Sun's Dark Companion 'Nemesis' Not So Likely 306
TravisTR passes along a story about the death of Nemesis. "The data that once suggested the Sun is orbited by a distant dark companion now raises even more questions... The periodicity [of mass extinctions] is a matter of some controversy among paleobiologists but there is a growing consensus that something of enormous destructive power happens every 26 or 27 million years. The question is what? ... another idea first put forward in the 1980s is that the Sun has a distant dark companion called Nemesis that sweeps through the Oort cloud every 27 million years or so, sending a deadly shower of comets our way. ... [Researchers] have brought together a massive set of extinction data from the last 500 million years, a period that is twice as long as anybody else has studied. And their analysis shows an excess of extinctions every 27 million years, with a confidence level of 99%. That's a clear, sharp signal over a huge length of time. At first glance, you'd think it clearly backs the idea that a distant dark object orbits the Sun every 27 million years. But ironically, the accuracy and regularity of these events is actually evidence against Nemesis' existence."
Re:How long since last time (Score:5, Informative)
Read the Fine Article.
We've got lots of time -- we're only 11 million years into this cycle.
Re:How long since last time (Score:5, Informative)
Re:Second comment debunks (Score:5, Informative)
that's the third comment.
Here is a bit from the second comment:
Debunked nicely in the comments (Score:5, Informative)
Re:Second comment debunks (Score:3, Informative)
The preprint has been peer reviewed and has been accepted for publication in Monthly Notices of the Royal Astronomical Society, one of the most prestigious astrophysics journals on this planet.
Re:There is worse... (Score:4, Informative)
Some more debunking in the second comment:
First off, there is likely no "growing consensus that something of enormous destructive power happens every 26 or 27 million years". It is an old idea, probably originated with the terrible paper by Raup and Sepkoski 1986, which I have criticized on the web several times; (...) [Not to poison the well, but Bambach published lately in Ruse and Sepkoski eds "Paleontology at the High Table." One must take a dim view with the abilities of anyone that choose to cooperate with "philosopher of biology" and known stealth creationist Ruse.]
Re:period of passing through the galaxy ecliptics? (Score:5, Informative)
The sun doesn't just orbit the center of the galaxy, though. It also moves up and down relative to the galactic plane. Some have suggested that whenever the solar system reverses direction in that oscillation, very bad things happen, possibly due to the Oort Cloud experiencing some lag in reversing direction relative to the rest of the system. The sun essentially winds up off-center in the Oort Cloud, and in comparison to normal periods a lot of comets get kicked into the inner solar system as a result of this imbalance.
Re:I am doubtful (Score:5, Informative)
Check out the Wikipedia article on the Oort Cloud. The Oort Cloud is thought to be well over a light year across. Out on its fringes the influence of the sun's gravity isn't much stronger than the pull of nearby stars, or the galactic core itself. So whenever the oscillation reverses direction and the sun begins moving back toward the galactic plane, a lot of stuff out on the fringes doesn't move neatly with it. Some of it will become gravitationally unbound from the solar system, but some of it will find its orbit perturbed and start heading inward. Whether that's enough stuff to lead to mass extinctions here on Earth is another matter.
This article mentions disk tides, encountered most strongly as the Sol system passes thru the galactic plane, as the possible culprit in disturbing the Oort Cloud on a regular basis:
http://www.americanscientist.org/issues/pub/perturbing-the-oort-cloud [americanscientist.org]
I don't accept the periodicity (Score:1, Informative)
IAAP (I am a paleontologist), and I've actually met Bambach, one of the guys mentioned in the article. He's a very good paleontologist.
However, these claims of periodicity in extinctions have been considered for a long time, ever since the pioneering work of Raup and Sepkoski in the 1980s. It is not an exaggeration to say that the question of periodicity is controversial.
A basic problem with the claims back then, and with the claims based on improved data sets now, is that this statistically difficult data to work with. It's trivial to do a Fourier transform of the data and pluck out dominant frequencies to a certain confidence level, but the underlying data has issues. Geologists are constantly revising the timescale to ever greater levels of refinement, which slightly changes the numerical timing of the events in question (it's like the time of your sample points keeps shifting around by a few percent). The extinction and other data is typically collected over intervals of geological time that are *not* equal in duration (i.e. it's unevenly aliased). Extinctions in the interval are assumed to have occurred at the end of the interval. This is most often the case for most of the extinctions, but not all of them. This is tricky stuff to work with when doing frequency analysis. And although the data compilations now are much better than a few decades ago, they inevitably have issues due to the vagaries of fossil preservation (some sites/times have much better fossil preservation than others), and of the rock record itself (e.g., the rise and fall of sea level that causes shifts in environment and depositional rates, so some times can be almost unrepresented whereas other times can have plenty of rock).
My guess is that if the frequency signal is real it will say more about some kind of cyclicity that affects terrestrial depositional and climatic processes which in turn affect fossil preservation (something analogous to Earthly Milankovitch cycles or maybe tectonic cycles), rather than some kind of entirely external process that specifically or only drives extinctions. Failure to preserve fossils at certain points in the rock record would be read as an "extinction", even if the creature survived for a bit longer. Either that, or it's some kind of numerical artifact/resonance frequency derived from hitting the practical limits of sample spacing. You're basically trying to pick a frequency signal out of data that is sampled only a few times higher than the signal (the mean sample spacing is ~3.6 to 11 million years depending on the datasets they're using).
Honestly, looking at their plot of extinction rates overlain with the 27Ma frequency (their Figure 2), I'm not all that impressed with the correspondence. Some major extinctions are 180 degrees out of phase (e.g., the Late Devonian extinction at ~372Ma and the Early Carboniferous extinction at ~326Ma) and some extinctions are doubled (e.g., the two towards the end of the Permian near 250Ma).
Re:11 million years (Score:4, Informative)
Well, average life-expectancy of a species is 5-million years. Homo Sapience has already doubled that putting us at the extreme end of the scale that gives this average.
How are you doing your math? The genetic evidence shows that Homo Sapiens can be traced back 200,000 years. Nowhere near the 5 million you are stating as an average for species longevity. If you are counting Australopithecus anamensis, that would get you back to 4 million years, but I would hardly consider it to be the same species as us.
Furthermore, the actual average longevity of a species is 1 million years, not 5 (as evidenced here [pbs.org]. Just because 10 million years appears to be an extreme upper limit does not make the average 5 million.
TFA doesn't think this is it either (Score:3, Informative)