Big Bang | EPFL Graph Search

Le Big Bang (« Grand Boum ») est un modèle cosmologique utilisé par les scientifiques pour décrire l'origine et l'évolution de l'Univers. De façon générale, le terme « Big Bang » est associé à toutes les théories qui décrivent notre Univers comme issu d'une dilatation rapide. Par extension, il est également associé à cette époque dense et chaude qu’a connue l’Univers il y a d’années, sans que cela préjuge de l’existence d’un « instant initial » ou d’un commencement à son histoire. La comparaison avec une explosion, souvent employée, est elle aussi impropre. Le concept a été initialement proposé en 1927 par l'astrophysicien et chanoine catholique belge Georges Lemaître, qui décrivait dans les grandes lignes l’expansion de l'Univers, avant que celle-ci soit mise en évidence par l'astronome américain Edwin Hubble en 1929. Ce modèle est désigné pour la première fois sous le terme ironique de « Big Bang » lors d’une émission de la BBC, The Nature of Things le (dont le texte fut publié

Cold gas in the cosmic web around galaxy clusters

Damien Louis Alexandre Spérone-Longin

Galaxies live in very different types of environment. Those different environments are relics from the evolution of the Universe since the Big Bang. This network of structures observed via large scale surveys is called Cosmic Web. Within this web, galaxies can be isolated, in a filament, in a group, or in a cluster. When part of a cluster of galaxies, their morphologies, gas content (cold and hot), and star formation activities is seen to be affected by other galaxies, or by the gas inside the cluster. However, the long-term effects of the local environment on galaxy evolution remain unclear. Also, it has been observed that a pre-processing is happening long before the galaxies enter the cluster cores, which could explain the build-up of the passive galaxy sequence, where galaxies have quenched star formation. The aim of this thesis is to see if and how the cold molecular gas reservoir of galaxies is modified when star formation activity is decreasing and if there is a link with the position of the galaxy in the CW around the cluster. We gathered unprecedented observations: the first and largest survey of the molecular gas content of galaxies in and around the same cluster environment at intermediate redshifts. These Atacama Large Millimeter Array observations are part of a larger survey called Spatially Extended ESO Distant Cluster Survey. We are focusing on two clusters at a redshift of 0.5, which benefit from previous deep photometric and spectroscopic observations. This will allow us to derive accurate stellar masses and star formation rates, in order to compare those galaxies to other surveys of the cold gas content of galaxies, either in the field or inside different clusters. We unveiled a new population of galaxies with normal star formation rates but less cold molecular gas than other galaxies at the same stellar mass. This could be an indication that the gas reservoir responsible for the star formation changes either in mass or properties, before the star formation is affected. We also noticed that galaxies with the lowest cold-molecular-gas-to-stellar-mass ratios are preferentially either in the cluster core or in overdensities away from the cluster, confirming the existence of a pre-processing in overdensities surrounding the cluster. But active galaxies inside or close to the cluster, do not always have depleted CO reservoir. This thesis opens up new prospects on the study of the cold gas content of galaxies at intermediate redshift. It showed the necessity to observe galaxies that are in a transition regime between star forming and passive, in order to understand how star formation is suppressed around clusters. Those results could be extended to much larger samples, with the help of the upcoming new generation of instruments such as the Square Kilometer Array.

EPFL2021

Combining Human Computing and Machine Learning to Make Sense of Big (Aerial) Data for Disaster Response

Julien Briant, Muhammad Imran, Stéphane Joost, Patrick Meier, Matthew Josef Parkan, Devis Tuia

Aerial imagery captured via unmanned aerial vehicles (UAVs) is playing an increasingly important role in disaster response. Unlike satellite imagery, aerial imagery can be captured and processed within hours rather than days. In addition, the spatial resolution of aerial imagery is an order of magnitude higher than the imagery produced by the most sophisticated commercial satellites today. Both the United States Federal Emergency Management Agency (FEMA) and the European Commission’s Joint Research Center ( JRC) have noted that aerial imagery will inevitably present a big data challenge. The purpose of this article is to get ahead of this future challenge by proposing a hybrid crowdsourcing and real-time machine learning solution to rapidly process large volumes of aerial data for disaster response in a time-sensitive manner. Crowdsourcing can be used to annotate features of interest in aerial images (such as damaged shelters and roads blocked by debris). These human-annotated features can then be used to train a supervised machine learning system to learn to recognize such features in new unseen images. In this article, we describe how this hybrid solution for image analysis can be imple- mented as a module (i.e., Aerial Clicker) to extend an existing platform called Artiﬁcial Intelligence for Disaster Response (AIDR), which has already been deployed to classify microblog messages during disasters using its Text Clicker module and in response to Cyclone Pam, a category 5 cyclone that devastated Vanuatu in March 2015. The hybrid solution we present can be applied to both aerial and satellite imagery and has applications beyond disaster response such as wildlife protection, human rights, and archeological exploration. As a proof of concept, we recently piloted this solution using very high-resolution aerial photographs of a wildlife reserve in Namibia to support rangers with their wildlife conservation efforts (SAVMAP project, http://lasig.epﬂ.ch/savmap). The results suggest that the platform we have developed to combine crowdsourcing and machine learning to make sense of large volumes of aerial images can be used for disaster response.

Mary Ann Liebert, Inc2016

Exploring the origin of low-metallicity stars in Milky-Way-like galaxies with the NIHAO-UHD simulations

Pascale Jablonka

The kinematics of the most metal-poor stars provide a window into the early formation and accretion history of the Milky Way (MW). Here, we use five high-resolution cosmological zoom-in simulations (similar to 5 x 10(6) star particles) of MW-like galaxies taken from the NIHAO-UHD project, to investigate the origin of low-metallicity stars ([Fe/H]