In the coming decade transmission spectra will be the best, if not only, way to spectroscopically study extrasolar planet atmospheres because the main challenge to observe extrasolar planet atmospheres is avoided: the 6 to 10 orders of magnitude brighter parent star. Transmission spectra make use of the bright starlight whereas direct imaging techniques null out the star light and are thus technologically challenging. Any nulling space missions in the next decade will almost certainly be limited to photometry on a limited range of semi-major axes of planetary orbits.

Many short-period transiting giant planets orbiting bright stars will have been found with ground-based transit searches. Follow-ups to radial velocity planet searches of additional stars (D. Fischer, private communication) will uncover more short-period planets including possibly "super-Earths". In addition, NASA's Kepler (2007 launch date) will discover over 100 transiting giant planets around relatively bright stars at a variety of semi-major axes. Thus, there will be many transiting planets available for follow-up study of transmission spectra. The transmission spectra method is not limited to any semi-major axis. The brightest stars will be accessible for follow-up with 6-10m class ground-based telescopes (e.g., Keck, Magellan). These giant planets transmission spectra will be accessible by NASA's JWST (launch 2011) and the next generation of 20-30m ground based telescopes. In contrast, obtaining planetary spectra via direct imaging is extremely challenging since it is hampered by the brighter parent star.

It is essential to test transmission spectra models. The transit of Venus presents an exceptional opportunity to do so. Observations of Venus during transit can be used together with transmission spectra models to "interpret" the data to determine Venus' composition, abundances, effective temperature, scale height, oblateness, size, and orbital inclination with respect to the solar equatorial plane. The "results" can then be compared to what is known about Venus. We ask, if Venus were an extrasolar transiting planet, what would we learn about its atmosphere and physical characteristics? What would we miss? What would we misinterpret? More importantly, we will be able to test our transmission spectra models---the same models that we plan to use to interpret extrasolar planet transmission spectra during the next decade---to see if they are both complete and accurate.

Following is an FAQ explaining the need for professional astronomers to observe the transit of Venus.

Venus is so well known, why not just compute a transmission spectra model and not bother observing the transit? It is essential to test models on real data. We should not miss the once-in-a-lifetime opportunity to measure the transit of Venus. It is important to realize that although the Venusian atmosphere has been well measured, Venus has never been observed spectroscopically as a transiting planet. Furthermore, model transmission spectra have not been thoroughly calculated for Venus since the only situations in which transmission spectra would be used are Venus occultations of background stars---very rare events (e.g., Menzel and de Vaucouleurs AJ 1960).

Are there really any parts of the Venus atmosphere model that are complex enough to seriously warrant this observation (beyond general model testing)? Refraction of starlight through the planet atmosphere is strong during the Venus transit. This refraction effect has been seen with the naked eye during ingress and egress when a luminous ring surrounded Venus (and was used by Lomonosov in 1761 to postulate the existence of a massive atmosphere on Venus). Refraction during the transit itself may in fact complicate abundance determination from transmission spectra because refraction affects continuum wavelengths, but not the absorption lines, and hence the continuum could be lowered causing line strengths to be overestimated. In addition, the magnitude of refraction with time during ingress and egress can tell us about the oblateness of the planet. We expect transiting extrasolar planets (with the exception of the short-period planets) to have refraction effects (Hui and Seager 2002). With very high precision photometry with Kepler and potentially JWST, we may be able to use refraction effects during ingress and egress to determine the planet's oblateness and atmospheric density (Hui and Seager ApJ 2002).

We might only expect low spectral resolution capabilities for future transmission spectra studies, so why use high spectral resolution? The best strategy is to obtain high spectral resolution data and to bin it to lower spectral resolution data. Future instrumentation designs are not known---in fact we may influence them if we find high spectral resolution more useful for interpretation.

Isn't an extrasolar Venus's atmosphere too small to detect amidst the entire background sun? Yes---and not to mention we do not yet know of transiting planets the size of Venus. We do know of two transiting giant planets and more will be discovered in the near future. The transit of Venus is a useful test for these cases as well as future detections of terrestrial-size transiting planets.

In addition to the questions discussed above, the following challenges must be taken into account when attempting to extract the transmission spectra from an unknown extrasolar planet’s atmosphere. These are effects caused by the dark planetary disk as it transits the star, and are viewed in the spatially integrated starlight. These can be observed for the sun, possibly using moonshine, for any observation which does not spatially resolve the sun.

Isn't the observational procedure and data analysis straightforward enough to render the transit experiment redundant? No, because the transit of the disk of Venus induces effects in the solar spectrum which may be comparable to the atmospheric signature. The reason why the planet-induced effects are so important is that the method used to detect extrasolar planet transmission spectra (Charbonneau et al. ApJ 2002) relies on the assumption that the integrated stellar spectrum is the same in transit and out of transit, and that the integrated stellar spectrum is the same during different stages of the transit. To describe the method in more detail: the planet transmission spectrum is so much weaker than the stellar spectrum that the planet signature must be searched for in the differenced in-transit minus out-of-transit measurement (Charbonneau et al. ApJ 2002). This subtracts out the stellar spectrum. Furthermore, for extrasolar planet transits, many observations during the entire transit must be co-added to increase the signal-to-noise ratio. Two of these effects are discussed below.

The Rossiter effect. Doppler effects arise because the transiting planet ``resolves'' the surface of the star. The integrated stellar spectrum is altered during a planet transit due to stellar rotation. If the planet is orbiting in the direction of stellar rotation, the planet will block out some of the blueshifted starlight during the first part of the transit; the integrated solar spectrum will be redshifted. Conversely, during the second part of the transit when the planet is transiting the opposite side of the star, the planet will block out some of the redshifted sunlight and the integrated starlight will be blueshifted. This "spectroscopic transit" was readily observed for HD209458b (Queloz et al. A&A 2000). This spectroscopic transit is very useful because it can indicate whether the planet is moving in the same direction as the stellar rotation, and can further be used to constrain the planet’s inclination with respect to the stellar equatorial plane. The changes in individual stellar lines could affect the extraction of the transmission spectrum.

Limb darkening effects in spectral lines also arise because the transiting planet resolves the stellar surface. The integrated stellar spectrum is altered by a transiting planet due to the fact that spectral lines form at different depths in the stellar atmosphere (e.g., the line core vs. the line wings). This phenomenon is known as limb darkening. For instance, looking at the edge of the solar disk, one sees the outer, cooler layers of the sun; when looking at the center of the solar disk one is looking at hotter, deeper layers. Because the dark disk of the transiting planet probes different depths of the stellar atmosphere as it transits different parts of the projected stellar disk, each stellar line spatially integrated over the sun will change. For the transiting extrasolar giant planet HD209458b, this "spectral line limb darkening" signature was smaller by an order of magnitude than the atmospheric signature (Charbonneau et al. ApJ 2002). The terrestrial planets, however, have atmospheres that are much smaller compared to their total planet radius than HD209458b, and such complicating stellar effects may play a more important role. While such an effect can be theoretically modeled, independent verification by observation is necessary.

Back to Transit of Venus .

Images from photographic plates of the Transit of Venus (Venus crossing the face of the Sun), a very rare phenomenon that last occurred in 1874 and 1882.
Courtesy, The Naval Observatory.