Giving Mars a magnetosphere

Thursday, September 20th, 2018

Brandon Weigel talks about the potential for supplying Mars with an artificial magnetic field:

By placing a satellite equipped with technology to produce a powerful magnetic field at Mars L1 (a far orbit around Mars where gravity from the Sun balances gravity from Mars, so that the satellite always remains between Mars and the Sun), we could encompass Mars in the resulting magnetic sheath.

Martian Magnetosphere

Earth’s magnetic field, originating at it’s core, has a strength of ~6*10^-5 Teslas at the distance of the Earth’s surface. This is the force which deflects compass needles. It is also the strength required to defend our atmosphere against deadly solar wind. However, a space-based magnetic field at Mars does not have to be quite this powerful. First of all, our goal is only to encompass Mars in the magnetosheath of the field; it does not need to extend as far as the Earth’s does. Earth’s magnetosheath extends to ~6 million kilometers. Mars L1 is only about 1 million km from Mars. Of course, we are going to want to allow some leeway for potential solar flare events, but extending the field ~1.5 million km is probably sufficient.

Another thing to take into account is the fact that the intensity of solar wind at Mars’ distance is less than half that at 1 AU. This means that we only need a magnetic field half as powerful as what we would have needed to defend a planet at Earth’s distance from the sun. Taking both of these factors into account, a space-based magnetic field around Mars only needs to have a strength of roughly 11% that of Earth’s. This will create a magnetosheath long enough to extend 500,000 kilometers beyond Mars.

Using the magnetic field magnitude equation, we can now solve for the amperage of the “wire” required to produce such a field. This yields a current of ~200 Mega-amperes. Any electrician knows right now that we are going to need a BIG ASS wire.


Some things to note are the exceptionally low voltage for the system of about 2 volts, and the dimensions/mass of the copper solenoid which come out to a torus with a total diameter of ~3.5 meters and a mass of ~57 tonnes. This is a big copper doughnut. It would fill the average living room area wall-to-wall and weigh more than 6x the legal mass of a loaded semi truck on the freeway. A magnetic field of ~81 Teslas is generated at the surface of the solenoid; nearly twice the strength of the strongest artificial continuous magnetic field ever produced to date. Another thing to note is the fact that a fission reactor of this size will require over 40 tonnes of uranium every two years to remain in operation. This may be the biggest problem for any future Martian-magnetosphere endeavor, seeing as a launch to Mars from Earth takes about 18 months and the abundance of uranium on Mars itself is unknown.


  1. Alistair says:

    That….seems astonishingly cheap and easy, compared to some other terraforming requirements discussed here.

    Could this be done cheaper with a constellation of smaller satellites rather than a single large one? They don’t have to be at L1, do they?

  2. Sam J. says:

    I asked if this could be done but I never thought about L1.

    I still say that some sort of statite would be a better answer. [A statite is a satellite over the north and/or south pole of a planet that balances it's gravitational pull down to the planet with a solar sail or other such propulsion. Maybe it could use electric fields to do some of the propulsion.] 400Mwatts is a shit load of power. Think of the radiator size needed to dissipate the power of this thing. Yuge…I think you would need something bigger and use a super conducting magnet.

  3. Alistair says:

    Well, 400 MW isn’t an unimaginable amount of power. It’s a small to mid-sized nuclear station. There’s plenty of engineering challenges here, but the raw numbers don’t have the intimidating feel that, say, importing nitrogen to Mars does.

    For a civilizational-scale endeavour, it seems very possible.

  4. Alistair says:

    So…at Mars orbit, we could receive 350W/m2 from Solar? Assume 22% PV efficiency…. we would need 5 million square metres of panels installed to drive this thing, not counting converter losses.

    Allow 50% engineering margin, and we’re looking at an array 3 km square. That’s not bad. By comparison Germany has got 41 GW installed capacity, which at German ground level irradiance and efficiency of ~200W/m2 must be about 200 Million square metres.

  5. Alistair says:

    Oops,my bad; Mars gets ~580W/m2. So much the better.

  6. Sam J. says:

    “…400 MW isn’t an unimaginable amount of power…”

    It is through one coil in the vacuum of space where you can only radiate away heat. It’s a lot. Spread out over a big city it’s not that much. From a article estimating power consumption of cities,”…A safe number for and industrialized city of 1 million would be 10,000 MWh…” So it’s the power of a city of roughly 400,000 people in one small space in one big ass coil of wire.

    Settling Mars is just silly. To do so you will have to move a great deal of dirt and process a great deal of material. Why not just do it on the Moon, magnetically launch it into space and build space habitats of vast size. I’m willing to bet the material requirements would similar and since the habitats are solar powered once you build them they mostly sustain themselves. One design was so big they estimated to see every space in it on foot would take a person most all of their life. One technique might to build some sort of AI to design different areas in different styles but have them generally follow the rules of design to make them pleasing to look at.

    A McKendree cylinder made of steel, of which there’s a vast amount in asteroids could be the living area of Russia. Made of carbon nano-tubes the living area would be the size of Euro-Asia.

    Smaller habitats could be built where people could live while these bigger ones are built. Most of this stuff would take a good deal of working out to get it online but after that a lot of the work could be automated.

    Really big ones.

Leave a Reply