A cone penetration test (CPT) is the most common geotechnical testing method used to estimate in situ the strength properties of soil. Although CPT provides valuable information, this information is restricted to the location of the measurement. We propose a new concept to integrate shallow S‐wave reflection seismic data with CPT data in order to obtain laterally continuous subsoil information. In this vein, a valid quantitative means to relate seismic reflections to CPT data is a primary requirement. The approach proposed here is based on the characterization of the scaling behavior of the local fine‐scale S‐wave velocity information extracted from the seismic reflection data and the same behavior of the CPT cone resistance. The local velocity contrast information is extracted by linearized Zoeppritz inversion of the amplitude‐preserved prestack reflection data. We have formulated a multiscale analysis approach employing the continuous wavelet transform in order to quantitatively characterize the nature of change at an interface of the local S‐wave velocity contrast and the CPT cone resistance and to illuminate any relation between these two. The multiscale analysis estimates the singularity parameter α, which indicates the nature of the interfacial change. The application of our method to the field data has uncovered a striking relation between the nature of variation of the local S‐wave velocity contrast and that of CPT cone resistance; otherwise, such a relation was not visible. Detailed analyses of two extensive field datasets have shown that the lateral fine‐scale variation of soil strength, as seen by CPT cone resistance, has a close resemblance with the variation of the local S‐wave velocity function as seen by angle‐dependent reflection measurements. This leads to a unique possibility to integrate two very different in‐situ measurements—reflection seismic and CPT—providing laterally continuous detailed information of the soil layer boundaries.

We present a new method for determining the azimuthal variation of ambient seismic noise sources, that combines the computational speed and simplicity of traditional approaches with the rigour of waveform-inversion-based approaches to noise-source estimation. This method is based on a previously developed theoretical framework of sensitivity kernels for cross-correlation amplitudes. It performs a tomographic inversion for ambient noise sources on the Earth's surface and is suitable for small (local) scale studies. We apply the method to passive seismic data acquired in an exploration context, and account for azimuth-dependent uncertainties in observed cross-correlation amplitudes. Our inversion results correlate well with the azimuthal distribution of noise sources suggested by signal-to-noise ratio analysis of noise cross-correlation functions.

We present a new method for determining the azimuthal variation of ambient noise sources, which combines the computational speed and simplicity of traditional approaches with the rigor of waveform-inversion-based approaches to noise source estimation. This method is based on a previously developed theoretical framework of sensitivity kernels for cross-correlation amplitudes. It performs a tomographic inversion for ambient noise sources on the Earth's surface and is suitable for small-(local-) scale studies. We apply the method to passive seismic data acquired in an exploration context and account for azimuth-dependent uncertainties in observed cross-correlation amplitudes. Our inversion results correlate well with the azimuthal distribution of noise sources suggested by signal-to-noise ratio analysis of noise cross-correlation functions.

Crosswell traveltime tomography is a common technique in the oil industry for determining the velocity function in the plane between two boreholes. However, the method suffers from the well‐known problem that the lateral resolution is far less than the vertical resolution because of the unfavorable illumination conditions for survey geometries comprising vertical wells. Consequently, it is very difficult to image sudden lateral changes in the velocity function accurately using this technique. We propose a method for determining such changes, which severely constrains the solution space by inverting for the position of a lateral velocity contrast only. The velocity model on each side of the contrast is derived from the well logs. The potential of the method is first demonstrated in two synthetic examples in which its properties are compared with those of an unconstrained 2-D tomographic inversion. As expected, the constrained method has much better convergence properties than the unconstrained one in these examples. We also compare the two methods in a real setting, in which a sudden lateral velocity change, associated with a geological transition, is expected. It turns out that a lateral contrast can be imaged with both methods in this real data example. However, the image obtained with the new method better explains the observed facts.

Given a 3D heterogeneous velocity model with a few million voxels, fast generation of accurate seismic responses at specified receiver positions from known microseismic event locations is a well-known challenge in geophysics, since it typically involves numerical solution of the computationally expensive elastic wave equation. Thousands of such forward simulations are often a routine requirement for parameter estimation of microseimsic events via a suitable source inversion process. Parameter estimation based on forward modelling is often advantageous over a direct regression-based inversion approach when there are unknown number of parameters to be estimated and the seismic data has complicated noise characteristics which may not always allow a stable and unique solution in a direct inversion process. In this paper, starting from Graphics Processing Unit (GPU) based synthetic simulations of a few thousand forward seismic shots due to microseismic events via pseudo-spectral solution of elastic wave equation, we develop a step-by-step process to generate a surrogate regression modelling framework, using machine learning techniques that can produce accurate seismograms at specified receiver locations. The trained surrogate models can then be used as a high-speed metamodel/emulator or proxy for the original full elastic wave propagator to generate seismic responses for other microseismic event locations also. The accuracies of the surrogate models have been evaluatedGeophysical Journal International 2 using two independent sets of training and testing Latin hypercube (LH) quasi-random samples, drawn from a heterogeneous marine velocity model. The predicted seismograms have been used thereafter to calculate batch likelihood functions, with specified noise characteristics. Finally, the trained models on 23 receivers placed at the sea-bed in a marine velocity model are used to determine the maximum likelihood estimate (MLE) of the event locations which can in future be used in a Bayesian analysis for microseismic event detection.