Science Literature

Ever since the first exoplanet was discovered, there has been an insatiable appetite for news of places with environments suitable for life. With over 200 exoplanets documented, none have yet satisfied this hunger. The next phase of research has now commenced, with two reports of planetary atmospheres.
The first is HD 209458b, at a distance of 150 light years from Earth. This gave a flat spectrum "showing nothing apart from a peak (Richardson] attributes to silicates, and possibly a molecule containing carbon-carbon bonds such as those seen in benzene." There was no trace of water, carbon dioxide or methane. The second report concerns HD 189733b, 60 light years from Earth. This also documents "a flat spectrum, with no water, methane or carbon dioxide" and no silicates either.
Richardson et al. comment: "All hot Jupiter spectra are expected to be shaped by water absorption because water is an abundant gas at the high temperatures of hot Jupiters (1,000 - 2,000 K)." Consequently, it is a real puzzle not to detect water in the spectra of these two planets. If water is present, it must be hidden by some means: "Unanticipated sources of opacity may be required to produce a temperature inversion at these altitudes, and thereby mask the effect of water opacity."
Theories of how planetary atmospheres formed will need to be reappraised. The findings create yet more problems for OOL research. In other contexts, finding water outside the Earth has been used to raise expectations of finding life, but at least that does not arise here. However, it is worth contrasting this point with some of the more sensational media reports: "Giant step in search for alien life" and "Nasa closer to discovering life on other planets". Beware of spin!

[Please note: this blog entry is written in the knowledge that some scientists think the earth's water came from comets, and recognises that their hypothesis is unaffected by the new finds. Also, the comments above do not imply that exoplanets containing water will not be found, nor that earth-size planets will not be discovered in the liquid water zone.]

A spectrum of an extrasolar planet
L. Jeremy Richardson, Drake Deming, Karen Horning, Sara Seager and Joseph Harrington
Nature 445, 892-895 (22 February 2007) | doi:10.1038/nature05636

Of the over 200 known extrasolar planets, 14 exhibit transits in front of their parent stars as seen from Earth. Spectroscopic observations of the transiting planets can probe the physical conditions of their atmospheres1, 2. One such technique3, 4 can be used to derive the planetary spectrum by subtracting the stellar spectrum measured during eclipse (planet hidden behind star) from the combined-light spectrum measured outside eclipse (star + planet). Although several attempts have been made from Earth-based observatories, no spectrum has yet been measured for any of the established extrasolar planets. Here we report a measurement of the infrared spectrum (7.5-13.2 [micro]m) of the transiting extrasolar planet HD 209458b. Our observations reveal a hot thermal continuum for the planetary spectrum, with an approximately constant ratio to the stellar flux over this wavelength range. Superposed on this continuum is a broad emission peak centred near 9.65 [micro]m that we attribute to emission by silicate clouds. We also find a narrow, unidentified emission feature at 7.78 m. Models of these 'hot Jupiter'5 planets predict a flux peak6, 7, 8, 9 near 10 m, where thermal emission from the deep atmosphere emerges relatively unimpeded by water absorption, but models dominated by water fit the observed spectrum poorly.

See also:

Sanderson, K. Direct view of a dark and distant world, Nature 445, 803 (22 February 2007)

Minkel, J.R., Water Mysteriously Absent from Extrasolar Planets' Atmospheres, Scientific American News, February 21, 2007
Contrary to predictions, two planets orbiting distant stars show no signs of water and other simple compounds; dark clouds or haze may hide them