Large-Scale 3D Anisotropic Inversion of Towed Streamer EM Data Using the Gauss-Newton Method

Abstract:
Presenting a newly developed 3D inversion code for Towed Streamer EM data utilising the Gauss-Newton method. The code is designed and implemented on a standard cluster of nodes to handle large-scale inversions of the order 25 million unknowns. The corresponding Jacobian matrix of about 2Tb is split over the subscribed computational nodes and processes in order to reduce the memory consumption on each node as well as to reduce the computational time. This enables the use of the full Gauss-Newton method on this amount of data. The forward modeling part is based on a parallelised integral equation formulation of the electric field. The forward and inversion grids cover the entire survey area including bathymetry and down to 5km below sea surface. The performance is demonstrated with an unconstrained anisotropic 3D inversion on Towed Steamer EM data over a survey area covering a total of 5206 sq. km, acquired in the Barents Sea in 2014. The survey consisted of 38 parallel acquisition lines with a spacing of 1.25km and an average length of 115km. The number of data observations was 505,820 with a total of 6 frequencies in the range from 0.2 to 2.2Hz, 7,226 shot points and 16 receiver offsets. The inversion and consisted of 13 million cells and extended from the water column down to 5km below the sea surface. The inversion ran 14 Gauss-Newton iterations and converged at a misfit of 37% at a geologically feasible anisotropic resistivity model.

Biography:

Johan Mattsson is currently head of the Towed Streamer EM R&D in Petroleum Geo-Services (PGS). He earned a PhD in engineering mechanics in 1996 from Chalmers University of Technology, Sweden. Dr. Mattsson has primarily worked with forward and inverse modeling in the areas of elastodynamics, ocean acoustics and marine electromagnetics.