Adjoint-state traveltime tomography: A new modality of seismic imaging

  • Venue:

    Online

  • Date:

    14.06.2022

  • Speaker:

    Ping Tong

    Nanyang Technological University, Singapore

     

  • Time:

    9:30 am

  • Source:

    Ping Tong is an assistant professor holding a joint position with the Division of Mathematical Sciences and Asian School of the Environment at Nanyang Technological University (NTU) Singapore. He is also a principal investigator at Earth Observatory of Singapore. Ping got his Ph.D. in computational mathematics from Tsinghua University in 2012. Before joining NTU in September 2016, he was a postdoctoral fellow at University of Toronto (2012-2014) and Stanford University (2014-2016). His research is primarily focused on seismic imaging, numerical modelling and inverse problems, recently with particular interests in seismic imaging of Southeast Asia and near-surface imaging techniques.

Abstract

Full waveform inversion has a theoretical advantage over traveltime tomography in resolving velocity variations with dimensions smaller than the dominant wavelength. In reality, this may not be (always) true, mainly due to the lack of an accurate initial model (including material properties and source mechanisms) and the high demand for computational resources. To use only reliable data in seismic inversions, we step back to common-source double difference traveltime data via cross-correlation measurement; however, it is still computationally prohibitive to model high-frequency data (>1 Hz in a regional scale), limiting the resolution of seismic images. We further simplify the mathematical-physical model for seismic wave modelling by using isotropic eikonal equation and anisotropic eikonal equations to model seismic wave propagation in various media. The associated inverse problems are solved by the efficient adjoint method. This gives a new modality of seismic imaging, named as adjoint-state traveltime tomography. In this talk, I will show how absolute traveltime data and differential traveltime data are used to reliably constrain velocity heterogeneity, seismic anisotropy and subsurface topography with the adjoint-state traveltime tomography method. The advantages of adjoint-state traveltime tomography over conventional ray-based traveltime tomography techniques and their differences will also be discussed. Adjoint-state traveltime tomography has the potential to become a routine seismic tomography technique.