Under the title “Subsurface exploration with HPC” on 2 June 2021 took place the second webinar of the ENERXICO project, a Europe-Mexico collaboration that aims to develop performance simulation tools that require exascale HPC and data intensive algorithms for different energy sources, among them oil & gas.
The webinar was chaired by Jose Maria Cela, Computer Applications in Science and Engineering (CASE) Director at the Barcelona Supercomputing Center, and it was divided in three parts: first, a section with three speakers presenting state-of-the-art results on this topic; later, a roundtable discussion about the importance and developments in HPC and Geophysics; and finally, a fun quiz for all attendees with questions about the presentations.
Josep de la Puente, Geosciences Applications Group Manager at the Barcelona Supercomputing Center, was the first speaker and he gave an overall view about HPC software and hardware developments for imaging in the last decade. Then it was the turn of Ursula Iturraran, Professor-Researcher at the National Autonomous University of Mexico (UNAM) with the presentation “Machine Learning for geophysical exploration”. In her talk she explained the current tools to apply Machine Learning and presented the “Gamma test”, which is a tool that allows to choose better inputs for a neural network. Finally, Romain Brossier, Associate Professor at Université Grenoble Alpes, presented FWI and explained the challenges and solutions for waveform inversion with reflected waves.
After the presentations, the three speakers participated also in the roundtable along with Taoufik Ait-Ettajer, Subsurface Manager and Technology Projects E&P at Repsol and Jose Maria Cela, Computer Applications in Science and Engineering (CASE) Director at the Barcelona Supercomputing Center. The discussion focused on the impact of HPC in geophysical exploration in the last 10 years and also the role that it will have in the future of oil & gas. On the other hand, the speakers also reflected upon the algorithmic developments that could make a breakthrough with the help of HPC and AI in the next months as well as the role of public R&D in the development of computational geophysics.
At the end of the webinar, there was a pop quiz for all attendees with questions related to the presentations in order to generate more interaction. Both speakers and participants had a good time and it was an enjoyable way to end the webinar.
Watch the video of the webinar: