1,721,107 research outputs found
High-resolution coherency functionals for improving the velocity analysis of ground-penetrating radar data
Full text linkWe aim at verifying whether the use of high-resolution coherency functionals could improve the signal-to-noise ratio (S/N) of Ground-Penetrating Radar data by introducing a variable and precisely picked velocity field in the migration process. After carrying out tests on synthetic data to schematically simulate the problem, assessing the types of functionals most suitable for GPR data analysis, we estimated a varying velocity field relative to a real dataset. This dataset was acquired in an archaeological area where an excavation after a GPR survey made it possible to define the position, type, and composition of the detected targets. Two functionals, the Complex Matched Coherency Measure and the Complex Matched Analysis, turned out to be effective in computing coherency maps characterized by high-resolution and strong noise rejection, where velocity picking can be done with high precision. By using the 2D velocity field thus obtained, migration algorithms performed better than in the case of constant or 1D velocity field, with satisfactory collapsing of the diffracted events and moving of the reflected energy in the correct position. The varying velocity field was estimated on different lines and used to migrate all the GPR profiles composing the survey covering the entire archaeological area. The time slices built with the migrated profiles resulted in a higher S/N than those obtained from non-migrated or migrated at constant velocity GPR profiles. The improvements are inherent to the resolution, continuity, and energy content of linear reflective areas. On the basis of our experience, we can state that the use of high-resolution coherency functionals leads to migrated GPR profiles with a high-grade of hyperbolas focusing. These profiles favor better imaging of the targets of interest, thereby allowing for a more reliable interpretation
Velocity model estimation by means of Full Waveform Inversion of transmitted waves: An example from a seismic profile in the geothermal areas of Southern Tuscany, Italy
No full textWe propose an FWI strategy that makes use of transmitted waves as input data and utilizes both global and local optimization methods to estimate the P-wave velocity model of the subsurface. We envisage that our approach may be applicable to difficult seismic land data, like those from geothermal areas characterised by complex geological structures. As a test case, we considered the CROP/18A seismic reflection profile that crosses the geothermal field of Larderello (southern Tuscany, Italy). The aim is to estimate the P-wave velocity model down to a few kilometres depth below the surface that could be used as complementary information to the standard seismic reflection image which, in this case, does not show interpretable reflections in a range of depths accessible to industrial drillings. One innovative aspect of the inversion we propose with respect to conventional FWI approaches is its independence of a starting model that, ideally, should reproduce the true long wavelength velocity structure of the subsurface and that may be rather difficult to obtain in case of low quality data and complex geology. We lessen the dependence on knowledge of a suitable starting model by performing a sequence of two inversions. First, we employ a genetic-algorithm (GA) based inversion, a global optimisation method that does not require any specific starting model, resulting in a long wavelength, low-resolution velocity model. This model then becomes the starting model for a second FWI, driven by a local optimisation algorithm, aimed at bringing in the fine details of the subsurface velocity structure. The reliability of the final model is checked by comparing observed and predicted waves for many common shot gathers along the seismic line and through the matching between the velocities measured by check shots in two nearby wells and the FWI velocities in the same locations. Many details of the velocity field, likely related to metamorphic and igneous formations, become apparent and may complement the interpretation of the standard reflection image. From these results, it appears that the use of transmitted waves and of the FWI approach discussed here may effectively improve the information for geophysical interpretation of challenging seismic land data, such as those that characterise many areas of geothermal exploration
A probabilistic full waveform inversion of surface waves
Full text linkOver the past decades, surface wave methods have been routinely employed to retrieve the physical characteristics of the first tens of meters of the subsurface, particularly the shear wave velocity profiles. Traditional methods rely on the application of the multichannel analysis of surface waves to invert the fundamental and higher modes of Rayleigh waves. However, the limitations affecting this approach, such as the 1D model assumption and the high degree of subjectivity when extracting the dispersion curve, motivate us to apply the elastic full-waveform inversion, which, despite its higher computational cost, enables leveraging the complete information embedded in the recorded seismograms. Standard approaches solve the full-waveform inversion using gradient-based algorithms minimizing an error function, commonly measuring the misfit between observed and predicted waveforms. However, these deterministic approaches lack proper uncertainty quantification and are susceptible to get trapped in some local minima of the error function. An alternative lies in a probabilistic framework, but, in this case, we need to deal with the huge computational effort characterizing the Bayesian approach when applied to non-linear problems associated with expensive forward modelling and large model spaces. In this work, we present a gradient-based Markov chain Monte Carlo full-waveform inversion where we accelerate the sampling of the posterior distribution by compressing data and model spaces through the discrete cosine transform. Additionally, a proposal is defined as a local, Gaussian approximation of the target density, constructed using the local Hessian and gradient information of the log posterior. We first validate our method through a synthetic test where the velocity model features lateral and vertical velocity variations. Then we invert a real dataset from the InterPACIFIC project. The obtained results prove the efficiency of our proposed algorithm, which demonstrates to be robust against cycle-skipping issues and able to provide reasonable uncertainty evaluations with an affordable computational cost
Comparison of object functions for the inversion of seismic data and study on the potentialities of the Wasserstein Metric
No full textA Bayesian approach to elastic Full-Waveform inversion: application to a semi-real near surface model
No full textEstablished methods for the inversion of surface waves, such as the inversion of dispersion curves, are limited to 1D subsurface structures. In contrast, full-waveform inversion has the potential to reconstruct high-resolution subsurface models even for 2D subsurface models. However, full-waveform inversion in near-surface seismic applications is very challenging due to the high nonlinearity of the optimization problem and the very high computational cost. Traditional methods solve the full-waveform inversion making use of gradient-based algorithms that minimize an error function, which commonly measures the distance between observed and predicted waveforms. This deterministic approach only provides a “best-fitting” model and cannot account for the uncertainties affecting the predicted solution. On the other hand, casting this inverse problem into a probabilistic framework must deal with the formidable computational effort of the Bayesian approach. We present a gradient-based Markov Chain Monte Carlo full-waveform inversion in which the posterior sampling is accelerated by compressing the data and model spaces through the discrete cosine transform, and by also defining a proposal that is a local, Gaussian approximation of the target posterior probability density. We demonstrate the applicability of this approach by performing a multiparameter inversion test on a near surface semi-real velocity model
A computationally efficient probabilistic full waveform inversion: application to the BP model
No full textA computationally efficient Bayesian approach to full-waveform inversion
No full textConventional methods solve the full-waveform inversion making use of gradient-based algorithms to minimize an error function, which commonly measure the Euclidean distance between observed and predicted waveforms. This deterministic approach only provides a ‘best-fitting’ model and cannot account for the uncertainties affecting the predicted solution. Local methods are also usually prone to get trapped into local minima of the error function. On the other hand, casting this inverse problem into a probabilistic framework has to deal with the formidable computational effort of the Bayesian approach when applied to non-linear problems with expensive forward evaluations and large model spaces. We present a gradient-based Markov Chain Monte Carlo full-waveform inversion in which the posterior sampling is accelerated by compressing the data and model spaces through the discrete cosine transform, and by also defining a proposal that is a local, Gaussian approximation of the target posterior probability density. This proposal is constructed using the local Hessian and gradient informations of the log posterior, which are made computationally manageable thanks to the compression of the data and model spaces. We demonstrate the applicability of the approach by performing two synthetic inversion tests on portions of the Marmousi and BP acoustic model. In these examples, the forward modelling is performed using Devito, a finite difference domain-specific language that solves the discretized wave equation on a Cartesian grid. For both examples, the results obtained by the implemented method are also validated against those obtained using a classic deterministic approach. Our tests illustrate the efficiency of the proposed probabilistic method, which seems quite robust against cycle-skipping issues and also characterized by a computational cost comparable to that of the local inversion. The outcomes of the proposed probabilistic inversion can also play the role of starting models for a subsequent local inversion step aimed at improving the spatial resolution of the probabilistic result, which was limited by the model compression
FWI of noisy seismic land data acquired for geothermal exploration
No full textWe present an experience of acoustic FWI on a noisy 2D land dataset to the end of providing a velocity model for direct geological interpretation. We first perform a dedicated processing trying to improve the data quality and we select as the input for the inversion direct, refracted and diving waves. In fact, notwithstanding the processing efforts, reflections are nearly absent. Next, we perform a sequence of FWIs. Since we wish to use the estimated velocity for the interpretation of the area, it is necessary that the estimation is not biased by unverified a-priori geological hypotheses. Therefore, to derive a low-resolution velocity model, we apply two runs of genetic-algorithm FWI (GA-FWI) with different data misfit functions based on envelopes and on waveforms. In fact, GA do not start from a given velocity model, thus risking to bias the final outcome, but from an ensemble of models randomly selected within large search ranges. The GA velocities are then used as starting model for a gradient-based FWI which yields an improved model, appropriate for evaluating different geological hypotheses. In two locations, the check-shot velocities of exploratory wells show a good matching with the 1D velocity profiles extracted from the final model
- …
