Multi-trace basis pursuit inversion with spatial regularization

Rui Zhang, Mrinal K. Sen, Sanjay Srinivasan

Research output: Contribution to journalArticlepeer-review

55 Scopus citations


Basis pursuit inversion (BPI) was developed to derive spiky reflection coefficients, which was applied on post-stack seismic data trace-by-trace for subsurface reflectivity. Such a trace-by-trace operation has no control on lateral continuity. To address this, we extend the former BPI to a multi-trace procedure with spatial regularization - a spatial derivative operator - to stabilize the lateral variation. The original objective function of BPI is reformed with a 'Z' shape spatial derivative imposed, which can balance the lateral continuity and vertical resolution. Synthetic tests show improved lateral continuity by spatial regularized BPI in the case of noise contamination. Real data applications demonstrate its effectiveness to enhance the lateral continuity of inverted spiky reflectivity.

Original languageEnglish (US)
Article number035012
JournalJournal of Geophysics and Engineering
Issue number3
StatePublished - Jun 1 2013

All Science Journal Classification (ASJC) codes

  • Geophysics
  • Geology
  • Industrial and Manufacturing Engineering
  • Management, Monitoring, Policy and Law


Dive into the research topics of 'Multi-trace basis pursuit inversion with spatial regularization'. Together they form a unique fingerprint.

Cite this