Projects completed in Spring 2017 Semester



The point spread function (PSF) is among the most critical signal to reconstruct for Weak Gravitational Lensing (WL) analysis. The PSF can only be sampled at the position of the stars, but should be known that the position where the WL is measured, i.e. the position of the galaxies. We want to apply state-of-the art machine learning tools (such as Auto-Encoders and Artificial Neural Networks) to the problem of PSF reconstruction and interpolation. We will work with the CosmoStat Laboratory at CEA (France) to make benchmark tests the methods we will develop.

Colour-Magnitude Diagrams (CMDs) are useful tools to study the stellar formation and evolution. They are constructed from the measurement of the magnitude of stars in two filter bands. We developed a tool based on detailed light profile of the stars that are members of the globular cluster. This profile is used to predict a magnitude in a band based on the image taken in the other band. With that scheme, we can build a CMD with only one image. The scheme can also be applied to future data to recover the spectral type of the object. This procedure was showed to be working in simple circumstances using Artificial Neural Networks. To goal of this work is apply the method to data without any simplification to test the limit of the scheme.

Future space cosmological surveys will measure the weak gravitational lensing effect of large scale structures on the image of distant galaxies, providing new clues on the nature of dark matter, dark energy, and on the content of the Universe. In practice, this requires to measure precisely the shape of billions of faint and small galaxies and will requires “image denoising”.  The work will consist in applying a denoising autoencoder (DA) to galaxy images with the objective of removing noise without altering the shapes of galaxies e.g. size  and ellipticity. An autoencoder is a specific type of Artificial Neural Network where the input and output layers are identical. The objective is to have the autoencoder learn how to build a new representation of the input data (the galaxy image) that has less noise while preserving the galaxy shape attributes. The DA will be applied to simulated galaxy images with noise of different levels and types (Gaussian, Poisson). The alteration of shapes before and after applying the DA will be assessed using a galaxy shape measurement tool.

Il s’agit de mettre au point une méthode rapide (C + python) pour la détermination automatique des paramètres atmosphériques d’étoiles: température effective, gravité, vitesse de micro-turbulence, et métallicité. Cette méthode s’appuiera sur des méthodes de minimisation robustes, inclura des prises de décision automatiques et des accès web à des bases de données extérieures.

Le code sera testé sur des données réelles et servira au plus grand survey actuellement en cours d’étoiles extrêmement déficientes en métaux dans le halo de notre galaxie pour comprendre les premières étapes dans la formation des galaxies.

Exemple de code existant:  http://www.aanda.org/articles/aa/pdf/2014/04/aa22430-13.pdf

Accurately reproducing the observed stellar properties of dwarf spheroidal galaxies (dSphs),

the smallest and faintest galaxies in our universe, represents a challenging task for the standard cosmological model. The work proposed here will consist in understanding the impact on the stellar properties of dSphs (star formation history, metallicity distribution, metallicity gradient, stellar populations, rotation) of the interactions and collisions that dSphs may suffer during a Hubble time.

The study will consist in extracting dwarf galaxies formed in cosmological numerical simulations and make them collide, running idealised N-body simulations based on the chemo-dynamical-C/MPI-code GEAR (Revaz & Jablonka 2012, Revaz et al. 2016) which includes

a complete treatment of the baryonic physics (gas cooling, star formation, chemical enrichment, supernova feedback). The aim will be to explore the effect of parameters like the mass of the progenitors, the impact parameter and the relative velocity on the final properties of the dwarf galaxies. In this purpose, a set of python routines will be developed to allow for a comparison between the model predictions and the observations. The final goal will be to try to reproduce peculiar features observed in local group dwarf galaxies, like distinct stellar populations and metallicity gradients.

The aim of this work is to study the influence of the Self-Interacting Dark Matter (SIDM) on the evolution of dwarf spheroidal galaxies (dSphs) and compare their properties with models run with normal Dark Matter. This is a unique occasion to constrain the cross section of SIDM  on a mass and density regime, the one of dSphs, poorly studied up to now.

The study will consist in running isolated models of dSphs formed out of a LCDM universe and compare their evolution with and without SIDM, focusing particularly on the inner density profile of the dark matter (Cusp/Core) but also on the evolution of merger galaxies. The work will rely on a modified version of the the chemo-dynamical-C/MPI-code GEAR (Revaz & Jablonka 2012, Revaz et al. 2016) which includes a complete treatment of the baryonic physics (gas cooling, star formation, chemical enrichment, supernova feedback) but also a recent implementation of (SIDM). Python routines will be developed to perform the reduction of the simulations.

Because the path of light is bent by gravity, massive galaxies act as “gravitational lenses” which distort the observed images of more distant galaxies. This gravitational lensing effect is a powerful tool to trace the location of dark matter, which cannot be directly seen by telescopes even though it constitutes most of the matter in the universe. The goal of this project is to combine two new data sets, eBOSS (a spectroscopic survey) and DECaLS (an imaging survey), to measure the gravitational lensing effect around distant, massive galaxies. In addition to learning about the dark matter surrounding these galaxies, these measurements will aid in validating the DECaLS imaging data. In the future, this type of measurement can also be combined with other observations to constrain the underlying cosmological model.

The clustering signal of galaxies is a primary cosmological probe for current and future surveys, including the Dark Energy Spectroscopic Instrument (DESI) and Euclid. In particular, interactions between photons and baryons in the early universe imprinted a preferred length scale, known as the BAO scale, that can be used to trace the geometry of the universe. These same interactions caused an initially supersonic flow of baryons, relative to the dominant dark matter, which may have impacted the formation of galaxies. The goal of this project is to construct a consistent model of this effect and include the model in a new software package designed to do cosmological calculations. We will use these results to forecast how sensitive DESI and other future surveys will be to these baryonic effects.

Strong gravitational lensing offers the most powerful way to constrain the mass distribution of massive cluster of galaxies. By combining information collected with the Hubble Space Telescope and the MUSE instrument on the ESO-VLT, we will explore what are the detailed mass models that can reproduce the most precisely the observational constraints of the Hubble Frontier Field clusters. Potentially this can also leads to new important measurement of the cosmological parameters or the properties of the dark matter particles.

Strong lensing happens when the path of the light coming from a background galaxy or quasar is distorted by a foreground massive and compact object. The project is to search for strong lensing systems in the ~2 millions galaxy and quasar spectra of the BOSS and eBOSS catalogs. Spectra are provided with PCA templates of galaxy or quasar automatically fitted by the BOSS pipeline. Strong lensing candidates are identified by looking for emission lines in the spectra not associated with the template, indicating the presence of an additional galaxy along the line of sight of the object. The goal is to conduct a systematic search for such emission lines in both galaxy and quasar samples and will use machine learning to obtain an automated classification of the lens candidates.

One of the main cosmological probes is the Baryon Acoustic Oscillation (BAO), which is measured through the clustering of a large number of galaxies. The on-going SDSS/eBOSS program – currently the largest BAO program – uses several tracers to measure BAOs, one of which are the Emission Line Galaxies (ELGs): star-forming galaxies at redshift ~0.9. The proposed master project is to work on the clustering analysis for the ELGs. This requires understanding from the ELG spectra observations down to the cosmological requirements. The developed tools will be tested against the complete, well-understood BOSS data and with preliminary ELG data, the observations of which has started in September 2016.

Quasars are supermassive black holes living at the centre of galaxies. The Dark Energy Spectroscopic Instrument (DESI) will obtain 700’000 quasar spectra to map the Universe at redshift z>2.1 using a recent technique called the Lyman-alpha forest. The detection of atypical spectra including broad absorption line and damped lyman-alpha systems will be essential for cosmological analyses. The project is to exploit the large quasar spectra database from BOSS and eBOSS (~100’000 object using machine learning algorithm to setup an automatic finder for atypical quasars to be used for DESI.

Il s’agit de tester et d’améliorer un algorithme d’identification de filaments, c’est à dire de galaxies corrélées spatialement en 3 dimensions, dans des champs dans lesquels nous avons les redshifts des systèmes. La méthode proposée s’appuie sur une technique de tesselation. Elle sera testée sur des simulations numériques et des données observées.

Code: http://www2.iap.fr/users/sousbie/web/html/indexd41d.html