Grant: Canada Research Chairs

Understanding the properties of nuclear matter requires a multi-faceted approach. One aspect of my research is focused on how nuclear matter behaves, using a modern formulation of fluid dynamics that I co-developed. The other involves penetrating probes, specifically electromagnetic radiation and high-energy nuclear probes known as jets, to confirm the expected nuclear matter behavior from fluid simulations. I plan to develop advanced numerical simulations for both fluid dynamics and penetrating probes to perform precise model-to-data comparisons using Bayesian inference, thus constraining nuclear matter properties such as its viscosity and equation of state (relating pressure to temperature). To understand the reliability of the constraints on nuclear medium properties, theoretical systematic uncertainties will be considered for both penetrating probes of the nuclear medium and its fluid simulation. These theoretical systematic uncertainties stem from different theoretical approaches used when describing the interaction between penetrating probes and the nuclear matter, and from mathematical approximations utilized to obtain the nuclear equation of state along with its dissipative properties such as viscosity. The theoretical frameworks used when calculating the interaction of penetrating probes with nuclear matter are also a source of theoretical systematic uncertainty that I will consider. Bayesian model-to-data comparisons in large-scale simulations, together with the methodical approach of including theoretical systematic uncertainties, provide a controlled and robust improvement of our understanding of the nuclear medium while providing an invaluable link between theoretical developments and experimental measurements. In turn, these Bayesian results highlight cases in which improved measurements are needed, while exposing poor parameter constraints, which entail more theoretical development. Both the physics explored and the approach used fit with the envisioned progress presented in the Canadian Subatomic Physics Long-Range Plan. The large-scale Bayesian comparisons planed herein cannot be achieved without a deep level of software optimization by leveraging the acceleration resources in modern processors. Optimized simulations are in turn useful for processor engineers, exposing bottlenecks in hardware design to which realistic simulations of physical systems are sensitive, thus pinpointing areas that necessitate additional hardware improvement.