At the very end of Jupiter's growth, the planet was surrounded by a disk of gaseous refractory dust and ice, from which the Galilean moons are thought to have formed. However, the physical and chemical properties of this disk are not well constrained, and the formation of the Galilean moons remains a major theoretical challenge.
Our knowledge of the physical properties of the circumplanetary disk has improved over the last decade thanks to three-dimensional hydrodynamic models of the protoplanetary disk, in which a forming Jupiter is introduced. They show a very massive and hot disk, which could be rapidly depleted to a cold and light disk depending on the accreting flow properties. Although very accurate, these models could only simulate evolution over a few thousand years, whereas one needs to study the disk over 100 kyr and up to 1 Myr of evolution to study the formation of the moons.
My work has therefore focused on adapting simpler models to take account of the new clues to the physics of the circumplanetary disk provided by hydrodynamic simulations. In this seminar, I will present a two-dimensional model that accounts for the evolution of Jupiter's surface temperature, which produces the unexpected effect of the disk's self-shadow. This shadow cast on the disk creates a cold region where volatile ices can form close to Jupiter. I will discuss the possible implications of this effect on the composition of the disk and whether it can have an impact on the moon formation process.

For the last 60 years, theoretical models have struggled to explain the observed level of tidal dissipation in bodies with substantial convective envelopes, such as giant planets and solar type stars.  In this talk, we review these models and show that they break down when tides are rapid,  that is,  when the tidal period is short compared to the convective turnover timescale.  We propose a new formalism applicable in this regime.  Assuming that tides transfer their energy to the convective flow, this formalism yields values for the tidal dissipation factor of Jupiter and Saturn, as well as for the circularization periods of solar-type binaries and extrasolar planets, in good overall agreement with observations. Lastly, we show that the phase lag and the tidal dissipation factor Q vary by orders of magnitude throughout  a gaseous envelope, and we define average quantities that can be directly compared with observational data.