1D inversion of multicomponent, multifrequency marine CSEM data: Methodology and synthetic studies for resolving thin resistive layers

Kerry Key
2009 Geophysics  
Numerical methods for 1D forward modeling and inversion of marine controlled-source electromagnetic ͑CSEM͒ data are used to examine the inherent resolution of various acquisition configurations to thin resistive layers simulating offshore hydrocarbon reservoirs. Synthetic data studies indicate that jointly inverting frequencies of 0.1 and 1.0 Hz offers better resolution than inverting either frequency alone. Further increasing the bandwidth or density of frequencies does not produce a
more » ... produce a commensurate increase in resolution. An inline horizontal electric dipole is found to provide better resolution than either broadside or vertical electric dipoles. The horizontal electric and magnetic fields for any transmitter orientation have better resolution than vertical fields. Separate inversions of electric and magnetic fields perform equally well at recovering the reservoir, and there is no resolution improvement from jointly inverting both fields. Smooth inversion for a multiple resistive layer model detects the presence of all resistive layers, and shallow thin resistive layers do not impact the ability to image deeper resistive layers. The accuracy of the inverted models is improved substantially by including the boundary depths of resistive layers as a priori structure in the inversion. Including the seawater resistivity profile as fixed structure in the inversion is found to be essential for obtaining optimal resolution of subseafloor resistivity. F9 function of measured field components and transmitter orientation would be helpful for survey-planning tasks such as choosing a receiver system and laying out an acquisition geometry. The purpose of this study is to present numerical algorithms for 1D CSEM forward modeling and inversion, and then to use inversions of synthetic data to examine the inherent resolution of 1D CSEM as a function of these survey parameters. Whereas previous studies have considered some aspects of 1D CSEM inversion for imaging offshore reservoirs ͑Constable and Hoversten et al., 2006; Christensen and Dodds, 2007͒ , there is a need for a systematic investigation of how the resolution depends on the frequency content, transmitter orientation, and recorded field components. This information could be useful to geophysicists planning offshore CSEM field campaigns. METHODOLOGY This section reviews the forward and inverse modeling methodology used for the synthetic inversions. Methods for computing the forward responses of dipoles embedded in multilayered 1D media are well studied ͑e.g., Stoyer, 1977; Chave and Cox, 1982; Ward and Hohmann, 1988 ; and many others͒. The approach used here follows the magnetic vector potential formulation described in Wait ͑1982͒, but generalizes this formulation to allow for multiple layers above the transmitter ͑in addition to multiple layers below͒, and uses exponential forms for the recursions rather than hyperbolic functions. Only isotropic conductivity is considered here, but readers interested in 1D methods for transversely isotropic and generally anisotropic media are referred to Xiong ͑1989͒ and Løseth and Ursin ͑2007͒, respectively. An overview of the 1D formulation is given below, and more specific details are provided in Appendices A-C. F10 Key 1D marine CSEM inversion and resolution studies F12 Key F14 Key F16 Key F18 Key F20 Key
doi:10.1190/1.3058434 fatcat:tdjaxgp5gjfstngy7vya5fxu2i