Some months back I made some from the hip comments about ‘Oumuamua, that strange little visitor from outside the solar system. Some of them were superseded by later events, others are certainly debatable. However, in the last week or so, ‘Oumuamua has been in the spotlight due to renewed speculation that it could be artificial, so it is time to revisit this object.

Most of the conclusions that have been reached about it are based on delicate calculations of its movement through space, made based on precise measurements of its position at different times. So it behoves us to understand this data and its implications.

The first thing to say here is that the data arc that is all we have to investigate the orbit, covers just the range from 14^th October 2017 to 2^nd January 2018. So, the last astrometric observation of ‘Oumuamua – that is, accurately measured position to determine its orbit – was taken on January 2^nd 2018. There will be no more data because, now approaching the orbit of Saturn (today, it is 8.5AU from the Sun), ‘Oumuamua is fainter than magnitude 32: that is far below the limit of any telescope on Earth, or the Hubble Space Telescope.

All our knowledge of this strange little body’s path through space is based on just 207 measurements of its position over 80 days, the first of them taken five weeks after it had passed perihelion.

Modern techniques of measurement are infinitely better than anything that was available thirty or forty years ago, but still have their limits. Astronomers like to have years, even decades of data to get a really precise orbit solution, particularly when we are treating the tiny effects that are causing such interest in ‘Oumuamua. Similarly, they like to have data both before and after the objects pass through perihelion. What we have is less than three months of data and all obtained as it moved away from the Sun.

What we know beyond any possible doubt is that ‘Oumuamua came from outside the solar system. It was travelling too fast when it passed by the Sun to have fallen from anywhere within our solar system. Technically, the orbit is strongly hyperbolic, or open. We can calculate that it entered the solar system at 26.4km/s. About half of that velocity can be explained as the Sun’s movement towards the Solar Apex: the point in the sky towards which the Sun and the Solar System are moving through space. If we trace ‘Oumuamua’s movement backwards, we discover that its point of origin is only about 4 degrees away from the Solar Apex, close to the star Vega:

RA (Origin) 18h 37m 55.0s, Dec. (Origin) 33⁰ 50ʹ 25″

The approximate point of origin of ‘Oumuamua, marked with the yellow cross.

Similarly, we can work out towards which point in the sky ‘Oumuamua is heading, which is in Pegasus:

RA (Final) 23h 51m 23.0s, Dec. (Final) 24⁰ 42ʹ 06″

‘Oumuamua takes 11500 years to travel one light year and travels 87 light years in a million years but, in a million years, constellation outlines change a lot. In less than the time than ‘Oumuamua takes to travel a distance equivalent to that from the Sun to Sirius, the stars of Lyra and the surrounding constellations will have moved quite substantially.

The North Pole of the sky and the Great Bear as we see it today.

And, as we will see the North polar area of the sky in the year 91000AD. Note how much the familiar form of the Great Bear has changed and how much Vega has moved away from the other stars of Lyra.

The North Polar area of the sky now and in the year 91000AD, compared. The sky and the constellation of Lyra would have looked very different a million years ago or, whenever in the past ‘Oumuamua set out, making it extremely difficult to work out which stars it may have been close to in the distant past.

Astronomers have analysed catalogues of nearby stars in an attempt to identify the origin of ‘Oumuamua. Two factors complicate this:

• Despite the immense progress that is being made with Gaia to measured precise positions of movements of nearby stars in space, this is still a work in progress – release of Gaia’s data is still in its early phases – not all candidate stars have good data yet so, their positions in space tens of thousands of years ago are uncertain.
• Each encounter of ‘Oumuamua with a star perturbs its orbit and adds an uncertainty to its future path. If the mass of the star and the exact distance at which ‘Oumuamua passes it are uncertain, the perturbation will be uncertain and thus the future path will be increasingly uncertain with each pass. Tens of stars would have influenced ‘Oumuamua’s trajectory in the last few million years and, with each encounter, the extrapolation of its path gets more uncertain.

Seven encounters with stars to a minimum distance of 1 parsec (3.3 light years[1]) were found in the last seven million years. In all but one of the cases, ‘Oumuamua would have passed the star much faster than its entry velocity in the solar system: for three stars, the encounter would have been at velocities from close to 200km/s up to more than 300 – implying that ‘Oumuamua flashed through the system very quickly. None of these look like likely points of origin. In three other cases, the velocity was between 60 and 70km/s, more than double the entry velocity in the Solar System. One star, HIP 981, with a rather low encounter velocity and an approach to half a parsec, just over six and a half million years ago, but the movement of this star is almost indeterminate, so the encounter circumstances are little more than a guess.

Although one star, the prosaically named UCAC4 535-065571, is an interesting candidate, its movement is not well enough known either at present to be certain about it. Playing with the numbers a little to obtain the best fit, they find that this encounter would have happened 2.1 million years ago.

It has to be said that most of the stars that ‘Oumuamua would have encountered are red dwarfs, with masses smaller than the Sun. None, at present, looks like a really good candidate. Some studies have suggested that ‘Oumuamua may have spent hundreds of millions of years wandering through space.

So, what we can say is that the point of origin of ‘Oumuamua is unknown, although there is some hope that the Gaia data release of 2021 will improve the situation by giving accurate data for all the possible stars in the solar neighbourhood. We must remember this: at present, it seems unlikely that ‘Oumuamua came from a star less than 100 light years away, when considering some of the more exotic theories about it.

The biggest killer to the study is that the orbit used in it is a quite old one, based on less than one month of data and it does not include the non-gravitational terms that were found later so, will inevitably have some cumulative errors when extrapolated into the far future that will make the conclusions unreliable, but much bigger errors when extrapolated into the distant past. Indeed, the orbit solution provided by NASA’s Jet Propulsion Laboratory (JPL) explicit states “the behavior … outside the observed data arc from 2017 October 14 to 2018 January 2 can only be assumed.  Predictions outside this time interval, especially prior to October 2017, could be much more uncertain than reported here.”

The difference between the orbit solution used to study past encounters of ‘Oumuamua with stars and the final orbit solution is nearly 2 arcminutes in the next 80 years, almost all in declination. If we go back in time, the difference is even bigger: 12 arcminutes in 1901 between the two ephemerides, which does very much invalidate conclusions about its point of origin.

It is this identification of non-gravitational terms of motion, a result, so controversial initially, that it was not released until thoroughly and absolutely checked[2] and other explanations excluded, that has re-awoken suggestions that ‘Oumuamua could be artificial.

As far back as the nineteenth century, astronomers realised that something other than just gravity was affecting the movement of comets. However accurate the observations, however careful the calculations, something was causing comets to advance or delay themselves in their orbits. It was only a tiny amount – usually only a few hours in an orbit of years, but it refused to disappear. Initially, astronomers tried to explain it as being the result of a “resisting medium”, something that slowed comets down like air resistance slows a ball thrown through the air. It was only about fifty years ago that is was realised it was the gas jets blasting off the nucleus of the comet that were pushing the it slightly off course.

With sufficient observations of sufficient quality, the effect of these forces could be measured as accelerations in the radial direction (away from the Sun, given as A1), the transverse direction (laterally, A2), and normal direction (perpendicular to the plane of the solar system, A3). Usually, A1 is the biggest and most easily calculated, while A3 is usually too small to measure.

As ‘Oumuamua moved away from the Sun, it became obvious that it was accelerating, or rather, the Sun’s gravity was slowing it less than it should. The difficulty was that no one could find any evidence of cometary activity that would cause the acceleration. That does not necessarily mean that such activity did not exist. When it was discovered, it was already outside the Earth’s orbit, at 1.1AU from the Sun and had reached 2.9AU when last seen but, at perihelion, it had been as close as 0.26AU from the Sun, well inside the orbit of Mercury. It is quite possible that there was activity when close to the Sun that was not detectable after discovery, although possibly still present at a low level.

The value of A1 that was calculated for ‘Oumuamua was rather large: A1=2.8 in units of 10^-7 AU/day².  This is the fifth largest ever calculated[3] (for comparison, the value for Comet Halley is 0.0027), which is an unexpected result for an object that is apparently inactive.

Despite the claim in the paper that the significance of the result is “some tens of sigma”, the formal value calculated by JPL is A1=2.79±0.36, making the formal significance 7.8 sigma – still totally beyond doubt, given that the probability that a 6 sigma result is obtained by chance is about one in a million. However, the values for A2 and A3 that are calculated for ‘Oumuamua are totally indeterminate (less than 0.5 sigma in both cases) and so, consistent with zero.

Argument now centres around whether these results can be obtained naturally, or not. Everything centres around interpretation.

If ‘Oumuamua is as elongated as is claimed – its light curve amplitude of two and a half magnitudes suggests a cigar-shaped object, ten times as long as it is broad – and if the jet activity is as great as is suggested, calculations indicate that it would make ‘Oumuamua start to spin so fast, like a windmill out of control, that it would rapidly break apart. This, obviously, has not happened.

So, there are least two possible interpretations:

1. ‘Oumuamua did have cometary activity close to perihelion, but is not as elongated as the light curve suggests. There are certainly researchers who believe that the very elongated shape has been exaggerated and that the difference between length and width is only a factor of 5, or less, making it less susceptible to catastrophic spin.
2. ‘Oumuamua is much less dense or thinner than believed and is being affected by light pressure.

The most dramatic manifestation of (2) is the recent suggestion that it might be a light sail, although the relatively low velocity of ‘Oumuamua is not consistent with a light sail, unless it has come from a star much less luminous than our Sun. Of course, a light sail implies artificial which, in turn, suggests deliberate targeting at our Solar System. In fact, the non-gravitational acceleration in the orbit is yet another parallel with the Rama or Arthur C. Clarke’s novel, “Rendezvous with Rama”, as Rama also accelerates as it moves away from the Sun.

More plausible than the light sail would be a monolith-shaped structure like TMA-1 from “2001 a Space Odyssey”, which would be susceptible to light pressure and, if tumbling, would produce the high-amplitude light curve. It is difficult though to understand how this could be a natural object and how it could come to visit us.

A maxim in science is “extraordinary claims require extraordinary proof”. You should only invoke a revolutionary explanation if you have bullet-proof evidence for it. So far, the evidence for a non-natural origin for ‘Oumuamua is circumstantial and alternative and, more likely, explanations exist.

What would have been extraordinary proof? That would be any evidence of manoeuvring in the inner solar system, or deliberate targeting towards a star if our Sun were being used for a gravitational assist. We have no evidence of the former and the latter seems not to be the case, given that there is no obvious target star.

I suggested to a colleague that if ‘Oumuamua is an alien probe – light sail or otherwise – it is going about exploring our solar system in an odd way. His reply, tongue in cheek, was “maybe it is a survey mission of solar-like stars”. Well, if it is, it is going about it in a very odd way because there are no nearby solar-like stars either in the direction that it has come from, or the one that it is heading and, unless aliens know some method of communication that breaks the laws of physics as we know them, would never be able to communicate its results.

No, it is far more likely that ‘Oumuamua is a completely natural object, but that our observations of it are so incomplete that we will probably never be able to understand it fully.

[Author’s Note: I have made a couple of small edits to this post since the original posting to add small details that I had previously overlooked.]

[1] 1 parsec is 3.26 light years, but the authors extended their search limit slightly to 3.3 light years to include a seventh encounter that seemed particularly interesting.

[2] Lest someone think that this was suppression of data to avoid “the public” finding out the true nature of ‘Oumuamua until it was too late, actually this was a situation in which a group of scientists wanted to be absolutely certain of their conclusions before risking embarrassment by announcing an incorrect result. The results were finally published in Nature in June 2018.

[3] The largest calculated values are: 316P/LONEOS-Christensen, 730; 205P/Giacobini-B, 78; 287P/Christensen, 8.3; 86P/Wild 3, 3.1; 1I/’Oumuamua, 2.8; 147P/Kushida-Muramatsu, 2.7; 74P/Smirnova-Chernykh, 2.5; 90P/Gehrels 1, 2.4; 117P/Helin-Roman-Alu 1, 2.4, 203P/Korlevic, 1.5.