Signal Correction to Reveal other Earths

Project Information

SCORE

Grant agreement ID: 851555

DOI

10.3030/851555

Project closed

EC signature date 10 October 2019

Start date 1 January 2020

End date 31 December 2024

Funded under

• EXCELLENT SCIENCE - European Research Council (ERC)

Total cost

€ 1 500 000,00

EU contribution

€ 1 500 000,00

Coordinated by

UNIVERSITE DE GENEVE
Switzerland

Periodic Reporting for period 4 - SCORE (Signal Correction to Reveal other Earths)

Reporting period: 2024-07-01 to 2024-12-31

Summary of the context and overall objectives of the project

Searching for life signatures on other planets is one of the key endeavours of astrophysics, and today,
we are in a unique position to make this possible. The TESS satellite has detected interesting Earth candidates
and PLATO in a few years from now will detect real Earth-twins orbiting bright stars, which will allow follow-up
studies with JWST, ELTs and next generation space telescopes (HWO, Life) to characterize the atmosphere
of those exoplanets.

However, TESS and PLATO will only measure the radius of the detected Earth-twins, which is not enough to interpret
the spectroscopic features in their atmospheres. The mass is also required, and it can be obtained using the
radial-velocity (RV) technique, which measures the gravitational influence of an exoplanet on its host star.

To measure the mass of the Earth-twins that TESS and PLATO will detect, the community have built incredible RV
instruments that can reach a RV precision of 0.3 m/s (ESPRESSO results on Proxima). Such an extreme precision
is required to measure the tiny signature of an Earth-twin, however, this is without considering the perturbing
signals induced by its host star, by Earth’s atmosphere and by instrumental noise. Indeed, we know that these
perturbing signals mask completely the signal induced by an Earth-twin, and now that the RV instruments have
the sensitivity to detect such planets, it is urgent to develop novel methods for mitigating the different perturbing
signals.

Understanding the different perturbing signals is extremely challenging and require incredible data. The PI
have built two telescopes that feed Sun-light into the best RV instruments. The obtained data are of
exceptional quality, and the goal of SCORE is to analyse them, explore novel promising methods for
mitigating the different perturbing signals and find the tiny signatures of Earth and Venus. This will
open the way towards the mass-measurement of Earth-twins, which is essential in the quest for finding
life elsewhere, but also to understand planetary formation and dynamics. SCORE will therefore benefit
the entire exoplanet community.

This project was successful in significantly improving the RV precision and demonstrating that on the solar
data owned by the PI, planets as small as 2 Earth-mass at one year of orbital period (20 cm/s) could be detected,
using a combination of spectra processing and machine learning techniques to mitigate perturbing signals. This is
a factor of two better that what other groups published (40 cm/s), and much better
to the situation before the SCORE project started. In addition, the developed framework also work for realistic stellar
observations, in which case habitable zone planets as small as 4 Earth mass could be detected. Thanks to the work
performed within SCORE, it was possible to confirm with collaborator 2 Super-Earth mass planets in the habitable zone
of the nearby stars HD48948 and HD20794, and two more similar planets are currently being
published. Such planets represent prime target for future ground and space-based missions that will allow an in-depth
characterisation of their atmosphere, for the search for life elsewhere in the Universe. As a conclusion, SCORE
demonstrated that the Earth-like planetary candidates that PLATO will find are within reach of the RV technique.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

We developed YARARA to probes perturbing signals at the level of spectral time series and correct for them using physically-driven or data-driven approaches. YARARA v1 produce spectra
corrected from the observed systematic and respective derived RVs show a gain of 20-30% in precision. In addition, we also developed YARARA v2, which mitigate further perturbing systematic
using a PCA-based correction using the RV information of each spectral line, and which demonstrate a further 20 to 30% improvement. When re-analysing the data of HARPS and HARPS-N with YARARA,
we discovered several very interesting small-mass planets. We recently published the discovery of a super-Earth planet in the habitable zone (HZ) of HD20794, and 3 super-Earths, including one in the HZ around HD48948.
More publications of similar planets are being prepared.

We also investigated the origin of stellar signals in RVs. By evaluating at which average physical depth in the photosphere each point of a high-resolution spectrum is formed, and then measuring a RV signal
per physical depth range, we demonstrated that stellar activity is more affecting spectral regions that are formed deep into the photosphere. We demonstrated that the activity signal seen in solar RVs,
but also on the RV of other stars, significantly changes from deep into the photosphere towards the surface of the star.

Based on the existing ESPRESSO data reduction software (DRS), we developed new DRS for HARPS-N and HARPS to remove at best the instrumental systematics observed at the level of the raw
data from those instruments. The new DRS applied to the HARPS-N solar RV data show an improvement of up to 20% in RV precision and results on HARPS demonstrate an improvement of up to 50% for data taken after mid-2015.
Both DRS are now part of the official ESO release of the ESPRESSO pipeline and are publicly available.

We also investigated the use of machine learning to enhance tiny planetary signal detection. Starting from YARARA v1 corrected spectra, we trained a CNN to
model how instrumental and stellar signals perturb stellar spectra. Analysis of the corrected spectra demonstrate on HARPS-N solar data a sensitivity to planets as small as 2 Earth-mass in the HZ.
This CNN framework can be applied to other stars, demonstrating a 4 Earth-mass sensitivity in the HZ. Such a CCN framework demonstrate that the detection of other Earths is within reach of the RV technique.

Finally, to test further stellar activity mitigation techniques at the spectral level without being perturb by instrumental signatures, we developed SOAP-GPU which can simulate stellar activity
at teh spectrum level. Comparison between SOAP-GPU RVs using as input the observed position of solar active feature and the HARPS-N solar RVs, show an excellent match down to a level of 0.9 m/s. We demonstrated as
well that the behaviour of individual spectral line is also properly modelled. This demonstrate that the main physics is included in SOAP-GPU and that the code can be used to simulate realistic solar activity effects at the
spectral level, and therefore can be used to test mitigation strategies.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The state of the art in the field of stellar and instrumental signal mitigation in RV consist in moving away from RV time series and looking for extra information in high-resolution spectra.
The SCORE project demonstrated that by carefully analysing and post-processing spectral time series we can gain 30-40% in RV precision and a better long-term stability (YARARA framework).
We also understand much better how stellar activity impact individual spectral lines depending on their formation depth (RV as a function of temperature, SOAP-GPU simulation). The SCORE project
already detected excellent (super-)Earth-like planets orbiting in the habitable zone of solar-like stars, that would be prime targets for further atmospheric characterisation with
space-based missions such as HWO, LIFE or ground based observations such as the ones conducted with extremely large telescopes. More detections are to come mainly in the era of the PLATO mission.

Comparison RV Sun HARPN old new DRS

Comparision SOAP-GPU and HARPN solar RVs

RV as a function of temperature of formation in stellar photosphere (from Al Moulla et al. 2024)

Spectral cleaning using YARARA (from Cretignier et al. 2021)