ASEG Extended Abstracts - ASEG2013 - 23rd Geophysical Conference, 2013

We present a multi-objective optimization approach to the subsurface geomodel updating problem using stochastic search techniques. This is a new approach to the geomodeling process for which a variety of direct and indirect measurements can simultaneously constrain the geomeodel. Due to the inherent uncertainties and noise in real data measurements, geological and geophysical datasets acquired in the same area may be in conflict with each other and a realistic subsurface model can only be obtained by simultaneously integrating the combined datasets in a reasonable manner. One approach to this problem is to perform joint inversion of multiple geological and/or geophysical datasets, where an optimal model is achieved by optimization of a linear combination of several objective functions measuring the match of the simulated datasets with the observed datasets. In this paper, we consider joint inversion of multiple datasets for geomodel updating, as a Multi- Objective Optimization Problem (MOOP), where separate objective functions for each subset of the observed data are defined. Then, a stochastic optimization technique is employed to find the set of best-compromise model solutions that fit the defined objectives along the Pareto front. We demonstrate that a customized initialization of the algorithm can speed up the convergence and result in a set of improved model solutions. We apply the proposed approach on a 3D reservoir litho-facies model that must honour a set of geological and geophysical attributes (e.g. log data and inverted seismic P- and S-wave impedances).

Towed transient electromagnetic (TEM) survey, coupled with resistivity modelling software is an effective method of detailing small scale groundwater conceptual models and assisting with near surface geological investigations. Practical investigation depth ranges from 1 m to 100 m or more given the restrictions of today’s electronics and practical trailer dimensions.

Towed TEM survey using loops on trailers behind land vehicles or boats may be conducted using various loop configurations. Due to the loop area and separation requirements of loops from each other and from towing vehicles, design of trailers and/or sleds must be tightly integrated with design of loop configurations. Although separated loops (slingram configuration) are good for avoiding mutual inductance problems and may permit exploration to maximum possible depth, they are difficult to tow, especially around corners. Alternative arrangements with overlapping loops or bucking coils, all on a single platform, permit design of more practical platforms. On such platforms, not only must mutual inductance of coils be minimized but practical means of minimization are limited by achievable dimensional accuracy and stability of towed platform designs. Design is further restricted by the need to avoid use of metallic materials in most places and the need to separate and/or de-couple the metal survey vehicle from the loops.

Case studies showing results of survey conducted with various platforms will be presented.

Estimating subsurface temperature and assessing the thermal structure in numerical models requires a vast database of measured values, a detailed geological model and the ability to identify, incorporate and constrain uncertainty in the parameters to provide a reliable and robust result. Sparse datasets with limited results required additional observables to be gathered. Using groundwater bores temperature in the shallow crust can be measured over a widely distributed area and in depth profiles. Calibration of the Sydney Basin thermal model has shown that using shallow groundwater bores strong constraints on parameters can be made, thus reducing overall model uncertainty. Deep measurements are limited therefore shallow groundwater bores are a good data substitute. The largest sources of uncertainty are the parameters governing temperature dependent thermal conductivity of the basement and Permian Coal Measures, as well as the basal temperature condition and unconstrained heterogeneities in the basement rocks. Variance in these parameters may significantly influence the resulting estimate of subsurface temperature. However through calibration the possible variance is limited due to the large number of available calibration points.

The Gippsland Basin is a potential site for CO2 storage which is dependent on the regional top seal in providing a secure subsurface containment. We present here some geological parameters derived with the aid of seismic attributes and inversion from 2D and 3D reflection seismic surveys predominantly along the southern flank of the basin. These parameters are potentially influential on the top seal efficiency and CO2 containment security in the basin.

An important factor in top seal efficiency is the spatial variation of its shale content (Vsh). The empirical relationship between acoustic impedance and shale content was used to estimate Vsh. Composite seismic amplitude and acoustic impedance traces were constructed to establish a tie with the well-derived reference Vsh. Using a multi-attribute regression analysis, a transformation was established from seismic attributes and impedance to Vsh and used to define Vsh pseudotraces. Eight vertical profiles were produced in the Southern flank of the basin and the Vsh data were interpolated to reveal the first order variation in shale content for the top seal.

Equally important in the assessment of containment risk is the distribution and density of faults in top seal. Seismic spectral blueing and attributes were used to increase the resolution of the 2D seismic data and a meta- attribute that sharpens the faults and suppresses non-fault discontinuities was coupled with similarity attributes to ensure a better imaging of low-displacement faults (<20 m). Automated mapping of the faults provided fracture density maps which depict the intensely deformed areas with potentially decreased seal efficiency on the southern flank of the basin.

A method of joint inversion of Magnetotelluric, seismic refraction and seismic reflection (JIMRR) is developed especially for typical hydrocarbon or hard-rock mineral exploration. JIMRR includes two parts: jointed seismic refraction and seismic reflection; and its combination with Magnetotelluric (MT) method. The objective of the research is to enhance spatial resolution of the three model parameters: electrical resistivity, seismic velocity and reflector depth. Since horizontal coordinates of reflector are not treated as model parameters in existing travel time inversion algorithm, seismic forward modelling may loss the true reflection point locations at the side edges of reflector with limited extension. We developed the technology of extensible reflector to overcome this problem. JIMRR is completed by employing the cross-gradient function as constraints which enforces the structural similarity between the resistivity and the seismic velocities, so as to reduce velocity-depth ambiguity. The cross-gradient constraints are incorporated into the solution through least squares and Lagrange multiplier method. This method results in integrated symmetric square linear matrix that is solved by bi-conjugate gradient method (BiCG). Two example synthetic models show that our joint inversion can significantly enhance the spatial resolution of inversion; and also the velocity-depth ambiguity caused by reflection travel time inversion can be notably reduced by constraints from shallow lithologies.

Under high in situ stresses, the excavation of underground openings generally causes the creation of a disturbed (EdZ: Excavation disturbed Zone) and/or damaged (EDZ: Excavation Damaged Zone) zone, resulting from the initiation and growth of cracks and fractures and from the pre-existing stress redistribution. The EdZ or EDZ changes the mechanical and hydromechanical properties which in return, constitute a potential risk for the performance of the geologic and/or engineered structures in the context of waste underground storage. Ultrasonic experiments have been implemented to characterize the EDZ extension around drifts and its evolution in time according to the structural support type (soft or rigid) and the environmental conditions. Those studies consist of two experimental components: (1) the prior auscultation of the floor and sidewalls of the gallery by ultrasonic transmission tomography, (2) the monitoring of the time-dependant evolution of EDZ and the analysis of measurable changes in the propagation of ultrasonic waves in the medium term. A code for computing these continuous in situ velocity measurements into the elastic has been developed. Then, the five dynamic elastic constants for the assumed transverse isotropic character of the rock are derived as a function of time and the distance from the drift wall. Performed a few months after the excavation of the galleries, the tomography shows that ultrasonic velocities are higher in the orthoradial direction (both in the concrete support and rock). This velocity field highlights clearly the damaged zone and induced stress shift.

A shallow high-resolution seismic reflection test survey was conducted in the Maryborough Basin in an area near Gundiah, Queensland. The purpose of the seismic reflection test survey was to evaluate the application of using high resolution seismic data acquisition methods to detect thin coal seams at between 30 m to 150 m depth for open-cut coal exploration.

The seismic test survey tested three seismic sources which included a 50 kg weight-drop from 2 m height, a 7 kg sledgehammer, and a 12 g (blank cartridge) in-hole shotgun. Seismic forward modelling indicated that thin coal seams (1 to 2 m thickness) should provide seismic reflections at shallow depths of 30 m, using 72 channel recording with 2 m geophone intervals. The seismic source tests showed that the 12 g in-hole shotgun produced the strongest seismic signal at the furthest geophone offsets. The seismic recording spread consisted of 96 seismic channels with single spiked 30 Hz geophones spaced at 2 m geophone intervals. Based on the seismic source tests, the 12 g in-hole shotgun was used to record a seismic test line with shot-points recorded from one end of the 96 channel spread through to the other end of the spread.

The recording arrangements provided seismic data to both test data quality with shot to receiver offset, and examine the optimum common depth point (CDP) fold coverage to use for a larger seismic survey.

Based on the seismic data processing results, a larger seismic survey could be conducted using a 72 channel recording spread with 4 m geophone and shot-point intervals and achieve production rates of 400 m per day with a 3-man crew.

We extend our previous work on 2D internal multiple attenuation without subsurface information to a 3D operation. We describe our implementation that involves selecting traces that honour the azimuth of acquisition for constructing multiple contribution gathers and then segmenting the chosen traces in a layer stripping fashion to predict internal multiple model without identifying the multiple-generating interfaces. We demonstrate through synthetic and field data examples that, by including crossline apertures in the prediction process and selecting traces with correct azimuths in the convolution and correlation processes, substantial improvement in image quality can be obtained for those data that exhibit the internal multiple problem.

Adjoint-state methods (ASMs) have proven successful for calculating the gradients of the functionals commonly found in geophysical inverse problems. The 3D ASM image-domain tomography (IDT) formulation of the seismic velocity estimation problem highlights imperfections in migrated image volumes and, using appropriate penalty functions (e.g., differential semblance), forms an objective function that can be minimized using standard optimization approaches. For time-lapse (4D) seismic scenarios, we show that the 3D ASM-IDT approach can be extended to multiple (e.g. baseline and monitor) datasets and offers high-quality estimates of subsurface velocity change. We discuss two different penalty operators that lead to absolute and relative 4D inversion strategies. The absolute approach uses the difference of two independent 3D inversions to estimate a 4D model perturbation (i.e. slowness squared). The relative approach inverts for the model perturbation that optimally matches the monitor image to the baseline image - even if migrated energy is imperfectly focused. Both approaches yield good 4D slowness estimates; however, we assert that the relative approach is more robust given the ubiquitous presence of non-repeatable 4D acquisition noise and imperfect model estimates.

In this paper we present a case history of multi-azimuth 3D PSDM processing. The datasets show strong HTI as well as VTI anisotropy. We show the processing workflow with emphasis on the construction of an imaging velocity model that correctly represents the orthorhombic anisotropy and short-wavelength velocity variations. The PSDM image is improved over earlier processings.

A 3D geological model for King Sound, in the offshore Canning Basin, Western Australia, was produced by interpretation of FALCON® high resolution airborne gravity gradiometry and magnetic data, aided by 2D gravity models, 2D seismic data and well logs. Pseudo- depth slices of the vertical gravity gradient (GDD) and magnetic data and interpreted seismic horizons were used to constrain the vertical distribution of sedimentary sequences, intrasedimentary intrusives and basement rocks. Basement depth was determined from the magnetic data using traditional profile-based automated magnetic depth estimation techniques with well control.

The 3D model indicates an elevated, fault-bounded platform of Archaean to Proterozoic basement in the north. The platform is rimmed by Late Devonian and Early Carboniferous carbonate reefs and carbonate breccias and in the south, contemporaneous siliciclastic submarine fans and turbidite deposits occur in a deep marine environment.

Density values derived from published literature and measured in wells at King Sound were assigned to units in the geological model. A forward model was calculated and compared to the observed GDD data. The assigned density values were then modified, within the expected range for each rock type, using property inversions until a good fit between the modelled and observed data was obtained.

Models derived from potential field data can be beneficial for petroleum exploration in frontier basins worldwide, where only limited well and seismic data are available. The 3D geological model provides a good framework for use in designing future exploration programs in the area and it aids data visualisation and interpretation.

Each layer of rock or sediment has its own velocity, that is, there are different velocities along the subsurface layers of the earth. Moreover, each layer has various values for different types of velocity. Therefore, the suggestion raised was to study the effects of vertical velocity heterogeneity on stacking velocity and depth conversion with different spread lengths, i.e. a small spread with a maximum offset of 2000 m and a large spread with a maximum offset of 4000 m. This study focused on the variation between stacking velocity and average velocity. In addition, the traveltime equation of Taner and Koehler (1969) for two terms and three terms was examined in order to find out which one provided better results.

Understanding the variations between the different types of velocities was crucial to this approach, which was carried out using data from the Tirrawarra-29 well in the Cooper Basin, South Australia. Well log data are used to calculate different types of velocities such as average velocity, root-mean-square velocity (for both short offset and three terms) and stacking velocity.

The results for both the T − X plots and the T² − X² plots for small (2000 m) and large spreads (4000 m) proved that the variation between average velocity and stacking velocity increases with offset. Furthermore, using the traveltime equation for three terms on the residual moveout plots for small and large offsets provided better results than using only two terms.

We show how refraction tomography (also called first arrival travel time tomography) helps to produce more accurate and detailed depth velocity models below a shallow seafloor. We do not use refractions by themselves to build a complete shallow velocity model. In our proposed workflow, refraction tomography complements standard reflection tomography and the priority remains with the reflections to guarantee stability of the solution and to avoid uncertainties associated with refracted or diving waves in complex media.

We use wave equation modelling to calculate synthetic gathers and estimate the travel time mismatch between real and synthetic first arrivals. It leads to a robust workflow which can be easily introduced into production depth-velocity processing.

We show how this joint reflection/refraction velocity inversion works using a real 1000sq.km 3D marine seismic dataset acquired in an area where the water depth varies from 20 m to 1100 m.

Over recent years, seismic methods have emerged as a potential imaging technique for delineation of ore-bodies and for mine planning. The application of surface seismic methods in hard rock environments is however challenging due to various effects such as energy attenuation and scattering. Borehole seismic methods can be used to reduce these effects. The methods offer higher resolution at target depths, thus allowing better delineation and understanding of reflections from ore deposits.

We present a synthetic study to understand the ability of the cross-hole seismic method to delineate ore bodies. Three variations of a simple scenario typical of nickel deposits found in the Yilgarn Craton were considered. Of the three models, two consist of volcanics overlying a granite body and a thin sulphide mineralized zone along the contact but at different locations relative to the source and receiver boreholes. The third consists of only the rock units with no sulphide mineralized zone along the contact. Synthetic shot records were produced and wavefield separated. Up-going wavefields were then used to create depth migrated images. The resulting images correlate well with the volcanic-granite contact and massive sulphide lens, showing the potential of using the cross-hole seismic method to delineate ore bodies.

Airborne TEMPEST electromagnetic (EM) and magnetic data was acquired over the Bulman project area in the McArthur Basin to identify the geological environment of stratabound carbonate-hosted Pb-Zn mineralisation. The Late Palaeoproterozoic to Early Mesoproterozoic sedimentary fill of the McArthur Basin in the Bulman area is intruded by Early Mesoproterozoic dolerite dykes and sills. The location of the dykes and sills was mapped using magnetic data. The depth of the intrusives was estimated by mapping resistive basement on Conductivity Depth Transforms (CDTs) generated from the TEMPEST B-field Z-component data. Additionally, a 3D conductivity voxel model was constructed from the CDTs to show the conductivity distribution in rocks.

There is no marked conductivity contrast between intrusives and sediments making up the resistive basement. Instead, the resistive basement horizon is interpreted to represent the base of the Cenozoic unconsolidated deposits or part of the Proterozoic sedimentary rocks. In places, the base of a sub-surface conductive zone is interpreted to be the top of the intrusives. The thickness of sediments above the resistive basement is variable, reaching up to 170 metres in the central - eastern part of the study area.

3D geological model was constructed to assist in visualising the distribution of interpreted geological units and the tectonic pattern

This study was carried out in Harat Rahat (south of Almadinah Almonwarah) using seismic reflection and resistivity methods. The main objectives of this study are to determine the extent of the basaltic layer and to define the subsurface faults and fractures that could affect and control the groundwater movement in the study area. A 2D seismic profile was acquired and the result shows that the subsurface in the study area has a major fault. We obtained a well match when the seismic result was compared with drilled wells. As a complementary tool, the resistivity method was applied in order to detect the groundwater level. The results of the resistivity method showed that six distinct layers have been identified. The interpretation of these six layers show that the first three layers, the fourth layer, the fifth layer and the bottom of the section indicated various subsurface structures and lithologies; various basaltic layers, fractured basalt, weathered basement and fresh basaltic layers, respectively. It is obvious that the eventual success of geophysical surveys depends on the combination with other subsurface data sources in order to produce accurate maps.

This study was carried out in Wadi Nisah to the south- west of Riyadh, Saudi Arabia, at latitude of 24° 14′ 28 N and longitude of 46° 29′ 59 E. The main purpose of this study is to investigate the depth of the shallow groundwater aquifer and the geological structures that could affect the ground water bearing layer in the area. For this purpose, the high-resolution seismic reflection technique was performed along Six 2D seismic lines and geophone spacing of one meter. The constructed seismic sections revealed that the depth of the water bearing formation lies in the range of 100 m to about 240 m and normal faulting affect the surface of this formation. The interpreted water-bearing formation is correlated with Biyadh Formation as confirmed by the drilled near water wells. The output of this research proved that the high resolution seismic reflection technique can be an effective method for determining water bearing layer depth in such arid area.

The purpose of this study is to evaluate the efficacy of using multichannel analysis of surface wave (MASW) to detect the near-surface cavities. The methods used in this study include interpreting dispersion curves and amplitude mapping of the multichannel analysis of surface wave technique and interpreting the delay in first arrivals of compressional waves. To test these methods, a seismic survey was conducted above a known near surface cavity in Al-Suman Area, Saudi Arabia. The cause of the cavity is carbonization in the area; there are many cavities similar to this one. The seismic data were collected using a seismograph system with 48 vertical geophones. Both techniques show a tangible result for detecting the cavity. The 2D section of shear wave velocity, which was obtained by inverting the dispersion curves from the MASW technique, leads us to determine the shape of the cavity, as described by a low-velocity zone. Frequency against relative offset is plotted and shows a significant frequency drop in the presence of the cavity, which also provides an indication to the presence of cavity underneath. This interpretation is matched by the interpretation of observed delays in first arrivals of compressional waves. The integration of both P-wave seismic refraction and MASW gives confidence in the result and matches observations of the existing cavity closely.

The 3DVSP technique becomes more popular with the emergence of multilevel 3 component borehole tools.

The value of the information derived from VSP is not always well understood.

In this paper we will present a case history of a joint surface and borehole seismic acquisition, with the goal to de-risk a shaft sinking location on a platinum mine.

TD was about 650 m and the VSP tool was 110 m long. On surface a dense grid of receivers and vibrator source points were laid out in a 1.2 km radius circle centred on the well head. A12 level 3C VSP digital tool was lowered in the borehole, in three successive depth positions.

The surface 3D cube was processed and interpreted independently from the 3D VSP data. On a near target reflector, the surface data structural interpretation showed mainly a clear E/W fault, and additional sub-seismic lineaments of differing azimuths, difficult to identify in terms of fault. The 3D VSP image limited to a short radius around the borehole confirmed the fault/dyke nature of these lineaments, separating monocline compartments.

As a consequence, the surface data was carefully re- interpreted and on the second structural images derived from two surface seismic reflectors and the near surface fault footprint from 3DVSP residual statics, a series of subtle faults were clearly assessed. Last, the few faults intersecting the borehole can be clearly recognized on the logs and the borehole radar logs.

This case study demonstrates the added value of a joint interpretation of surface and borehole data in a decision making process for shaft sinking.