Exploration Geophysics - Volume 49, Issue 2, 2018

This paper presents the application of a novel trans-dimensional sampling approach to a time domain airborne electromagnetic (AEM) inverse problem to solve for plausible conductivities of the subsurface. Geophysical inverse field problems, such as time domain AEM, are well known to have a large degree of non-uniqueness. Common least-squares optimisation approaches fail to take this into account and provide a single solution with linearised estimates of uncertainty that can result in overly optimistic appraisal of the conductivity of the subsurface. In this new non-linear approach, the spatial complexity of a 2D profile is controlled directly by the data. By examining an ensemble of proposed conductivity profiles it accommodates non-uniqueness and provides more robust estimates of uncertainties.

We apply a novel trans-dimensional Bayesian approach using a wavelet parameterisation to airborne electromagnetic (AEM) inversions using data from the Broken Hill region. This approach allows exploration of a range of plausible subsurface conductivity models and provides more robust uncertainty estimates while accounting for potential non-uniqueness.

In 2012, the National Iranian Oil Company conducted an electromagnetic survey in the Gachsaran oil field. The Bahr’s skew and Mohr diagrams were used to perform the dimensionality analysis. Magnetotelluric data were modelled using the smoothness-constrained least-squares method. The resulting model revealed the main anticline and overthrust zone in the region.

The magnetotelluric method is most often considered in hydrocarbon exploration in cases which are difficult for seismic imaging. In 2012, the National Iranian Oil Co. (NIOC) conducted an electromagnetic survey which included magnetotelluric and time domain electromagnetic (TEM) methods in the Gachsaran oil field in the south-west of Iran to delineate the reservoir formation of the region. Magnetotelluric data were collected at 5215 sites and a regular site spacing of 200 m was utilised. We chose a 16.5 km profile perpendicular to the main geological strike direction in the study area. The Bahr’s skew and Mohr diagrams were used to perform the dimensionality analysis of the magnetotelluric (MT) data and indicated that the subsurface structures are one dimensional or two dimensional at shallow depths, whereas they are mainly three dimensional at lower depths. The phase tensor showed that the dominant geoelectrical strike in the survey area is in the NW–SE direction. Two dimensional inversion was utilised to acquire a realistic resistivity model that was compromise between the spatial smoothness of the inversion model and the MT data fit. Apparent resistivity and phase data were modelled using the smoothness-constrained least-squares method. Models obtained of the TM, TE and TM+TE mode data were examined to have the best possible interpretation. The resulting 2D model revealed the main anticline and overthrust zone in the region. The near surface layer in the model which has a low resistivity, was identified as the cover rock of the region. The formation of the top of the reservoir in the region is estimated to be located at the depth of 1400–1900 m below sea level. The resistivity model is in good correlation with the geological features and the results of well drilling.

Adequate estimation of shear-wave (VS) and P-wave (VP) velocity profiles is one of the significant objectives in the course of seismic surveys; however, the main problem in obtaining VS and VP is the non-uniqueness of surface waves and refracted seismic inversion results. Moreover, the hidden-layer problem often exists in the case of the refraction seismic method. The main purpose of this study is to cope with the above problems and reconstruct subsurface structures by joint inversion of Rayleigh wave dispersion curve and refraction traveltimes. The proposed joint inversion is based on a multi-objective particle swarm optimisation (MOPSO) strategy as a new tool for joint inversion of seismic datasets. The Pareto front concept was applied in the proposed joint inversion scheme. Using the Pareto front, the presented inversion algorithm provided a useful tool to evaluate the results (i.e. the number of layers, thicknesses or Poisson ratio values for the estimated models). The proposed algorithm was tested on two synthetic datasets and also on an experimental dataset. Furthermore, the joint inversion results were compared with the results of individual datasets inverted using PSO inversion algorithm. To verify the applicability of the proposed method, it was applied at a sample site located in Tabriz city, north-western Iran. For a real dataset, the refraction microtremor (ReMi) was used to obtain Rayleigh wave dispersion curves. Moreover, sourced from a vertical-incident seismic source (sledgehammer), seismic refraction data were recorded via vertical component geophones. The results showed that the proposed joint inversion technique can considerably reduce uncertainties of the inverted models.

This study introduces the multi-objective particle swarm optimisation (MOPSO) strategy as a new tool for joint inversion of Rayleigh wave dispersion curve and refraction traveltimes. The proposed algorithm was tested on two synthetic datasets and an experimental dataset. The results showed that the applied joint inversion technique can considerably reduce uncertainties of the inverted models.

To determine the seismic site conditions and microzonation of Chuncheon, Wonju, and Gangneung in Korea, shear-wave velocities were derived at 313 sites by the ESPAC method. The proxy-based Vs30 indicated that these cities were mainly categorised into NEHRP classes B, C, and D, with a minor proportion of A.

To determine the seismic site conditions and microzonation of Chuncheon, Wonju and Gangneung cities in the Gangwon Province, Korea, the dispersion curves of Rayleigh waves were derived at 313 sites by the extended spatial autocorrelation (ESPAC) method. Using the shear-wave velocities (Vs) determined from dispersion curves, average depth to the bedrock (Db) and Vs at the top of the bedrock (Vsb), the overburden layer (Vso) and the top 30 m depth layer (Vs30) were determined. The resonance frequencies (fr) were then computed using both Db and Vso. The estimated averages of the three cities were 13 ± 7 m for Db, 472 ± 109 m/s for Vsb, 248 ± 44 m/s for Vso, 411 ± 157 m/s for Vs30 and 5.8 ± 2.8 Hz for fr. Microzonation maps based on the proxy-based Vs30 indicated that the three cities were mainly categorised into National Earthquake Hazards Reduction Program (NEHRP) classes B, C and D, with a minor proportion of A. Although no area was estimated to be in class E using the proxy-based Vs30, the Vs30 values derived from the recorded Rayleigh waves at 13 sites in Gangneung were less than 180 m/s. This indicates a greater vulnerability to seismic amplification during large earthquakes in this city, which had the smallest Vso, Vs30 and fr, and the greatest Db of the three cities. Microzonation maps, together with information for fr, can be effectively used for seismic risk assessments, urban planning, and disaster management.

The finite difference (FD) method exhibits great superiority over other numerical methods due to its easy implementation and small computational requirement. We propose an effective FD method, characterised by implicit spatial and high-order temporal schemes, to reduce both the temporal and spatial dispersions simultaneously. For the temporal derivative, apart from the conventional second-order FD approximation, a special rhombus FD scheme is included to reach high-order accuracy in time. Compared with the Lax-Wendroff FD scheme, this scheme can achieve nearly the same temporal accuracy but requires less floating-point operation times and thus less computational cost when the same operator length is adopted. For the spatial derivatives, we adopt the implicit FD scheme to improve the spatial accuracy. Apart from the existing Taylor series expansion-based FD coefficients, we derive the least square optimisation based implicit spatial FD coefficients. Dispersion analysis and modelling examples demonstrate that, our proposed method can effectively decrease both the temporal and spatial dispersions, thus can provide more accurate wavefields.

We propose a new finite difference (FD) scheme that is implicit in space (adopting the Taylor-expansion-based or least-square based FD coefficients) and has high-order accuracy in time (adopting a combination of rhombus and cross stencils). Dispersion analysis, stability analysis and modelling examples validate the superiority of our method.

For reverse-time migration, high-resolution imaging mainly depends on the accuracy of the velocity model and the imaging condition. In practice, however, the small-scale components of the velocity model cannot be estimated by tomographical methods; therefore, the wavefields are not accurately reconstructed from the background velocity, and the imaging process will generate artefacts. Some of the noise is due to cross-correlation of unrelated seismic events. Interferometric imaging condition suppresses imaging noise very effectively, especially the unknown random disturbance of the small-scale part. The conventional interferometric imaging condition is extended in this study to obtain a new imaging condition based on the pseudo-Wigner distribution function (WDF). Numerical examples show that the modified interferometric imaging condition improves imaging precision.

The artefacts caused by random fluctuations in the background velocity model will contaminate the final image, which makes the work of interpretation difficult to implement in reverse-time migration. The imaging condition proposed in this paper, which exploits the pseudo-Wigner distribution function, can reduce this noise, and also help improve the continuity of complex layers.

To alleviate the effect of phase in waveform classification, the adaptive phase k-means is introduced for unsupervised seismic facies analysis. This method improves the traditional k-means algorithm by using an adaptive phase distance for waveform similarity measure, and is thus robust to phase variations caused by horizon interpretation.

Waveform classification is a powerful technique for seismic facies analysis that describes the heterogeneity and compartments within a reservoir. Horizon interpretation is a critical step in waveform classification. However, the horizon often produces inconsistent waveform phase, and thus results in an unsatisfied classification. To alleviate this problem, an adaptive phase waveform classification method called the adaptive phase k-means is introduced in this paper. Our method improves the traditional k-means algorithm using an adaptive phase distance for waveform similarity measure. The proposed distance is a measure with variable phases as it moves from sample to sample along the traces. Model traces are also updated with the best phase interference in the iterative process. Therefore, our method is robust to phase variations caused by the interpretation horizon. We tested the effectiveness of our algorithm by applying it to synthetic and real data. The satisfactory results reveal that the proposed method tolerates certain waveform phase variation and is a good tool for seismic facies analysis.

A new Q estimation approach, called improved frequency weighted exponential (IFWE), is presented by combining the advantage of the FWE method and the power spectrum. Tests of synthetic and field data show that the IFWE is more robust and the bandwidth selection for the IFWE is more tolerant than the FWE.

Seismic waves propagating in the subsurface suffer from attenuation, which can be represented by the quality factor Q. Knowledge of Q plays a vital role in hydrocarbon exploration. Many methods to measure Q have been proposed, among which the central frequency shift (CFS) and the peak frequency shift (PFS) are commonly used. However, both methods are under the assumption of a particular shape for amplitude spectra, which will cause systematic error in Q estimation. Recently a new method to estimate Q has been proposed to overcome this disadvantage by using frequency weighted exponential (FWE) function to fit amplitude spectra of different shapes. In the FWE method, a key procedure is to calculate the central frequency and variance of the amplitude spectrum. However, the amplitude spectrum is susceptible to noise, whereas the power spectrum is less sensitive to random noise and has better anti-noise performance. To enhance the robustness of the FWE method, we propose a novel hybrid method by combining the advantage of the FWE method and the power spectrum, which is called the improved FWE method (IFWE). The basic idea is to consider the attenuation of the power spectrum instead of the amplitude spectrum and to use a modified FWE function to fit power spectra, according to which we derive a new Q estimation formula. Tests of noisy synthetic data show that the IFWE are more robust than the FWE. Moreover, the frequency bandwidth selection in the IFWE can be more flexible than that in the FWE. The application to field vertical seismic profile data and surface seismic data further demonstrates its validity.

We have conducted numerical and field experiments to investigate the applicability of electrode configurations and line layouts commonly used for 2D resistivity surveys to 3D inversion. We propose that parallel lines are useful to highlight areas of particular interest where further detailed work with an intersecting line could be carried out.

We have conducted numerical and field experiments to investigate the applicability of electrode configurations and line layouts commonly used for two-dimensional (2D) resistivity surveys to 3D inversion. We examined three kinds of electrode configurations and two types of line arrangements, for 16 resistivity models of a conductive body in a homogeneous half-space. The results of the numerical experiment revealed that the parallel-line arrangement was effective in identifying the approximate location of the conductive body. The orthogonal-line arrangement was optimal for identifying a target body near the line intersection. As a result, we propose that parallel lines are useful to highlight areas of particular interest where further detailed work with an intersecting line could be carried out. In the field experiment, 2D resistivity data were measured on a loam layer with a backfilled pit. The reconstructed resistivity image derived from parallel-line data showed a low-resistivity portion near the backfilled pit. When an orthogonal line was added to the parallel lines, the newly estimated location of the backfilled pit coincided well with the actual location. In a further field application, we collected several 2D resistivity datasets in the Nojima Fault area in Awaji Island. The 3D inversion of these datasets provided a resistivity distribution corresponding to the geological structure. In particular, the Nojima Fault was imaged as the western boundary of a low-resistivity belt, from only two orthogonal lines.