Taking a star’s temperature

The scientists first turned their attention to stellar spectra to improve exoplanet detection using radial velocity. This method measures slight oscillations in a star generated by the gravitational pull of an orbiting planet. The greater the oscillations, the larger the planet.

But it’s hard to detect very small oscillations and therefore low-mass planets. To overcome this problem, Artigau and his team developed a technique exploiting the radial velocity method that analyzes a star’s full spectrum and not just a few portions, as previously done with this method.

This makes it possible to detect planets as small as the Earth orbiting around small stars. Artigau then came up with the idea of using a similar strategy to detect not only variations in a star’s oscillations but also in its temperature.

Temperature measurements are critical in the search for exoplanets, which are mostly observed indirectly by closely tracking their star. In recent years, astronomers have faced a major hurdle: how to distinguish between the observable effects of a star and its planets.

This is a problem in both the search for exoplanets using radial velocity and the study of their atmospheres using transit spectroscopy.

“It’s very difficult to confirm the existence of an exoplanet or to study its atmosphere without precise knowledge of the host star’s properties and how they vary over time,” explained Charles Cadieux, a doctoral student at IREx who contributed to the study.

“This new technique gives us an invaluable tool for ensuring that our knowledge of exoplanets is solid and for advancing our characterization of their properties.”

A star’s surface temperature is a basic property that astronomers rely on because it can be used to determine the star’s luminosity and chemical composition. At best, a star’s exact temperature can be known to an accuracy of about 20 degrees Celsius.

However, the new technique measures not exact temperatures but temperature variation over time, which it can determine with remarkable precision.

“We can’t tell whether a star is 5,000 or 5,020 degrees Celsius, but we can determine if it has increased or decreased by a degree, even a fraction of a degree – no-one’s ever done this before,” said Artigau.

“It’s a challenge to detect such minute temperature changes in the human body, so imagine what it’s like for a gaseous ball with a temperature in the thousands located dozens of light-years away!”

To demonstrate that their technique works, the researchers used observations taken with the SPIRou spectrograph in the Canada-France-Hawaii Telescope and the HARPS spectrograph in the ESO’s 3.6-m telescope.

In the data captured by these two telescopes for four small stars in the solar neighbourhood, the team could clearly see temperature variation, which they attributed to either the star’s rotation or to events at its surface or in the surrounding environment.

The new technique made it possible to measure large variations in temperature. For the star AU Microscopii, known for its high stellar activity, the team recorded variations of almost 40 degrees Celsius.

With this technique, they were able to measure not only very rapid changes in temperature associated with short rotation periods of a few days, such as those AU Microscopii and Epsilon Eridani, but also those occurring over much longer periods of time, a difficult feat for ground-based telescopes.

“We were able to measure changes of a few degrees or less occurring over very long periods, such as those associated with the rotation of Barnard’s star, a very quiet star that takes five months to complete a full rotation,” explained Artigau. “Before, we would have had to use the Hubble Space Telescope to measure such a subtle and slow variation.”

The new technique also made it possible to detect very fine temperature changes at the surface of the stars. For example, the team detected subtle temperature changes in star HD 189733 coinciding with the orbit of its exoplanet HD 189733 b, a giant “hot Jupiter” planet.

The UdeM researchers point out that the technique works not only with SPIRou and HARPS, but with any spectrograph operating in the visible or infrared range.

The innovative technique will be directly applicable to observations from NIRPS, a spectrograph installed last year in the ESO telescope in Chili. According to the researchers, it would also be possible to use this technique with space-based instruments, such as the James Webb Space Telescope.

“The power and versatility of this technique means we can exploit existing data from numerous observatories to detect variations that were previously far too small to be perceived, even on very long timescales,” said Artigau.

“This opens up new horizons in our study of the stars, their activity and their planets.”