Kepler, CoRoT and future space missions should give us an estimate of how common small, rocky planets are in the galaxy. But there is much we can do from Earth, as Jay Farihi told the Royal Astronomical Society’s 2010 meeting today. Farihi’s team used data from the Sloan Digital Sky Survey to conclude that rocky worlds emerge around at least a small percentage of A- and F-class stars. The method: Analyze the position, motion and spectra of white dwarfs found in the SDSS survey. Farihi was interested in the presence of elements heavier than hydrogen and helium in the stellar atmospheres.
Finding calcium, magnesium or iron in the atmosphere of a white dwarf is, Farihi believes, evidence of rocky debris, and the new work shows that at least 3 percent and as much as 20 percent of all white dwarfs may be contaminated in this way. Such elements should have sunk below the photosphere in the high gravity of a white dwarf, leading to the belief that any visible contamination must be the result of external causes. Farihi sees the heavier elements as debris left over from what once may have been planetary systems containing terrestrial worlds around these stars.
Assuming this is the case, then, we can take this debris as evidence for the existence of such systems around A- and F-class stars. Moreover, the composition of the debris shows that many of these stars are polluted with material containing water. Says Farihi:
“In our own Solar System with at least one watery, habitable planet, the asteroid belt – the leftover building blocks of the terrestrial planets – is several percent water by mass. From our study of white dwarfs, it appears there are basic similarities found among asteroid-like objects around other stars; hence it is likely a fraction of these white dwarfs once harbored watery planets, and possibly life.”
Why focus on planets or planetesimals as the source of the heavy metal contamination? Farihi demonstrates in the paper on this work that there are no correlations between calcium abundances and the distribution of these stars in relation to interstellar materials that could have been the source. Two thirds of the stars under study are located above the galactic gas and dust layer, and study of their motion shows long residence in areas where interstellar materials are all but absent. Assuming planetary materials as the cause of this signature, the paper adds:
…at least 3.5% of white dwarfs appear to be polluted by circumstellar matter, the remains of rocky planetary systems. This translates directly into a similar lower limit for the formation of terrestrial planets at the main-sequence progenitors of white dwarfs, primarily A- and F-type stars of intermediate mass. While this fraction is likely to be significantly higher…, it is difficult to quantify without a commensurate examination of all the cool SDSS white dwarfs, which is beyond the scope of this paper. The appearance of hydrogen in DZA stars suggests a common origin for both heavy elements and hydrogen, and indicates DZA star pollution by water-rich minor planets may be semicontinuous on Myr timescales.
White dwarf spectral classification schemes use an initial letter D followed by letters describing secondary features of the spectrum, which is what the DZA reference above is all about. More on white dwarf classifications here.
The significance of this work is that it rules out the interstellar medium as the source for the metal pollutants in white dwarfs like these. Two conclusions follow: At least a small percentage of A- and F-class stars build terrestrial planets (‘terrestrial’ meaning small, rocky worlds) whose debris supplies the heavy elements, and “…the pattern of hydrogen abundances in DZ stars is likely a reflection of the diversity of water content in extrasolar planetesimals.”
The paper is Farihi et al., “Rocky planetesimals as the origin of metals in DZ stars,” accepted by Monthly Notices of the Royal Astronomical Society (abstract).
Interesting statistics – 3.5% of A & F stars with terrestrial planets. Does that hold for colder stars? If so, then what are the odds for exo-Earths then? If an exo-Earth can exist around G & K stars – themselves roughly 20% of all stars – then just 0.007% of stars will be suitable. Just how many then have planets that are Earth-like is anyone’s guess, but if we assume they’re similarly endowed like the Sun, there will be around ~700 million exo-Earths in our Galaxy. However Earth was only Earth-like – with a breathable amount of oxygen – for the last ~700 myr and may only remain habitable for another ~500 myr. Thus Earth will be ‘Earth-like’ for (a minimum of) 10% of the Sun’s 12 Gyr as a fusion-powered star. Thus, averaging again, there’s about 70 million Earth-like planets presently in our Galaxy.
Odds are that it’s a totally meaningless number. Exoplanet studies are showing us to always expect the unexpected. Consider Europa – covered in ice, but probably an oxygen-rich ocean beneath. Habitable? Depends on who by.
Adam- Why do you say that earth will only remain habitable for another 500 million years? We still have about 5 billion years until the sun becomes a red giant…
“Oxygen-rich” ocean on Europa? Is that true? Who says?
Hi Joel
The Sun is fine but the geophysical processes that keep carbon dioxide levels low will do their job too well by then, leading to the extinction of all life on land when carbon dioxide is too low for photosynthesis. Thus the end of the habitable Earth. James Kasting & Ken Caldeira computed that minimum figure of 500 myr in a 1992 paper. The oceans begin leaking away into space as the stratosphere gets too hot in about ~1 billion years, so all life will be extinct in c.1.5-2 Gyr from now.
However we might get a reprieve for some kind of life if nitrogen’s partial pressure goes down as CO2 declines, but eventually the greenhouse takes over in c.2.3 Gyr. Another life extension might come if the oceans are subducted and Earth dries out – that reduces the instability against a runaway greenhouse. Lakes might persist near the poles in that scenario right up to the Sun leaving the Main Sequence in c.5.5 Gyr. But Earth won’t be “Earth-like” in such a dry planet scenario.
Incidentally the Sun’s departure from the Main Sequence is a leisurely affair, with a lengthy ‘sub-giant’ phase before it really starts ballooning as a Red Giant.
Adam,
I had thought carbon dioxide is low in the atmosphere because it is being fixed very efficiently by plant life. No plant life, I thought, and CO2 will come back, via geological processes (vulcanism, mostly). Now you seem to be saying the opposite. Where can I read about this? http://en.wikipedia.org/wiki/Carbon_cycle seems to confirm my view, saying volcanic eruptions produce carbon dioxide, but I am not sure I am reading it correctly and how it applies in the REALLY long term.
Eniac: I believe Adam is referring to the carbonate-silicate cycle, which is also an important process for removing carbon dioxide from the atmosphere. I’m not sure how important the relative contributions of the biological vs geological processes are for removing carbon dioxide from the atmosphere are though.
Here’s the paper in case anyone is interested – “The life-span of the Earth revisited”
http://www.nature.com/nature/journal/v360/n6406/pdf/360721a0.pdf
Unfortunately there is a paywall on this paper, but the abstract does not mention the geological release of CO2, only the removal. I assume this is addressed in the text? It would also be interesting to know why they think the same thing is not happening on Venus and Mars, both of which have plenty of CO2 in their atmospheres. Anyone who read it want to comment?
Andreas
Gaseous Debris Disks around White Dwarfs
Authors: B.T. Gaensicke
(Submitted on 20 Jan 2011)
Abstract: This is a short and rather narrative summary of our ongoing efforts in identifying white dwarfs with gaseous debris disks, and developing an understanding of the structure and origin of these disks.
Comments: Part of Planets beyond the Main Sequence 2010 proceedings this http URL
Subjects: Earth and Planetary Astrophysics (astro-ph.EP)
Cite as: arXiv:1101.3946v1 [astro-ph.EP]
Submission history
From: Boris Gaensicke [view email]
[v1] Thu, 20 Jan 2011 16:29:45 GMT (46kb)
http://arxiv.org/abs/1101.3946