Linearized inversion of data of multi-offset data for vertically inhomogeneous background

Abstract

One of the most widespread problems in seismology is the necessity to adjust some a priori given medium structure. As a rule, such a priori information is vertically inhomogeneous component of wave propagation velocity. Such theoretical aspects of the problem as its uniqueness estimates of conditional stability are studied rather well. There axe also a variety of algorithms for its numerical solution. Therefore, in this paper the main attention is paid to numerical analysis of resolving ability and information contents of linearized inversion of multi-offset data.*Domaine de Voluceau, The work was supported by RFBR grants No. 96-01-01515 and 97-05-65280; and by the France-Russian Center of Appl. Math, and Inf. grant "Reliable numerical methods of solutions of nonlinear continually correct problems in application to inverse problems of wave propagation theory". Brought to you by | University of Arizona Authenticated Download Date | 5/31/15 6:15 AM * As we shall see from the presentation below, the case of half-space z > 0 does not have serious changes except the calculations which are more tedious. Brought to you by | University of Arizona Authenticated Download Date | 5/31/15 6:15 AM

Deterministic sensitivity analysis for a model for flow in porous media

“…Equations (5) and (6) are now used to compute the i th row of F . Let gỹ ,i denote the i th row of the Jacobian matrix of the measure operator O:…”

“…With this method "first order" uncertainties are computed using the derivatives of the function F that associates the output parameters to the input parameters. The hierarchization of the inputs of the function F according to their influence on the outputs of F is provided by the Singular Value Decomposition (SVD) of the Jacobian matrix F (x) [21,6,7,1]; see also [35] for the history of the SVD and [16] for a detailed description. It is interesting to note that the SVD, the main tool for the deterministic approach described here, is also commonly used in statistical data analysis, see [8,9,25,36].…”

Full‐waveform inversion for macro velocity model reconstruction in look‐ahead offset vertical seismic profile: numerical singular value decomposition‐based analysis

“…Silvestrov and Tcheverda (2011) used SVD to compare different parametrizations in elastic inversion of surface and cross-well data. Cheverda et al (1998) showed that SVD for surface reflection data is strongly related to trend/reflectivity decomposition of the model. They used 1D model assumptions and infinite surface acquisition and found that smooth velocity perturbations correspond to singular values with relatively large indices.…”