Computational Geosciences

We offer leading expertise on a wide variety of methods for discretizing and solving systems of nonlinear partial differential equations. We excel at developing new computational algorithms and turning these into efficient, reliable, and robust software of professional quality. We are internationally oriented and combine high academic quality in research with a strong focus on industrial relevance:

• 16 permanent scientists, all with a PhD in mathematics, physics, or geophysics; in addition: several postdocs, PhD, and master students
• 20+ years of continuous involvment in industrial and academic research
• 20+ publications in leading journals and conferences anually
• 15+ years track record in sharing open data (e.g., CO₂ DataShare) and developing high-quality, open-source, community software (MRST and OPM) having hundreds of users world wide

Applications in the Computational Geosciences

Most of our research aims at developing enabling simulation technology for subsurface applications like hydrocarbon recovery and carbon storage.  The MRST website contains descriptions of many of the different methods and problem types we have worked with, including novel ideas like flow diagnostics and multiscale methods. For more in depth information, we refer you to two our open-access textbooks on MRST (book 1, book 2), or you could watch one of our recent overview lectures that explain the history and the capabilities of the MRST software (90 min version, 30 min version). The GeoScale webpages document activities from the early years of the research group. 

4-5 years ago, we started to transfer our specific expertise developed for the oil and gas industry into other geoscience and geoenergy applications like geothermal energy, gas and energy storage, and water-related problems. Norwegian-speaking readers can get some idea from the following popular-science articles:

• Kunstig intelligens mot flom og styrtregn (in English: Fighting flooding and torrential rain using AI)
• Norske teknologer og forskere har mål om å skaffe vann til hele Somalias befolkning (in English: Finding water for the people of Somalia)
• Hindrer flom med kunnskap fra olje- og gassbransjen (in English: Preventing urban flooding using expertise developed in the O&G industry)

We also do work related to ocean forecasting, microfluidics, battery technology, and digital manufacturing.

Mathematics in Technology: Generic Expertise

The Computational Geosciences group originates from a long tradition of more generic research on numerical methods and simulation tools. Although much of our work is geoscience centric, we keep a strong focus on generic expertise and think of ourselves as applied mathematicians and computer scientists rather than application experts. We believe that this key characteristic is highly important to be able to develop efficient, flexible, and robust simulation technology. In line with this, we have for the past two decades organized the highly popular Geilo Winter Schools in eScience and computational mathematics. The leader of the group is also responsible for SINTEF Digital's prioritized research area, Mathematics in Technology, which functions as an incubator for new research directions and development of new research talents. Examples of activities include work on differentiable simulators, optimization under uncertainty, and physics-informed neural networks, to name a few. 

Emerging Technologies: Quantum Computing

Scientists from the applied mathematics groups in SINTEF have a long tradition for utilizing emerging computer hardware in computational modelling. For instance, starting in the early 2000s, we were among the pioneers in using GPUs for solving systems of conservation and balance laws. We still maintain a high level of expertise in this type of heterogeneous computing. Continuing this tradition, our group is currently leading a new Gemini center on future applications of quantum computing. This impending technology offers a huge potential for societal and business disruption. However, quantum computers work in a fundamentally different way than classical computers, and utilizing them requires a deep understanding of the underlying quantum-physical principles and how these can be used to rethink how algorithms are designed. The purpose of the Gemini center is to mobilize awareness and interest in Norway and start building the necessary expertise so that we are “quantum ready” when the technology becomes a commerical reality for widespread adoption.

Gemini Center on Quantum Computing Making Norway quantum ready!