A Bayesian Method for Sediments Sound Speed Structure Inversion with Seabed Multipath Time Delay in Deep Sea

Wang Hao, Rui Duan, Kunde Yang, Yue Guo

Research output: Contribution to journalConference articlepeer-review

Abstract

When the broadband pulse signal of a point source is in contact with the deep-sea unconsolidated seabed, it often obtains two kinds of bottom signals: seabed interface reflection and sediment refraction. Normally, the refraction arrives later than the reflection. According to the ray theory, the travel time difference between them is closely related to the sediment sound speed structure. Therefore, this paper proposes a Bayesian method using travel time difference data for the sediment thickness and compression sound speed inversion. To improve the efficiency, a simple forward model of travel time difference based on a sediment-basement two-layer seabed is introduced and the ensemble Markov chain Monte Carlo sampling with fewer hand-tuning is used to estimate the posterior distribution. The method is demonstrated in the deep-sea experiment in the South China Sea in 2018, and the results indicate that the sediment sound speed structure of this area is well estimated.

Original languageEnglish
JournalOceans Conference Record (IEEE)
DOIs
StatePublished - 2022
EventOCEANS 2022 - Chennai - Chennai, India
Duration: 21 Feb 202224 Feb 2022

Keywords

  • Bayesian inference
  • deep sea
  • inversion
  • seabed multipath

Fingerprint

Dive into the research topics of 'A Bayesian Method for Sediments Sound Speed Structure Inversion with Seabed Multipath Time Delay in Deep Sea'. Together they form a unique fingerprint.

Cite this