Long-Term Optical and Near-Infrared Spectroscopic Monitoring of epsilon Aurigea During the 2009-11 Eclipse
I just posted a blog about a poster on eps Aur titled, "Long-Term Optical and Near-Infrared Spectroscopic Monitoring of epsilon Aurigea During the 2009-11 Eclipse" that was presented at the most recent AAS meeting - our visual data were used in one of the figures. I asked the primary author, John Barentine, to share a bit more about the project. Here is his response:
It's 2012 and the latest primary eclipse of Epsilon Aurigae is now in the recordbooks. Thanks in part to an extensive and unprecedented pro-am collaboration, we may have learned more about this system than at any previous eclipse since the variable nature of its stars was first noticed in the early 19th century. I was just a kid at the time of the last primary eclipse, but I remember reading about eps Aur in elementary school; specifically, that "by the time of the next eclipse in about 2010, we may finally understand this mysterious star". To a six-year old in 1983, the year 2010 might as well have been a hundred years away. But here we are; that year has come and gone, and we really do understand this system in a fundamental way. No more is it the "mysterious" eps Aur I grew up reading about.
I'm now a graduate student at the University of Texas at Austin, finishing up my Ph.D. in astronomy next year, and I happened to work on eps Aur during the recent eclipse quite by accident. I was on the staff at Apache Point Observatory in New Mexico for several years before returning to grad school. After moving on, I kept up my staff contacts there, in particular with my former coworker Bill Ketzeback, now the Chief Telescope Engineer on the Astrophysical Research Consortium's 3.5-meter telescope. In early 2009, Bill and I wondered aloud in a series of email exchanges how we might bring the observatory's resources to bear in tackling the upcoming eclipse, knowing it was a once-in-three decades opportunity. The ARC 3.5m is a remarkable telescope with a fleet of modern instruments for collecting and analyzing the light of everything from the Moon to some of the most distant quasars in the Universe. Among its advantages is the ability to quickly switch between instruments, a process that can take hours at other facilities. Once an observer settles on an instrument configuration at a typical telescope, it's an all-night commitment. Not so with the 3.5m.
Objects as bright as eps Aur usually don't attract the attention of researchers using large-aperture telescopes; in fact, it would be too bright to observe at many facilities. However, Bill and I quickly realized that, given its brightness, we could obtain high signal-to-noise spectra of eps Aur and all the necessary calibration images in about half an hour with the visible-light Echelle Spectrograph. This instrument takes very high resolution spectra and is sensitive to Doppler shifts of light as small as 8 km/s, ideal for studying the subtle motions expected in the disk of dark material surrounding the unseen companion to eps Aur. We also observed the system frequently with TripleSpec, a lower-resolution instrument that takes spectra in the near-infrared. We realized we could schedule the observations routinely and have APO staff collect the data; given the quick instrument change time, we could insert eps Aur observations into the middle of many observing nights with minimal impact on other, scheduled observations. With the blessing of ARC leadership, we took spectra of eps Aur roughly once a week for the entire two-year primary eclipse. Our weekly sampling was only interrupted by infrequent spells of bad weather and the summer months when the system is at solar conjunction.
Our dataset is almost unique in the frequency of observations and the high resolution of the visible-light spectra. Now that the eclipse is over, work really begins: analyzing all that data, consisting of hundreds of individual spectra. We've just recently collected all the calibrated spectra together and begun to survey the landscape broadly, noting the most basic changes and details we see. Already evident are tantalizing suggestions of structure in the dark disk around the secondary star and significant changes in the environment of the entire system throughout the eclipse as revealed by the time-varying shape and size of hydrogen absorption and emission lines. A full treatment of the data will involve modeling the modification of light as it passes through the various system components before it reaches telescopes on Earth, from which we can deduce more information about the physical characteristics of the stars and the nature of their interaction.
An example of our spectra is shown below. In this plot, we show the region around an absorption line of neutral titanium during the last few months of the eclipse. Individual spectra are stacked one atop the other with time increasing from top to bottom; the "rest wavelength" of the line, in the absence of any Doppler shift, is indicated by the vertical line. The numbers at right are a form of counting time in units of days, and the day number for every other spectrum is shown to avoid confusion. The blue shaded area highlights the region of the line, which is seen to strengthen and shift towards shorter wavelengths, finally disappearing just after fourth contact when the eclipse "officially" ended. Our current guess is that this line represents absorption due to a clump of cold gas and dust in the outermost regions of the disk; after the disk finished transiting the visible face of the primary star, the absorption abruptly vanished. This observation gives us information about the extent of the disk and physical conditions inside it. There are thousands of individual absorption lines in our spectra of eps Aur, though not all show such dramatic changes.
In January, we presented a poster detailing our basic conclusions at the recent meeting of the American Astronomical Society in my current hometown of Austin, Texas. We wanted to "advertise" the dataset in hopes of recruiting additional collaborators who may bring expertise to our analysis we don't currently have. The scope of the data is such that it can likely be mined for information in ways we haven't yet anticipated. As part of our presentation, we tried to put eps Aur in context for poster viewers who know about stars, but aren't familiar with either this system in particular or eclipsing binaries in general. The AAVSO was indispensable in helping to provide that context, in the form of very well-sampled light curves obtained from the Organization's public website. The light curves help articulate the events of the eclipse: in the plot below, the points in the upper panel signify the dates of our observations relative to the phase of the primary eclipse seen in the light curves underneath. Here, one picture really is worth a thousand words: at a glance, the viewer gets a good sense of how well we sampled the eclipse in time during 2009-11.
What discoveries are lurking in our data? That part of the story remains to be written. However, participating in an effort that ran the gamut of professional and amateur astronomers alike, and literally spanned generations to shed light on a dark cosmic mystery has been thoroughly rewarding. All those involved in this effort can pat themselves on the back for a job truly well done.