Dr. Guillaume Guiglion
Senior Postdoctoral research associate at AIP (Feb. 2017 - present)
- Chemo-dynamical evolution of the Milky Way structures, using chemistry and kinematics (Guiglion et al. 2015)
- Chemical evolution and stellar nucleosynthesis , e,g, neutron-capture elements and lithium (Guiglion et al. 2018, Guiglion et al. 2019a)
- Stellar spectroscopy (Guiglion et al. 2016),
- Machine-learning for astronomy (Guiglion et al. 2020)
- Survey strategy (Guiglion et al. 2019b)
- Variable stars (Guiglion et al. 2013)
40 publications, including 24 refereed, from which 6 as first author, with a total of 893 citations. H-index=16. (full list here)
Surveys & Collaborations
- Gaia-ESO Survey (GES): WG15 (2012-present)
- Gaia-Benchmark project: chemical abundance working group (2013-2014)
- The RAdial Velocity Experiment (RAVE): DR6 core-team (2017-present)
- 4-metre Multi-Object Spectroscopic Telescope (4MOST): member of the LR & HR Milky Way Halo (S1, S2), and Milky Way Disc and Bulge (S3, S4) Surveys. Co-leader of the very metal-poor stars survey in S3. Infrastructure Working Groups: Galactic pipeline iWG7 and Survey Strategy iWG2 (2017-present)
- Maunakea Spectroscopic Explorer (MSE): Milky Way working group (2019-present)
- Sloan Digital Sky Survey (SDSS): Milky Way (Halo+Disc) & chemical pipeline working groups (2020-present)
- GAlaxy MApping at Calar Alto (GAMACA): member of the M31 ISM working group (2020-present)
- Member of ChETEC: "Chemical Elements as Tracers of the Evolution of the Cosmos" (2017-present)
- GAUGUIN (standard stellar abundance pipeline, Guiglion et al. 2016): integrated to Gaia-RVS DPAC Apsis pipeline; RAVE DR6 abundance pipeline; Gaia-ESO DR2&DR3.
- CNN (Convolutional-Neural-Network for stellar parametrization, Guiglion et al. 2020): developed, and made public in the context of RAVE. Currently under integration to the 4MOST Galactic Pipeline 4GP.
Some more details on my research
Revealing the lithium evolution in the metal-rich regime
Together with Cristina Chiappini, we investigated the decrease of the lithium upper boundary in metal-rich dwarfs stars (Guiglion et al., 2019a). Coupling lithium abundances from Guiglion et al. (2016) and chemical evolution models including stellar radial-migration(Chiappini 2009), we proposed that the downturn of the lithium upper boundary is due to a significant fraction of old stars migrated from the inner-regions; those stars depleted their photospheric lithium abundance during their travel-time.
Maximizing the scientific output of the RAdial Velocity Experiment (RAVE) with Machine-Learning
In order to perform chemo-dynamical investigations of the Milky Way structures and put solid constraints on their formation, precise stellar chemistry is needed. In Guiglion et al. 2020, I developed a Convolutional-Neural Network method to parametrize the spectra of the RAdial Velocity Experiment, using high-resolution APOGEE DR16 stellar labels. The ultimate goal was to prepare the ground for Gaia-RVS analysis in 2022. I showed that CNN is a powerful method that can unify different observables: RAVE spectra, magnitude from Gaia DR2, 2MASS and WISE, and Gaia DR2 exquisite parallaxes. It allows to lift pass over the spectral degeneracies faced by standard pipelines when analyzing RAVE narrow-range and intermediate resolution RAVE spectra (see figure below). I also derived chemistry for more than 400000 stars, well beyond RAVE DR6 (Steinmetz et al. 2020).
Preparing the ground for the 4-metre Multi-Object Spectroscopic Telescope (4MOST)
The ultimate goal of 4MOST is to provide an unprecedented chrono-chemo-dynamical map of the Milky Way and Magellanic Clouds, and to unveil the nucleosynthesis history of a large variety of elements. 4MOST will provide both LR and HR spectra (first light in 2023) allowing to study for instance the chemical evolution of lithium for several 100000 stars, and will allow to answers key questions, such as the melting of the Spite plateau. In this context, a well tailored pipeline for spectral analysis in needed, and I am working as an active member of the Galactic Pipeline Infrastructure Working group 7 (iWG7) of 4MOST. I lead the development of a Convolutional-Neural Network approach for the FGK stars of 4MOST. The aim is to infer precise atmospheric parameters and chemical abundances of a large number of species (>20) at both high- and low-resolutions, over a large wavelength range and variety of noise in the data. I am currently integrating my CNN approach to the 4MOST Galactic pipeline 4GP, and the current CNN implementation showed its efficiency to deal with such richness of 4MOST spectra.
In my free time, I enjoy long bike rides in Brandenburg. When I am not riding my bike, I guess one can find me in the Alps. Mountaineering literature and bouldering are also my cups of tea.
PublicationsLatest refereed publications, retrieved from NASA ADS:
Astronomy and Astrophysics, 653, A72; published September 2021
Astronomy and Astrophysics, 651, A84; published July 2021
Astronomy and Astrophysics, 644, A168; published December 2020
The Astronomical Journal, 160, 2, 83; published August 2020
The Astronomical Journal, 160, 2, 82; published August 2020
Astronomy and Astrophysics, 635, A101; published March 2020
The Astronomical Journal, 158, 4, 155; published October 2019
Monthly Notices of the Royal Astronomical Society, 487, 3, 3946; published August 2019
Astronomy and Astrophysics, 623, A99; published March 2019
Astronomy and Astrophysics, 622, A59; published February 2019