Earth Has a 27.5M-Year 'Heartbeat', but We Don't Know What Causes It | Hacker News Reader

Earth Has a 27.5M-Year 'Heartbeat', but We Don't Know What Causes It

sciencealert.com

88 pointsCarCooler30 comments

I saw two independently published papers the other day showing a 27.5M year cycle. One was this paper talking about cycles in volcanic activity, another was talking about mass extinctions. Article here: https://phys.org/news/2020-12-mass-extinctions-land-dwelling... . I'm going to hazard a guess that the former is very likely a direct cause of the latter. We suspect this to be true of most of the worst mass extinctions, like the Permian extinction. With the dinosaurs it might have been the Chicxulub asteroid impact, or that might have merely been the finishing blow to a system under severe stress from the Decan Traps flood basalt eruptions in India that had been underway for millions of years by that point. There are plenty of indications that species were declining well before the impact.

The solar system takes 230 million years to make one complete orbit around the Milky Way, so I don't think this is likely caused by our passage through the galaxy. It's more likely an internal rhythm in activity deep within the Earth.

I wouldn't be too quick in dismissing the Galaxy. The galaxy isn't uniform so something could easily be related to one of the harmonics. And 27.5M years happens to be quite close 1/8th of a galactic year, so something to do with the earth passing in/out galactic arms or similar is not out of the question.

To be slightly pedantic: our sun is a star in the galaxy, so as the entire galaxy rotates we would not pass in and out of a galactic arm. We would rotate with it.

Maybe not. https://en.wikipedia.org/wiki/Density_wave_theory

The galaxy isn't a rigid spinning object though, it's composed of many stars acting under gravitational influences. Like our solar system, objects closer in will orbit faster than objects further out. Rather than fixed structures, the arms are more like a standing wave of greater density.

Some related links

https://en.wikipedia.org/wiki/Density_wave_theory

https://en.wikipedia.org/wiki/Galaxy_rotation_curve

https://www.nature.com/articles/s41550-018-0627-5

As I understand it, galaxies don't rotate like rigid bodies: stars further from the centre take longer to go round than ones closer in, much like planets further from the sun take longer to complete an orbit.

If stars that start off inside an arm remain so, but the outer part of the arm goes round more slowly, the arm would get wound 'tighter' as time goes on, and this doesn't seem to happen.

I think what's really going on isn't yet well understood, but it's likely to be that the arms are something more like a standing pattern, and stars do indeed move in and out of the arms as time goes on (something like the way a traffic jam can move more slowly than the individual cars that it's made of).

Fair point. So maybe we should say that our sun would occasionally be a part of the galactic arm as the pattern morphs and re-forms. A fun concept.

It turns out that the stars further from the center don't take as long as one might think. Observations of this phenomenon led to the development of the theory of dark matter.

https://en.wikipedia.org/wiki/Galaxy_rotation_curve

Not pedantic at all, I for one have never speculated about galactic dynamics

So in the solar system the inner planets have faster orbits than the outer planets, I assume it would be the same on a galactic level - if we’re a slow poke out on the edge, could there be some region of the galaxy that orbits the core 8x faster than us such that we’re periodically effected by it?

Edit LOL my sibling comments have filled in some info for me on the mean time

Most stars would not pass in and out of each spiral arm of a galaxy.

Perhaps it’s just our Sun that does it.

I believe the solar system actually cycles from below the plane of the galaxy to above the plane of the galaxy. I've heard numbers between 25M years and 40M years. Debris on that plane could shake comets loose from the oort cloud. Other theories on dark matter support the idea that crossing through dark matter could increase the temperature in the cores of planets, leading to increased seismic activity.

Those are good points. Maybe it does have to do with our galactic orbit.

In addition to the galactic orbit the disk itself has it's own local gravitational pull that keeps stars in the plane by pulling up/down when a stars trajectory takes it below/above the local disk. When we are out of the plane we loose the protective cover all the shielding the disk provides from the galactic core (electrons go zing). When in the middle of the disk there is a higher density of mater to interact with (rocks go boom). Our typical exposure to "orbits" earth & moon tends to be that they are roundish flat & smooth and we might be tempted to think the suns orbit around the galactic core is smooth, but it may be better to think of us as also bobbing up and down the highth of the disk as we go.

That's an interesting way to visualize it. As usual, every phenomena is more complex the more you dig into it.

> Other theories on dark matter support the idea that crossing through dark matter could increase the temperature in the cores of planets, leading to increased seismic activity.

Source? That’s fascinating, almost beyond belief, and I’d love to dig into related research.

Seen various versions of this around: https://astronomy.com/magazine/2019/07/dark-matters--shadowy...

230 million years? perhaps our Solar system is traversing each spiral arms of our Milky Way.

Just reaching.

Looks like bogus statistics to me. (Caveat: I only read the sciencealert link not the underlying publication). But with a sample of 89 "events" over a purported 10-cycle period we are in the small sample realm where it is easy to get spurious apparent relationships.

People love to see patterns, and if you aren't really careful you can "prove" a pattern in essentially random data.

Underlying:

https://www.sciencedirect.com/science/article/pii/S167498712...

Section 3.3 there, the Fourier analysis, looks really weak.

If you just plot 89 random numbers between 0 to 270 it'll have strong spectral peaks. Their method:

> After permuting the order of the intervals, a new time series was re-calculated based on the new set of intervals. This method ensures that the test time series are between 0 and 260 Ma and the number of events remains at 89. Ultimately, only ~ 4% of the test datasets produced a higher spectral power at 27.5 Myr, indicating a confidence level of at least 96% for that period.

... seems very fishy to me. I could believe that the spectral power at that frequency is the mode in the data. I don't see a proof that the particular spectral power associated with 27.5 is statistically significant given a random set of events.

It is incorrect to take 27.5 Myr as a prior fixed quantity when there was no prior theoretical reason to expect that period. Correct analysis would ask, given a randomly generated data set, what are the odds as strong a signal as this is generated for any recurring period in some range (like 3Myr-100Myr)

If you had read the article you would have been informed that this publication is one in a long line of similar ones going back as far as 1920 who all came to similar numbers.

You are right, if there was only this one paper it would be entirely unremarkable and informing the general public would be wrong. But that is not the case because the article fortunately does not limit itself to pointing to just that publication.

Just because everyone's seeing the same pattern in random data doesn't mean there's a pattern at all. For very good evolutionary reasons, humans are extraordinarily good pattern-recognition machines: we see ghosts in data when nothing exists. Our collective false positives are quite high.

Going back to about 1920, everyone saw canals on Mars, too.

For mass extinctions there is the common explanation that it takes a long time after each extinction before enough diversity is built up again for another major extinction.

For example after the current Anthropocene one we likely won't have another one any time soon because there is not enough diversity left for a true mass extinction.

That gives a natural effect where mass extinctions are spaced with some minimum interval. They're not independent events.

on the other hand, it is well-known in medical science that people who have say, a serious traumatic brain injury (TBI), are far more likely to get many other serious injuries soon afterwards and also at high risk for a long time afterwards for another TBI.

That’s a great point. But it’s not necessary for the events to be periodic.

Could this be related to our solar system orbiting through our galaxy?

not a bad one! I guess we would find out, when we find the heartbeat of other planets in our solar system. If all have same fingerprint, that would be a good indicator of a system-wide effect.

I like the idea that it is a slow charge buildup from galactic ambient high energy rays.

These would cause a very slow charge build up, and the piezo electric effect of the earth would come into play. E.g. a very slow quartz crystal clock cycle.

Reply on news.ycombinator.com