Mathematics Colloquium
Yanzhi Zhang
Missouri University of Science and Technology
Title: Nonlocal modeling and computations in seismic data inversion
Date: Friday, January 28, 2022
Place and Time: LOV 101, 3:05-3:55 pm
Abstract.
In this talk, I start with a brief survey of my research areas, and then focus on the topic on nonlocal modeling and computations for seismic data inversion. Seismic waveform inversion is a widely used technique in the oil and gas industries. Existing seismic imaging methods are implemented in the classical framework of elasticity, which mostly ignores the viscous (attenuation) property of the Earth's subsurface. The recently proposed fractional viscoacoustic wave models have proved to be successfully accounting for both seismic attenuation and dispersion in wave propagation. However, their nonlocality introduces considerable challenges for both analysis and computations. In the talk, I introduce our newly developed computational methods for the forward seismic modeling in both homogeneous and heterogeneous media. For homogeneous media, finite difference method will be introduced, and fast algorithms will be discussed for its efficient implementation. While meshfree methods will be the focus for heterogeneous media. Numerical analysis and experiments will be provided to demonstrate the effectiveness of our methods for forward seismic modeling. Nonlocal effects on wave propagation are also discussed.