No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Epsilon Aurigae - Polarization, light curves, and geometry of the 1982-1984 eclipse
Abstract
Polarization and light curves of the recent eclipse of ε Aur are presented and modeled, using a tilted, opaque secondary disk and a model atmosphere for the F Ia primary star. That model qualitatively explains the striking asymmetries in the observed curves. But out-of-eclipse polarimetry shows the primary to be a nonradial pulsator, and this will have to be included in a truly quantitative picture of the eclipse and the system.
... Shapley (1928) concluded that these variations arose from a ~355-day quasi-periodic variation; however, the exact period was poorly constrained by these data. After the 1983–1985 eclipse, Kemp et al. (1986) proposed that a ~100-day period may exist in polarimetry data. Later, Henson (1989) showed there was little to no wavelength dependence in the variations, implying that the source of polarization is Thompson scattering from free electrons. ...
The lure of a 50% reduction in light has brought a multitude of
observers and researchers to epsilon Aur every twenty-seven years, but
few have paid attention to the system outside of eclipse. As early as
the late 1800s, it was clear that the system undergoes some form of
quasi-periodic variation outside of totality, but few considered this
effect in their research until the mid-1950s. In this work we focus
exclusively on the out-of-eclipse (OOE) variations seen in this system.
We have digitized twenty-seven sources of historic photometry from
eighty-one different observers. Two of these sources provide
twenty-seven years of inter-eclipse UBV photometry which we have
analyzed using modern period finding techniques. We have discovered the
F-star variations are multi-periodic with at least two periods that
evolve in time at DP ≈ -1.5 day/year. These periods are detected when they
manifest as near-sinusoidal variations at 3,200-day intervals. We discuss our
work in an evolutionary context by comparing the behavior found in epsilon
Aur with bona-fide supergiant and post-AGB stars of similar spectral
type. Based upon our qualitative comparison, we find the photometric
behavior of the F-star in the epsilon Aur system is more indicative of
supergiant behavior. Therefore the star is more likely to be a
"traditional supergiant" than a post-AGB object.
We encourage continued photometric monitoring of this system to test our
predictions.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
An ongoing collaboration between the authors and an international team of professional astronomers has demonstrated the scientific potential of using modest aperture, commercially produced, semi-robotic telescopes situated under steady dark skies and affordable off-the-shelf astronomical cameras to reveal extremely dim, diffuse structures on the outskirts of distant galaxies that sheds light on galactic evolution. In this paper, we share our techniques, experiences and highlights of our investigations thus far.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
In a previous paper, the author discussed the construction of an automated dual beam imaging polarimeter and of observations made in the November 2009 to February 2010 period. Here, we discuss observations and instrumental improvements that span the period from late August 2010 through third and into fourth contacts in Spring 2011. Approximately 930 linear polarization measurements of the target star in BVR bands were obtained during 99 nights of observation. Additional measurements were made of both known polarization standards and zero polarization stars to verify instrument calibration. The polarization of Epsilon Aurigae was observed to vary by nearly 0.4% peak to valley during this period. These variations occurred in several major cycles of varying duration. Measurement error is estimated to be on the order of +/-0.05%. The observed variations resemble excess polarization seen during the 1984 eclipse egress, but may show some differences in detail. During this project, a new optical rotator was developed in conjunction with Optec, Inc., and used for the last two months of observations. This project was initiated at the suggestion of Dr. Robert Stencel at the May 2009 SAS meeting to extend measurements done during the 1984 eclipse by Dr. Jack Kemp and followed up by his student, Dr. Gary Henson thereafter.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
A new generation of point and shoot digital cameras, when combined with open source firmware enhancements can operate as astrographs. This paper explores the research and astro-photographic opportunities and capabilities offered by this pairing of mass production optics and open source functional extensions that retail for as little as $200.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
The Silicon Photomultiplier (SPM) offers sensitivity comparable to conventional photomultipliers with the added advantage of small size, low operating voltages, and robust tolerance to excess/ambient light. A Peltier cooled SPM running at -30°C was used in conjunction with wideband electronics and a 17-inch Astrograph to collect photometric data without a reference star. High speed photometric trials were conducted on eclipsing binary AW UMa demonstrating fast data rate capability. Data shows the SPM exhibits excellent sensitivity, acceptable signal to noise, and bandwidth with sampling times as short as 1 millisecond for brighter targets. Automated digital data acquisition is discussed along with digital signal processing techniques for noise reduction, spectral analysis, and data mining. The SPM demonstrated acceptable signal to noise for fast photometric studies for 8-10th magnitude targets depending on scintillation and background conditions. Future SPM study topics are also discussed.
... The non-agreement implies that either the F star diameter is different or that the eclipser is not a simple spherical object. While we cannot exclude the possibility of an extended atmosphere of the F star as described by Cha et al. (1994) and Kemp et al. (1986), the interferometric image of the disk makes it clear that more careful study of the light curve production with non-spherical shapes is needed. Table 9 shows the original pre-eclipse predicted first contact times with the difference between the average and predicted dates. ...
The mysterious star system epsilon Aurigae undergoes an eclipse every 27.1 years that lasts nearly two years. The most recent eclipse started during the late summer of 2009. An international campaign for observing this eclipse was created in 2006, with a web site for information and, to-date, 17 periodic newsletters for details, as well as a Yahoo forum List for immediate announcements and comments. Photometric data in the UBVRIJH bands have been submitted. Ingress occurred with first contact in the V band estimated at the second week of 2009 August and second contact estimated at 2010 mid-January. Spectroscopic data were also obtained during ingress. Spectroscopic data have been provided in the potassium I region, hydrogen alpha and beta regions and sodium D line region of the star system's spectrum. In this paper we describe details of observations and preliminary analysis during ingress and second contact. We introduce the observers and discuss plans for observing throughout totality and the end of the eclipse in 2011.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
Tycho had rudimentary instrumentation (by modern standards), but he also had a first-rate mind (by any standard). With them, he was able to demonstrate that a comet and a nova were more distant than the Moon. I have investigated the feasibility of replicating similar measurements using very common modern equipment (a DSLR, a spreadsheet, and a star catalog). It appears to be feasible for a science student to replicate - and extend - the great scientist's observations and analysis. I'll describe how the distance to the Moon and the distance to an asteroid can be determined from a single fixed site, through the use of diurnal parallax.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
In this paper the authors discuss the flat field systems utilized in the AAVSO Photometric All Sky Survey (APASS) to include the various stages of construction and details of the testing from each system studied and used. The need for well-calibrated images is obvious when conducting a photometric survey intended for use as a catalog of secondary standards covering the entire sky. Several systems for flat field calibration were investigated.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
Pictorial astronomical imaging, derisively known as "pretty picture" imaging, has become the wild west of astronomy where objects are often portrayed with garish, oversaturated and unreal colors. While false color representation of celestial objects may have its place in science, and can display familiar objects in new and exciting ways, these images are misleading to the general public as a true representation of the universe. A new approach to color rendering is described that attempts to assign colors, calibrated to the visible spectrum, to RGB and narrowband images for a more realistic representation of celestial objects.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
This paper examines a series of measurements made to determine the proper motion and parallax of Barnard's Star using differential astrometry. In differential astrometry, the same reference stars are used for every coordinate determination. Although the absolute coordinate solution is offset by a constant due to the imprecisely known positions of the reference stars, relative coordinate solutions can be precise. For this study, observations of the field of Barnard's Star were made in 2009 and 2010 using an 8-inch f/4 Newtonian telescope. The singlenight standard error in each coordinate can be as low as 10 mas. We discuss the errors, error distribution, and error sources in the astrometric solution, as well as the derived proper and parallax motions of the star.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
Be spectral classification stars typically show variable and complex emission detail in the H alpha spectral line. These emission phenomena are not from the star itself but are thought to originate from a circumstellar disk of material that has formed around the star due to a combination of mass loss, rapid rotation, a magnetically torqued disk of the surrounding material and nonradial stellar pulsation. To analyze these H Alpha line characteristics a high resolution spectrograph is required to display the subtle changes over time. The construction of an inexpensive high resolution Littrow spectrograph for that analysis process is herein described.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
When things go bump in the night (sky), it's opportunity knocking. This paper discusses observing programs particularly suitable to amateurs, and geared to provide rapid response to unusual celestial events. With more and more large-scale surveys coming online, the opportunities for follow-up observations are exploding. Since such observations can be of great value to professionals, amateurs need to know what is required and how report their observations.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
The release of light curves from the NASA Kepler mission, allows an examination of previous V band observations made by the author of the star field surrounding EV Lyr. Five field stars were indentified in the Kepler observational field that shows sufficient variability to be detected and measured in the EV Lyr observations from 2006. Additional light curves for these stars were created from the author's observations, and in one case, the Northern Sky Variability survey. These additional light curves were compared against the Kepler light curves. Where possible, a period analysis was performed.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
The objective for this study will be to explore various time series methods using CCD photometry for use with amateur telescope observations of the dense globular cluster M15. Amateur telescopes are defined here as having an aperture of less than 40 cm, and are collecting photometric filtered observations over time. Specifically, we attempt to determine the light curve of the core of M15 as a "single star". This requires selection of comparison and check stars to perform differential photometry; i.e. subtraction of flux density measures between a nonvariable (reference star) and the variable "single star" of the M15 core as it changes in magnitudes over time. We explore the possibility of measuring the M15 periodicity as an aggregate of many standard stars as identified in the Stetson catalog for NGC7078 (2010). In this paper we'd like to propose methods and techniques for aggregating different cluster region's flux densities (luminosity) and periods.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
Although many astronomers have long been sensitive to issues of light pollution and deteriorating sky quality it is only in recent years that such interest has extended to other groups including, among others, ecologists, health professionals, and urban planners. Issues of light pollution and loss of dark skies are starting to appear in the scientific literature in the context of health and behavior impacts on both human and animal life. Nonetheless, a common deficiency in most such studies is the absence of historical or baseline data against which to compare sky brightness trends and temporal changes. To address this deficiency we have begun to collect a variety of types of quantitative sky brightness data for insertion in an international sky brightness archive that can be accessed for research projects which are dependent upon an understanding of the nature of local light pollution issues. To aid this process we have developed a mobile sky brightness meter which automatically logs sky brightness and observation location. The device can be stationary for long periods of time or can be easily transported for continuous sky brightness measurement from ground vehicles, boats, or aircraft. The sampling rate is typically about 0.25Hz. We present here examples of different modes of sky brightness measurement, various means of displaying and analyzing such data, ways to interpret natural astronomical phenomena apparent in the data, and suggest a number of complementary scientific projects that may capture the interest of both professional and amateur scientists. Finally, we discuss the status of the archive and ways that potential contributors may submit their observations for publication in the archive.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
We describe and demonstrate an alternative approach for finding and applying the transformations required to convert instrumental magnitudes to magnitudes on a photometric standard system including correction for atmospheric extinction. The method is aimed at those using differential photometry with small-field CCD cameras and working in climates where photometric conditions occur only rarely. Both primary and secondary standard stars are used to enable observers at geographical latitudes far from the equator to measure the required parameters over a wide air mass range.
... If real, such modulations would each be of about 40 days duration. These results resemble those reported by Kemp et al. (1986), who also found a polarization excess that was a maximum (2.6%) close to third contact. ...
This article illustrates the use of an internal aperture stop in a high speed, wide field Cassegrain optical system. The astrograph features a well corrected and uniformly illuminated focal plane. Opto-mechanical design and manufacturing issues are also discussed.
The results of spectral observations of eclipsing binary system Epsilon Aurigae are presented. Observations were carried out using the coude echelle spectrograph on 2-m telescope Zeiss-2000 at the Terskol Peak Observatory with spectral resolution R = 45000. Epsilon Aurigae is long periodic eclipsing binary system that consists of F0 supergiant star as first companion and huge gas dust disk as second companion. The uniqueness of this system is that eclipses of star by a gas-dust disk occur with 27.1 years period and continues two years. Presented spectral observations Epsilon Aurigaе were carried out in period May 2008 – May 2009 outside eclipse, December 2009 at the beginning of eclipse and March 2010 within eclipse. Based on the results of observations, an analysis of the variability for Hα line was made and some other spectral lines at the pre-eclipse and eclipse stages. Long-term variations of Hα line profiles were detected
The eclipsing binary ϵ Aur is unique in being a very-long-period binary involving an evolved, variable F star and a suspected B main-sequence star enshrouded in an opaque circumstellar disk. The geometrical arrangement is that the disk is viewed almost perfectly edge on, with the alignment leading to a partial eclipse of the F star. Despite a global observing campaign for the 2009–11 eclipse, there remain outstanding questions about the nature of the binary, its components, the disk, and the evolutionary state of the system. We analyze optical-band polarimetry in conjunction with broadband color variations to interpret brightness variations across the surface of the F star. We model this both during and after the 1982–84 eclipse for which an extensive and dense data set exists. We develop a model in terms of surface temperature variations characterized by a small global variation overlaid with a temperature variation described with low-order spherical harmonics. While not providing a detailed fit to the data set, our modeling captures the overall characterization of the color and polarimetric variability. In particular, we are able to recover the gross behavior of the polarimetric excursion in the Q – U plane as observed during eclipse of the F star when compared to posteclipse behavior.
Worldwide interest in the recent eclipse of epsilon Aurigae resulted in the generation of several extensive data sets, including high-resolution spectroscopic data sets. This lead to the discovery, among other things, of the existence of a mass transfer stream, seen notably during third contact.We explored spectroscopic facets of the mass transfer stream during third contact, using high-resolution spectra obtained with the ARCES and TripleSpec instruments at Apache Point Observatory. One hundred and sixteen epochs of data were obtained between 2009 and 2012, and equivalent widths and line velocities of high- and low-eccentricity accretion disc lines were measured. These data sets also enabled greater detail to be measured of the mideclipse enhancement of the He I 10 830 Å line, and the discovery of the P Cygni shape of the Pa β line at the third contact. We found evidence of higher speed material, associated with the mass transfer stream, persisting between third and fourth eclipse contacts. We visualized the disc and stream interaction using SHAPE software, and used CLOUDY software to estimate that the source of the enhanced He I 10 830 Å absorption arises from a region with nH = 10¹¹ cm⁻³ and temperature of 20 000K, consistent with a mid-B-type central star. Van Rensbergen binary star evolutionary models are somewhat consistent with the current binary parameters for their case of a 9 plus 8 solar mass initial binary, evolving into a 2.3 and 14.11 solar mass end product after 35 Myr. With these results, it is possible to make predictions which suggest that continued monitoring prior to the next eclipse (2036) will help resolve standing questions about the mass and age of this binary.
Disks in binary systems can cause exotic eclipsing events. MWC 882 (BD-22 4376, EPIC 225300403) is such a disk-eclipsing system identified from observations during Campaign 11 of the K2 mission. We propose that MWC 882 is a post-Algol system with a B7 donor star of mass $0.542\pm0.053\,M_\odot$ in a 72 day period orbit around an A0 accreting star of mass $3.24\pm0.29\,M_\odot$. The $59.9\pm6.2\,R_\odot$ disk around the accreting star occults the donor star once every orbit, inducing 19 day long, 7% deep eclipses identified by K2, and subsequently found in pre-discovery ASAS and ASAS-SN observations. We coordinated a campaign of photometric and spectroscopic observations for MWC 882 to measure the dynamical masses of the components and to monitor the system during eclipse. We found the photometric eclipse to be gray to $\approx 1$%. We found the primary star exhibits spectroscopic signatures of active accretion, and observed gas absorption features from the disk during eclipse. We suggest MWC 882 initially consisted of a $\approx 3.6\,M_\odot$ donor star transferring mass via Roche lobe overflow to a $\approx 2.1\,M_\odot$ accretor in a $\approx 7$ day initial orbit. Through angular momentum conservation, the donor star is pushed outward during mass transfer to its current orbit of 72 days. The observed state of the system corresponds with the donor star having left the Red Giant Branch ~0.3 Myr ago, terminating active mass transfer. The present disk is expected to be short-lived ($10^2$ years) without an active feeding mechanism, presenting a challenge to this model.
The bright star ε Aurigae, an eclipsing binary with the longest known orbital period and eclipse, has been slow to reveal it secrets. The most recent eclipse, 2009-2011, was monitored with an array of new and old astronomical techniques, leading to new evidence concerning the nature of the eclipsing body and its contents. While this has led to new insights about the state of the F superigant star and to more clues about the evolutionary state of the system and components, a high-mass versus low-mass model dilemma persists. This chapter examines some of the recent developments and attempts to place the system among the pantheon of binary types. It also offers suggestions for observational tests prior to the next eclipse in 2037.
We report on a total of 106 nights of optical interferometric observations of
the $\epsilon$ Aurigae system taken during the last 14 years by four beam
combiners at three different interferometric facilities. This long sequence of
data provides an ideal assessment of the system prior to, during, and after the
recent 2009-2011 eclipse. We have reconstructed model-independent images from
the 10 in-eclipse epochs which show that a disk-like object is indeed
responsible for the eclipse. Using new 3D, time-dependent modeling software, we
derive the properties of the F-star (diameter, limb darkening), determine
previously unknown orbital elements ($\Omega$, $i$), and access the global
structures of the optically thick portion of the eclipsing disk using both
geometric models and approximations of astrophysically relevant density
distributions. These models may be useful in future hydrodynamical modeling of
the system. Lastly, we address several outstanding research questions including
mid-eclipse brightening, possible shrinking of the F-type primary, and any
warps or sub-features within the disk.
Well-calibrated spectropolarimetry studies at resolutions of $R>$10,000 with
signal-to-noise ratios (SNRs) better than 0.01\% across individual line
profiles, are becoming common with larger aperture telescopes.
Spectropolarimetric studies require high SNR observations and are often limited
by instrument systematic errors. As an example, fiber-fed spectropolarimeters
combined with advanced line-combination algorithms can reach statistical error
limits of 0.001\% in measurements of spectral line profiles referenced to the
continuum. Calibration of such observations is often required both for
cross-talk and for continuum polarization. This is not straightforward since
telescope cross-talk errors are rarely less than $\sim$1\%. In solar
instruments like the Daniel K. Inouye Solar Telescope (DKIST), much more
stringent calibration is required and the telescope optical design contains
substantial intrinsic polarization artifacts. This paper describes some
generally useful techniques we have applied to the HiVIS spectropolarimeter at
the 3.7m AEOS telescope on Haleakala. HiVIS now yields accurate polarized
spectral line profiles that are shot-noise limited to 0.01\% SNR levels at our
full spectral resolution of 10,000 at spectral sampling of $\sim$100,000. We
show line profiles with absolute spectropolarimetric calibration for cross-talk
and continuum polarization in a system with polarization cross-talk levels of
essentially 100\%. In these data the continuum polarization can be recovered to
one percent accuracy because of synchronized charge-shuffling model now working
with our CCD detector. These techniques can be applied to other
spectropolarimeters on other telescopes for both night and day-time
applications such as DKIST, TMT and ELT which have folded non-axially symmetric
foci.
Results of the 2010 eclipse campaign are described, and preliminary
interpretations proposed. These include photometric, interferometric,
spectroscopic, astrometric and polarimetric observational results. Next
steps, along with continued monitoring, include simulations and other
future work. Numerous acknowledgements are appropriate for the many
participants in making this international effort a success.
Epsilon Aurigae is an eclipsing binary system consisting of an F0 supergiant
(200 R⊙) and a cool, opaque edge-on disk (2000 R⊙) possibly
encompassing a second star. V-band polarimetry data for Epsilon Aurigae's
2009-2011 eclipse is being obtained to better understand the system's
geometry and the nature of its two components. Several polarized and
non-polarized standard stars are also being observed in order to characterize
the instrumental polarization and sources of measurement error in the
polarimeter configuration and the measurement procedures for observations.
We report here on the status of these observations at the midpoint of the
eclipse and include a preliminary analysis of noise and instrumental effects
in the measurements.
During the 2010-2012 eclipse of epsilon Aurigae, the
author obtained linear polarization measurements during 200 nights of
observation over three observing seasons. These observations began
before second contact and have extended some six months into the
post-eclipse period. Measurements were made in V, B, and R photometric
bands. The polarization of epsilon Aurigae was observed to vary by
nearly 0.6% peak to valley during this period in cycles of varying
duration. These variations resemble, at a qualitative level, those seen
by Kemp and Henson during the 1984 eclipse egress. In particular they
show evidence of local polarization activity extending well past 4th
contact.
The recent eclipse of the enigmatic binary star system, epsilon Aurigae,
offered a special opportunity to explore the role of spectropolarimetry
in discovery of unknown facets of the objects involved. Here we present
spectropolarimetric results for H-alpha, H-beta, Ca I (422.6 nm), and K
I (769.9 nm) based on more than 50 epochs of high dispersion spectra
obtained with the ESPaDOnS instrument at CFHT during
2006-2012.
We present and analyze epsilon Aurigae data concerning the evolution of
the Halpha line on the occasion of the 2009 International observation
campaign launched to cover the eclipse of this object. We visually
inspect the dynamical spectrum constructed from the data and analyze the
evolution with time of the EW (Equivalent Width) and of the radial
velocity. The spectroscopic data reveal many details which confirm the
complexity of the epsilon Aur system. The object is far from being
understood. In particular, according to our measurements, the eclipse
duration has been underestimated. A complete analysis of details
revealed by our data would require much time and effort. Observers are
encouraged to continue monitoring the Ha line out of eclipse in the hope
that it will provide further important information.
Evidence is provided from the array of observations amassed during the
2009-2011 eclipse, that defines the enigmatic binary
epsilon Aurigae as comprised of an unstable F0-1 Iab star in orbit
around a comparable mass upper main sequence star (or stars) enshrouded
in a disk resulting from F star mass loss. In this picture, the F star
may be undergoing rapid evolutionary changes, and the recent 67-day
primary quasi-period may make it suitable for asteroseismic studies. The
hidden star(s) may have gained mass from the F star, and the disk itself
provides opportunities for study of accretion, dust evolution, and
dynamics.
A dual-beam imaging polarimeter has been built for use on the 14 inch Schmidt-Cassegrain telescope at the ETSU Harry D. Powell Observatory. The polarimeter includes a rotating half-wave plate and a Wollaston prism to separate light into two orthogonal linearly polarized rays. A TEC cooled CCD camera is used to detect the modulated polarized light. We present here measurements of the polarization of polarimetric standard stars. By measuring unpolarized and polarized standard stars we are able to establish the instrumental polarization and the efficiency of the instrument. The polarimeter will initially be used as a dedicated instrument in an ongoing project to monitor the eclipsing binary star, Epsilon Aurigae. This project was funded by a partnership between the National Science Foundation (NSF AST-0552798), Research Experience for Undergraduates (REU), and the Department of Defense (DoD) ASSURE (Awards to Stimulate and Support Undergraduate Research Experiences) programs.
ε Aur is an eclipsing binary with the longest known orbital period (27.1 years). It is a very unusual system with an interesting and complex behavior and its structure is still not fully established. The next primary eclipse, which begins in August this year, gives us a great chance to reveal many secrets of this system. The first part of this paper shortly resumes known facts about ε Aur and introduces its current model. The second part then introduces my running spectroscopic analysis of this system and presents the first results.
The eclipsing binary Epsilon Aurigae consists of an F0 supergiant and a
cool, mysterious eclipsing companion with an orbital period of 27.1 yr.
The light curve of this system reveals two sources of variability: the
eclipses themselves and the variation of the supergiant. Photoelectric
observations were made with the 38 cm reflector at the Villanova
University Observatory. The bright star undergoes semiregular light
variations both inside and outside the eclipse, with a characteristic
time scale of a few months which are found to correlate extremely well
with changes in color index. It appears that these light and color
variations arise from pulsations of the supergiant. The light variations
are similar to those found for other luminous A-F supergiants. A
computer code has been developed to model the eclipse and explore
possible configurations of the disk. The properties of the disk appear
more consistent with an interpretation as a protoplanetary system than a
remnant of mass transfer from the supergiant.
The geometry of the eclipsing component of Epsilon Aurigae was
investigated using the synthetic light curve method. It is found that
the eclipsing body represents a system of rings, forming a thin disk
with a central aperture. The system appears as a composite ring, flat,
extended, and consisting of concentric bands with different
transparencies. Features such as large central aperture, semitransparent
outer edge of the disk, and transparent annular zone splitting the main
opaque ring into two concentric bands are recognized. The inclination of
the disk plane with respect to the sky is found to be equal to 82 deg.
An analysis of the secular variability of the ring system shows that
minor features such as the relative extension of concentric bands and
the sharpness of the external edge of the disk may be variable and that
the rings may consist of dust, formed by accretion of particles onto
stable orbits.
Twisted thin accretion discs are of interest in explaining long-term
periodicities in X-ray binaries, the eclipse in Epsilon Aurigae, and
perhaps precessing radio jets and possible warped molecular outflows in
star-forming regions. Earlier results used isothermal discs to determine
the alignment radius, Ra. Now the author sees how polytropic
discs affect earlier results for discs around compact objects and around
close binaries. He finds that Ra for polytropic discs can be
up to an order of magnitude larger, depending on the polytropic index,
the viscosity parameter and the precession mechanism. There is little
change in the earlier conclusion that the case for alignment in Her X-1
is marginal, while there is substantial alignment in ɛ Aur. A
lower limit is put on α for isothermal discs, which is expected to
hold for polytropic discs as well. A bound is put on the energy of the
perturbed flows.
Twisting and alignment in a steady-state circumbinary accretion disc is
studied. It is then used to account for observed features in the
scenario of Epsilon Aurigae as a triple. The alignment depends on
viscosity in the disc, but it is always substantial and leads to a
tilted slab-like profile when viewed edge-on.
H alpha spectroscopic observations have been made by using a
high-dispersion and high-resolution thick CCD system, when the orbital
phase of epsilon Aur was near 0.3. During two years of out-of-eclipse
observations, the H alpha profile varied significantly. The
circumstellar gas circling the F0 superginat is possibly a ring-like H
II structure, which consists of moving gaseous clouds. H alpha profile
variation outside eclipse may be due to the apparent motion of gaseous
clouds within the ring following the rotation of the primary, and to its
intrinsic physical changes.
Among the longest known eclipse durations and binary periods is that of the star Aurigae, which exhibits 2-year long eclipses every 27.1 years. Oddly, the nature of the secondary in the system continues to elude ready identification. In 1965, Huang proposed a massive disk as the eclipsing body, and study of the 1984 eclipse led Lissauer and Backman to suggest an embedded B star binary in the disk to maintain it. A collaboration of observers allows me to present recent optical photometry and spectroscopy, near-IR spectroscopy and Spitzer space telescope IRS and MIPS observations of Aurigae as it approaches its next eclipse. These data argue for current detectability of the embedded binary, and precession of the disk axis, suggesting a radical change is possible for the next mid-eclipse brightening. An international monitoring campaign for the 2009-2011 is being organized, and participation invited via website http://www.du.edu/~rstencel/epsaur.htm.
Epsilon Aurigae is a complicated binary star that undergoes optical eclipses every 27 years, including the present year. An update is given here on the array of photometric and spectroscopic observations underway, thanks to the eclipse observing campaign and its dedicated participants. In addition, breakthrough results have emerged from (1) infrared and ultraviolet spectral energy distribution observations, and (2) especially with interferometric imaging that revealed the long suspected dark disk in transit, plus (3) new optical spectra that are revealing substructure inside the disk itself. Implications of many of these observations is discussed, but as the eclipse data are still being collected, I anticipate that additional discoveries are still to come, throughout 2010, and beyond. Comment: 8 pages, 3 figures
Eclipses of the single-line spectroscopic binary star, epsilon Aurigae, provide an opportunity to study the poorly-defined companion. We used the MIRC beam combiner on the CHARA array to create interferometric images during eclipse ingress. Our results demonstrate that the eclipsing body is a dark disk that is opaque and tilted, and therefore exclude alternative models for the system. These data constrain the geometry and masses of the components, providing evidence that the F-star is not a massive supergiant star. Comment: As submitted to Nature. Published in Nature April 8, 2010.
We report new and archival K-band interferometric uniform disk diameters obtained with the Palomar Testbed Interferometer for the eclipsing binary star ∈ Aurigae, in advance of the start of its eclipse in 2009. The observations were intended to test whether low-amplitude variations in the system are connected with the F supergiant star (primary), or with the intersystem material connecting the star with the enormous dark disk (secondary) inferred to cause the eclipses. Cepheid-like radial pulsations of the F star are not detected, nor do we find evidence for proposed 6% per decade shrinkage of the F star. The measured 2.27 ± 0.11 mas K-band diameter is consistent with a 300 solar radius F supergiant star at the Hipparcos distance of 625 pc. These results provide an improved context for observations during the 2009-2011 eclipse. © 2008. The American Astronomical Society. All rights reserved. Printed in U.S.A.
Epsilon Aurigae (epsilon Aur) is a visually bright, eclipsing binary star system with a period of 27.1 years. The cause of each 18-month-long eclipse has been a subject of controversy for nearly 190 years because the companion has hitherto been undetectable. The orbital elements imply that the opaque object has roughly the same mass as the visible component, which for much of the last century was thought to be an F-type supergiant star with a mass of approximately 15M[symbol:see text] (M[symbol:see text], mass of the Sun). The high mass-to-luminosity ratio of the hidden object was originally explained by supposing it to be a hyperextended infrared star or, later, a black hole with an accretion disk, although the preferred interpretation was as a disk of opaque material at a temperature of approximately 500 K, tilted to the line of sight and with a central opening. Recent work implies that the system consists of a low-mass (2.2M[symbol:see text]-3.3M[symbol:see text]) visible F-type star, with a disk at 550 K that enshrouds a single B5V-type star. Here we report interferometric images that show the eclipsing body moving in front of the F star. The body is an opaque disk and appears tilted as predicted. Adopting a mass of 5.9M[symbol:see text] for the B star, we derive a mass of approximately (3.6 +/- 0.7)M[symbol:see text] for the F star. The disk mass is dynamically negligible; we estimate it to contain approximately 0.07M[symbol:see text] (M[symbol:see text], mass of the Earth) if it consists purely of dust.
Photometric data of the once-every-27 yr eclipse of the Epsilon Auriga F supergiant are reported. The data were taken on 28 nights from July 1982-May 1984 and covered the 1.25-20 microns region. The secondary object was identified at 5, 10, and 20 microns when the eclipse removed 48 percent of the F star's light. The data indicated that the secondary is a compact object surrounded by a large optically thick cloud, with a luminosity under 0.01 of the primary. The secondary's mass is, however, more than 14 solar masses.
A workshop proceedings concerned with the new data collected during the 1982-1984 eclipse period of the 27-year system Epsilon Aurigae is presented. This binary star has been a classic problem in astrophysics because the opaque eclipsing object is nonstellar, and probably disk shaped. Invited papers concerning the history of the system, optical, infrared and ultraviolet photometry, optical polarimetry and ultraviolet spectroscopy are included. An invited paper concerning comprehensive theoretical interpretation in the context of stellar evolution also is included.
Adopting the photometric period and a circular orbit for Epsilon
Aurigae, the Sproul astrometric data over the interval 1939-1977 yield a
relative parallax of -0.0010 arcmin + or - 0.0010 (p.e.), an orbit with
a scale (a1) of 0.0227 + or - 0.0010 arcsec, an inclination
of 89 deg, and a mode of 92 deg; these results satisfy the epoch of
mideclipse within one year. Equating the astrometric angular and the
spectroscopic linear value of a1, a parallax of 0.00172 + or
- 0.00008 arcsec (distance 580 + or - 30 pc) is found, which yields an
absolute visual magnitude of -5.9 for the visible component, not
allowing for galactic absorption.
The first detection of the stellar limb polarization predicted by
Chandrasekhar in 1946 in an eclipsing binary system is reported.
Observations of linear polarization of the bright star Algol were made
in unifiltered light over a passband of roughly 1000-5500 A,
concentrated around the times of primary eclipse from December 1980 to
February 1983. An eclipse polarization was isolated that has a full
amplitude of 0.01 percent, and a phase-dependent pattern in agreement
with that expected for the Chandrasekhar mechanism. Results also support
the contention that the binary AB and third-star AB-C orbits in Algol
are nearly perpendicular to each other.
From slightly before the 1982-1984 eclipse of Epsilon Aurigae to the present observers from around the world have been making photoelectric photometry observations of this star system. Over 2000 UBV observations have been reported as well as observations in the R, I, J, H, K, L, M, N, and Q bandpasses plus the y, b, v, and u bandpasses. Twenty nine observers from 9 countries submitted photometry data to the campaign. The data have shown many interesting features of the star system including a Cepheid-like pulsation, flare activity, mid-eclipse brightening, post egress brightening, plus other strange activity.
The major characteristics in the ultraviolet of the 1982-84 eclipse of Eps Aur as observed with IUE by various workers are summarized. This star can be observed over the entire IUE wavelength range, from 1200 to 3200 A, in low dispersion, allowing eclipse light curves to be obtained in broadband regions, but due to its steep spectral gradient and the sensitivities of IUE cameras, high resolution exposures adequately cover only the regions from 1700-1900 and 2400-3200 A. In many ways, the UV data confirms or expands upon interpretations of the stem made from observations in other wavelength regions, but in other respects the system remains as enigmatic as before.
Basic observational data of Epsilon Aurigae are summarized and used as the basis of a discussion of possible evolutionary states of the system. Constraints posed by the presence of a cold disk surrounding the secondary star are also outlined. Possible evolutionary models of the F0 Ia supergiant range from pre-main sequence contraction through shell hydrogen burning, core helium burning, to shell helium burning, depending on the absolute luminosity of the system, for models in which no mass transfer has taken place. Models invoking binary interaction include core and shell helium burning stars, and pre-white dwarfs, again depending on the absolute luminosity of the system. A massive shell helium burning star or a pre-white dwarf mass transfer remnant would appear the most likely of these models at present. Observational tests of these models are briefly outlined.
Intermediate band blue (4530A), far red (7790A) and H-alpha intermediate and narrow band photoelectric observations of the peculiar, 27 year eclipsing binary, Epsilon Aurigae were made from December 1981 through the present (December 1984). BD +42 1170 served as the primary comparison star because of its angular proximity to the variable star. The analysis of this data along with other available photometry was undertaken to study the characteristics of the low amplitude, semi-regular light variations that appear inside and outside of eclipse. It appears that these short term light variations arise from nonradial pulsations of the luminous f supergiant in the system. Furthermore, the semi-regular light variations found for Epsilon Aurigae are similar to those found for other luminous A-F supergiants. Also, the preliminary results from the analyses of the light variations produced by the eclipse of the F-supergiant by the mysterious cooler component is discussed.