Project description
Understanding the origin and role of mesoscale convection
Mesoscale convection is often found in the form of ordered patterns that persist for several hours. These systems are not present at larger and smaller scales, and evolve in unpredictable ways. They are important sources of uncertainty in the prognosis of climate change. In climate models describing the long-term evolution of Earth's atmosphere, these processes are greatly simplified and described as additional heat fluxes. The EU-funded MesoComp project will conduct parallel simulations to deepen understanding of their dynamics, origin and role in the transport of heat and momentum. These high-fidelity simulations will guide the design of classical and quantum machine learning models to predict the evolution and statistics of mesoscale convection and quantify the transport fluxes beyond the mesoscale.
Objective
Turbulent convection flows in nature display prominent patterns in the mesoscale range whose characteristic length in the horizontal directions exceeds the system scale height. Known as the turbulent superstructure of convection, they are absent on both larger and smaller scales and evolve in ways not yet understood; but they are an essential link in the heat and momentum transport to larger scales, an important driver of intermittent fluid motion at sub-mesoscales, and one major source of uncertainty in the prognosis of climate change and space weather. In MesoComp, I will investigate the formation of superstructures in massively parallel simulations of compressible turbulent convection in horizontally extended domains, aiming for a deeper understanding of their dynamical origin and role in the transport of heat and momentum. I will then use these high-fidelity simulations to build recurrent machine learning models to predict the evolution and statistics of the superstructure and thus quantify the transport fluxes beyond the mesoscale. I will also analyse the impact of the mesoscale structures on the highly intermittent statistics at the small-scale of the flow and reveal the resulting feedback in the form of improved subgrid parametrizations by means of generative machine learning. MesoComp opens additional doors to the application of quantum algorithms in machine learning which significantly improve the statistical sampling and data compression properties compared to their classical counterparts. From a longer-term perspective, my research results in a quantum advantage for the numerical analysis of classical turbulence, which accelerates the parametrizations of mesoscale convection and increases their fidelity. This work will finally lead to more precise predictions of the on-going climate change and global warming. The results will also improve solar activity models and thus solar storm prognoses with impacts on satellite communication and electrical grids.
Fields of science
- natural sciencescomputer and information sciencesartificial intelligencemachine learningsupervised learning
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectrical engineeringpower engineeringelectric power distribution
- natural sciencesphysical sciencesquantum physics
- engineering and technologymechanical engineeringvehicle engineeringaerospace engineeringsatellite technology
- natural sciencesphysical sciencesclassical mechanicsfluid mechanicsfluid dynamicscomputational fluid dynamics
Programme(s)
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Topic(s)
Funding Scheme
ERC - Support for frontier research (ERC)Host institution
98693 Ilmenau
Germany