Observatoire de Paris Institut national de recherche scientifique français Univerité Pierre et Marie Curie Université Paris Diderot - Paris 7

Sauf exception, les séminaires ont lieu sur le site de Meudon, dans la salle de conférence du bâtiment 17.

S’abonner aux mises à jour de cette page :

Prochains séminaires

  • Jeudi 1er juin 2023 à 16h00 (Salle de réunion du 1er étage du bâtiment 16 et visioconférence)

    NA-SODINN : a deep learning algorithm for exoplanet image detection based on residual noise regimes

    Carles Cantero, Université de Liège

    As of today, there exists a plethora of post-processing algorithms for exoplanet imaging. Their performance has been assessed using different data sets and metrics, which caused confusion in the HCI community when comparing their detection ability. In order to homogenize the comparison of these algorithms, the Exoplanet Imaging Data Challenge (EIDC) was born. With twenty eight algorithm submissions, the first EIDC phase (exclusively dedicated to exoplanet detection) provided two interesting conclusions :

    1. detection algorithms that exploit the local behavior of image noise obtained the highest detection score ;
    2. supervised machine learning algorithms, such as the SODINN deep neural network developed at ULiège, produced a relatively high number of false positives.

    With the aim of improving the robustness of SODINN against false alarms, we built a more advanced version, referred to as Noise-Adaptive SODINN, which relies on two new strategies that help the training to capture stronger local image noise correlations.

    First, unlike its predecessor, NA-SODINN trains an independent classification model per image noise regime in the processed frame.

    Second, its network is fed with S/N curves, local discriminators that contain additional physical-motivated features and help the trained model to better disentangle an exoplanet signature from speckle noise.

    NA-SODINN is evaluated against SODINN through a Receiver Operating Characteristics (ROC) analysis, in which we observe a clear improvement in both sensitivity and specificity. Then, it is submitted to EIDC, where we observe that it is ranked at the top (first or second position) of the challenge leaderboard for all considered evaluation metrics.

    Lien pour assister au séminaire :

Séminaires passés

  • Jeudi 25 mai 2023 à 16h00 (Salle de réunion du 1er étage du bâtiment 16 et visioconférence)

    Measuring the ELT petal mode in presence of AO residual turbulence with pyramid wavefront sensor

    Nicolas Levraud, LAM (Laboratoire d’Astrophysique de Marseille)

    The next generation of Extremely Large Telescope (24 to 39m diameter) will suffer from the so-called "pupil fragmentation" problem. Due to their large spiders, differential pistons will appear in the wavefront between the part of the pupil separated by these spiders during observations.

    The Adaptive Optics (AO) system necessary to compensate atmospheric turbulence appears unable to sense this differential piston leading to bad control by the loop. Hence, such differential pistons, a.k.a petal modes, will prevent the AO system from reaching the diffraction limit of the telescope and ultimately will represent the main limitation of AO-assisted observation with an ELT. All the future single conjugated AO systems for the ELT have a PyWFS (Pyramid Wavefront Sensor) that is sensible to differential piston unlike the Shack-Hartmann, but it is not trivial to get a good enough sensitivity.

    This is particularly true for high contrast observing modes. These differential pistons can evolve quickly, so we are looking for an AO loop scheme able to measure both the atmospheric turbulence and the petal modes. Solutions have been proposed such as the Holographic Dispersed Fringe Sensor (HDFS) for the Giant Magellan Telescope but they are not fast enough to be implemented as WFS of the AO loop and require longer AO sensing wavelength.

    In this talk we want to study how to make the PyWFS sensitive to petal mode with visible light. We show that a small modulation radius makes the PyWFS sensitive to petal but unable to measure atmospheric turbulence due to the PyWFS non-linearities. We therefore propose to add dedicated petal sensor as a 2nd path and we study the unmodulated PyWFS as a candidate for this role.

    We study the reconstruction of the petal mode present in the residuals by this petalometer. We show that the petal mode, due to its spatial frequency distribution beeing infinite, can be confused with other high spatial frequency modes present in the residual turbulence. We propose a focal plane spatial filter to reduce high frequency residuals. The spatial filter helps in reducing this confusion, improving the petal measurement. In this talk we perform E2E simulations to demonstrate the validity and performance of this new concept.

    Lien pour assister au séminaire :

  • Lundi 15 mai 2023 à 14h30 (Salle de conférence du bâtiment 17)

    Applications of Isotopic Geochemistry in Understanding the Origin of Phobos

    Frédéric Moynier (IPGP)

    The MMX mission, which aims to collect and return to Earth more than 10 grams of sample from Phobos, one of Mars’ moons, has a key objective of investigating whether Phobos (and its sibling moon Deimos) are capture asteroids or were formed through a giant impact event. In this presentation, I will show how the isotopic composition of these samples can provide valuable insights into the origin and history of Phobos, and will make a comparison with what we have learned about the origin of our Moon using similar methods.

  • Jeudi 4 mai 2023 à 11h00 (Salle de conférence du bâtiment 17)

    The Mars Soundscape and beyond

    Baptiste CHILDE (Los Alamos National Laboratory)

    On February 18, 2021, NASA’s Perseverance rover landed in Jezero Crater carrying the two first microphones operating on the surface of Mars : the Supercam microphone, positioned on top of the rotating rover’s mast and the EDL microphone fixed on the body of the rover. Working flawlessly since then, they provide the first characterization of Mars’ acoustic environment in the audible range and beyond, from 20 Hz to 50 kHz. Detected sounds originate from three main sources : the atmosphere (turbulence, wind), the shock-waves generated by the Supercam laser ablating rocks, and mission-induced artificial sounds such as the signal generated by the high-speed rotating blades of the Ingenuity helicopter.

    After one Martian year, the Perseverance playlist features more than 12 hours of Martian sounds. In addition to provide an unprecedented short timescale characterization of the wind, temperature fluctuations, and the turbulence dissipative regime, this dataset highlights the unique sound propagation properties of the low-pressure CO2-dominated Mars atmosphere : acoustic impedance varying with the season, large intrinsic attenuation of the high frequencies, and the dispersion of the sound speed in the audible range. This presentation will review these results to date and extend them to the exploration of the acoustic environment in our Solar system.

  • Jeudi 27 avril 2023 à 16h00 (Salle de réunion du 1er étage du bâtiment 16 et visioconférence)

    Deep learning-based focal-plane wavefront sensing : principle, simulations, and in-lab tests

    Maxime QUESNEL (Université de Liège)

    High-contrast imaging instruments are today primarily limited by non-common path aberrations appearing between the scientific and wavefront sensing arms. These aberrations can produce quasi-static speckles in science images that are difficult to distinguish from exoplanet signatures. With the help of recent advances in deep learning, we have implemented convolutional neural networks (CNN) to estimate pupil-plane phase aberrations from point spread functions (PSF). In this talk, I will show results with simulations obtained behind a vortex coronagraph, exploiting its properties to provide an alternative type of phase diversity with a 100% science duty cycle. I will also introduce an autoencoder-based method, that uses a deep CNN as the encoder and a differentiable simulator of the instrument as the decoder. This enforces the latent space to represent phase aberrations, and because the approach is unsupervised, it is not necessary to know the true aberrations to train the models. This is particularly promising for on-sky applications, and results on laboratory data using the Subaru/SCExAO instrument are first presented.

    We will meet in the meeting room on the first floor of building 16 or on zoom.

    Lien pour assister au séminaire :

  • Lundi 27 mars 2023 à 10h30 (Salle de conférence du bâtiment 17)

    Diurnal Variability of Martian Water Ice Clouds As Viewed by the Emirates Mars Mission

    Mike WOLF (Space Science Institute)

    - Attention, horaire inhabituel -

    The orbit of the Emirates Mars Mission (EMM) has created an opportunity for systematic studies of diurnal variations in the Martian atmosphere across seasons (and years). We provide an overview of the mission and the three instruments onboard, with an emphasis on the Emirates eXploration Imager (EXI).

    With EXI, we explore the diurnal, seasonal, and spatial behavior of aphelion cloud belt during Mars Year 36. Building from previous work with the Mars Color Imager (MARCI) onboard the Mars Reconnaissance Orbiter, water ice optical depth has been derived, and zonal and meridional diurnal trends are highlighted. In essence, the retrievals show large morning-evening asymmetries about a minimum near 12 hr, with striking latitudinal distributions in the early morning. We discuss comparisons to the Mars Planetary Climate Model, and also introduce some results of the Emirates Mars InfraRed Spectrometer (EMIRS) for the same epoch.

  • Vendredi 16 décembre 2022 à 11h00 (Salle de conférence du bâtiment 17)

    Searching for singularities in fluid turbulence

    Thierry Lehner (Luth)

    We study hydrodynamical turbulence (for Euler and Navier Stokes=NS equations, mainly for the moment in incompressible cases with 3D homogeneous and isotropic turbulence) by two original ways.
    1) by postulating singularities (of the self similar Leray kind) and by detecting them indeed in real experiments data , which may lead to explain intermittency and dissipation of the turbulent kinetic energy. These predictions go beyond the Kolmogorov scaling laws. An in course proposal model on the possible nature of these singularities (finite time explosion) makes use of non linear Schrödinger equations (NLSE).
    2) by applying the scale relativity theory (born and developed in Meudon by L Nottale) we transform the NS equation into a macroscopic Schrodinger one , which allows to derive new results in turbulence , in particular on Lagrangian intermittency.
    3) we shall compare these current attempts with recent mathematical results on the existence of singularities for the related kind of Euler and NS equations.

    Nous étudions la turbulence hydrodynamique (équations de Euler et de Navier-Stokes=NS , surtout déjà dans des cas incompressible et 3D homogène et isotrope) par deux approches originales .
    1) en postulant des singularités (de type self-similaires à la Leray) et en les recherchant avec succès dans des data réels d’expérience , qui peuvent conduire à une explication de l’intermittence et être la source de la dissipation de l’énergie cinétique turbulente ; ces prédictions diffèrent des lois d’échelle à la Kolmogorov. Une modélisation proposée (en cours) de la nature possible de ces singularités (explosion en temps fini) est faite à l’aide d’équations de Schrodinger non linéaires .
    2) en appliquant la théorie de la relativité d’échelle (née et développée à Meudon par L Nottale) on transforme l’équation de NS en une équation de type Schrodinger au niveau ‘macroscopique’ , ce qui nous permet d’obtenir des résultats nouveaux en turbulence dont sur l’intermittence lagrangienne.
    3) on confrontera ces 2 approches avec des résultats mathématiques récents obtenus sur l’existence de singularités pour les équations de type Euler et de NS.

  • Vendredi 9 décembre 2022 à 11h00 (Salle de conférence du bâtiment 17)

    Bayesian modelling of beta-meteoroid observations with Solar Orbiter

    Samuel Kociscak (LESIA)

  • Jeudi 1er décembre 2022 à 16h00 (Salle de réunion du bâtiment 16 et visioconférence)

    Photonics, astrophysics, and mid-infrared glass

    Harry Kenchington Goldsmith (LESIA)

    Direct imaging of exoplanets promises the detection of new worlds in the habitable zone around distant stars. Interferometry uses multiple, or one-segmented, telescopes to make direct observations of exoplanets. In the mid-infrared the light of the star outshines the exoplanet by 8 orders of magnitude and interferometry can overcome this by nulling the star. New technologies in photonics may provide a pathway to do this. Photonics is the study of light guided and manipulated within transmissive materials. It has roots in the telecommunication industry and is now being used by astronomers in visible light and near-infrared light, with mid-infrared light to follow in the near future. The basic applications in astronomy are to move away from telescopes with large lab benches and environmentally sensitive fibre optics, to small and stable photonic chips. In this talk I will detail my work in mid-infrared photonics, highlighting the production of evaporatively deposited photonic chips and including specific issues with adapting photonic chips for astronomy applications, with the goal to create a nulling interferometer to detect exoplanets.

    Lien pour assister au séminaire :

  • Lundi 28 novembre 2022 à 16h00 (Salle de conférence du bâtiment 17)

    Caractérisation des phases d’altération dans les roches et les sols de Mars par LIBS avec l’instrument ChemCam (Mars Science Laboratory/Curiosity)

    Gael DAVID (LESIA)

    Les conditions environnementales passées de la surface de Mars sont accessibles à travers les enregistrements géologiques de la planète, et plus particulièrement via les minéraux d’altération qu’ils contiennent. Ces phases secondaires peuvent constituer de véritables marqueurs géochimiques des conditions durant l’altération aqueuse.

    L’instrument ChemCam, à bord du rover Mars Science Laboratory (Curiosity) est le premier instrument de spectroscopie sur plasma induit par laser (LIBS) déployé à la surface de Mars. Il offre un moyen inédit de caractériser in situ et à une échelle d’analyse submillimétrique, la composition chimique de la surface de la planète. Durant ce séminaire, je vais présenter quelques travaux réalisés avec cet instrument sur la détection de phases d’altération (i.e. sulfates et oxydes de fer) dans les sols et les roches sédimentaires du cratère de Gale. Leurs processus de formation seront également discutés ainsi que l’histoire géologique qui en découle.

  • Lundi 21 novembre 2022 à 16h00 (Salle de conférence du bâtiment 17)

    Directly imaging cold and temperate exoplanets in reflected starlight : atmospheric characterization and target selection

    Oscar Carrion (LESIA)

    With more than five thousand exoplanets discovered to date, the next milestone is the atmospheric characterization of a significant number of these worlds. The observing techniques that are currently available have biased the discovery and atmospheric characterization of exoplanets towards hot, giant planets on short-period orbits. Direct-imaging observations will be needed to overcome these biases and enable the analysis of cold and temperate long-period exoplanets and their atmospheres.

    Studying this population of planets, which has remained out of reach thus far, is a key to understand the diversity of exoplanets and, in particular, the existence of habitable conditions beyond the Earth. Direct-imaging observations of temperate exoplanets have indeed been identified as a scientific priority for the coming decades both by the US Astro 2020 Decadal Survey and by the ESA Voyage 2050 report. In this talk we address the physical fundamentals of direct-imaging observations in reflected starlight, and derive practical conclusions for the planning and interpretation of the upcoming measurements.

    We also analyse the prospects for atmospheric characterization of exoplanets from reflected-starlight spectra with different observing strategies. Finally, we compute which of the known exoplanets are potentially observable with several upcoming direct-imaging space telescopes, and discuss some interesting science cases.