[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Evidence (or otherwise) for Nemesis

For a dinosaur list, there sure are frequent items to provoke astronomers.

The latest scoop on "Nemesis" (yes, the quotations imply my assessment)
seems to be as follows:

Whitmire and Jackson (1984 Nature 308, 713) followed the Raup and Sepkoski
periodic-extinction claims by asking what sort of astrophysical clock would
have the asserted 26-million-year period and be able to induce mass
extinction (i.e. drive up the impact rate). A disturbance in the Oort
cloud would certainly fill the bill (the Oort cloud being the postulated
reservoir for long-period comets, needed to maintain the presence of any
such in the inner solar system against their trapping into short-period 
orbits and subsequent destruction by successive solar passages - the Oort
cloud has yet to be directly observed, but objects are known to populate
its inner counterpart the Kuiper belt). They suggest that a solar companion
in a 26-million year, elongated orbit could trigger such influxes of comets.
The required mass, given that we have no companion identifiable as a normal
star, would make it a "brown dwarf" - formed in the same way as ordinary
stars, but with insufficient mass to sustain core hydrogen fusion. Such an
object would be very difficult (and correspondingly interesting) to detect.
A 26-million year period implies a semimajor axis to the orbit of Nemesis
given by Kepler's third law as (26,000,000)^2/3 = 88000 AU, or 1.4
light-years; to see a strong periodic signal in gravitational disturbance
requires that the orbit be highly eccentric, so the maximum distance
from the Sun would be close to twice this (round numbers, halfway to the
next stellar system though of course not necessarily in that direction).
The same issue of Nature also contains preliminary dynamical analysis of the 
idea, and discussion of crater ages and periodicities.

Some problems with the dynamics of "Nemesis" were soon addressed - both as 
the effects of any such close passage on the planetary systems (see
Hills 1984 Nature 311, 636) and the long-term stability of its orbital 
period. The latter is a significant problem; on a scale of 2 light-years,
the Sun's gravitational environment is grainy in both space and time.
Passing field stars would vary the period of such an orbit by 10-20% since 
early Mesozoic (Hut 1984 Nature 311, 638). Furthermore, the tidal
forcing by the overall potential of the Galaxy limits the range of
orbital orientations that are stable over this time over and above
variations due to passing stars (Torbett and Smoluchowski 1984 Nature
311, 641). Additional perturbations by giant molecular clouds (with masses
of order 10^6 solar, concentrated along spiral arms) would exacerbate 
the lack of stability.

A 1986 conference on "The Galaxy and the Solar System" (proceedings
published by the University of Arizona Press) brought together 
many of the workers both pro and con. Evidence for some kind of disturbance
to the Oort cloud was claimed from the distribution of orbits of 
"fresh" long-period comets. Muller summarized the evidence pointing to 
a distant solar companion. Scott Tremaine (a serious celestial-mechanics pundit)
"This chapter contains a brief review of the major astronomical explanations 
for a
possible 25 to 35 Myr period in the cratering and extinction records. The only 
theory is that the sun has a distant companion star; however, the required 
orbit for
this companion is an improbable one, since the companion will probably escape 
in a
time short compared to its present age. The most likely resolution of this
uncomfortable situation is that there is no periodicity: in both the cratering 
extinction records, the statistical evidence for periodicity is very weak."

There is, of course, no substitute for seeking the beast. The best hope of
finding Nemesis directly is in the far-infrared, as it should have a 
temperature only tens of Kelvin (formerly known as degrees K) - imagine
finding Jupiter only be its intrinsic luminosity, sans reflected sunlight.
The data from the Infrared Astronomical Satellite were combed for such an
object among the nearly 10^6 detected sources. The problem is that the
orbital motion would have been too slow to detect over the mission's
1-year cryostat lifetime, and similarly the parallax would have been
below the IRAS threshold - so one would look for a fairly bright far-IR
source with at most a very faint optical counterpart, not identified
with something else like a galaxy or dust cloud (and spectral evidence could
also separate the classes independent of what is seen in visible light).

Optical observations were also of service. Saul Perlmutter's
dissertation at Berkeley involved measuring parallax-based
distances to 2770 candidate red dwarfs/brown dwarfs, of which none
proved to be very nearby. Statistics of wide binary stars (Peter Garnavich
dissertation, Washington University 1991) show an apparent cutoff
near 0.1 parsec=0.3 light-year, with the statistical properties of
the sample suggesting that if anything too many bogus companions
remained in the sample.

There you have it to this point. No solid candidate for Nemesis, and
celestial mechanics makes it unlikely that any such distant companion
would have kept a constant enough period long enough for a periodic
signal in the fossil record. I do wish to end by stressing that comet
impacts are potentially very catastrophic things - you were all watching
last summer, I trust, when some of the Shoemaker-Levy 9 fragments
blanketed regions larger than the Earth's surface area with debris,
against 2.5 Earth gravitational accelerations?

And now back to dinosaurs...

Bill Keel                       Astronomy, University of Alabama