Tristan van Leeuwen (Mathematics) was presented with the Clarence Karcher Award by the Society of Exploration Geophysicists in Houston. The award is a recognition for his development of theories and algorithms to process measurement data into images of geological strata for a specific area much more quickly. Van Leeuwen also works on similar problems for other applications, such as MRI scans and X-ray images. His goal is to eventually create images that are sharp enough to lead to entire new scientific insights.
The jury of the Clarence Karcher Award praised him for “his extremely thorough background in mathematics and his in-depth knowledge, understanding and insights for applications in geophysics”. Van Leeuwen had already received a prize for his research two years ago from the Society for Industrial and Applied Mathematics. He is pleased to discover that his research has been recognised and appreciated both in the fields of mathematics and in geophysics.
In his own research, Van Leeuwen looks for problems that other people are facing. “The challenge is to come up with smart math to help solve their problems. The best thing to me is to link that solution back to the real-world situation”, Van Leeuwen explains.
Months of calculations
Gaining an understanding of the sub-surface strata is relevant for all sorts of applications, such as oil and gas exploration, building bridges, and scientific research. This is done by transmitting sound waves into the earth and measuring how they are reflected; the same principle used to see unborn children inside the uterus. The problem is that the measurements provide so much data that the computer needs months to calculate an image of what is underneath.
Thanks to Van Leeuwen’s research, scientists can reduce the time needed to make these calculations by a factor of 10 or 20. His greatest breakthrough was his approach, which requires a simulation of only a portion of the millions of measurement data, while still producing a reliable image. “There is often a lot of overlap in the measurements, which means that some measurements are more important than others. Unfortunately, you can’t know in advance which ones are important. By processing a small portion of the measurements first, and then adding a few new ones bit by bit, you can gradually get an idea of how it looks underground, without having to calculate all of the data”, Van Leeuwen explains.
The problem of how to process an increasing amount of data is one researchers face in every field of science. Measurement equipment is developing rapidly, and are providing ever-more data. To Van Leeuwen, the solution is not to build faster computers or ‘supercomputers’, but to use smarter calculation methods. “My goal is to eventually convert measurement data into the sharpest possible images, in which you can observe things that you can’t see today, and gain completely new scientific insights. Finding the mathematical solutions you need to do that is my own personal Holy Grail.”
About Tristan van Leeuwen
Tristan van Leeuwen studied Computational Science at Utrecht University and earned his PhD in Geophysics at TU Delft. After working as a postdoc at the University of British Columbia in Vancouver in Canada and the Centrum Wiskunde & Informatica in Amsterdam, he returned to Utrecht, where he is now an Associate Professor at the Mathematical Modeling research group. Two years ago, he was presented with the SIAM Activity Group on Geosciences Junior Scientist Prize. SIAM, the Society for Industrial and Applied Mathematics, presents this award every two years to a promising young researcher in the field of mathematics, with an application in the geosciences. >> Read more