Division of Electronics, Detectors and Computing for Physics (DEDIP)

The Sun's activity is modulated according to an average 11-year magnetic cycle, with the next maximum expected in 2025. This increase in activity implies greater temporal variability for our star, both in terms of its magnetic field, with intense structures appearing and disappearing at a higher rate, and in terms of its atmosphere, which will produce a wind of charged particles that varies in speed and density. These variations have major consequences for the Earth, as it becomes more difficult to predict their impact on our technological society, such as radio blackouts or electrical surges. One of the greatest challenges facing space weather forecasting today is to provide reliable forecasts for the most variable events, which are often also the most extreme.
This thesis proposes to take advantage of the unprecedented conjunction of observations available for the next solar maximum with the Parker Solar Probe and Solar Orbiter space probes, in order to significantly improve the available solar wind models. The student will be able to calibrate the Wind Predict-AW 3D MHD model, one of the most advanced in Europe, to characterize its ability to reproduce conditions of maximum activity. This characterization will involve automated comparisons with different solar datasets, on highly parallel simulations (HPC) producing Big Data-scale results. He will also participate in the development of a new model capable of evolving magnetograms over time, based on the magneto-frictional approach and the evolution of the photospheric electric field - the most advanced techniques for the temporal evolution of magnetic structures - and will use them to quantify the information missing at maximum solar activity, and thus improve space weather forecasts.

In little more than one year of scientific operations, JWST has already revolutionized our understanding of exoplanets and their atmospheres. The ARIEL space mission, to be launched in 2029, will contribute in due course to this revolution. A main finding that has been enabled by the exquisite quality of the JWST data is that exoplanet atmospheres are in chemical disequilibrium. A full treatment of disequilibrium is both complex and computationally expensive. In a first step, our project will numerically investigate the extent of chemical disequilibrium in the atmospheres of JWST targets. We will use towards that end an in-house photochemical model. In a second step, our project will explore Machine Learning (ML) techniques to emulate the outputs of the full photochemical model at a reduced computational cost. The performance of the ML-based emulator will be analyzed with the ultimately goal of its integration into atmospheric retrieval models. The proposed project combines the sophisticated physics and chemistry of exoplanet atmospheres with developments in new numerical techniques.