In-depth description: Heavy stellar traffic, deflected comets, and a closer look at the triggers of cosmic disaster
Strikes by large asteroids or comets are a global danger to be taken seriously – they have occurred in Earth's past, and they will occur again. The first piece of good news is that impacts with regional or even global consequences are exceedingly rare, and occur at a rate of no more than one per million years. A random person's risk of dying in a plane crash is an estimated 25 times as great as the risk of perishing due to a cosmic impact event. The second piece of good news is that current asteroid monitoring systems yield a fairly complete picture of the larger asteroids and comets – those more than a few hundred meters in diameter – that can be found in our solar system neighborhood, and indicate reliably that none of these currently known ``Near-Earth Objects'' pose a concrete threat to the Earth.
Understanding cosmic collisions
Still, the threat is so fundamental that scientists are eager to understand the underlying mechanisms. When it comes to impacts by comets, the chain of events leads even further into the depths of space, far beyond our solar system. Our Sun is only one of an estimated 200 to 300 billion stars in our home galaxy, the Milky Way. Viewed from afar, you would see the Milky Way as a stately disk, so large that it takes light 100,000 years to travel from one side to the other. Within this disk, the stars orbit the Milky Way's center; our Sun completes one orbit in about 225 to 250 million years. Look more closely, and stellar motion becomes more complicated, with the stars following individual orbits that can cross each other, bringing certain stars into close proximity now and then in (cosmically speaking) brief encounters.
These close encounters play an important role when it comes to cometary impacts on Earth. The solar system is believed to be surrounded by a gigantic cloud of small, icy bodies, namely comets. This "Oort cloud" is a roughly spherical shell extending from 2000 to 50,000 times the Earth-Sun distance from the Sun, so its outer edge is about one light-year from the Sun. There are likely to be billions of these comets with sizes up to a few or even a few dozen kilometers.
Given their great distance, these comets feel only a very slight pull of the Sun's gravity; only just enough to keep them in an orbit about the Sun. Thus, the gravity of a star that passes within a few light-years of the Sun can be strong enough to deflect them markedly from their original paths. The amount of deflection depends not only on how close the star passes, but also on how massive it is and how fast it is moving.
From encounters to collisions
Some of the comets can be deflected in a way that carries them into the inner solar system. As they approach the inner regions, the Sun's light as well as its particle winds will strip material from the icy object, creating the distinctive long tail of a comet. After its closest approach to the Sun, the comet will head back out towards the Oort cloud, ready to repeat its orbit as long as it remains intact. In a few, rare cases, a comet could instead collide with a body in the inner solar system.
The existence of the Oort cloud and occasional disturbances are thought to be the explanation for long-periodic comets, whose journey around the Sun takes between 200 and thousands of years per orbit. In fact, the gigantic size of these comets' orbits, as inferred from observations of their passage through the inner solar system, was the motivation for postulating the existence of the Oort cloud in the first place. What we know so far about long-period comets supports this hypothesis, even though the Oort cloud has not yet been observed directly.
Thus, there is a direct connection between close encounters with stars and comet impacts on Earth, and to understand the latter, you need to research the former: How often do close stellar encounters occur? What encounters have there been in the past how have they influenced the frequency of impact events, and do they have a bearing on our estimates for the present?
Reconstructing stellar motions and close encounters
Answers to these questions depend crucially on the available data for stellar motion in our direct cosmic neighborhood, and in particular on how precise those data are. Now, Coryn Bailer-Jones, a staff scientist at the Max Planck Institute for Astronomy, has published the first systematic estimate of the probability for such near stellar encounters. Bailer-Jones' calculations are based on the first data release (DR1) from the ESA astrometry satellite Gaia, which was published on 14 September 2016.
Gaia's mission is to measure the position, distance, and velocity of over one billion stars in our Galaxy, with an accuracy which has never before been achieved for so many stars. The final results will contain precisely what is needed to describe the orbits of stars in our galactic neighborhood: where these stars are in the surrounding space, and in which direction they are moving.
While the full reduction and analysis of the Gaia data is still in progress, DR1 published preliminary results on a special data subset that goes at least part of the way. This is the so-called TGAS catalogue of 2,057,050 stars, which makes use of both the first Gaia data and the data of the ESA-satellite Hipparcos, twenty years earlier, to yield the best stellar distances and motions to date.
Using this catalogue, Bailer-Jones identified 468 stars that, at their current rate of movement, would seem to come within 32.6 light-years (10 parsec) of the Sun, either in the past or in the future. For these stars he performed a computer simulation of their orbits – taking into account the gravity of our home galaxy – to determine more precisely their closest approach to the Sun. He found that 16 stars will pass, or have already passed within less than 6.5 light-years (2 parsec) of the Sun. (The light-year values are not round numbers since Bailer-Jones chose his sample values to be round in another distance unit used by astronomers, 1 parsec = 3.26 light-years.)
The closest future encounter
The closest encounter found is for the star Gl 710 ('Gliese 710'), which has been known for some time to be heading for a close encounter with the Sun. The new data and calculations show that this encounter, which will take place in 1.3 million years, comes much closer than was thought before DR1: just a quarter of a light-year (or 16,000 times the Earth-Sun distance), well within the Oort cloud. This confirms similar calculations by two Polish astronomers, Filip Berski and Piotr Dybczyński, in a 2016 article, also based on the first Gaia data release . Although Gl 710 has a comparatively low mass, just 60% the mass of the Sun, its velocity is also low, giving it plenty of time for exerting its gravitational influence on the Oort cloud. Given that Gl 710 may bring its own Oort cloud, this raises the intriguing possibility that our Sun could even swap comets with passing stars!
Deriving the rate of encounters
But beyond identifying the closest encounters, Bailer-Jones went an important step further. Astronomical surveys are never complete; they will only detect their targets down to some minimal brightness, and miss light sources dimmer than that. Bailer-Jones modelled the encounters that DR1 detected, compared them to the survey's limitation, and used statistical tools to estimate how many stellar encounters his DR1-based evaluation was likely to have missed. In addition, Bailer-Jones took into account the uncertainties of the Gaia data, which are known from systematic studies by the Gaia team. For each star, he calculated not only the orbit corresponding to the nominal Gaia values for distance, position and motion, but the orbits for a whole swarm of virtual stars. The swarm represents the (sometimes large) uncertainties in the data – and hence the fact that, with a certain probability, the derived parameters for the encounter could be different from the nominal estimates. This gives a more reliable estimate than relying on the nominal data alone.
As a result, Bailer-Jones obtained the first systematic estimate of the average stellar encounter rate for the past and future 5 million years. (The model reconstruction is not accurate enough to extrapolate to encounters further in the past or future with the DR1 data.)
The result, which meshes with earlier, less systematic estimates, is that within each period of a million years, between 490 and 600 stars will pass the Sun within a distance of 16.3 light-years (5 parsec) or less. This covers stars of all masses, although the most common ones are low mass stars, like Gl 710. Within a smaller distance of 3.26 light-years (1 parsec), some 19 to 24 encounters are expected per million years. Given that it takes several million years for disturbances to abate, our Oort cloud seems to be in fairly constant upheaval, with no extended periods of calm in between.
Not quite to the dinosaurs
This is valuable information for the scientist attempting to calculate the rate of cometary impacts on Earth. Gaia's next data release – DR2, in April 2018 - should allow an extension of these reconstructions to 25 million years from the present. The release after that, DR3, will contain estimates of the masses and radii of the observed stars, based on the data from Gaia's on-board spectrometer. Bailer-Jones is, in fact, in charge of the group of Gaia data analysts tasked with deriving these and other astrophysical quantities from Gaia's huge treasure trove of measurements. Information about masses and radii will allow the astronomers to estimate how large the disturbances in the Oort cloud will be on average, allowing for more precise estimates of their consequences for the impact rate.
Extending the reconstructions further will be difficult; as the simulations reconstruct longer orbits, uncertainty about the mass distribution in the Milky Way becomes a major limiting factor (although the Gaia data themselves will help us improve knowledge of this). Astronomers in search of stars that might be responsible for sending a comet to Earth that, 66 million years ago, caused or at least hastened the demise of the dinosaurs, will need to know our home galaxy much better than we currently do.