Airborne Gravity Gradient Data Research Articles

Three-dimensional cooperative inversion of airborne magnetic and gravity gradient data using deep-learning techniques

Using multiple geophysical methods has become a prevailing approach in numerous geophysical applications to investigate subsurface structures and parameters. These multimethod-based exploration strategies have the potential to greatly diminish uncertainties and ambiguities encountered during geophysical data analysis and interpretation. One of the applications is the cooperative inversion of airborne magnetic and gravity gradient data for the interpretation of data obtained in mineral, oil and gas, and geothermal explorations. In this paper, a unified cooperative inversion framework is designed by combining the standard separate inversions with a deep neural network (DNN), which serves as the link between different types of data. A well-trained DNN takes the separately inverted susceptibility and density models as the inputs and provides improved models that will be used as the initial models of deterministic inversions. A two-round iteration strategy is adopted to guarantee the reasonability of the recovered models and overall efficiency of the inversion. In addition, this deep-learning (DL)-based framework demonstrates excellent generalization abilities when tested on models that are entirely distinct from the training data sets. The framework can easily incorporate multiphysics without necessitating any structural changes to the network. Synthetic experiments validate that our DL-based method outperforms conventional separate inversions and cross-gradient-based joint inversion in view of the accuracy of the recovered models and inversion efficiency. Successful application to field data further verifies the effectiveness of our DL-based method.

Method for assessing the three-dimensional density structure based on gravity gradient inversion and gravity gradient curvature

AbstractCompared with gravity data, gravity gradient data can reflect the structural features of density anomalies with higher sensitivity and resolution. Gravity gradient inversion is a common method for interpreting gravity gradient data, but the non-uniqueness of gravity gradient inversion solutions makes it difficult to quickly determine the effectiveness and reliability of the inversion results. To solve this problem, this paper combines the gravity gradient inversion and gravity gradient curvature methods and applies them to different sets of synthetic models. The gravity gradient inversion results are useful primarily for determining the spatial distributions and physical properties of density anomalies, while the gravity gradient curvature results have a prominent advantage in judging the geometric features of the density anomalies. In addition, one form of gravity gradient curvature, the contour lines of the shape index, can be compared with the top view of the inversion results to rapidly confirm the distribution of density anomalies, which is especially valuable for unexplored regions. On the basis of the respective advantages and features of these two methods, the gravity gradient curvature can be used as an important criterion to judge the results of gravity gradient inversion. The two methods are then combined to analyse airborne gravity gradient data measured in the field at the Kauring Test Site in Australia. Compared with the findings of previous gravity inversion research, this study shows that several anomalous blocks exist in the vicinity of the central anomalies, thereby demonstrating that the combination of these two methods is effective and reliable in identifying density anomalies.

A fast imaging method for airborne gravity gradient data based on tensor invariants

A fast imaging method for airborne gravity gradient data based on tensor invariants

Reconstructing airborne gravity gradient data based on Fourier transform

Reconstructing airborne gravity gradient (AGG) components will improve the signal-to-noise ratio by extracting the coherent signal. Fourier transformations are routinely used to reconstruct a consistent multicomponent AGG data set for noise suppression purposes. The crucial step is calculating the potential from multicomponent data based on the Fourier transform. Given that solving a least-squares (LS) problem is involved in most cases, the methods based on the Fourier transform have no essential difference. However, existing methods ignore the different noise levels of the components. To tackle this problem, we have used a weighted LS method instead and adopted a general expression for calculating the potential from multicomponent data corrupted by noise. With the wavenumber shift technique, this expression, used for AGG reconstruction, achieves the highest accuracy in theory. We finally develop a practical procedure for raw AGG data reconstruction that iteratively evaluates the noise level by updating the weights. In addition, we examine the errors within denoised data, and an improved simulation is adopted to estimate the statistics of noise levels and residual errors after denoising. We evaluate the effectiveness of our approach through synthetic examples, and the results find the ability of noise removal and estimation for data containing noise.

Potential field data interpolation by Taylor series expansion

Data interpolation is critical in the analysis of geophysical data when some data are missing or inaccessible. We interpolate irregular or missing potential field data using the relation between adjacent data points inspired by the Taylor series expansion (TSE). The TSE method first finds the derivatives of a given point near the query point using data from neighboring points and then uses the Taylor series to obtain the value at the query point. The TSE method works by extracting local features represented as derivatives from the original data for interpolation in the area of data vacancy. Compared with other interpolation methods, the TSE provides a complete description of potential field data. Specifically, the remainder in TSE can measure local fitting errors and help obtain accurate results. Implementation of the TSE method involves two critical parameters — the order of the Taylor series and the number of neighbors used in the calculation of derivatives. We find that the first parameter must be carefully chosen to balance between the accuracy and numerical stability when data contain noise. The second parameter can help us build an over-determined system for improved robustness against noise. Methods of selecting neighbors around the given point using an azimuthally uniform distribution or the nearest-distance principle are also presented. Our approach is first illustrated by a synthetic gravity data set from a single survey line, and then it is generalized to the case over a survey grid. In both numerical experiments, the TSE method demonstrates an improved interpolation accuracy in comparison with the minimum curvature method. Finally, we apply the TSE method to a ground gravity data set from the Abitibi Greenstone Belt, Canada, and an airborne gravity gradient data set from the Vinton Dome, Louisiana, USA.

Unveiling the 3D undercover structure of a Precambrian intrusive complex by integrating airborne magnetic and gravity gradient data into 3D quasi-geology model building

Mineral exploration under a thick sedimentary cover naturally relies on geophysical methods. We have used high-resolution airborne magnetic and gravity gradient data over northeast Iowa to characterize the geology of the concealed Precambrian rocks and evaluate the prospectivity of mineral deposits. Previous researchers have interpreted the magnetic and gravity gradient data in the form of a 2D geologic map of the Precambrian basement rocks, which provides important geophysical constraints on the geologic history and mineral potentials over the Decorah area located in the northeast of Iowa. However, their interpretations are based on 2D data maps and are limited to the two horizontal dimensions. To fully tap into the rich information contained in the high-resolution airborne geophysical data, and to further our understanding of the undercover geology, we have performed separate and joint inversions of magnetic and gravity gradient data to obtain 3D density contrast models and 3D susceptibility models, based on which we carried out geology differentiation. Based on separately inverted physical property values, we have identified 10 geologic units and their spatial distributions in 3D which are all summarized in a 3D quasi-geology model. The extension of 2D geologic interpretation to 3D allows for the discovery of four previously unidentified geologic units, a more detailed classification of the Yavapai country rock, and the identification of the highly anomalous core of the mafic intrusions. Joint inversion allows for the classification of a few geologic units further into several subclasses. We have demonstrated the added value of the construction of a 3D quasi-geology model based on 3D separate and joint inversions.

A fast equivalent source method for airborne gravity gradient data

Airborne gravity gradiometry measures several gravity gradient components unlike conventional gravimetry taking only the vertical gravity component into account. However, processing of multicomponent airborne gravity gradient (AGG) data without corrupting their internal consistency is often challenging. Therefore, we have developed an equivalent source technique to tackle this challenge. With a combination of Gauss-fast Fourier transform and the Landweber iteration, we have developed an efficient way to compute equivalent sources for AGG data. This method can handle two components simultaneously. We first examined its viability by applying this approach to a synthetic example. Afterward, we applied our method to real AGG data collected in the area of Karasjok, Norway. Our result is almost the same as the results that meet the industry standard but with great efficiency.

Geophysical expression of buried range-front embayment structure: Great Sand Dunes National Park, Rio Grande rift, Colorado

Great Sand Dunes National Park and Preserve (GRSA, Colorado) lies along the eastern margin of the San Luis Basin and the tectonically active Sangre de Cristo fault system that are part of the northern Rio Grande rift. GRSA lies within a prominent embayment in the range front where two separate sections of the Sangre de Cristo fault system intersect. Fault scarps are observed along both intersecting fault zones within older basin-fill alluvium, but have been obscured by the actively migrating dunefield. The dune sand is also strongly magnetic, locally limiting the usefulness of aeromagnetic methods for mapping concealed structure. This study uses airborne geophysical methods, primarily airborne gravity gradient data, along with constraints from geologic mapping and limited subsurface data and groundwater modeling, to interpret the subsurface basin geometry and range-front structure of the embayment. Using forward modeling of the gravity gradient data and locations of faults inferred from gravity gradient and aeromagnetic lineaments, several previously unrecognized tectonic elements are interpreted adjacent to the range front. Some of the largest rift-­related fault offsets are demonstrated to be basinward of the normal fault zones mapped at the surface along the range front of the Sangre de Cristo Mountains, along faults concealed under the dunefield and subparallel to the two fault sections. A fault-bounded structural bench, likely composed of Proterozoic rocks, underlies most of the high dunefield at depths of 500 m to 1 km. The bench is truncated on its southwest margin by a northwest-trending, southwest-dipping normal fault. A northeast-trending, northwest-dipping normal fault with ∼600 m of estimated relief lies under the southern margin of the dunefield and bounds a structurally higher bench of Proterozoic rocks concealed at <400 m depth near the range front. The northwest- and northeast-trending geophysical lineaments generally correspond well with the trends of faults mapped at the surface, and with both pre- and syn-rift structures in the Sangre de Cristo Mountains. Aeromagnetic anomalies are explained by variations in the magnetization of pre-rift rocks, and the strongly magnetic dune sand.

The 3D Resolution Power of the Full Tensor Gravity Gradient

Airborne gravity tensor (or gravity gradient) surveying is one of the newest techniques for geophysical exploration. The rise in the acquisition of this data type is partially due to the fact that these data provide much more complete, extensive and higher resolution information of rock density distribution than conventional land gravity data. This has positioned Airborne Gravity Gradient (AGG) surveys among the geophysical services requested by mining companies, alongside aeromagnetic and radiometric surveys. The use of these data has been primarily to support geological mapping given the limited access to commercial software for quantitative depth interpretation. In particular, there has been a scarce development of AGG data inversion software, which has limited our understanding of the significance of these data for resolving three-dimensional subsurface targets. In this work we hypothesize that AGG data can provide more detailed information of the multidirectional variations of subsurface density. To prove this hypothesis, we developed conjugate gradient AGG-data inversion software for three-dimensional targets. This software was applied to synthetic data generated by several test assemblages of three-dimensional bodies and used to perform a Singular Value Decomposition (SVD) sensitive analysis to explore the actual resolution power of the different tensor data components and whether they are indeed superior to the conventional vertical gravity.

3D unconstrained and geologically constrained stochastic inversion of airborne vertical gravity gradient data

We present an inversion tool for airborne gravity gradient data that yields a 3D density model using stochastic methods i.e. cokriging and conditional simulation. This method uses geostatistical properties of the measured gravity gradient to estimate a 3D density model whose gravity response fits the measured gravity gradient anomaly. Linearity between gravity gradient data and density allows estimation of the model (density) covariance using observed data, i.e. we adjust iteratively the density covariance matrix by fitting experimental and theoretical gravity gradient covariance matrices. Inversion can be constrained by including densities known at some locations. In addition we can explore various reasonable solutions that honour both the estimated density covariance model and the gravity gradient data using geostatistical simulation.The proposed method is first tested with two synthetic datasets generated from a sharp-boundary model and a smooth stochastic model respectively. The results show the method to be capable of retrieving models compatible with the true models; it also allows the integration of complex a priori information. The technique is then applied to gravity gradient survey data collected for the Geological Survey of Canada in the area of McFaulds Lake (Ontario, Canada) using the Falcon airborne gravity system. Unconstrained inversion returns a density model that is geologically plausible and the computed response exactly fits the observed gravity gradient anomaly.

Consistency investigation, vertical gravity estimation, and inversion of airborne gravity gradient data — A case study from northern Sweden

For airborne gravity gradient data, it is a challenge to distinguish between high-frequency intrinsic and dynamically produced noise caused by the aircraft and small-scale effects from shallow density variations. To facilitate consistent interpretation, techniques that include all of the measured gravity gradient components are particularly promising. We represented the measurements by a common potential function accounting for lateral and height variations. Thus, it was possible to evaluate the internal consistency between the measured components and to identify components with bias or particularly strong noise. As an extra benefit for data sets that contain terrain-corrected and nonterrain-corrected gravity gradient measurements at flight altitude, we estimated terrain-corrected anomalies on the topographic relief using downward continuation and retrieved nonterrain-corrected gravity gradient data suitable for inversion using upward continuation. For a field data set from northern Sweden, the largest differences (up to 50 eotvos) between the measured and estimated components of the gravity gradient data were found in areas of high topographical relief. But the average residual standard deviations of the individual components were between 3.6 and 7.4 eotvos, indicating that the components were consistent in an average sense. We have determined the successful conversion of terrain-corrected airborne gravity gradient data to Bouguer gravity data on the topographic relief using ground-based vertical gravity data as a reference. A 3D inverse model computed from the nonterrain-corrected data clearly showed the depth extent of the geologic structures observed at the surface, but it only produced a weak representation of the shallow structure. In contrast, a 2D surface density model in which only lateral variations of density in the topographic relief was allowed exhibited more realistic density distributions in fair correlation with geology.

3D geophysical model of the Glyde Basin, Northern Territory, based on curvatures derived from airborne gravity gradient data

SUMMARY This paper presents automatic 3D geophysical model generation based on equivalent pseudodepth slicing of the shape index of the equipotential surfaces derived from airborne gravity gradient data. The method is carried out in three steps. First, the pseudodepth slices of the vertical gravity gradient and the magnitude of the differential curvature components are generated. Second, the equivalent pseudodepth slices of the shape index are generated. Finally, 3D interpolation is carried out to obtain the final model. The method is applied to FALCON airborne gravity gradiometer data from the Glyde Basin, Northern Territory and compared to an independently interpreted, integrated 3D geological Earth model.

Processing gravity gradients to detect kimberlite pipes

A modelling and pattern recognition-based approach is applied to processing airborne gravity gradient data for kimberlite exploration. The carrot-like bodies with low density crater facies that typ...

Lithologic characterization using airborne gravity gradient and aeromagnetic data for mineral exploration: A case study in the Quadrilátero Ferrífero, Brazil

We present a study on utilizing airborne gravity gradient and magnetic data to characterize an iron ore formation in Minas Gerais, Brazil. The target iron ore bodies have a distinctly high density contrast and produce well-defined anomalies in airborne gravity gradiometry data. The high-grade hematite iron ores are associated with low and moderate susceptibility, making magnetic data useful in distinguishing potential ore bodies from the host iron formation. The airborne gravity gradient and magnetic data over part of the Gandarela Syncline iron formation in the Quadrilátero Ferrífero are independently inverted to obtain a 3D susceptibility and density contrast model. These detailed 3D physical property distributions of subsurface features are then used for geologic characterization and interpretation purposes through lithologic associations. We outline two approaches to link the two physical property distributions and identify representative geologic units in the study area. The geologic units are then organized into a 3D lithology model to help characterize subsurface geologic structure and ore distribution. The lithologic models provide an intuitive representation of the geology and can assist in future exploration plans or in assessment of resource distribution and quality. Our study demonstrates that such approaches are feasible on the deposit scale.

Investigations related to scientific deep drilling to study reservoir-triggered earthquakes at Koyna, India

AbstractArtificial water reservoir-triggered earthquakes have continued at Koyna in the Deccan Traps province, India, since the impoundment of the Shivaji Sagar reservoir in 1962. Existing models, to comprehend the genesis of triggered earthquakes, suffer from lack of observations in the near field. To investigate further, scientific deep drilling and setting up a fault zone observatory at depth of 5–7 km is planned in the Koyna area. Prior to undertaking deep drilling, an exploratory phase of investigations has been launched to constrain subsurface geology, structure and heat flow regime in the area that provide critical inputs for the design of the deep borehole observatory. Two core boreholes drilled to depths of 1,522 and 1,196 m have penetrated the Deccan Traps and sampled the granitic basement in the region for the first time. Studies on cores provide new and direct information regarding the thickness of the Deccan Traps, the absence of infra-Trappean sediments and the nature of the underlying basement rocks. Temperatures estimated at a depth of 6 km in the area, made on the basis of heat flow and thermal properties data sets, do not exceed 150 °C. Low-elevation airborne gravity gradient and magnetic data sets covering 5,012 line km, together with high-quality magnetotelluric data at 100 stations, provide both regional information about the thickness of the Deccan Traps and the occurrence of localized density heterogeneities and anomalous conductive zones in the vicinity of the hypocentral zone. Acquisition of airborne LiDAR data to obtain a high-resolution topographic model of the region has been completed over an area of 1,064 km2centred on the Koyna seismic zone. Seismometers have been deployed in the granitic basement inside two boreholes and are planned in another set of six boreholes to obtain accurate hypocentral locations and constrain the disposition of fault zones.

Automatic generation of 3D geophysical models using curvatures derived from airborne gravity gradient data

A method for automatic 3D model generation derived from airborne gravity gradient data was illustrated. The proposed method computed a volumetric distribution of curvatures that form a 3D image of possible density distributions. The method relies on spectral analysis and the equivalence of the power spectra of the classical mean and differential curvatures of equipotential surfaces to create pseudodepth slices of a quantity that describes the geometry of the surfaces: the shape index. The method was carried out in three steps. First, the pseudodepth slices of the vertical gravity gradient and the magnitude of the differential curvature components were generated. Second, equivalent pseudodepth slices of the shape index were generated. Finally, 3D interpolation was carried out to obtain the final model. The synthetic model data indicated that the vertical density contrasts were well modeled. A 3D model derived from FALCON airborne gravity gradiometer data from the Canning Basin, Australia, was compared to an independently interpreted integrated 3D earth geologic model.

Application of curvatures and Poisson’s relation to airborne gravity gradient data in oil exploration

SUMMARY The application of equipotential surface curvatures and Poisson’s relation to airborne gravity gradient data is presented. The mean and differential curvature of the equipotential surface, the curvature of the gravity field line, the AGG Geometry Map, the zero contour of the Gaussian curvature of the equipotential surface and the zero contour of the determinant of the gravity gradient tensor should improve the understanding and geological interpretation of gravity gradient data.

Geometric shapes derived from airborne gravity gradiometry data: New tools for the explorationist

We display airborne gravity gradient data using curvature attributes of the equipotential surface to derive geometric shapes that can be associated with structural and geologic features. Concepts used in differential geometry that describe all geometric aspects of a surface are applied to airborne gravity gradiometry data. Gravity gradient components are ideal for this application because they are related to curvature of the gravitational potential. We also emphasize that with gravity methods, the surfaces for which the curvature attributes are derived represent the succession of equipotential surfaces, which, under favorable circumstances, correspond to one or more geologic surfaces that dominate the signature of the airborne gravity gradiometry data.

Basin architecture from high-resolution gravity gradient, magnetic, and seismic data, King Sound, Canning Basin, Western Australia

An interpretation of geologic structure at King Sound in the Canning Basin was completed using airborne gravity gradient, magnetic, and seismic data. During the Late Devonian and Mississippian periods, the elevated part of the basement in the north was rimmed by carbonate reefs and redeposited carbonate debris, whereas in the south, siliciclastic submarine fans and turbidites were deposited along the margin of the basement in a deep-marine environment. Three principal lithologic units were identified from the vertical gravity gradient (GDD) in the basin: (1) the Fairfield Group carbonates of high density are interpreted to be the source of prominent positive gravity anomalies; (2) forereef debris and carbonate clastics reworked from carbonates higher up the slope or from the carbonate platform are interpreted to be the source of medium-density responses; and (3) turbidites, debris flows, and associated clastic basinal sequences of low density are interpreted to be the source of prominent negative gravity anomalies. Depth slices of GDD indicate the channelized nature of turbidite flows. In the lower section of the basin, intrasedimentary intrusives were identified from magnetic, GDD, seismic, and well data. Depth to magnetic basement calculation indicates that the surface of the Archean to Paleoproterozoic basement ranges from 3200 to 130 m (10,499–427 ft) below sea level. The northwest- and northeast-oriented south-dipping faults cut the basement and propagate upward into the sediments. A three-dimensional geologic model constructed for King Sound satisfies all known geologic constraints and is consistent with the gravity, magnetic, seismic, and well data.

Analysis and reprocessing of airborne gravity gradiometer data over the Strange Lake rare Earth deposit, Québec-Labrador

A high-resolution airborne gravity gradiometer survey was flown in March 2012 over the Strange Lake intrusive complex, located on the border of Québec and Newfoundland and Labrador. This survey provided an opportunity to better understand the performance and limitations of this technology and also investigate potential improvements to the associated data processing. Airborne gravity gradient data low-pass-filtered at 75 m, rather than the standard 300 m, were provided by the contractor for this purpose. One flight line was reflown several times and used to assess the repeatability of the data, showing that in this particular area, the airborne gradient gravity system has a full-sine (wavelength) resolution of about 700 m. The system can therefore detect anomalies having a width of about 350 m. We also found that for gradiometer data, kriging is a better interpolator than low-pass filtering combined with minimum curvature gridding. Using kriging, line-to-line correlation (and in some cases, resolution) appears slightly improved.