Imagine an Earth-like planet orbiting a Sun-like star in the inner halo of the Milky Way. As a halo star, it will likely be somewhat metal-poor, having formed early in the life of the galaxy, but other than that, we can assume the planet and star to be like our own. The system lies about 4 kpc from the Galactic Center; as such, it completes one orbit of the galaxy in somewhere between 100 million and 200 million years.

While in the halo, the star is unlikely to be near other stars, but during each orbit, it passes through the disk of the galaxy twice. If the star is traveling at $\sim$300 km/s, then it should take it about 3.3 million years to travel through the disk, where it will pass by numerous other stars and other objects. I'm trying to determine if these passages will show up in the planet's geologic record many millions of years in the future. Ideally, alien geologists (with the same tools as human geologists today) would be able to use recurring signs of interactions in the disk to figure out the period of the star's galactic orbit.

I only have one vague idea: the planet would be more likely to be near a supernova while in the disk, which would cause changes in isotopic abundances in certain rock layers. However, I have no idea whether or not this is plausible, and if it would be detectable.

Would the trips through the galactic disk be apparent in the planet's geologic record? If so, how would they show up?

  • 1
    $\begingroup$ When you say metal poor, you mean metal for a chemist or metal for an astronomer? $\endgroup$
    – L.Dutch
    Feb 6, 2020 at 16:48
  • 2
    $\begingroup$ @L.Dutch-ReinstateMonica Metal in the astronomical sense. $\endgroup$
    – HDE 226868
    Feb 6, 2020 at 16:48
  • $\begingroup$ You want a metal-poor star with a metal rich planet like Earth? $\endgroup$
    – Alexander
    Feb 6, 2020 at 18:17
  • $\begingroup$ @Alexander Relatively metal-poor, yes, but that's certainly not a problem. Kaptyen's star is an excellent example of a metal-poor halo star with a possible terrestrial planet. $\endgroup$
    – HDE 226868
    Feb 6, 2020 at 18:19
  • $\begingroup$ @HDE 226868 I wouldn't call it "certainly", but I agree that it is possible. $\endgroup$
    – Alexander
    Feb 6, 2020 at 18:29

3 Answers 3


Space dust deposition might be different in the disk.

Cosmic dust of extraterrestrial origin rains down on the earth all of the time - thousands of tons of it. In your scenario, as the planet passed through the disk, one would expect a change in the quantity and composition of dust accumulating. The dust might be comprised in part of dense elements unusual to find in the crust, like iridium. Or the dust particles might contain some with microscopic shapes characterizing an extraterrestrial origin.

It would be easiest to study this someplace like the moon where dust can pile up undisturbed. Comparable accumulators on the Earth would be the deep ocean and Antarctic ice. I looked to see if anyone has studied cosmic dust accumulation in the ice. I found this.

Interstellar 60 Fe in Antarctica

Earth is constantly bombarded with extraterrestrial dust containing invaluable information about extraterrestrial processes, such as structure formation by stellar explosions or nucleosynthesis, which could be traced back by long-lived radionuclides. Here, we report the very first detection of a recent 60 Fe influx onto Earth by analyzing 500 kg of snow from Antarctica by accelerator mass spectrometry. By the measurement of the cosmogenically produced radionuclide 53 Mn , an atomic ratio of 60 Fe / 53 Mn = 0.017 was found, significantly above cosmogenic production. After elimination of possible terrestrial sources, such as global fallout, the excess of 60 Fe could only be attributed to interstellar 60 Fe which might originate from the solar neighborhood.

If you had strata from an accumulator dating back a very long time (for example a deep ocean core)you might note a period difference in dust constituents corresponding the passage thru the disk, or maybe just a thicker layer of dust corresponding to the greater amount of material in the disk.

No supernovas. Nothing flashy. Just dust, and more dust.

  • $\begingroup$ Side note for persons who study cosmic dust in real life: the linked article used accelerator mass spectroscopy to look for the 60 Fe isotope. But if you wanted to look for macroparticles of a shape characterizing extraterrestrial formation you could mechanically filter and then use flow cytometry (flow dustometry?) to sort by shape, enriching your sample for the extraterrestrial fragments. A little like panning for gold though that sorts by weight, not shape. $\endgroup$
    – Willk
    Feb 12, 2020 at 14:40

Assuming an Earthlike planet orbiting a halo star and passing through the galactic disk, let's look at the case for geological evidence of supernovae. Because the planet is Earthlike we can assume the geological evidence will be effectively the same for the Earth. If its passage through the galactic disk takes it close to more supernovae than the Earth has experienced in its past, then that evidence will give a stronger geological signal.

Past supernovae might be detectable on Earth in the form of metal isotope signatures in rock strata. Subsequently, iron-60 enrichment has been reported in deep-sea rock of the Pacific Ocean by researchers from the Technical University of Munich.[11][12][13] Twenty-three atoms of this iron isotope were found in the top 2 cm of crust (this layer corresponds to times from 13.4 million years ago to the present).[13] It is estimated that the supernova must have occurred in the last 5 million years or else it would have had to happen very close to the solar system to account for so much iron-60 still being here. A supernova occurring so close would have probably caused a mass extinction, which did not happen in that time frame.[14] The quantity of iron seems to indicate that the supernova was less than 30 parsecs away. On the other hand, the authors estimate the frequency of supernovae at a distance less than D (for reasonably small D) as around (D/10 pc)3 per billion years, which gives a probability of only around 5% for a supernova within 30 pc in the last 5 million years. They point out that the probability may be higher because the Solar System is entering the Orion Arm of the Milky Way. In 2019, the group in Munich found interstellar dust in Antarctic surface snow not older than 20 years which they relate to the Local Interstellar Cloud. The detection of interstellar dust in Antarctica was done by the measurement of the radionuclides Fe-60 and Mn-53 by highly sensitive Accelerator mass spectrometry, where Fe-60 is again the clear signature for a recent near-Earth supernova origin.

This suggests the amount of Fe-60 present in geological strata may be taken as evidence of close supernovae events.

However, there are more extreme possibilities for the impact of supernovae on Earthlike planets. These will occur when the supernovae are a bit too close for comfort.

Gamma ray bursts from "dangerously close" supernova explosions occur two or more times per billion years, and this has been proposed as the cause of the end Ordovician extinction, which resulted in the death of nearly 60% of the oceanic life on Earth.

The good thing is dangerous close supernovae will be, in general, rare. Unless, of course, the galactic disk is rich in stars undergoing supernova. Then passage through the galactic disk will be somewhat fraught.

In conclusion, the impact of supernovae will appear in geological evidence both in terms of the isotopic abundance of Fe-60 and, possibly, in mass extinction events.

Please note: This has confined itself to considering the evidence for supernovae on the hypothetical Earthlike planet orbiting a halo star.


Near-Earth supernova

  • $\begingroup$ I was also thinking of gamma ray bursts when reading the question. A reasonably close gamma ray burst will completely eradicate all life on a planet and properly sterilize it for a while. Far away from the galactic center where our sun is, these kind of events happen only once every few billion years so it is plausible that earth was never hit in the last 10 billion years. Closer to the galactic centers these are a lot more frequent. So it has been argued that there can be no highly evolved life much closer to the galactic center because it gets wiped out every 100 million years or so. $\endgroup$
    – quarague
    Feb 7, 2020 at 12:42
  • $\begingroup$ Honestly, I'll take 5%. Also, it looks like the probability quoted is proportional to the supernova rate (which I'd assume is proportional to the star formation rate), so I suspect that by changing the setting to a star-forming galaxy, I should be able to improve those odds substantially. $\endgroup$
    – HDE 226868
    Feb 7, 2020 at 18:20
  • $\begingroup$ @quarague Mass extinctions every 100 million years! Now that's what I call a natural selection pressure. In fact, it's possible to argue the reverse: that evolution near the galactic centre will run 'faster'. Expect a proliferation of very exotic lifeforms. By 'faster' doesn't imply a direction for evolution, simply increased variation being necessary for survival in an extreme environment. $\endgroup$
    – a4android
    Feb 7, 2020 at 22:31
  • $\begingroup$ @HDE226868 Very sensible. Adjusting the star formation rate to the proportional to supernova rate and choosing the right sort of galaxy is a good move. Knowing how to tweak your world is part of worldbuilding, so you can build a world to meet specific needs. $\endgroup$
    – a4android
    Feb 7, 2020 at 22:35
  • $\begingroup$ @quarague no, it wouldn't. The actual radiation wouldn't have any significant immediate impact thanks to to the atmosphere (and since it's a short burst, only one hemisphere would be exposed), only causing a brief elevation in UV. The long term effects comes increased UV due to ozone depletion, but that would work itself out after a few years and wouldn't affect subterranean or deeper aquatic life. So no, even a close GRB wouldn't "eradicate all life on a planet and properly sterilize it for a while". $\endgroup$ Feb 11, 2020 at 18:26

Signatures aren't your problem, Uranium is

Most dating is done with Uranium and if you can't get absolute ages on surrounding rocks you might be able to figure out this pattern exists but you need corresponding strata to have minerals like zircons (mostly found in volcanic ash) with uranium imperfections to get an accurate age. They could easily isolate the strata from the pass-throughs as long as there are impacts or such deposits with different isotopic signatures, but I can't come up with a dating method that works on long enough time scales without using heavier elements. (We come up with way more subtle effects all the time). Maybe with a list of element concentrations against a list of radiometric dating techniques?

An alternative

Maybe once the fact a pattern is found to exist geologically astronomical models could work out the likely time frame


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .