The Planet Paparazzi: Earth through the Lens of Interplanetary Spacecraft

E ver since Voyager 1 took a snapshot of the fading Earth and Moon shortly after launch in the fall of 1977, the resourceful strategy of using interplanetary spacecraft for Earth observations has become routine. Shrinking science budgets make pinning such observations to the tails of other missions good economic sense, and pixilated, whole-disk images of our home planet are valuable from a scientific perspective because they accurately simulate the appearance of an unresolved—and as yet hypothetical—image of an extrasolar Earth-like planet.

More than a dozen spacecraft (cf. Planetary Society, 2011), including EPOXI featured in Robinson et al. (2011, this issue), have now photographed the distant Earth in a variety of phases. Voyager 1 for mainly aesthetic reasons snapped the classic 1990 portrait of Earth as a “pale blue dot” from the edge of the Solar System. More recent planet paparazzi have included Cassini at Saturn, the Mars Exploration Rover Spirit on the surface of Mars, Venus Express, and the Galileo orbiter, which had an exceptionally productive encounter with Earth en route to Jupiter in 1990.

In an experiment designed to test whether life would be evident on a planet known to be inhabited (Sagan et al., 1993), Galileo successfully identified the spectral “red edge” of chlorophyll (Seager et al., 2005) on Earth surfaces covered by plants; measured O₃, CH₄, CO₂, and H₂O in the atmosphere; and observed sunlight glinting off the oceans. Each of these discoveries points directly to a planet with water and conditions amenable to life. The key follow-up question from this mission was whether life or even water could be detected on a more distant planet observed at substantially lower resolution.

The question is taken up here by Robinson et al. in analyzing data from NASA's EPOXI spacecraft, which is on a heliocentric circular trajectory similar to Earth's. This has resulted in several close encounters with comets as well as Earth since 2007. EPOXI took∼24 hours of data in the visible spectrum (0.4–0.7 μm) in 2008 with Earth in a variety of phases. In the Robinson et al. (2011) paper, a rigorous computer model is used to simulate Earth's appearance and spectra at the time of the EPOXI observations so as to validate the model as a tool for characterizing the spectral and photometric variability of extrasolar planets.

The algorithm employed for this study is the state-of-the-art three-dimensional Earth spectral model developed by the University of Washington Virtual Planetary Laboratory (VPL) team and described by Tinetti et al. (2006). The model calculates Earth spectra spanning the near-UV to the mid-IR for arbitrary viewing geometries, phases, and spatial resolutions. This sizeable spectral window includes both scattered sunlight and thermal energy radiated by the planet. The latest version of the model includes the effects of atmospheric Rayleigh scattering and specular reflection of light from wavy, watery surfaces. A key feature of the new VPL model is its robust treatment of clouds, which includes both ice and water clouds of two different optical thicknesses. Clouds are a modeler's nightmare because they are difficult to predict and they significantly affect the disk-integrated brightness. For this study, the clouds were known from the MODIS satellite data for the EPOXI observing windows, and the agreement between the model results with clouds and both the photometric and spectral data is excellent.

However, the addition of highly parameterized clouds and other Earth-specific detail raises the complexity of the problem and underscores the uncertainty in using models tuned to Earth for broad characterization of extrasolar planets. The Robinson et al. fit to the EPOXI observations is a triumph of planetary modeling, but it is also a sobering reminder of how difficult it will be to glean and interpret light specifically from the surface of an unresolved terrestrial planet.

Abbreviation

VPL, Virtual Planetary Laboratory.