In this study, the geophysical strata rating (GSR) is calculated from petrophysical data using the equations developed for clastic rocks. The region being investigated is the South Pars gas field in the Persian Gulf Basin, where the Permian–Triassic Dalan and Kangan reservoirs host the largest accumulations of gas in the world. A 3D GSR model is estimated from 3D poststack seismic data by using a probabilistic neural network model. In this study, two methodologies are used to obtain GSR values at the different scales of wireline logs and 3D seismic data. Strong correlations between neural network predictions and actual GSR data at a blind well prove the validity of the intelligent model for estimating GSR. The GSR results are also in good agreement with porosity and elastic moduli of these carbonate rocks. Discrimination between the reservoir and non-reservoir shaly units can easily be obtained by comparing GSR and well logs. Very low GSR values with high gamma ray log responses indicate shaly intervals. These can cause washouts, casing collapse and other related drilling problems. Intervals with low GSR values and low gamma ray log responses indicate the presence of good reservoir units. In this study, the geophysical strata rating (GSR), which is an empirical measure of rock competency, is calculated from petrophysical data. The GSR is then extended to the whole South Pars gas field in the framework of 3D seismic data through an acoustic impedance poststack seismic inversion.

Faults are the most important geological structures that need to be detected in any modern underground coal mining project. Even a fault with a throw of a few metres can create safety issues and lead to costly delays in mine production. While seismic surveys are usually successful at locating faults with throws greater than 5-10m, reliable techniques for resolving more subtle faults, dykes and other minor features are yet to be developed.In seismic data, there are two types of seismic energy signatures that can be used for subsurface imaging: specular reflections and diffractions. Specular reflections are created by relatively smooth surfaces between layers of differing impedance contrasts such as at coal-rock interfaces. Specular reflections are used in conventional seismic reflection surveys. Diffractions are generated by local discontinuities caused by surface roughness, faults, dykes and fractures. In this paper, we describe the use of a moving average error filter which, along with a process of reflection event flattening for accommodation of dips and undulations of coal seam reflections, can be used to effectively extract diffraction signals from post-stack reflection seismic data. By identifying diffractions, small faults and other minor features that are elusive using conventional seismic reflection processing can be detected, suggesting that diffraction imaging may be a superior method to seismic reflection imaging. Both synthetic and real (2D and 3D) coal seismic data are used to illustrate the feasibility of diffraction imaging for small fault detection. It is demonstrated from synthetic examples that the extracted diffractions can be used to detect faults with a throw of 1m while a real data example shows that it is possible to detect igneous dykes with widths of just 4m.

In exploration programs for coal mining, geophysical logging of boreholes is mainly undertaken to reveal seam locations and the basic lithological section. Through the introduction of the Geophysical Strata Rating (GSR), this paper develops new geotechnical applications for geophysical logs. The GSR is an empirical rating scheme based on P-wave velocity data from sonic logs and estimates of the clay content and porosity derived from natural gamma and density logs. In coal measure strata, GSR values range between 0 and 100, with values of <15 indicating very poor rock and values over 80 representing extremely good rock.Given that geophysical logging data are obtained at closely spaced intervals within boreholes and that coal measure strata are usually laterally persistent between boreholes, geophysical results such as the clay content and the GSR can be modelled in two- and three-dimensions. This allows geotechnical information to be viewed in a geological context. Examples are provided of the log analysis and modelling. Geotechnical applications in risk identification, hazard assessment and design optimisation are also discussed. These applications exist in both underground and open cut mining.

The mining of stratiform ore deposits requires a means of determining the location of stratigraphic boundaries. A variety of geophysical logs may provide the required data but, in the case of banded iron formation hosted iron ore deposits in the Hamersley Ranges of Western Australia, only one geophysical log type (natural gamma) is collected for this purpose. The information from these logs is currently processed by slow manual interpretation. In this paper we present an alternative method of automatically identifying recurring stratigraphic markers in natural gamma logs from multiple drill holes.Our approach is demonstrated using natural gamma geophysical logs that contain features corresponding to the presence of stratigraphically important marker shales. The host stratigraphic sequence is highly consistent throughout the Hamersley and the marker shales can therefore be used to identify the stratigraphic location of the banded iron formation (BIF) or BIF hosted ore.The marker shales are identified using Gaussian Processes (GP) trained by either manual or active learning methods and the results are compared to the existing geological interpretation. The manual method involves the user selecting the signatures for improving the library, whereas the active learning method uses the measure of uncertainty provided by the GP to select specific examples for the user to consider for addition.The results demonstrate that both GP methods can identify a feature, but the active learning approach has several benefits over the manual method. These benefits include greater accuracy in the identified signatures, faster library building, and an objective approach for selecting signatures that includes the full range of signatures across a deposit in the library. When using the active learning method, it was found that the current manual interpretation could be replaced in 78.4% of the holes with an accuracy of 95.7%.

• Monitor-while-drilling and geophysical logging data have been acquired for blast holes in an open cut coal mine • Comparison of results show that MWD results respond to the variations in geology revealed by the geophysical logs. • MWD data can be used to predict sonic velocity . • MWD data can be used for geological and geotechnical analysis including blast design.

Modern underground coal mining requires certainty about geological faults, dykes and other structural features. Faults with throws of even just a few metres can create safety issues and lead to costly delays in mine production.In this paper, weuse numerical modelling in an ideal, noise-free environment with homogeneous layering to investigate the detectability of small faults by seismic reflection surveying. If the layering is horizontal, faults with throws of 1/8 of the wavelength should be detectable in a 2D survey. In a coal mining setting where the seismic velocity of the overburden ranges from 3000 m/s to 4000 m/s and the dominant seismic frequency is similar to 100 Hz, this corresponds to a fault with a throw of 4-5 m. However, if the layers are dipping or folded, the faults may be more difficult to detect, especially when their throws oppose the trend of the background structure. In the case of 3D seismic surveying we suggest that faults with throws as small as 1/16 of wavelength (2-2.5 m) can be detectable because of the benefits offered by computer-aided horizon identification and the improved spatial coherence in 3D seismic surveys.With dykes, we find that Berkhout's definition of the Fresnel zone is more consistent with actual experience. At a depth of 500 m, which is typically encountered in coal mining, and a 100 Hzdominant seismic frequency, dykes less than 8 min width are undetectable, even after migration.

Seismic reflection surveying in basalt-covered areas often fails to image underlying reflectors. To gain insights into the nature of the problem and obtain potential solutions, we have conducted experimental 2D seismic reflection and offset VSP surveys at two coal mines in the Bowen Basin of Australia. At the first mine, the basalt is relatively deep (114 m) and relatively thin (20 m). Conventional seismic acquisition and processing of a 2D seismic line provide poor results. However, upgoing reflections from layers below the basalt are clearly evident in the VSP survey and prestack depth migration is able to improve the continuity of the reflectors beneath the basalt. At the second mine, the 360 m wide basalt is at a depth of 40 m and has a thickness of about 40 m. It is fresh and unweathered and consists of multiple flows which are interlayered with unconsolidated sediments. Long-offset data acquisition combined with prestack depth migration was expected to produce satisfactory results but this is not the case. The associated VSP survey suggests that the problems at this mine are due to (1) the generation of complex downgoing and upgoing wave-fields within the basalt and (2) significant scattering of surface waves from outside the basalt at the margins of the basalt. Another problem is that the target coal seams are at about 300 m depth and the muting required to remove refraction events limited the effectiveness of the prestack depth migration. Reducing the strength of the surface waves through selection of an appropriate source and placement of shots at the base of the low-velocity zone (as had been the case at the first mine) will therefore improve the chances for a successful outcome. A Vibroseis survey subsequently undertaken at the second mine, which produced shot records with reduced surface waves, shows this to be the case.

Computer modelling of coal seams and their properties is standard geological practice for both underground and open cut mining. It is based on correlations of coal seams made between boreholes and the interpolation of relevant coal seam properties on the basis of the inferred coal seam boundaries. In the case of geotechnical studies, this same approach is not followed because there are insufficient geotechnical test results to form a basis for modelling and the geological models do not typically create boundaries for interburden rock types. Instead, geotechnical models tend to be based on coal seam geological models with test results shown as point data in the interburden intervals. This situation can be improved if geophysical logging data are used as the basis for geotechnical modelling. Appropriately analysed, these logs provide continuous measurements of lithological and geotechnical properties. In the case of natural gamma data, an analysis to show the variations in clay content allows sandstones to be separated from finer grained siltstones. If geophysical strata rating values are determined from the geophysical logs, they provide a measure of rock quality. From these analyses, 3D models showing interburden properties as well as boundaries of the relevant rock types can be created and used as a basis for mine design and control of geotechnical hazards.

Faults are the most important geological structures which need to be detected in any modern underground coal mining project. Even a fault with a throw of a few metres can create safety issues and lead to costly delays in mine production. While locating faults with throws greater than 5-10 m is quite successful by seismic survey, techniques to resolve the more subtle faults, shears and features, which exploration programs should also locate are needed. Faults cause breaks in continuity of seismic horizons. These discontinuities generate diffraction patterns. Before the days of seismic migration and generation of very high fold data, diffraction patterns were sought by seismic interpreters as an indication of faulting, especially for small faults where the discontinuities of the seismic reflections are less evident. Most processing is now aimed at suppressing these diffractions. However, in recent years, techniques for diffraction imaging developed for petroleum seismic data processing, makes small fault detection possible by separating the diffraction events from the reflection seismic events. In this paper, we apply the diffraction imaging techniques to enhance the detectability of the small faults for coal seismic environments and demonstrate the feasibility of a new fault imaging method with numerical examples. The effects of noise and migration velocity to fault imaging will also be discussed.

Since their introduction to the coal mining industries of the United Kingdom and West Germany in the 1970s, geophysical methods are now utilised in coal mining around the world. The range of applications in both surface and underground mining is extensive. Applications include coal seam mapping and geological fault detection, lithological mapping, geotechnical evaluation, assessment of the rock mass response to mining, detection of voids, location of trapped miners and guidance of drills and mining equipment. The range of techniques that can be employed is also extensive and includes geophysical borehole logging, the potential field methods, seismic reflection (2D and 3D), resistivity, electromagnetics and microseismic monitoring using active and passive sources. This paper discusses the major applications and the geophysical methods that can be applied. It also discusses future trends and suggests that the future motives for applying geophysics will not only include the current motivations of mine safety and productivity but will also include an increased emphasis on environmental management, the monitoring of sequestration activities and the provision of sensors to enable autonomous mining.

In geological investigations, drill hole information can come from core and chip samples, the monitoring of drill performance and various forms of down-hole testing such as geophysical logging. Fusion of some of these forms of measurements is possible and can improve the geological understanding. In this paper we conduct fusion of drill monitoring data and geochemical assays of drill hole samples using Multiple Task Gaussian Processes (MTGPs). MTGPs are a popular statistical supervised learning and fusion technique in the machine learning community. The proposed algorithm autonomously learns the intrinsic interconnections between rock strength parameters and geochemistry and uses these interconnections to improve the quality of the geological model.We demonstrate the principles of our approach by fusing drill monitoring data from closely spaced blast hole drilling at an open pit iron ore mine and assay results from more widely spaced exploration drill holes. The drill monitoring data are represented by a parameter we call the Adjusted Penetration Rate (APR) and we observe a strong correlation between APR and iron grade from the assays. Fusion allows a more detailed geological model to be obtained.

Automated rock recognition is a key step for building a fully autonomous mine. When characterizing rock types from drill performance data, the main challenge is that there is not an obvious one-to-one correspondence between the two. In this paper, a hybrid rock recognition approach is proposed which combines Gaussian Process (GP) regression with clustering. Drill performance data is also known as Measurement While Drilling (MWD) data and a rock hardness measure - Adjusted Penetration Rate (APR) is extracted using the raw data in discrete drill holes. GP regression is then applied to create a more dense APR distribution, followed by clustering which produces discrete class labels. No initial labelling is needed. Comparisons are made with alternative measures of rock hardness from MWD data as well as state-of-the-art GP classification. Experimental results from an actual mine site show the effectiveness of our proposed approach.

Summary Modern underground coal mining requires certainty about geological faults and other structural features. Even a fault with a throw of a few metres can create safety issues and lead to costly delays in mine production. In this paper, we investigate the detectability of small faults by the seismic reflection method through numerical modelling in an ideal noise-free environment with homogeneous layering. We find that 1) the smallest faults that can be identified in a 2D survey have throws of 1/8 of the wavelength; 2) faults are more difficult to detect when they occur within other structures. In typical seismic exploration for coal mining, the dominant seismic frequency is about 100 Hz and the seismic velocity of the overburden ranges from 3000m/s to 4000m/s. The corresponding wavelength is 30m to 40m. This suggests that the detectability limit for faults is about 4 – 5m. However, in the case of 3D seismic surveying we suggest that this can be redefined to 1/16 of wavelength (2 - 2.5m) because of the benefits offered by computer-aided horizon identification and the improved spatial coherence in 3D seismic surveys. In all cases, the actual fault detectability will depend on the quality of the seismic data and the geology of the area under investigation.

Autonomous operation of blast hole drill rigs requires monitoring of drilling parameters known as "Measurement While Drilling" (MWD) data. From these data, rock properties can be inferred. A supervised classification scheme is usually used to map MWD data inputs to rock type outputs given some labeled training data. However, the geology has no definite ground truth that can allow a reliable labeling of the training data, nor is there a clear input-output pair connection between the MWD data and the rock types. In this paper, an adaptive unsupervised approach is proposed to estimate the rock types in a data driven way by minimizing the entropy gradient of the characterizing measure "Optimized Adjusted Penetration Rate" (OAPR). Neither data labeling nor fixed model parameters are required because of the data driven nature of the algorithm. Experimental results illustrate the effectiveness of our solution.

This work is motivated by the need to develop new perception and modeling capabilities to support a fully autonomous, remotely operated mine. The application differs from most existing robotics research in that it requires a detailed world model of the sub-surface geological structure. This in-ground geological information is then used to drive many of the planning and control decisions made on a mine site. This paper formulates a method for automatically detecting in-ground geological boundaries using geophysical logging sensors and a supervised learning algorithm. The algorithm uses Gaussian Processes (GPs) and a single length scale squared exponential covariance function. The approach is demonstrated on data from a producing iron-ore mine in Australia. Our results show that two separate distinctive geological boundaries can be automatically identified with an accuracy of over 99 percent. The alternative approach to automatic detection involves manual examination of these data.

The Lapstone Structural Complex is an association of monoclines and high-angle reverse faults that occupies the frontal ridge of the lower Blue Mountains, west of Sydney. Permian-Triassic movements have been proposed on the basis of unpublished seismic profiles reportedly showing abrupt stratigraphic thickness variations in the Sydney Basin succession across the Lapstone Structural Complex and indicative of syn-depositional deformation. In a companion paper, we present an interpretation of recently reprocessed seismic data and show that the Lapstone Structural Complex played no role during Sydney Basin sedimentation. Paleomagnetic data collected from two monoclines are interpreted to bracket development of the Lapstone Structural Complex within the mid-Cretaceous to Miocene interval. Cenozoic uplift of the frontal ridge of the lower Blue Mountains is shown by uplifted semi-consolidated gravels (Rickabys Creek Gravel). A Paleogene timing of uplift was proposed from modelling of landscape development. Association of structures of the Lapstone Structural Complex with the uplifted lower Blue Mountains is most simply explained by a Paleogene age for these structures. Evidence for additional neotectonic activity along the Lapstone Structural Complex reflects the modern compressive stress regime associated with the transpressional plate boundary in the South Island of New Zealand. A Paleogene age for the uplift of the Blue Mountains is consistent with the high relief of this region in contrast to more stable regions of the highlands of eastern Australia characterised by low local relief, reflecting antiquity.