What is it about?

This research is about improving the way we simulate how seismic waves move through underground rock and fluid layers. Understanding these waves helps scientists learn more about what’s beneath the Earth's surface, which is useful for things like oil exploration and earthquake studies. The problem is that fully 3D simulations (which model waves in all directions) require a lot of computer power and memory. Instead, the researchers use a efficient approach called 2.5D modeling, which works well when underground structures are mostly 2D (like layered rock formations). We developed a new method that makes these simulations both more accurate and faster by: 1) Using a special math technique (Chebyshev differentiation) to better calculate how waves move in space. 2) Applying a Taylor-series recursion to handle wave energy loss over time. 3) Transforming complex mathematical calculations into simpler ones that are easier for computers to handle. 4) Running the simulations in parallel on multiple processors to speed things up. When we tested their method against full 3D simulations, we found that it gave very accurate results while using much less computing power. This makes it a great tool for studying seismic waves in complex underground environments efficiently.

Featured Image

Why is it important?

Understanding how seismic waves move through the Earth is important for many reasons. It helps in oil and gas exploration by identifying underground reservoirs, improves earthquake studies by revealing how waves travel through different rock types, and supports geothermal energy and carbon storage by evaluating underground structures. However, accurately simulating these waves in 3D is very demanding, requiring huge amounts of computer power and memory. If researchers can achieve similar accuracy with a more efficient method, they can save time and resources while still getting valuable insights. This study introduces a new method that balances accuracy and efficiency. Instead of running a full 3D simulation, it uses a 2.5D approach, which simplifies the problem while still capturing important 3D effects. The method is unique because it applies advanced mathematical techniques to improve both precision and speed. It uses Chebyshev differentiation to better calculate how waves move in space, Taylor-series recursion to efficiently handle energy loss over time, and a real-domain transformation to simplify complex calculations. Additionally, the method is designed for parallel computing, allowing it to run much faster on modern computers. The result is a highly efficient and accurate way to model seismic waves in complex underground environments. This makes seismic studies more practical and accessible, benefiting industries and researchers working on energy exploration, earthquake analysis, and underground resource management.

Perspectives

This publication presents an important advancement in seismic wave modeling by balancing accuracy and computational efficiency. Full 3D simulations are often too costly, but this 2.5D approach captures essential 3D effects while reducing computational demands. What stands out is the integration of advanced techniques—Chebyshev differentiation, Taylor-series recursion, and real-domain transformation—which improve precision and speed. The use of parallel computing further enhances efficiency, making large-scale simulations feasible. Beyond its technical strengths, this method has real-world impact, making seismic imaging more accessible for applications like oil exploration, earthquake analysis, and underground resource management. It’s a well-designed and practical contribution to geophysical modeling.

Moosoo Won
Korea Institute of Geoscience and Mineral Resources

Read the Original

This page is a summary of: AN ACCURATE AND EFFICIENT NUMERICAL METHOD FOR 2.5-D TIME-DOMAIN VISCOACOUSTIC AND VISCOELASTIC WAVE MODELING, Geophysics, February 2025, Society of Exploration Geophysicists,
DOI: 10.1190/geo2023-0324.1.
You can read the full text:

Read

Contributors

The following have contributed to this page