Printer Friendly

Eta carinae's year of glory.

Astronomers still don't know if this famous supergiant is one star or two --or why it hasn't already blown itself apart.

Enviably situated in one of the Milky Way's most photogenic neighborhoods, Eta Carinae is the dramatically variable star that briefly outshone the supergiant star Canopus in 1843--even though Eta, at 7,500 light-years, is two dozen times as distant. Along with a select few others, Eta Carinae is one of the most massive stars known, weighing in at perhaps 100 solar masses. It also belongs on a short list of stars that could appear as supernovae in our night skies anytime.

With its extreme characteristics, Eta Carinae holds the key to numerous cosmological and astrophysical puzzles. "We have so little understanding about how massive stars evolve," says astrophysicist Theodore R. Gull (NASA/ Goddard Space Flight Center). "Eta Carinae is our test case." With their prodigious, chemically complex winds and eventual explosions, supermassive stars like Eta Carinae also may teach us how the early universe was seeded with heavy chemical elements.

What's more, the very existence of supermassive stars challenges fundamental tenets. Eta Carinae is uncomfortably close to the so-called Eddington limit: the light it emits exerts so much pressure upon its own constituent atoms that the star is very close to blowing itself apart. Finally, theorists still don't know just how such massive stars manage to assemble themselves in the first place.


Eta Carinae has merited astronomers' attention since the dawn of astrophysics--and possibly since the dawn of written history. However, new technologies have recently enabled the star to be scrutinized as never before. Detectors reaching beyond the visible portion of the electromagnetic spectrum have been particularly important, since dust turns much of Eta Carinae's light output into infrared radiation, while X-rays are largely unaffected by it.

In the early 1990s Eta Carinae brightened at visual wavelengths by nearly a factor of 2 from marginal naked-eye visibility near 6th magnitude. At the same time, astronomical instrumentation was rapidly progressing on the ground and in orbit. Together these forces motivated a flurry of observations across the electromagnetic spectrum.

The findings were profound. At visual wavelengths, new proper-motion measurements from the Hubble Space Telescope and ground-based imagery confirmed that the Homunculus, a dusty twin-lobed nebula surrounding the star, was produced in the 1840s--presumably as a result of the event that made Eta Carinae outshine Canopus (S&T: October 1996, page 13).

Elsewhere across the electromagnetic spectrum, telescopes capturing everything from radio waves to X-rays documented rapid variability in the brightness, form, and spectral signature of Eta Carinae and/ or the material surrounding the star. By the mid-1990s, some astronomers were taking these findings as signs that Eta Carinae was readying itself for another great eruption.

In 1996 Brazilian astronomer Augusto Damineli (University of Sao Paulo) added a new, puzzling twist to the saga of Eta Carinae. That March, Damineli published evidence for periodicity in the star's spectrum: every 5 1/2 years, he claimed, certain emission lines disappeared just as the star brightened at near-infrared wavelengths. This reportedly clockwork behavior seemed at odds with the star's seemingly random light curve.


Damineli put forth two possible causes for Eta Carinae's 5 1/2-year period. In one scenario, Eta Carinae's photosphere was undergoing recurrent episodes of expansion and cooling (S&T: August 1996, page 13). In the other, Eta Carinae was two stars, not one, with an eccentric 5 1/2-year orbit that somehow modulated its various emissions.

In the now widely favored binary-star picture, gas blobs scattered a few thousand astronomical units from the binary system produce emission lines because they are energized by ultraviolet radiation from the smaller, hotter secondary star. The exception is during closest approach (periastron), when the secondary star (itself a supergiant with perhaps 30 solar masses) enters the densest, dustiest portion of the primary's wind and that material absorbs the secondary's intense ultraviolet light. Gull has found new support for an ultraviolet-bright secondary star with a 5 1/2-year orbit in the International Ultraviolet Explorer satellite's archived data.

Damineli's 5 1/2-year period also found support from observations by the Rossi X-Ray Timing Explorer satellite, or RXTE. Launched in 1995 and operating well past its design lifetime, RXTE watched Eta Carinae's X-ray output brighten throughout 1997--precisely when the spectral changes analyzed by Damineli were due to occur if his 5 1/2-year period were real. Then, just as Damineli's emission lines faded from Eta Carinae's visible-light spectrum, confirming his provocative hypothesis, the star's X-ray output dove off a cliff.


Astronomers have worked hard to account for this dramatic X-ray dropoff, which University of Minnesota astronomer Kris Davidson predicted just weeks beforehand in this magazine's January 1998 issue (page 36).

With millions of times the Sun's luminosity, a star like Eta Carinae casts off copious high-speed particles--a furious stellar wind--as would its putative partner. Michael F. Corcoran (NASA/ Goddard), Kazunori Ishibashi (Massachusetts Institute of Technology), and their colleagues have concluded that when the two stars approach one another every 512 years, their respective particulate winds collide, producing copious X-rays as well as a fresh round of dust.

As 1997 drew to a close, however, the X-ray-producing shock wave either was eclipsed (perhaps by the densest portions of the primary star's wind) or became unstable and fell apart. In the eyes of Corcoran and his colleagues, this took place when the two stars were at or very near closest approach. That would explain why the eagerly anticipated changes in the system's visible-light spectrum recurred simultaneously: when the hotter secondary star was buried within the primary's dense wind, its ultraviolet radiation couldn't get out to stimulate the normally luminous gas blobs scattered about the neighborhood.

While many loose ends remain, the binary-star model "seems to explain most of the things we see," says Gull. In addition, it opens up some exotic possibilities.

Take as one example the following suggestion by University of Amsterdam astronomer Pat W. Morris and several colleagues. Noting that the stars come within a few astronomical units of one another at periastron, the astronomers have proposed that extreme tides may have stripped the outer atmosphere from the now-smaller of the two stars and transferred it to the dominant partner during the mid-19th-century flare-up (S&T: April 2000, page 16).

Mass transfer also may have taken place during the 1997-98 periastron event, albeit on a far less dramatic scale, say several astronomers. Corcoran and his colleagues first inferred this from Eta Carinae's X-ray light curve, with its unexpectedly long-lasting "eclipse." More recently, radio astronomers Robert A. Duncan (Australia Telescope National Facility) and Stephen M. White (University of Maryland) bolstered the notion with a decade-long series of radio images from the Australia Telescope Compact Array. Published earlier this year, the images suggest that some new material enveloped at least one of the two stars shortly after periastron and bottled up the ultraviolet radiation that ultimately energizes surrounding plasma, powering the radio emission.

It's still conceivable that Eta Carinae is a solitary star. And even if it is a binary, its mid-19th-century eruption may have occurred simply because one of the two stars became unstable on its own, brightening spectacularly while shedding the several solar masses of material that now constitute the Homunculus.

In addition, the case for a binary star hasn't been helped by what amounts to a retraction of one key piece of evidence.

In January 2000, a ground-based spectroscopic study led by Damineli buttressed the binary-star model. It plotted Doppler shifts that seemed to reflect the more luminous member's acceleration "around the corner" of its orbit (S&T: April 2000, page 17). But Damineli now freely admits that the spectral signatures he documented may not reflect orbital Doppler motion after all. Along with several other specialists, he now attributes the radial-velocity variations not to Doppler shifts but to changes in spectral-line shape. Nevertheless, he holds out the possibility that those changes are caused when the secondary star's ionizing radiation carves holes in the primary star's wind --and that they indeed reflect the system's 5 1/2-year period.


Given all the uncertainty in Eta Carinae's basic nature, aficionados have commandeered telescopes on Earth and in orbit to scrutinize the star as never before. Most notably, the Hubble Space Telescope is taking detailed spectra of the star and its surroundings at carefully selected intervals and will continue doing so throughout the X-ray event.

If the suspected 5 1/2-year cycle holds, Eta Carinae's X-ray brightness will flare, then drop abruptly, in late May or early June, approximately at periastron--and approximately when this issue of Sky & Telescope reaches most readers. If the binary-star model is right, the rapid velocity changes that take place during periastron in an eccentric orbit may just be discernable in the Hubble spectra, obtained with the Space Telescope Imaging Spectrograph, or STIS. (STIS has far finer angular resolution than the ground-based spectrographs that produced Damineli's ambiguous radial-velocity curve.)

A multitude of astronomical instruments will complement Hubble's high-resolution spectrograph. RXTE will continue to monitor daily the star's X-ray brightness, which obligingly has ramped up on cue in recent months. The orbiting Chandra satellite will obtain detailed X-ray spectra of the star(s) at several instances throughout the anticipated X-ray "eclipse." Using Chandra, Corcoran is particularly keen to track Doppler shifts in spectral lines produced by the 100-million-degree shock front. "If we can do that, we can put a limit on the mass of the companion [star]," he says.

The Far Ultraviolet Spectroscopic Explorer (FUSE) spacecraft has gotten into the act, monitoring the system's ionizing radiation and revealing the temperatures, densities, and velocities of various chemical elements. Duncan continues to watch Eta Carinae's circumstellar plasma with the Australia Telescope National Facility's six-antenna array. Patricia Whitelock (South African Astronomical Observatory) is acquiring frequent near-infrared photometry to monitor what may be dust production at the interface of the two stars' winds.

Finally, Damineli and his collaborators are obtaining visible-light spectra on a

near-nightly basis with a 1.6-meter reflector in southern Brazil. He hopes to watch spectral lines from argon, iron, neon, nitrogen, and sulfur fade as they did 5 1/2 years ago, when ionizing radiation from the putative secondary star was presumably prevented from reaching the gas blobs surrounding the binary. Substantial changes are expected by the time this issue reaches readers. In fact, according to Damineli, a March 15th spectrum at slightly longer near-infrared wavelengths shows a long-expected change in the shape of the 1083-nanometer neutral-helium emission line--the line that first revealed the 5 1/2-year period in 1996.


Amateur astronomers and educators with small telescopes may be able to play a significant role in this unprecedented star-watch --provided, of course, that they live and work south of the equator. "This is going to be a real opportunity for amateur astronomers," says Corcoran, "especially if they can get spectra." Gull agrees. After all, he notes, pioneering Argentine astronomer Enrique Gaviola chronicled variability in one of Eta Carinae's helium lines in the early 1950s, using spectra recorded on photographic plates. Armed with CCDs, today's "amateurs could indeed replicate--exceed--what he did," Gull notes.

Granted, amateurs can work only with visible light and cannot come anywhere near the Hubble Space Telescope's spectral or angular resolution. Thus they almost certainly are not in a position to measure Doppler shifts induced by the stars' orbital motions, for example. However, they can provide nearly continuous time coverage and plug gaps that will no doubt occur even in intensive ground-based campaigns like Damineli's.

Well-calibrated broadband photometry of Eta Carinae remains of interest, says Davidson. After all, he says, a sudden 1999 brightening "came as a shock" to professional astronomers. South African amateurs Raymond Windsor Jones and Fanie de Villiers are monitoring Eta Carinae with photoelectric photometers under the aegis of the American Association of Variable Star Observers (AAVSO). Visual observers like David Frew in Australia and Sebastian Otero in Argentina have carefully monitored the star for years.

But spectroscopy would be particularly valuable, says Damineli, who offers to correspond with potential observers by e-mail (see the box on this page). Although he has near-nightly access to the Brazilian 1.6-meter telescope, weather will preclude some observations, and key spectral lines could brighten or fade suddenly. Furthermore, says Damineli, institutions have shut down many of the small instruments that could have monitored Eta Carinae from other sites, like La Silla, Chile. Still other opportunities were lost when Australia's Mount Stromlo Observatory burned on January 18th (April issue, page 18).

Damineli suggests that would-be spectroscopists use telescopes with apertures of at least 12 inches (30 centimeters) and CCD-based spectrometers putting no more than 0.2 nanometer (2 angstroms) on each pixel. "The fading of the [emission] lines" tabulated on the facing page "is easy to detect with an amateur spectrograph and a CCD," he says. Any would-be contributor should have experience establishing his or her instrument's wavelength scale and should be able to orient its grating(s) so the desired spectral lines fall on the CCD chip.

Brazilian amateurs Tasso Napoleao and Rogerio Marcon have geared up to monitor Eta Carinae's emission lines with a homebuilt spectrograph, an SBIG ST-7E CCD camera, and a Meade 12-inch LX200 telescope. They had just completed a successful first observing run in northern Chile as this issue of Sky & Telescope went into production.

Also of interest is the so-called Balmer jump at a near-ultraviolet wavelength of 364.7 nm, just past the blue end of the visible spectrum. A photon whose wavelength is shorter than this can ionize a hydrogen atom whose electron is in its second-lowest energy state. The extent to which Eta Carinae's spectrum diminishes at wavelengths shorter than this is a measure of how thoroughly the star's surroundings are ionized.

One potential complication, cautions Napoleao: the sensitivity of most commercially available CCD cameras declines steeply with wavelength at the ultraviolet end of the spectrum, and some CCDs may be unable to straddle the Balmer jump. If your chip can operate around 370 nm, Damineli stresses the importance of calibrating your spectrometer's sensitivity to light of various wavelengths. He suggests doing so in two steps: first by putting Eta Carinae and 8th-magnitude HD 303308 (1' to the north) on the spectrograph slit at the same time, and then by observing spectroscopic standards like 4.7-magnitude HR 4468 (Theta Crateris).

Corcoran frequently posts updated RXTE light curves on his home page (listed in the box above) and maintains close contact with Eta Carinae enthusiasts through VSNET, the Japan-based network of variable-star observers. Although he and the other architects of the Eta Carinae campaign already are working around the clock to schedule, monitor, and interpret their observations, Corcoran will, like Damineli, make every effort to correspond with potential amateur collaborators via the Internet.


The Sun aside, few stars spawn entire scientific conferences every couple of years. Few stars command the attention of billions of dollars of space hardware and dozens of ground-based research telescopes for months at a time. Few stars can underpin a scientist's entire career. And few stars promise in mere months to flood astronomers with data that will take years to analyze and interpret. But Eta Carinae is no ordinary star.

Will this year's unprecedented observing campaign establish once and for all whether Eta Carinae is one star or two? "That is my fervent hope," says Corcoran, who favors the binary hypothesis. However, there's much more at stake, from the theory of stellar evolution to the physics of matter in extreme environments.

And then there's the unforeseen. The Hubble Treasury Program on Eta Carinae is "the most ambitious spectroscopic job ever undertaken with HST," says Davidson, and he expects it to answer questions astronomers haven't yet asked. Some of the unexpected answers may be only months away. Stay tuned!

Key Web Pages and E-mail Addresses






REA (Brazilian amateur network) SPECTROSCOPE


Emission Lines to Watch

Balmer jump *               --     350-380 nm
doubly ionized neon       Ne III   386.8 nm
doubly ionized iron       Fe III   465.6 nm
singly ionized nitrogen    N II    575.5 nm
doubly ionized sulfur     S III    631.2 nm
doubly ionized argon      Ar III   713.5 nm

Source: Augusto Damineli

* A discontinuity in the continuum rather than an
emission line

Marooned nearly halfway between the equator and the North Pole, senior editor JOSHUA ROTH can only vicariously experience Eta Carinae's long-awaited antics.
COPYRIGHT 2003 All rights reserved. This copyrighted material is duplicated by arrangement with Gale and may not be redistributed in any form without written permission from Sky & Telescope Media, LLC.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Roth, Joshua
Publication:Sky & Telescope
Date:Jul 1, 2003
Previous Article:Space infrared telescope facility.
Next Article:Mysteries of Mars.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters