Université Côte d'azur

UE Observational Science

Code de l'ECUE : RMUSP10

Terre solide : géodynamique des enveloppes supérieure, paléobiosphère , Astronomie, astrophysique
Campus Valrose
Master 1 , Master 2
Semestre impair


This course offers an overview of the methodology traditionally employed in the context of observational sciences. Applications will cover the realms of Earth (Geology & Geophysics) and Space (Astronomy & Astrophysics) science. The course will attempt to cover the cycle of production of scientific knowledge in four major phases: it starts with the individual curiosity and the origin of scientific inquiry, which leads to the design and engineering of a scientific observing program, switfly followed by the art of data collection and production, and goes full circle with the interpretation of data and how it leads to explanations to the phenomena and objects that we experience and see. Albeit strongly rooted in several hundreds of years of tradition, we will also look at how the coming of age of very large data-sets calls for complementary approaches to the traditional model of scientific inquiry.

This ought to be a fun ride that will take us from the deepest parts of our Earth all the way to distant stars!

Responsable(s) du cours

, Frantz Martinache


  • 24h de cours magistral


Avant le début du cours, je dois ...
  • Licence


A la fin de ce cours, je devrais être capable de...
  • to translate one's curiosity for an object and/or a phenomenon into the test of a scientific hypothesis
  • to exploit data as input for inverse problems
  • to recognize where the big data framework can contribute to one's topic of inquiry
  • to collect data and their auxilliary meta-data in the lab or in the field
  • to design or decide of an experiment, an instrument or an observing program that can put a hypothesis to the test


  • This session will introduce the overall arc of the course and focus on how to translate everyday curiosity into testable hypothesis that are at the core of the scientific method. Two complementary approaches will be introduced:

    • the opportunic use of a one-time event for research purposes
    • the pursuit of refinement in our understanding of fundamental physics


    This session will be taught by: Frantz Martinache, Gueorgui Ratzov & Gilles Métris.

  • This session will teach you how one goes about specifying an instrument to reach a specific scientific objective. The general method of translating scientific top-level requirements (TLR) into technical specifications will be illustrated by the example of the stellar interferometric recombiner MATISSE at the focus of the VLTI.


    This session will be taught by Sylvie Robbe-Dubois

  • Specification of an experiment embarked onboard a satellite: the MICROSCOPE experiment.

    Testing the Universality of Free Fall (or similarly the Equivalence Principle), is fundamental to provide guidelines in order to unify the gravitation theory wit the quantic interactions. During this session we will see how we can realize this test much more precisely than Galileo, Newton and their successors thanks to a space experiment and will study the limitations.

    This session will be taught by Gilles Métris.

  • This session will teach you how one can use the chemical and mineralogic analysis of meteorites to learn more about the story of our Solar System.

    You will learn:

    • to characterize meteorites and their constituents
    • to carry out macroscopic and microscopic observations
    • to disentangle a complex system in a time-physico-chemical space
    • to elucidate their modes and locations of formation
    • to constrain the evolution of the solar system

    and even to see how what we learn from meteorites can lead space missions to visit asteroids!


    This session will be taught by Guy Libourel & Clément Ganino.


    This session includes lab work that will require the presence of students on the Mont-Gros campus of the Observatoire de la Côte d'Azur!


  • This session will teach you how you can use seismometers and geodetic instruments to measure the deformation of planets and to sound their interior.

    Our journey will start right where we are, on the surface of our familiar but very dynamic planet Earth, and we will see how we can use the aforementioned instrumentation to reveal the secrets of its inner structure.

    Armed with that knowledge, we will be ready for a journey to the even more mysterious Mars with InSight, a mision of the NASA Discovery program ("faster, better, cheaper") designed to catch marsquakes and learn about the interior of the red planet.

    This session will be taught by Lucie Rolland.


  • Whereas you can directly weigh yourself every morning on a bathrooom scale, many objects and phenomena are just beyond the reach of direct measurements. Instead, in many cases, we must use data collected to indirectly infer otherwise inaccessible properties like the inner structure of our Earth or the fundamental parameters of stars.

    In this session, you will learn:

    • to formulate an inverse problem, whose resolution will lead to infer parameters describing a physical process.
    • to identify the tools and methods to use to solve specific inverse problems

    • to carry out the analysis and to present its results: values for parameters and their uncertainties


    This session will be taught by Jean-Paul Ampuero & Florentin Millour.

  • This session will reveal what are the strengths and limitations of numerical simulations that are often used to make sense of complex physical objects and phenomena. We will apply these tools to the study of astrophysical plasmas in supernovas (radiative shocks) as well as in the interplanetary medium (non-collisional shocks in the Earth - Sun interactions).

    You will learn:

    • to develop faster and systematic treatment of analytic equations
    • to find solutions for systems without a known analytical solution
    • to find solutions to non-linear, non-intuitive problems

    You will also learn:

    • to choose the (physical) model for a given problem
    • to choose the numerical methods to implement
    • to choose the data visualisation and interpretation of the simulation results


    This session will be taught by Claire Michaut & Pierre Henri.

  • In most of the scenarios that were covered thus we saw data as being the product of a deliberate process necessary to answer a specific scientific question.

    In this session we will acquaint ourselves with a very different situation that is becoming more and more common: the availability of pre-existing very large data-sets, either acquired on purpose or as by-products of other types of inquiries or general monitoring activities.

    We will see how to take this reality into consideration to adapt the cycle of scientific production and learn how to choose from:

    • a (supervised) model driven approach, or
    • a (unsupervised) data driven approach


    Big data brings other kind of big problems and we will introduce a few technical tools to keep things under control with:

    • dimensionality reduction
    • clustering & regression


    This session will be taught by Benoit Carry.

Accéder au Syllabus complet (Authentification requise)
Ce syllabus n’a aucune valeur contractuelle. Son contenu est susceptible d’évoluer en cours d’année : soyez attentifs aux dernières modifications.