Research
Our project is focused on improving the way we visualize what’s beneath the Earth’s surface, which is crucial for finding critical minerals and tapping into geothermal energy. Currently, the main challenge in this area is the time and computational resources required to process 3D data from Magnetotelluric (MT) surveys using traditional methods like Markov Chain Monte Carlo simulations.
To address this, our research involves the development of Reduced-Order Models (ROMs) enhanced with a technique called hyper-reduction. This approach significantly speeds up the data analysis process, making it more efficient and less resource-intensive.
By enabling faster and more effective subsurface imaging, we can better locate and assess mineral deposits and geothermal resources. This not only aids in the exploration and sustainable extraction of these resources but also supports the broader goal of utilizing Earth’s natural resources more effectively.
Host
Universidad Politécnica de Cataluña
Expected Results
Implement a solver able to deal with the tight time constrains required by inversion algorithms. Knowledge of the performance of different hyper-reduction techniques in the context of MT. Three publications in high-impact indexed journals in the field of applied mathematics, numerical methods or applied geophysics.