compensate.

How does "slowly dumping energy into a magnetic field" work?

Magnetic fields are a form of potential energy. Increasing them requires energy. And decreases release energy, to some mix of electrical current and heat.

For a deeper cut, see https://physics.stackexchange.com/questions/94273/what-force...

> is that an immediate effect (10 years or so) or century-scale effect?

We should be so lucky. Planetary scale can be rather large and slow. For instance the oxygenation of the Earth's atmosphere took hundreds of millions of years


We don't know how well cyanobacteria fared against its microbial competitors, how efficient the early photosynthetic pathway was, or how large the oxygen sinks were pre-GOE saturation so we have no idea whether planetary oxygenation actually needs to take that long. Without an existing biome that captures oxygen or large bodies of water to dissolve oxides, the process can be many orders of magnitude easier (especially since Mars' atmosphere will take significantly less volume and we might be able to survive just fine with lower air density with a higher concentration of oxygen).

It's extremely unlikely that we can pull it off in a human lifespan but I would bet that planetary oxygenation is possible given a few millennia

We have managed to warm the earth considerably in less then a century and that's without trying. I reckon if we gave it our best we could change planetary scale things even faster than that.

That was a byproduct of economically useful work though. If you can think of economically useful work that this project could generate, on an ongoing basis while it's in progress, that's greater than it's ongoing costs that would be great.

> in less then a century

The industrial revolution has been going since before 1800 https://en.wikipedia.org/wiki/Industrial_Revolution

Yes, a magnet has both a peak magnetic field (which is what is usually quoted) and a magnetic dipole moment that is roughly proportional to diameter. The magnetic field at a distance is proportional to the moments, but falls off proportional to the inverse cube of the distance.

So, yes if the magnet dipole were the size of Mars and it needed to be 5mT=500,000nT at 320 Mars diameters, then roughly (ignoring higher order effects) you would need an average 16kT around the surface poles, which isn't possible with current technology.

At 1 Tesla you would want to cover an area 50 times the diameter than Mars? That's pretty fanciful.

OK, either my simple calculations are bunk, or this is a truly crazy idea.

It's not located on/at Mars, it's located at the Lagrange point - which means much like your hand can cover entire mountains at a distance, a smaller magnetic field could shield an entire planet when positioned between mars and the sun.

Based on the drawing that they made the size is at least 1-2 diameters of mars (but there is no field there?). Since the diameter of Mars is ~4000mi that would >10,000km at 5mT.

So with the mentioned 1-2 Tesla starting field, we end up and a cube law reducing the strength 200x at 6 times the distance, and a 1.5km moment. That would require at least 2-3km^2 of 1T magnets. Ignoring the support structure that would have a volume of ~10^10 cm^3 and weigh 100ktons, but only cost $100B on Earth at todays Samarium Cobalt pricing, if the market wasn't highly distorted.

Sounds more reasonable...

sounds surprisingly realistic if you consider that this is a proposal to create an artificial magnetic shield for an entire planet.

Also, isn't there a shockwave effect?

Your hand does not cover entire mountains, only your eyes. The cross section of effective magnetic field has to cover the whole planet.

high altitude nuclear tests have created artificial electromagnetic radiation belts (similar to the Van Allen belt);


Those also damage electronics far below. In fact, that is the characteristic way of using high altitude nuclear weapons to damage communications over a continental region. But far enough away it would push a little solar wind around, for a short time.

This is where I am confused too the speaker in the video said Earth's magnetic field is 31,000 nanoTesla.

He said the best plan for a shield for Mars is a 500,000 nanoTesla field at L1 which is 320 Mars radii away from Mars.

Maybe I am misunderstanding what he is saying or pronouncing, nanoTesla i.e. billionths of a Tesla?

I imagine there is also a size component involved. The Earth's magnetic field might be 31,000 nanoTesla, but it's spread over an enormous volume. This magnetic shield would have to be much smaller than a planet, so it makes sense that it should have a much strong field.

How do they keep the magnet in place, though? Won't the solar wind pushing against it push it out of there?

I mean, I know that's a stable Lagrange point, but you have to wonder if it could end up moving and or spinning around after enough solar flares hit it and what that would do.

I think the general idea to terraform in a reasonable time is to nuke the poles to release CO2.

Even if one did this, any atmosphere on Mars would be temporary as the solar wind would eventually strip it away. The magnetic field is required for long term viability.

You're not wrong, but that's kind of a non-issue.

It takes on the order of hundreds of millions of years to lose that much atmosphere.

So you are saying we could crash some comets, boil the CO2/ice caps and solve the magnetosphere issue later?

well, this is a serious proposal to solve the magnetosphere-problem! We could build this AND nuke the plant.

Or we could at least use this as an indicator that the problem is solvable over a few centuries and start nuking. Better sooner than later!

Porque no los dos?

"can generate a magnetic dipole field at a level of perhaps 1 or 2 Tesla (or 10,000 to 20,000 Gauss)"

I'm just amused by this conversion. Who is this for? Are there people who know one unit of magnetic flux and not the other?

It is probably for the "non-scientific" reader who thinks "1 Tesla seems very low to me, that's not really impressive. Now 10.000 Gauss: I can imagine that to be pretty ... much"

I'm not sure, but everyone in atomic physics talks gauss for some reason.

Well think about it... 1 Tesla = 10,000 Gauss. When you're talking about magnetic fields in MRI coils, or planetary magnetosphere it helps to use a larger scale. When you're talking about EM forces on the scale on an atomic nucleus though, it's more helpful to speak in terms of Gauss.

But that is why we have metric prefixes, different scales call for the appropriate prefix. Magnetic field of a neutron star ~1MT, magnetic field of a human brain ~1pT

Sure, but when you actually have to do math with them, and use them every day in practical language, it starts to make a lot less sense. After all, the goal of terms of art isn't to make the language of a given pursuit less opaque to "outsiders", but to facilitate work within the pursuit.

Who is this for? The kind of person who doesn't stop to consider how much energy it would take to run this hypothetical device, and what that energy/money/resources could do on Earth. "The common clay of the new West..."

Yes, there is that. But it's also rather like off-site backup. Or at least, somewhat off-site.

Edit: Anyway, I do agree that the priority ought to be displacing carbon emissions, and removing CO2 from the atmosphere.

Yes, but if you have zettawatts of energy to pump into a space station, you probably don't need to be doing something so crude.

They'd be using local PV for the field. With PV at that scale, they'd need to beam it back to Earth. And that could be quite the weapon system.

Sure, but you'd be talking about a civilization that could afford to build it, and operate it... a civilization so advanced that they can afford more than double our current energy production just to electromagnetically shield Mars.

I saw estimates here that 10% of current US energy use would suffice to start. And this would be entirely automated, using methods not that different from JWST, except in scale. And arguably, it's less complicated than beaming energy back to Earth. But still, those could be parallel efforts.

Is there a good entry-level article or video explaining planetary magnetics? I'm an aerospace engineer and I struggle with the difference between magnetospheres, magnetotails, magnetosheaths, magnetopauses and Magneto's psychological struggles.

As another layman I find the Wikipedia article on the subject to be really good: https://en.wikipedia.org/wiki/Magnetosphere#Structure

There's nothing inherently different about a planet's magnetic field pushing the stellar/solar wind out of its way and e.g. a boat on water pushing water out of the way. Terms like magneto(sheath|pause|tail) are just labels for different parts of the "wake".

"There's nothing inherently different "

Yes there is. Magnetohydrodynamics are more complex than hydrodynamics because of the electromagnetic effects.


Yes, there is the fluid component, but in addition to that, there are the bits coming from electromagnetics. Arising from Maxwell's "a moving charge creates a magnetic field..." and "a force is induced on a charge moving through a magnetic field "etc.

Is that really inherently different, or just a more complex manifestation of the same basic principles?

I.e. Atoms repel each other because of their magnetic fields, and depending on their polarity and spin in the aggregate they can produce bigger and more complex magnetic fields, such as the ones analyzed by magnetohydrodynamics.

magnetopause sounds like a bad menopause pun

Magneto goes through "the change"

X-men comics will provide you with an intro-level understanding of at least one of those.

The article and paper lacks any indication of the amount of energy needed to run such a magnetic field. What amount would it take to run it, and what sources of energy would be viable for it?

Research NMR magnets the size of several refrigerators already exceed 8-10T; they are superconducting and could after initial run up likely run for years with no power at all. They drift down over time but it likely not be a show stopper, and the cold of space would help them minimize cryogen maintenance times - it would be an activity on the way to or from the planet maybe? Edit: though the magnet itself wouldn't require continuous power, others added there would be pressure from the sun and say it could be counteracted by orbit geometry or active force being added.

So you thave ouched on an important limitation of electromagnets; they need to be cooled. In space, the vacuum surrounding this power hungry object would be inferior to air for heat transfer; heat merely seeps out slowly radiatively.

So there would be a major improvement in keeping heat low/improving system efficiency or to transfer heat away without cooking the crew after the turn the thing on.

Is that true? A superconductor magnet needs to be cooled down to temperature (say 4K), but does it generate heat once operating? Seems like heat would come from resistance in the coil, but this resistance is zero when at operating temperature.

It would consume power by fighting against incoming particles. This would end up as heat inside it.

A permanent magnet also fights against incoming particles, and doesn't seem to consume any energy. Why would an electromagnet be any different?

Edit: the permanent magnet will feel a force applied against it in the opposite direction, of course. In space, that's have to be countered by rockets though, not electrically in the current running through the superconductor, right?

> Why would an electromagnet be any different?

Because the incoming electrically charged particle induces current in the coils of the electromagnet.

A permanent magnet does not have to deal with that - even if there is current it doesn't affect the magnetism in a permanent way.

Think about an electric solenoid: A stationary electromagnet does not consume power, a moving one does as it acts on the external magnets.

Space is not cold, at least in the inner solar system. To say nothing of internal resistance, however small, sunlight would fall on it and warm it up.

Im not sure of your context; but yes an electromagnet is a non-starter here from the get go; dealing with all the mass and power issues, bad spacey thermodynamics.

Wouldn't you be able to use somewhat normal refrigeration, then put a thermoelectric cooler on the heat exchanger side? If you could get the TEC's high temperature side hot enough, would it radiate heat more efficiently? (I'm assuming that hotter items radiate more joules of energy than cooler)

How about using something like a Dyson-Harrop satellite https://en.wikipedia.org/wiki/Dyson–Harrop_satellite as the power source

Presumably PV panels. Although a really large RTG (nuclear thermal reactor) might work too. It really depends on the amount of power we're talking about.

Does building a magnetosphere for the purposes of protection require an atmosphere, or is it all about capturing and shaping solar wind?

If it doesn't require an atmosphere, could this approach be used to build a magnetospher around the moon? It feels like colonizing the moon first is a much easier and more useful problem to tackle. Once we have a moon colony you can crack water to make fuel and then go "where ever" you'd like - other asteroids, or mars. The moon is much closer and easier to get to, though maybe people need the "excitement" that travel to Mars connotates.

IIUC, the proposal is to build a magnetic shield at a stable Lagrange point between Mars and the sun. Mars would be far away, in the "shade" created by the shield. So there's no magnetosphere around Mars. It's more like a magneto-cylinder. :)

This image explains it better than I can: https://3c1703fe8d.site.internapcdn.net/newman/gfx/news/2017...

I don't think an approach like this would work for the moon, because there's no stable Lagrange point between the moon and the sun. (And even if we did build a magnetic shield for the moon somehow, I think it's still too small to accumulate an atmosphere.)

L1 isn't a _stable_ Lagrange point. I think only L4, L5 are.

It's a common misconception that the moon is easier to get to. As far as rocket launches are concerned, getting to the moon is as difficult as getting to Mars. Worse, the moon has no atmospheric drag to assist with landing. The only way the moon is better is it's closer, which is almost a moot point because all the difficulty is tied up in getting off of Earth. And unless dramatic changes in our space flight capabilities take place, it's no easier to mount a timely rescue mission to the moon than it is to Mars - by the time we launch anything it'll be weeks or months after any accident takes place and well beyond the length of time stranded astronauts could survive.

Beyond that, the moon is generally much less suited for long term habitation than Mars is for several reasons. I won't get into details about that right here (unless you want me to) and Mars isn't exactly a cakewalk either, but the moon poses a more significant challenge in many ways.

That said, I welcome any kind of manned missions/outposts/colonies etc, whether the be on the moon or Mars or somewhere else entirely. I just believe that the payoff for starting on Mars could be much greater than doing the same on the moon.

The atmospheric drag on Mars is a bitch. Too little to help much, too much to ignore. Go too low and you burn up. Too high and you skip on the surface.

> Does building a magnetosphere for the purposes of protection require an atmosphere ...

No, it's the other way around -- for a sustainable atmosphere, one needs a magnetic field.

> If it doesn't require an atmosphere, could this approach be used to build a magnetospher around the moon?

That would have an effect on surface radiation from charged particles, but the moon can't have an earth-like atmosphere because of its low gravity. So the payoff is smaller.

More details: http://arachnoid.com/restoring_mars

The problem with using this approach with the Moon is that there is no stationary orbit that would keep the magnetic shield properly between the Moon and the Sun. Mars can use the Sun-Mars L1, but no such point exists for the Moon.

Wouldn't it be a halo orbit around the Sun-Earth L1?

Where does the water come from?

There's more than a trillion pounds of water ice on the moon.

Mixed in with rock at low concentrations. Are you proposing shipping massive mining machines to the Moon?

This is a thread about building a magnetic shield to protect an entire planet. Shipping some gear to the poles of the moon to melt ice doesn't seem that crazy in comparison.

Mars has a lot of water, it is just solid.

I don't know the specifics of solar radiation well enough to say how much power the deflection would require, but there is an easy order of magnitude calculation to figure out how much energy setting up a field that size would require. It's not the same as consumption but it's a start.

The energy in a uniform field is very simple to calculate: a sphere with a radius of 6371 km (radius of earth, diameter of mars) and 50,000 nT would store about a 10^18 joules, about 6.5% of US annual electricity consumption. At the extremal 500,000 nT that would be 10^20 joules, around 2x the global electricity consumption.

A dipole field would require ~10x more energy, a zetajoule. That's around 2x the global human annual energy consumption, including for heat/transport/industry/etc.

I don't think this is a uniform field, also this is at L1 so it could likely be much weaker than earths magnetic field and still work.

For comparison earths magnetic field is 31,000 nanoTesla and they are putting it at L1 so a weaker/smaller field to do the same job.

Correct on both points. I started with a uniform field because it requires 10% of the energy- it would mean building a distributed network of satellites, but if we're inserting exajoules of energy into near-mars orbit, making a lot of satellites is probably within our means.

50,000 nT was their suggestion of what actually starts providing relevant protection. 5,000 nT was where they started seeing it at all, which would be 10^16 joules/2 megatons of TNT. I believe the massively stronger field is needed because the protection they are looking at for terraforming would need to be much better than Earth's. We still lose 94,608 tonnes of hydrogen annually to atmospheric escape, despite having 263% stronger gravity.

Could this be reduced, by putting the shield further out?

No, the Martian Lagrangian point L1 is the only stable point that can be used. If you tried to move the blocker closer to the sun, the device would have to orbit faster than Mars. L1 is the point where the gravity of the sun and Mars cancel out.

The only other place you could have it would be directly on mars, where it would have to be much larger.

Can you use the solar power collector as light sail to nudge the orbit closer to the sun-side of L1. I would assume every km would help.

That actually might make it require more energy.

Here's a thought: Could we put one of these at the L1 point of the Earth-Sun system to protect against solar storms? If something like the Carrington Event happened to our current society, it would be absolutely devastating.

Bonus: Could we make it big enough to encompass the moon?

Would doing this reduce or eliminate northern lights? That would be a little sad.

Unfortunately this will not work. That isn't to say that the construction, placement, and powering of an artificial magnetosphere at Mars-Sun L1 is infeasible. It can totally be done with today's tech and modest meter scale superconducting rings. Mini-Magnetospheric Plasma Propulsion (M2P2): High Speed Propulsion Sailing the Solar Wind (http://earthweb.ess.washington.edu/space/M2P2/STAIF2000.PDF) probably represents the core concept. Except instead of trying to go somewhere you stay where you are. The M2P2 paper says a 10cm diameter superconducting coil can divert solar wind in a bubble up to 20 km in diameter. It wouldn't take too much to scale that up.

Of course diverting that much solar wind would create a force on the object. In the paper above that force is used to accelerate a craft. It'd be some handful of newtons at 20 km and linearly more to do what this project wants. One way you might get around that is to "lean into" the wind by going down the sun-side of the L1 halo orbit and allowing the force from the diverted solar wind to counter the sun's gravity's acceleration.

But for the vast majority of the lost Mar's atmosphere the kinetic energy needed to achieve escape velocity is not from impact with solar wind ions or other solar wind related/magnetic field means. All those hereafter referenced under the umbrella term "jeans escape".

Instead the majority of the kinetic energy needed comes from the ejected electrons from the sun's light ionizing the upper atmosphere. That ejected electron has quite a bit and it is distributed to the ions it later interacts with and as those ions interact with others. If there was no solar wind at all the rate of atmospheric loss at Mars would drop but not significantly if the sun still shone upon it.

That isn't to say that, having no magnetosphere, Mars (or Venus) does not lose an additional small amount of it's atmosphere to jeans escape mechanisms. But that amount is limited due to the currents created in the upper atmosphere by photoionization creating their own local magnetic field. That creates a bow shock about the ionopause which slows the incoming solar wind down such that it's constituent ions no longer have the energy needed to deliver the boost required for escape. And because of the induced magnetic field other solar wind magnetic field based mechanisms which would pick-up the ions ionized by the sun's light are mitigated.

I love the idea of this but it isn't going to make Mars have a decent atmospheric pressure in just some years.

On a (much) lighter note, I was thinking if you put these artificial magnetospheres all over the inner system and then coordinated turning them on and off you could "paint" the termination shock surface of the heliosphere with different scales of tubulence in charge density. It'd be a multi-hundred AU wide screen visible only from very far away with sensitive polarimeters (detecting the changes in lines of sight charge density through faraday rotation). Might be a decent way to METI since it'd not require much energy or high angular resolution at the other end.

tldr: It's technically and economically feasible. But it doesn't work like suggested for atmospheric protection because almost all the mass loss is from light caused photoionization, not the solar wind (and other Jeans escape mechanisms).

> But for the vast majority of the lost Mar's atmosphere the kinetic energy needed to achieve escape velocity is not from impact with solar wind ions or other solar wind related/magnetic field means.

Source? That's in direct contradiction with the article stating:

> In addition to determining that solar wind was responsible for depleting Mars' atmosphere, these probes have also been measuring the rate at which it is still being lost today.

You seem to be forgetting that Mars has a very thin atmosphere, so there is no appreciable bow shock created. The upper atmosphere will protect itself only if it's dense enough to create sufficiently strong eddies to deal with peak solar wind. Mars' atmosphere is not dense enough. Venus' is, maybe that's what you're thinking of? Adding a synthetic magnetosphere to Venus will not accomplish much due to the effect you're describing.

And besides, do you really think the Director of Planetary Science of NASA, who recently launched MAVEN to study exactly this topic, somehow missed the core assumption of this project?

You're making a very strong claim, so please provide strong proof of the following:

- Show that "But for the vast majority of the lost Mar's atmosphere the kinetic energy needed to achieve escape velocity is not from impact with solar wind ions or other solar wind related/magnetic field means" -- NASA seems to disagree

- Show that photoionization accounts for "the majority of kinetic energy needed"

- Show that bow shock in Mars' thin atmosphere is sufficiently strong to deflect peak solar wind activity

I don't believe any of those three points are provable for Mars. I believe you can show that bow shock is sufficiently strong on Earth and Venus, but not Mars. I believe you can show that Jeans escape is small on Venus because of the composition of the atmosphere, but not Mars.

> Might be a decent way to METI since it'd not require much energy or high angular resolution at the other end.

I really don't think this is a good idea, unless we want a massive object accelerated to a significant percentage of c to get lobbed our way. :)

Disregarding the effects on atmosphere loss, wouldn't a machine like this eventually be more or less essential if humans ever decide to move in long term? At least, if we want to hang out on the surface (in bubbles, of course).

The good news is "to get lobbed our way" doesn't mean much to a culture that can do distributed projects all over a solar system. A problem along the lines of whats the ideal cannonball size to genocide the common mosquito?

There is also a really nice MAD effect where the window where you can hit a solar system without getting hit back is extremely short, like a century perhaps. Longer than that and we send a probe to the cannonball and turn it into diffuse gray goo that the solar wind blows away when it enters the system, and we return fire 1000 times over, so any logical civilization would send in the asteroid weapons much earlier, like the first time they sniff RF transmissions or the first dark side of the planet electric lights.

Well, the attackers in The Killing Star, by Charles R. Pellegrino and George Zebrowski, did more than just lob an object our way. And there's the issue: "One of the iron rules of relativistic bombardment was that if you could see something approaching at 92 percent of light speed, it was never where you saw it when you saw it, but was practically upon you...". So sending probes is rather unworkable.

In the The Killing Star, our doom was already sealed by "Star Trek" :(

Sounds like something straight out of Cixin Liu's "The Dark Forest"

Indeed, it's a pretty strong argument that the most prudent (in the context of the continuing survival of your own civilization) reaction to discovering a developing civilization in another solar system is to preemptively destroy it. Also, The Algebraist by Iain M. Banks. :)

Unless you assume that cooperation is more likely than conflict. Maybe conflict is more common than usual on Earth. Imagine for example a group of worlds that were able to survive only due to cooperation. On encountering us they would likely lean towards cooperation as that was their experience. We might lean towards conflict but our experience tells us little about the average experience of other inhabitants of our galaxy.

The scary thing about the xenophobic solution to the Fermi paradox is that there might be lots of nice civiliations around that do want to cooperate - but it could potentially only take a few very aggressive, advanced xenophobic bullies lobbing large objects at anyone they spot before the threshold for everyone else to stay in their metaphorical foxholes hiding until/unless they have very good reason to assume they have established a clear advantage.

While we only have a sample of 1 (Earth), we have do millions of samples of interspecies behavior that lean strongly to conflict, and often the cooperation isn't necessarily pleasant for one species either.

What if the first message you receive is from a peaceful and advanced civilization, and you reply with "Hi, we are earth, send your teleportation devices here please", and within the 60 years that the round trip message has taken to be received, a military junta has taken over the civilization and decided to exterminate all humans because our exponentially accelerating technology will make us a threat to their hegemony in another 60 years? :)

I enjoyed that trilogy greatly, but damn, it was depressing :(

> Disregarding the effects on atmosphere loss, wouldn't a machine like this eventually be more or less essential if humans ever decide to move in long term?

I think it's more practical in the long term to create a magnetic field at Mars itself. Using future technology on a scale unimaginable at present (and similar to that proposed in the linked article), we could magnetize Mars' (now-solid) ferrous materials by the onetime application of a very strong magnetic field.

Mars once had a dynamic magnetic field like ours, but Mars' liquid ferrous materials solidified as the planet cooled, which caused its field to disappear. A permanent field could be created by the deliberate application of a strong temporary field that would magnetize the planet's ferrous materials.

Details: http://arachnoid.com/restoring_mars/

It'd be a multi-hundred AU wide screen visible only from very far away with sensitive polarimeters.

ISTR the Swordholder did something similar in Death's End. The purpose wasn't "light", however.

Yeah, I just think the concept is intrinsically cool which may be a bit of a http://www.smbc-comics.com/comic/2011-08-08 issue.

Taking it even further (and more absurd) the interface between the heliosphere and the interstellar medium (ISM) lets in matter mostly along a curved path at the bow who's size and orientation is defined by the relative magnetic vector orientations of the heliosphere and the ISM. This was a recent discovery by the IBEX interstellar neutral atom monitor. If you could get the artificial magnetospheres to influence the magnetic field around the region in the bow where this happens (by controlling the magnetic field in the plasmoids your artificial magnetospheres sends off with the wind) you might be able to control the direction interstellar neutral atoms come in at. And that'd allow you to apply arbitrary (small) force to the entire heliosphere and control direction over very long time periods. But that's so back of the napkin I doubt it's actually feasible.

I think you're thinking of Project Snowflake which was featured more heavily in The Dark Forest. Awesome trilogy!

Of course you're right. It was this project that justified his "promotion" from Wallfacer to Swordholder.

> Instead the majority of the kinetic energy needed comes from the ejected electrons from the sun's light ionizing the upper atmosphere.

IIRC, at martian surface temperature, hydrogen atoms have escape velocity. That goes a long way to explaining why there's no water on Mars: there's lots of Oxygen, but most of the hydrogen has fled.

It is unclear what kind of technology are they planning to use for this shield. If it is usual electromagnets, it's unclear where would they take energy from, if superconducting ones than it's unclear that it would be possible to keep them in superconducting state for long.

Wish someone more knowledgable about this kind of tech comment on this.

Definitely superconductors; that's the only viable way to generate a field that strong. Design the spacecraft properly and you may not need an active cooling system. The James Webb Space Telescope aims to keep the shaded side below 50K… while at the same time solar panels and all its solar-facing equipment will operate 250K hotter near room temperature. The trick is maintaining this temperature gradient.

Apparently the design of the JWST is sufficient to reach temperatures as low as 37K on the cold side passively — below the superconducting limit for some materials. They're additionally using a cryocooler to take the temp down to below 7K for one specific instrument.


Solar works okay on Mars. I don't think the power requirements for keeping your liquid helium liquid is not so great.

> The current scientific consensus is that, like Earth, Mars once had a magnetic field that protected its atmosphere. Roughly 4.2 billion years ago, this planet's magnetic field suddenly disappeared, which caused Mars' atmosphere to slowly be lost to space.

What kind of event could cause the loss of a planet's magnetic field?

It once had a liquid core (like Earth) which generated the magnetic field. Being a smaller planet, it cooled faster and now the core is solid, hence no more magnetic field.

> What kind of event could cause the loss of a planet's magnetic field?

Planetary magnetic fields require a liquid ferrous mantle or core. As Mars cooled, its liquid ferrous material solidified, after which its magnetic field disappeared. The same thing will eventually happen here, and the reason our field still exists is because Earth is a larger planet than Mars, requiring more time to cool down.

More details here: http://arachnoid.com/restoring_mars

Curious why Mars and not the Earth's moon, seems to be the prime target for colonization?

Because of the gravitational influence of Earth there's no stable Moon-Sun L1 point. The Earth would keep pushing anything there one way and then the other.

Also, the extant atmosphere of Mars is a big advantage. It's not much but it helps at lot relative to vacuum with soaking up heat, letting you synthesize oxygen, etc. And while Mars doesn't have much free hydrogen it has a lot more than the Moon does.

Oh, and then there's daylight. A day on Mars is about a day on the Earth. A day on the Moon is a month. So you've got to have lots and lots of batteries if you're going to last out a lunar night. On Mars you just need a bit more mass of batteries than you have of solar panels.

And there's more gravity on Mars which might be important. We know what 1G does to the body and we know what 0G does to the body but nobody knows if there's a difference between the .4G you get on Mars and the .15G you get on the Moon.

Well, L1 is not stable anyways so you would also have to correct for that. Also, due to the fixed rotation Moon has mountains of ethernal light that could provide constant solar power.

And last but not least, the lower gravity on the moon (according to wikipedia) makes space elevators possible even with current technologies. As those would make space travel a lot easier, thats a huge argument to colonize the moon first.

And there's also the shorter transit time from Earth. Really, there are a lot of good reasons to think about establishing a settlement on the Moon before Mars. But I still think that trying to terraform the Mooon is a bit unreasonable.

And yes, L1 is unstable. But the lower bound on how little fuel you have to spend staying at the Mars-Sun L1 is much, much smaller than the Moon-Sun L1. I'd guess 10 m/s of delta-v every year for the Mars/Sun one and 10,000 m/s every year for the Moon/Sun.

Mars has atmosphere (not much, but enough to process into useful gases), water (in the form of ice), and a larger variety of surface materials than the moon.

Mars is substantially easier to establish a long-term operation on than the moon since you can create many of the materials you need onsite.

An interesting theme in KSR's mars trilogy was robot miner machines making a mohole pit and "a couple KM down" the martian atmosphere is either sea level earth air pressure or at a somewhat different depth the atmosphere is comfy warm. I don't remember the modeling of how deep or how far apart but it didn't seem terribly unrealistic compared to the grand scheme of teraforming mars. Obviously a thickened atmosphere would not require moholes quite as deep. Possibly to a useful level. True, increasing surface pressure on mars by a factor of ten means you'll still die on the surface, but to a first approximation you just made your mohole 1/10th as deep and 1/10th the construction time and cost.

The atmosphere on Mars is actually pretty nice at an altitude of negative a couple kilometers, which doesn't exist naturally but we're quite talented at building horizontal tunnels that long so pointing downward can't be that hard.

If you need raw material, robots that never leave the mine don't have scalability problems that human deep mines have and on mars liquid water would be a valuable commodity and robots don't mind water anyway, and solar powered robots can be left alone to generate iron ingots for construction, and for environmental reasons there's no reason to make 100 shallow pits when one deep mine produces just as much and is valuable real estate after digging is complete.

On the earth you can't point a large TBM downward because it'll flood or overheat in just a couple miles or less, but on mars it might be quite useful.

The problem with moholes is eventually you end up with the submarine problem where no matter how tough your metal hole liner is in an abstract sense, it will collapse under rock pressures. So you can't bore a hole clean thru a planet, however interesting that idea sounds.

(edited to add, I looked up data more recent than KSRs old book series, looks like bare rock mines are only possible on mars due to rock blowing out to a depth of 10 KM (better than earth, lower gravity), of course the sky's the limit with steel liners. And at a mere 5 deg/km temp increase on mars, the temp will be cold but more habitable than the surface)

Also, it's easier to build a circular roof than an entire dome.

Mars also has more gravity than the moon.

Is it fully known yet what minerals can be found on the moon? I thought it is still somewhat unsure what is below the few meteres or regolith.

Mars has more metals and the (frozen) water is just an inch bellow surface.

Mars has water and more gravity. It's a better place to colonize despite the distance.

I think the science fiction movie "Spaceballs" was head of its time. Planet Druidia had an enclosure to protect its oxygen, but Lord Helmet tried to suck it out with MegaMaid.

NASA might have to think about how to counter such threats, possibly from a future and more hostile Earth civilization.


Best idea I've seen in years on space solutions, specially at colonization. No marketing bullshit, just plain science. And it may be the first practical step on terraform a planet before we try to colonize it.

"While it might seem like something out of science fiction, it doesn't hurt to crunch the numbers!"

Now we need those numbers.

Would a distributed constellation of satellites at L1 work? A bunch of smaller magnetic dipole umbrellas working together.. Easier to repair and replace individual satellites without having the whole system go down.

"As a result, Mars atmosphere would naturally thicken over time, which lead to many new possibilities for human exploration and colonization.." How long would it take though?

That depends on how involved humanity is in cultivating that atmosphere. There have been proposals to, for example, redirect comets/near-Mars objects that are icy/have oxygen or other desirable materials to burn up in the Martian atmosphere, adding to it. There are likely other ways we can affect the atmosphere.

From a terraforming standpoint, a viable, long-term magnetosphere is necessary in the medium and long term. In the short term, humans would be stuck in shielded suits and indoors, both to protect from radiation and to ensure a breathable atmosphere.

It's a nice idea, and it's one I proposed some years ago:


I am curious why we don't focus more efforts on Titan, which already has a thick nitrogen atmosphere.

It's very far away from the Sun.

Plus Jupiter has a pretty intense magnetosphere that Titan transits. Jovian space has a lot of self generated radiation that makes it less than hospitable.

Edit: Titan is a moon of Saturn. Oops.

Pretty sure everywhere but earth is going to be extremely inhospitable

And us?

Jupiter is 5.2 times farther from the Sun than the Earth. Thanks to the inverse square law, it gets 3.7% of the sunlight that Earth does. You won't be growing any plants there, and solar electric is going to be terrible.

Saturn, which Titan orbits, is 9.6 times farther, and receives 1% of the sunlight that Earth receives.

Mars is much better at 1.5 times farther, and receiving 44% of the sunlight that Earth does.

Also, Titan averages -179C compared to Mars' range from -125C up to 20C.

Go other way to Mercury and one has virtually limitless energy. The polar craters contain ice and a thermal reservoir.

Still very far away from the Sun, but closer.

I believe this is the workshop referenced, and they do have a LiveStream that you can watch.


Skip to 1:36:00 for the start, 1:40:00 for the actual details.

The live stream I found only went up to 1:20:00 or so - is there a longer one?


There's also this very short pdf release which has no interesting detail. http://www.hou.usra.edu/meetings/V2050/pdf/8250.pdf

Would something like this be viable as part of a geoengineering solution to climate change?

E.g. could we partially cancel out earth's magnetic field in order to control atmosphere loss and thereby control the greenhouse gas effect?

Maybe that's a viable solution but to me controlling green house gases by shedding our atmosphere sounds like a terrible idea. I'm pretty sure we would want to keep the oxygen around. Finding ways to capture CO2 from the atmosphere and sequester the carbon should hopefully also be easier than controlling the magnetic field of the earth.

I think losing our atmosphere to control the greenhouse effect would be ... a very bad idea. It would work, but we would end up wishing we hadn't done it.

> NASA proposes a magnetic shield to protect Mars' atmosphere

That's officially called a "deflector", anyone who's seen Star Trek knows.

Cool, let's build the Death Star there.


A variation put into the Earth-Sun L1 and used to direct focused steams of solar particles onto an earthward target....

That won't work without shutting down Earth's magnetic field.

which it does on a regular basis during natural field reversals. The geologists imply its a very sharp transition like a century or less from normal field, no field, reversed field. If you donate to the UN fund that pays for the solar shield, you get shielded when the time comes. If not, well... life does survive reversals, but its likely not fun.

> If you donate to the UN fund that pays for the solar shield

You jest, but look at the difficulty of countries agreeing on carbon pricing. Which is a similar prisoners dilemma

Actually, it appears that that the field does not actually disappear completely during the switch.

Instead of terraforming, would it be possible to build a giant glass ceiling over most of the surface? Humans don't really need an entire atmosphere, just a few hundred feet of it.

This is still infeasible now, but it seems much easier than terraforming. With robotic labor and automation, it may not even be that expensive.

And what about an artificial moon instead?

This could lead to life elsewhere in the galaxy to protest climate change on Mars. We might get visitors if we try messing up a second planet.

how about just moving there an polluting the atmosphere as a side-product of just being there?

Not enough trees, peat, coal, or oil to burn there.

maybe there's enough uranium

Good luck they have that money.

You hated Trump that much huh

Please don't do this here.

Why not a shield for earth, when this atmosphere is clearly in need of help, and already much closer to liveable than Mars'?

Because Earth already has a magnetic field.

For now yes. We are overdue for a magnetic field reversal so this would be a very helpful tool in our planets arsenal

Yes, but is a reversal really a bad thing? Magnetic field reversals don't seem to be a serious threat–there's no evidence of mass extinctions caused by them.

I'm not sure how much we know, but it could easily lead to massive disruption of our technology if nothing else. We might not be wiped out, but we could be in for a lot of pain depending on how things go.

only mid-reversal is a bad thing.

The reversal could take hundreds of years, where our planet has no electromagnetic shielding. All of our technology will be toast.

I call that bad.

I guess you're right. Bad for humans, maybe good enough to reverse some of the damage humans have done.

I could be mistaken, but I understand that bombarding the earth with radiation will do plenty of damage to other species as well. According to Wikipedia, a pole reversal could lead to vigorous convection leading to widespread volcanism, and that subsequent airborne ash could cause mass extinctions.

Ok, but what about having a S.H.I.E.L.D. for earth? Wouldn't that work?

So let me get this straight - science can't currently beat flipping a coin in terms of predicting the weather more than 2 days in advance, but he can accurately simulate space weather for many years and how it will affect other planets?

We don't need to simulate the weather, we just need to know the long running average and deviations to estimate how much of that we need to block. Short time of high activity won't blow the atmosphere off.

This is the difference between weather and climate. I can't tell you what will be precise weather in one year, but I can tell you that January will be probably cold.

There is weather, and there is climate. The latter we have very good models for, the former is notoriously difficult to predict.


I'm not convinced that we have a good model for climate, either. I think we do not have any clue about large-scale cycles in the climate. Nobody predicted the long drought in California, but the Southwest has had a bunch of cyclical droughts over the past 1000 years. This year was supposed to be a La Nina year for California, but instead something else happened and there has been three years' worth of rain in northern California instead.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact