MIT Earth Resources Laboratory presents Prof. Rongrong Wang of MIchigan State University on "Full waveform inversion with unknown acoustic sources."

"Full Waveform Inversion (FWI) has been widely used for seismic velocity inversion under active seismic data. The usage of FWI under passive seismic data is much more challenging, since aside from the velocity, the source location and activation time are also unknown.

In this talk, we consider a joint recovery problem of the acoustic source and the medium velocity. As this problem is conjectured to be ill-posed in the most general setting, we focus on two practical special cases and propose their corresponding reconstruction algorithms. The first case considers recovering a sparse source under unknown velocity and partial data, the second case considers recovering the velocity model under general sources and full boundary data. Numerical results on synthetic data are used to demonstrate the efficacy of the proposed method. This is joint work with Shuyang Qin and Yang Yang."

Dr. Rongrong Wang received her Ph.D in Applied Mathematics from the University of Maryland College Park. From 2013 to 2017, she was a postdoc at the University of British Columbia's Department of Mathematics and Department of Earth, Ocean, and Atmospheric Sciences. In 2017, Rongrong joined the Department of Computational Mathematics, Science and Engineering and the Department of Mathematics at Michigan State University. Her current research interests include applied and computational harmonic analysis, inverse problems and machine learning.