A little while ago the first images of exoplanets were released and caused quite a stir in the astronomical community, as well as on the blogosphere! Here at UVa, like good little grad students, we read and discussed the discovery papers shortly after they were released. I read the papers and took notes of our discussion and just wanted to share some highlights of those to go beyond the press releases.
First, Kalas et al. (astro-ph and Science) present optical images from the Hubble of the star Fomalhaut and its planetary companion, cleverly named Fomalhaut b. From the initial press release, a few of us couldn’t help but think, are you sure that’s a planet in those two images and not some artifact of the imaging? Sure enough the paper lists all six epochs of data in which the planet is imaged. Some are in 2004, others are in 2006, and this allows us to see the motion of the planet in its orbit. The change in position is much larger than the errors, showing it to be a real movement in an orbit! Not only that, but Fomalhaut is moving with respect to us, and this proper motion across the sky can be measured. If the speck of light is really a planet orbiting the star, then its proper motion must be the same. It must be moving along with the star. These observations show that, indeed, it is!
It had previously been hypothesized that a planet must exist in that region for the debris disk around the star to have the particular shape that it has. The planet’s gravity affects the dusty disk such that it has a sharp inner edge, and the center of the disk is 15 AU away from the star! Given these characteristics of the debris disk and the actual observations of the planet’s location, models show that it is probably around 3 times the mass of Jupiter. The models quite definitively rule out very massive objects, close to brown dwarf masses. What we have here is definitely a planet.
Two points are not enough to constrain the orbital parameters of this planet, especially when the orbit’s inclination, or tilt with respect to us, is not well known. However, the planet is approximately 115 AU away from the star. Remember, 1 AU defines one “astronomical unit” or the mean distance between the Earth and the Sun (roughly some 93 million miles.) The furthest planet in our solar system, Neptune, orbits at an average of 30 AU from the Sun. The Kuiper Belt, populated with icy bodies such as Pluto, is at roughly 50 AU from the Sun. Fomalhaut b orbits at more than double that! This is reminder that planetary system formation scenarios must take in a wide range of solar systems as end products from their models. For decades, we only had a sample size of one, our dear Sol and its puny companions, on which to base our models. Now, theorists must take into account such large systems as Fomalhaut, and all those super-Jupiters that orbit incredibly close to their stars (what would be within Mercury’s orbit.) I suspect that will be a fruitful field to come!
There is a lot more that the paper covered that I won’t get into here. Various hypotheses are raised about the planetary atmosphere from the brightness of the planet in different color filters, but this is highly model-dependent and on shaky ground without an actual spectrum. What really fascinated me, as usual, was in the description of HOW the observations were made. Seeing a planet next to a star is a difficult task! Planets are teeny, tiny little guys giving off (more likely reflecting) just a small amount of light compared to the whopping bright furnaces that are their parent stars. The device they used (the the Advanced Camera for Surveys High Resolution Channel in its coronagraphic mode, for the curious) has two occulting spots that literally block the light from the parent star. Images were tried with one, then the other, occulting spot covering the star. However, an occulting spot still leaves a bright halo of light in the image. One way to remove this halo is to image another bright star in this mode, such as Vega, and subtract that halo from the one around Fomalhaut. Another method, and I really thought this was brilliant, was to rotate the instrument for different images. In each successive image, the actual astronomical objects will rotate around your frame, but the errors from the device itself and the halo will stay fixed! Thus errors can be separated from objects. Although these are certainly not new techniques, they have been applied in a way that has let us see exoplanets for the first time, which is just cool. Again, there are more great descriptions of the methods used in the supplement sections of the article, and I highly encourage reading them if you are interested in such things.
Needless to say, I am impressed at the work of these astronomers in gathering and analyzing all of these data. And, the Hubble Space Telescope, as if you weren’t already impressed with its awesomeness, has added another scientific achievement to a very long list. Since writing this took longer than I thought, I’ll have to summarize my thoughts on the other planetary system to be imaged, HR 8799, some other time!
Many thanks to the lovely Jo for presenting on this topic in a timely fashion, and the journal club attendees for the great discussion!
P. Kalas, J. R. Graham, E. Chiang, M. P. Fitzgerald, M. Clampin, E. S. Kite, K. Stapelfeldt, C. Marois, J. Krist (2008). Optical Images of an Exosolar Planet 25 Light-Years from Earth Science, 322 (5906), 1345-1348 DOI: 10.1126/science.1166609