By David Antao

In this article, I will try to show you that amateur astronomers can achieve remarkable results in the astrophysical study of the stellar world. For some years now, we have had powerful tools at our fingertips. We can, thanks to them, have access to the dynamics of the sky and to carry out our own studies, either for our greatest pleasure, or also to participate, at our modest level, in the progress of science. The ideal is to combine the two, which I could do in the study of this spectroscopic binary star. Indeed, many collaborations between professional astrophysicists and amateur astronomers have started with encouraging results.

During a meeting between professionals and amateurs, was born my collaboration with David Valls-Gabaud, astrophysicist at Paris Meudon observatory. He suggested me to follow a spectroscopic binary. I was finally able to combine the pleasure of amateur observation with the utility for research! And for an amateur astronomer it’s great!

What is a spectroscopic binary?

It’s a pair of stars linked by gravity that revolve around each other. The closer they are, the shorter their period, the faster they spin around each other and the less they can be separated in a telescope, until they can no longer be used even by the largest telescopes today. What is great is that we will be able, in these cases, use a spectroscope and thanks to the Doppler-Fizeau effect, we will see the 2 stars turn around each other. We have access to the sky in motion!

Composition of the orbital binary velocities seen from the Earth.
The red arrow symbolizes the proper speeds of the stars. On the other hand, only the projection of these velocity on our axis of observation, symbolized by the black arrow, is accessible to us. It’s called radial velocity. This velocity is maximum when the stars are in position 2 and minimal or even zero when they are in position 3.

The target was HD23642, a star in the Pleiades of magnitude 6.9 and A0V spectral type. This magnitude is low for the LhiresIII 2400 gr/mm and a 250 mm telescope. It was necessary to choose to carry the study on the line H beta, because the flux in this wavelength is superior to H alpha for the hot stars. The unit exposure was 20 minutes.

This couple of stars has a period of about 2.5 days. It took me several nights to cover almost the entire period. This led to 319 high resolution spectra of which only 245 were exploitable. This low efficiency is due to some disturbances (clouds, wind, humidity too strong …) that degrades the signal-to-noise ratio and the spectrum is no longer usable. This represents a total of about 122 hours of observation to capture 91% of the revolution. This is also the strength and usefulness of amateur astronomers: having a lot of telescope time to be able to devote to a long background work.

My first spectra

Here are 3 spectra at different times of the orbital period. It is easy to see that the profile of the H beta line changes with time.

What is important to understand is that our spectrum is the addition of the light of 2 stars. Depending on the position of the stars in their orbit, seen from Earth, a star moves away from us, while the other approaches. So when we integrate the Doppler-Fizeau effect, the H beta line of the approaching star will be shifted to the blue (that is to say towards the left part of the spectrum) and the one that moves away will be shifted towards the red (that is to say towards the right part of the spectrum).

Decomposition of the observed spectrum

We will have the impression that the line is split. This applies to H beta line, but also for all other lines, except the atmospheric lines, as can be seen on the stack of spectra as a function of the phase.

It’s very good think to make all these amateur spectra, but what quality can they have? After asking the question to David Valls-Gabaud, he replied that it was very good, but the doubt is still present … until the famous evening before going to bed around 11 pm, I look at one last times my emails. Here is an email from David: “Here is a spectrum of HD23642 acquired by the UVES instrument of the VLT”. What was my surprise when I opened the attachment …. It was one of the greatest emotions of my life as an astronomer. Can not find sleep before late at night! My spectra are superbly overlapping with those of the VLT, but of course, I am far from competing with the VLT with my 250mm telescope. To compare, the VLT realized in a single exposure a spectrum over the entire visible range and even a little beyond. The exposure time on the UVES instrument of the VLT was 110 seconds, against 1200 seconds for me, only around the H beta line. But still, the result is impressive!

Superposition of amateur spectrum with a spectrum of VLT

Note on the second superposition, that the blue part of the amateur spectrum (left side of the spectral line) is slightly less intense. This is due to a bad correction of the atmosphere effect. On the other hand, it does not affect the shape, the position of the lines, and therefore neither the radial velocities.

After switching to SBS software

It’s exciting that from the study of radial velocities we can get all this information on the physics of a couple of stars.
THANK YOU KEPLER!

Being president of an astronomy club in the Tarn http://www.astrosurf.com/apam/ Of course, I did a preview of my work to my friends from the Astronomy Association Montredonnaise of Photography, A.P.A.M. This gave some people the desire to go further and especially with additional measurements in photometry, by Vivien Pic. This was possible because this spectrometric binary is also an eclipse binary. In spectroscopy we will be able to make the measurements above, but the inclination is inaccessible to us; precisely photometry will fill this gap. And in the end, we can go even further in the study of this pair of stars, and even estimate its distance and suddenly, that of the cluster of which it is a part.

Vivien followed this star with a 100mm telescope and the results are well compatible with those of the Kepler spacecraft! This precision is remarkable!