Untitled Document

So Elon Musk's Tesla roadster is now in an orbit that takes it right out to the asteroid belt not far from Ceres. And what a thrilling launch it was with the two boosters landing so perfectly choreographed. It was nearly flawless. Only one minor hiccup, that the core booster missed the barge and crashed into the water, probably at around 300 mph, scattering the barge with shrapnel. They will surely fix that too with future launches.

If any of you haven't seen it yet, here is the launch video archived by SpaceX.

And here is Elon Musk talking about it in a press conference afterwards

But what is going to happen to his roadster in the long term? On the webcast of the Falcon Heavy launch they said it would stay in its orbit possibly for a billion years. 

However if you are a keen follower of research on Near Earth asteroids, you will know that that can't be. Part of the IAU definition of a planet is that it clears its neighbourhood. Whatever you think about that as a way of defining a planet - it's certainly true of Earth. The only reason we have Near Earth Asteroids is because we have a constant input of them from the asteroid belt, perturbed by Jupiter.

The current population of asteroids that fly past Earth will nearly all be gone 20 million years or so from now (and replaced by new ones). Over millions of years they are almost bound to eventually hit something or be ejected from the solar system. 

So, what will happen eventually to Elon Musk's Tesla cherry red roadster?

Immediately after the launch it went into a temporary orbit with maximum altitude 7,000 km, and that's what the Starman video shows:

Five hours later they did a final burn to put it on an interplanetary orbit out to beyond the orbit of Mars. They could have done that right away without going into the orbit at all - but as a requirement for future military launches, apparently they had to prove that their final stage could withstand spending a long time in the Van Allen belts. That’s why they deliberately kept it going through the Van Allen belts over and over for five hours before sending the payload into its interplanetary orbit.

It’s in a Hohmann transfer orbit to Mars from Earth - except that they launched it at the wrong time in Mars’ orbit to go anywhere near Mars (intentionally for planetary protection reasons). And actually it’s more than enough for Mars. At first he thought it would end up not far off Ceres’ orbit. He tweeted|:

“Third burn successful. Exceeded Mars orbit and kept going to the Asteroid Belt.”

Since then we’ve had a revised orbit. 0.99 x 1.71 AU x 1.1 deg. That's the nearest and furthest point, and inclination, so it's sqrt(((0.99+1.71)/2)^3) for the period. or 1.5686 years, by Kepler’s third law that the square of the period is proportional to the cube of the semimajor axis. This means it will cross Earth’s orbit frequently, every 1.5686 years or so, and from time to time will do close flybys of either Earth or Mars or both.

So - it's not traveling quite so far out into the asteroid belt, but still, at 1.71 au it is well beyond Mars' orbit (ranges from 1.38 au when closest to the sun and 1.66 au when furthest away).


(UPDATE - someone did a study forward to three million years and found an 11% chance of hitting Earth rather higher than the 3% of V type asteroids - see the end of this article for details. 3% for Venus and minute for Mars. But thy didn't take the study far enough to see what happens to it on the very long term so perhaps the 50% chance of hitting the sun and 15% chance of going interstellar in this section is as good an estimate as any until they do)

It is sure to be perturbed by Jupiter, and may have flybys of Mars too, so it's not going to be in that orbit for long. Its period will change slightly with every orbit as well as the other orbital elements such as eccentricity, inclination, semimajor axis and orientation in space of the ellipse.

But to show why it is bound to do many close flybys of Earth with a simple calculation - assuming it has a constant period of exactly 1.5686, then it would do flybys of Earth at 11, 69, 80, 1109, 6734, and 7843 years, using this Fast Decimals to Fractions:

(approximations of 1.5686 )

11/7 = 1.5714285714285714 … 7* 1.5686 = 10.9802

69/44 = 1.5681818181818181 …44* 1.5686 = 69.0184

80/51 = 1.5686274509803921 … 51* 1.5686 = 79.9986

1109/707 = 1.5685997171145687 … 707* 1.5686 = 1109.0002

6734/4293 = 1.5686000465874679 … 4293* 1.5686 =6733.9998

7843/5000 = 1.5686 … 5000* 1.5686 =7843

In an orbit like that, it could hit Earth. But it is far too small to do any harm if it hits us. It hasn’t got an aeroshell, so it would just burn up in the upper atmosphere harmlessly and not much would reach the ground.

But Earth and Mars are both tiny targets. It is easy to get close enough for a flyby to change the orbit. But we are such tiny planets compared to the vastness of interplanetary space that it is really unusual for any asteroid to hit us.

Its orbit will also be constantly changing, as is the orbit of Earth and Mars. It will be constantly precessing around the sun, and the eccentricity, and tilt constantly changing also. The eccentricity of the orbits of Mars and Earth change in cycles, sometimes more circular and sometimes more elliptical. But in this orbit, right into the asteroid belt, under Jupiter’s influence, Jupiter is likely to make its orbit more and more eccentric until it becomes a sun grazer - unless it is perturbed by the other planets along the way.

As a first approximation by looking at studies of similar asteroids in the past - it's far more likely, surprisingly, that it reaches the sun and evaporates (about 50% chance), and next most likely, that it is ejected from the solar system (about 15% chance; this can happen if it gets into an orbit that takes it as far as Jupiter from the Sun).

Techy details: This is a paper with statistics from 117 test objects from 2000. I will see if I can find a more recent study. Meanwhile though - It suggests the most likely fate is that eventually it hits the sun (more than half so more likely than not), probably within 60 million years.

The next most likely fate is that it is ejected from the solar system (15% chance), and after that, that it might hit Venus, or Earth, or Mars in that order.

It has a remote (perhaps less than 1%) chance of hitting Mercury or Saturn. Jupiter surprisingly doesn't feature - but that may be just a result of using a small sample size.

Half of the test objects hit something or are ejected from the solar system within ten million years. Surviving in that orbit for a billion years seems very unlikely.

Of their 117 objects, 19 survive for 60 million years out of an initial 117. 5 hit Earth, 2 Mars ,12 hit Venus, 1 Mercury, surprisingly none hit Jupiter and 1 hits Saturn, 12 ejected and 65 end their life as sun grazers - get so close to the sun they would eventually evaporate away.

You might wonder how that’s possible. After all it’s a 29.8 km / sec delta v to reach the sun, and 42.1 km/s to escape the solar system. Isn’t it far more likely it hits Earth or Mars given that it is already in an orbit going back and forth between the two?

But, it is surprising what a difference a few flybys can do. Here is how Rosetta got to a Jupiter crossing orbit from an initial orbit not as far as Mars with three flybys of Earth and a flyby of Mars.

Elon Musks’ Tesla could end up flying past Jupiter in a similar way indeed it starts with a more eccentric orbit than this. Though it is not deliberately put into such an orbit, still, it is likely to do Earth flybys eventually. These could easily send it outwards to Jupiter or inwards to Venus, Mercury and the sun.

Once it is a Jupiter crosser, Jupiter’s powerful gravity assists can easily eject it from the solar system with just one flyby, or if it goes past Jupiter on the other side, deflect it into an orbit that takes it close to the sun eventually with more flybys becoming a sun grazer.

It could also get to the Sun via Venus flybys. The Parker solar probe will get within 5.9 million km of the sun with seven Venus flybys - I know it already starts with an orbit that gets within Mercury but just as for the flybys boosting it to Jupiter - with additional flybys of Earth and Venus it could get into an orbit like that first and then through these seven more flybys get close enough to touch the sun,

Parker Solar Probe - Wikipedia

Of course those are optimized to do the flybys quickly over a period of years. A non optimized randomized orbit would take thousands through to millions of years to just happen to do the right flybys to hit the sun, but it seems that is its most likely end fate, with ejection from the solar system as the next most likely.

Jupiter can also act on its orbit through resonances that make its orbit more and more eccentric until it becomes a sun grazer, even if it doesn’t get out as far as Jupiter. This seems to be the usual way it ends up becoming a sun grazer according to the study:

“Therefore, to a first approximation, the general picture of NEO evolution consists of these objects being dominantly controlled by close encounters while interior to 1.8 AU, which then deliver them to the region of powerful resonances exterior to 2 AU, where they are efficiently disposed of, mostly into the Sun.”

(page 179 of this paper)

I would have thought most, and quite possibly all of that 15% that get ejected from the solar system do it via a Jupiter flyby - and get to Jupiter via boosts of their orbits from flybys of Earth, Mars and Venus. The delta v to hit the Sun is 30 km / sec so a lot higher than Mars or Venus. But it is a much larger target. It doesn't have to actually hit the Sun - that figure was for sun grazers. If it gets close to the Sun then even if made of rock or metal it will just evaporate away over several flybys.

I’d be interested to know what happens to the 19 that survived 60 million years. There are other possible end states.

For instance, it could end up under Mercury’s influence becoming a Vulcan, orbiting very close to the sun. These are hypothetical asteroids, none yet detected, but there is a chance there are asteroids a few kilometers through to tens of kilometers (but no larger or we’d have spotted them by now) in a sort of ‘asteroid belt’ stability region between Mercury and the sun.

It could also end up in the asteroid belt presumably. And also as a long period comet with an orbit of billions of years. And just possibly also though minute chance, a trojan of Jupiter or some other planet, a third of an orbit ahead or behind.

There is another study here that comes to a somewhat similar conclusion, this time integrated over 10 instead of 60 million years. It’s for V type asteroids - originating probably from Vesta and with orbits not that different from Elon Musk’s Tesla.

“ The mortality or loss rate in 10 million years is equal to~43%:

  • 5.3% impacts with terrestrial planets,
  • 26.4% impacts with the Sun and
  • 11.4% escapers.
  • A not significant number of V-NEAs can have impacts even with Jupiter (<0.1% in 10 Myr).”

(I added the bullet points for clarity). They explain earlier in the paper that of that 5.3%, then 3% hit Earth, 1.5% hit Venus and the remaining 0.8% hit Mars or the Moon.

A likely fate from that paper (see section 3.5) is to become more and more eccentric in its orbit under influence of Jupiter until it becomes a sungrazer. Alternatively, if the semi-major axis is greater than 2.5 then as it becomes more eccentric it gets into the vicinity of Jupiter and can get flung into a very long period orbit or ejected from the solar system before it is so eccentric it gets lost as a sungrazer.

I made it a bit under 2 au from the diagram, which would suggest that becoming a sun grazer is quite likely - but flybys of Earth which are quite likely could change that easily enough as for Rosetta.

I’m interested if any of you know of more recent papers on the topic. I think I have read something else at least, in addition to that paper I cited, but can’t find it just now.

Also do say if you see anything to correct in this answer, thanks!

See also this article in Scientific American:

But don’t take it from me—I certainly wouldn’t, so I asked a real expert. Musk’s interplanetary Roadster “will remain in solar orbit for a long time, but on a timescale of millions of years,” not billions, says Jonathan McDowell, a spacecraft-tracking astronomer at the Harvard-Smithsonian Center for Astrophysics. “After perturbations from planets and differential thermal effects from solar radiation it will most likely fall into the sun, with less likely long-term outcomes being ejection from the solar system, impact on a planet, or relocation to the Musk City Tesla Automotive Museum on Mars.” Not a bad fate for a Starman. Elon Musk Does It Again


No it’s not, not any more. The Tesla car had no batteries. That's no surprise really given the possible fire hazards of Li-ion batteries catching fire. Don't suppose they are rated for the gs and rattling of a launch or the vacuum of space.

Video of Tesla batteries on fire (though it's probably no worse than a gasoline fire, you wouldn't want this happening on a rocket):

It remains attached to the second stage and was transmitting back via the second stage which had power for only twelve hours. It’s just a dead object now with no way of communicating home and a small object so will be impossible to detect except for the occasional close flyby of Earth or better telescopes than we have now.

I did wonder if they thought of attaching retroreflectors which would have made it a bit easier perhaps to spot it during Earth flybys and also permit very precise laser ranging but AFAIK they didn’t do that.


It may seem a waste to do this - why send a useless car into space?

However, the thing is that with the first launch of a rocket, then they usually have a dummy payload. They have to prove its ability to launch mass into space. It might be a lump of cement, or it might be barrels of water and lumps of metal. He just chose to send his Tesla roadster instead of a lump of concrete.

If it was me I’d have sent a big chunk of material to impact into the lunar poles as for LCROSS and got more information about the composition of the poles and whether there is lunar ice there.

But I’m not a CEO of a space company :). He is and it’s his choice what to do. He gives as his reason that he thinks it stimulates people’s imagination and gets them interested in space travel. And who can say he is wrong there? Given that he has to send something into space. If he had just sent a barrel of water, a lump of cement or a big chunk of iron, nobody would have complained at all. So why does it matter if he sends a Tesla Roadster?

You might also wonder if the Tesla roadster would still be usable. Answer is no - but it may survive in better shape than you expect - well until / unless of course it starts to do close flybys of the sun. At present it’s perihelion is at the same distance from the Sun as Earth - rather extreme conditions because of the huge temperature variations you get in the vacuum of space, and in a hard vacuum. But it’s not going to melt away and evaporate any time soon. See If You Could Bring Elon Musk's Tesla Back From Space, Would it Still Work?


The chance of hitting Mars is probably very low, fraction of a percent in 10 million years. From those figures for V type asteroids ,probably less than 0.8% in 10 million years.

It is a small object with lots of exposed surfaces and tumbling - most of the surface would be quickly sterilized with UV, and there's the heat / cold cycle and then cosmic radiation will deal with most of the rest over a timescale of thousands of years - and then also it's not aerodynamic and most of it would burn up in the Mars atmosphere even though it is thin.

Also it’s not unique either. Nearly every mission to Mars sends unsterilized final stages on Mars crossing orbits, carefully biased to miss Mars on the first flyby. (The Indian MOM is a rare exception). There are also missions that missed Mars by accident and flybys. They are all in independent heliocentric Mars orbit crossing orbits around the sun.

So the risk of it contaminating Mars must be pretty tiny - and also there must be a high chance of it eventually getting recovered by our descendants or at least them keeping a close eye on it with technology that would make it easy to protect Mars from it in the very remote chance it has a hit in the next few thousand years and we also decide that we need to keep Mars protected from Earth life indefinitely (e.g. discover some form of life there that is an early precursor to Earth life and very vulnerable to introduced microbes).

So that's a lot to go wrong for it to be an issue at all.

So I don't think this is a significant issue for his Tesla motor. But it does highlight something about the current planetary protection requirements. He would only be required to show that it can't hit Mars for the next century - and not an absolute proof either, only that there is a very low probability of it hitting Mars. The original assumption was that we'd have an "exploration phase" during which Mars has to be kept free of Earth microbes and after that it wouldn't matter. And - that might well be what happens.

Maybe within the next century we prove that it doesn't matter if we introduce Earth microbes to Mars. But we don't know that for sure and they no longer talk about an 'Exploration phase' but still have this rule that you only have to prove that it is unlikely to hit for a century. Again I think it isn't that big of a deal so long as the number of objects are small and kept track of, as we should if necessary be able to do something to protect Mars with the improved technology we have, whatever it is, a century from now, should that be needed (a lot of hypotheticals there).

However, the final stages as well as the spacecraft that were headed for Mars and missed often don’t have well determined orbits.

If we get to the point where we have thousands of these objects with not very well determined trajectories, and each of them having a small but non zero chance of hitting Mars in the next century, and also a situation where for some reason we think it is vitally important to continue to keep Mars free of Earth life to preserve its science value as a heritage of some unique form of biology - say early life precursors to modern DNA - then it might be more of an issue.

The rules for planetary protection of Mars were drawn up on the basis that we have to accept some probability of contaminating Mars and to keep that risk low enough to be tolerable. Later as we explore Europa and Enceladus and further afield we have a chance to set out new rules.

I happen to think myself that for Europa and Enceladus the rule should be 100% sterility for any mission that has even a remote chance of introducing Earth microbes to habitats there, including landers. That wasn't possible at the time of Viking, but it is nowadays. We already have one idea for a Europa mission with 100% sterility by using high temperature electronics heated to 500 C throughout the coasting phase for the mission out there. The same approach could also be used for 100% sterile missions to any liquid water surface habitats on Mars if desired. This technology is being developed for surface missions to Venus and is already in use in high temperature applications on Earth.

Meanwhile we can find out a lot about their subsurface oceans without risk of contaminating them at all, by studying the geysers. Our spacecraft can fly through them and directly sample them, as Cassini already did. We may even be able to directly study microbes from their oceans this way, without any planetary protection issues at all. I think it is  the obvious place to start myself, both from a practical approach and for reasons of planetary protection. I cover this topic in detail in my “Touch Mars?” book.

This originated as my answer to Will the roadster Elon Musk just sent to Earth orbit with the Falcon Heavy be left in orbit? If yes, do we expect it to re-enter the atmosphere on its own? on Quora.

If you are on Facebook you may like to 'like' the page Robert Walker - Science20 Blog Alerts to get alerts for these blog posts.


Oh, so someone has actually done the proper computer modeling of the long term fate of the Tesla roadster. Not unlike what I got by my analogy with the V type asteroids.

From the paper itself, hjgher chance of hitting Earth and Venus, 11% chance of hitting Earth after 3 million years (V type asteroids 3% after 10 million years) - not too surprising given that it starts off from Earth which you'd think would increase its chance of hitting Earth. Chance of hitting Venus about 3% just reading from their graph of hitting Venus already after 3 million years. Tiny < 1% chance of hitting the sun by then. But longer term it could end up hitting the sun or other planets. Chance of hitting Mars is minute, again similar to V type asteroids (asteroids probably deflected from Vesta).

They don't integrate it far enough into the future to work out the probability of hitting the sun or being put into an interplanetary orbit. They only briefly discuss the sun - which takes millions of years typically to hit from such an orbit, and got only one out of their over 200 runs to hit the sun after 3 million years.

They don't mention interplanetary orbits at all. But do note the possibility of it escaping the zone of the terrestrial planets in this passage in the paper:

"As can be seen in Fig. 3, most orbits remain at inclinations of less than 15◦ in our integrations. We expect that the orbits that reach a high enough inclination will have longer lifetimes and are more likely to escape the terrestrial planet zone through resonant and secular interactions. The relevant timescales for this effect would be significantly longer than a few million years, but we leave more detailed investigations to future work."

And an interesting detail, there is a close flyby closer than the Moon in 2091. It's not a super accurate model for short term studies - they model the Earth Moon system as a single object - but pretty good until we get a projection from the CNEOS which we should do at some point presumably.

I'm quite pleased to have got so close based on such elementary arguments in my article :).

It's been followed by amateur astronomers at increasing distances as it leaves Earth, which also gives promise of spotting it again as it flies past Earth in the future. See Elon Musk's Tesla Roadster photographed at 10 lunar distances. Video too and very bright in the photo taken with a 0.5 meter telescope so surely this record will get beaten as they continue to observe it.

Actually this must be the second stage + roadster and it would be interesting to know how big the combination is It was mounted at such an angle that you don't see the second stage in the live feed of the Tesla during its first orbit .