Eta Carinae Spectroscopic Observations
INTRODUCTION :
The "regular" Eta Carinae spectra show many strong high excitation emission lines such as those of [NII], [NeIII], and several HeI emission lines, plus the Balmer series of hydrogen in emission. This state is generally known as the "high state" spectra. Every 2020 days (5.53 years), however, when the star approaches the cyclical "spectroscopic event" discovered by the Brazilian astronomer Augusto Damineli in the mid-1990s, the high excitation lines fade and even disappear. The spectra taken during the event (usually called "low state" spectra) show low excitation lines such as the FeII and hydrogen emission lines.
It seems reasonable to believe that the high excitation lines are not generated in the photosphere or chromosphere of the central star (or stars), but instead they come from a nebular region usually called "The Homunculus", which surrounds Eta Car and which was generated by the great eruptions of the nineteenth century.
Theories explaining the 5.53-year cycle range since the binary model developed by Prof. Damineli by late 1990s to pulsating models based on a hypermassive single star with associated instabilities, as proposed by other professional researchers. An overall consensus on the "best" theory for Eta Car seems still to be reached, although the binary model has been increasingly accepted over the last few years.
Damineli's model describes Eta Car as a binary system with components embedded in the Homunculus and separated by a distance close to the Jupiter orbital radius. Both stars follow excentrical orbits and the periastron takes place every 5.53 years. Both stars produce strong stellar winds, and the tremendous collision between them - particularly when close to periastron - heats up the gases to about 63 million K, generating strong X-ray emission. When the secondary star approaches the periastron, it ploughs into the primary's stellar wind, generating the blanketing of the high-excitation lines.
One way to clarify the discrepancies between the theorical models is to identify by spectrometric techniques if the event follow precisely the 2020-day cycle, or, instead, if the line blanketing takes place with significant time deviations as compared to the binary model forecasts. In other words, if the next events take place precisely on schedule, it might be more difficult to sustain the hypermassive single star model.
The 1997 event appears to have come right on schedule with the binary theory. For the 2003 event, a complex international monitoring network has been established among several professional observatories, including the HST, the Brazilian LNA, and various telescopes in Chile, Australia and Argentina, besides X-ray satellites such as Chandra and RXTE.
OUR WORK :
By March 2003, we decided to follow the Eta Car spectroscopic event with our
homemade spectrograph described in this site and concentrating in two different
spectral regions to follow the development of some lines: (1) The HeI line at
6678 Angstroms and (2) the lines of [NII] at 5755 A, and the HeI 5874 A in its
neighborhood. Obviously, we did not expect that our observations could add
anything to what was being already done by professionals with high resolution
instruments at the large observatories. Our aim, instead, was to demonstrate
that the event could be detected in small amateur instrument - and monitoring,
if possible, the transition from high state to low state spectra.
Within our instrumental limitations, our results have indicated that the spectroscopic event could perfectly be detected by our detectors and that it has come as scheduled: the lines we followed have shown a sharp decrease in intensity, particularly by the second half of June 2003. A summary of our results may be seen in Figs 1 through 4.
Figs 1 and 2 show the development of the HeI 6678 line in three nights: April 03, 2003 (high state), June 11 and June 26 (low state).
In all measurements, we used a 12-in Meade LX-200 scope plus our spectrograph and a ST-7E CCD camera. All spectra were taken with 30-min integration time (sum of three exposures of 10 min each). Processing was done with the excellent freewares IRIS (by Christian Buil) and VisualSpec (by Valérie Desnoux), both available for download at the Web.
Fig. 1
Figure 1 has aesthetic purposes only. The processed spectra is shown for the three nights described above. The strongly saturated emission line on the far left is hydrogen alpha at 6563 A. Even in this kind of representation, the weakening of the HeI 6678 line (midway from left to center) with time can be perceived.
On the other hand, Fig 2 show the real calibrated spectral profiles in the same dates. Clearly we can see the weakening of the HeI 6678 line as time goes from April 03 (pink curve) to June 11 (blue curve) and to June 26 (green curve).
Fig. 2
Figs 3 and 4 are the same as Figs 1 and 2, but now for the [NII] 5755 and HeI 5874 lines. Spectra were taken on the same nights with same equipment and same integration times. The "aesthetic" Fig 3 shows those two lines respectively close to the center, and midway between center and the right end. Once again, the weakening of both lines can be visually perceived.
Fig.3
Fig 4 shows the calibrated spectral profiles for the second region and again demonstrates the weakening of the [NII] 5755 and HeI 5874 lines with time. The pink curve is for April 03, the green curve for June 11, and the blue curve for June 26.
Fig.4
Under our viewpoint, it became evident that the Eta Car spectroscopic event can perfectly be recorded even with the instrumental limitations of our amateur equipment. As far as we know, this is the first time that amateurs have followed this event - and it shows also the huge potential of spectroscopy for amateur astronomy.
Return
to REA Spectroscopy Page