Most of the ambient-noise studies are performed with sensor arrays located at the surface. Passive recordings containing seismic arrivals from subsurface sources could be seen as having a geometry resembling reverse vertical seismic profiling (RVSP). In such scenarios, the intersource seismic-interferometry technique can be used to redatum the surface receivers to the subsurface source positions resulting in virtual shot gathers at depth. The success of the interferometric processing primarily requires that a correlation panel created in the interferometric process contains stationary-phase regions that, when summed, retrieve events with correct timing, whereas nonstationary contributions are canceled. We have addressed the combination of the RVSP configuration and ambient-noise measurements. We develop a prototype of a data-driven technique allowing us to adapt the summation process such that changes in the stationary-phase requirements imposed by changes in the noise-sources distribution can be adaptively satisfied without the need for array redeployment. We develop to identify the receivers located in the stationary-phase regions by scanning for stationary contributions in the correlation panel prior to stacking. Our method uses the correlation coefficient and time windowing to distinguish between stationary and nonstationary arrivals. The improved Green's function estimate is obtained by limiting the summation to only those receivers that, when summed, enhance the stationary and attenuate the nonstationary contributions. We test this using a simple 2D numerical example to find a practical way to alleviate insufficient receiver coverage. We determine the theoretical possibility to improve the intersource Green's function estimates without explicit knowledge of the target and the source distribution. However, our data-driven approach has the disadvantage of being limited to scenarios in which the correlation panel from all receivers contains identifiable stationary-phase arrivals.

We apply a full-scale 3D seismic virtual-source survey (VSS) for the purpose of near-mine mineral exploration in the Kylylahti sulfide deposit, Finland. Based on the ambient-noise (AN) characterization including beamforming results, we created a 10-days subset of AN recordings that were dominated by multi-azimuth high-velocity arrivals. We use an illumination-diagnosis and location procedure to show that the AN recordings associated with the high apparent velocities are related to body-wave events. Next, we produce 994 virtual-source gathers by applying seismic-interferometry processing by crosscorrelating AN at all receivers resulting in a full 3D VSS. We apply standard 3D time-domain reflectiondata processing and imaging using the subset and the full AN data, and validate both results against a pre-existing detailed geological information and 3D active-source data processed in the same way as the passive data. The resulting post-stack migrated sections show agreement of reflections between the passive and active data and indicate that VSS provides images where the active-source data are not available. In particular, the previously unknown extent of the ore-bearing complex was captured exclusively by passive data, which added a new geological insight into the Kylylahti formation. The methodological approach developed can be used in other areas in mineral exploration context.

For the first time, we apply a full-scale 3D seismic virtual-source survey (VSS) for the purpose of near-mine mineral exploration. The data were acquired directly above the Kylylahti underground mine in Finland. Recorded ambient noise (AN) data are characterized using power spectral density (PSD) and beamforming. Data have the most energy at frequencies 25-90gHz, and arrivals with velocities higher than 4gkmgs-1 have a wide range of azimuths. Based on the PSD and beamforming results, we created 10gd subset of AN recordings that were dominated by multi-azimuth high-velocity arrivals. We use an illumination diagnosis technique and location procedure to show that the AN recordings associated with high apparent velocities are related to body-wave events. Next, we produce 994 virtual-source gathers by applying seismic interferometry processing by cross-correlating AN at all receivers, resulting in full 3D VSS. We apply standard 3D time-domain reflection seismic data processing and imaging using both a selectively stacked subset and full passive data, and we validate the results against a pre-existing detailed geological information and 3D active-source survey data processed in the same way as the passive data. The resulting post-stack migrated sections show agreement of reflections between the passive and active data and indicate that VSS provides images where the active-source data are not available due to terrain restrictions. We conclude that while the all-noise approach provides some higher-quality reflections related to the inner geological contacts within the target formation and the general dipping trend of the formation, the selected subset is most efficient in resolving the base of formation.

Despite the unrivalled spatial resolution and depth penetration of active-source seismic methods used for mineral exploration in hardrock environment, economic and environmental restrictions (e.g., source permitting) may preclude its full-scale application. In such a case, 2D passive reflection seismics can be considered a cost-effective way to perform reconnaissance-type survey and provide body-wave structural imaging using ambient-noise seismic interferometry (ANSI). This is, however, conditional to the presence of noise sources in the subsurface, for example produced by underground mining activity. Here, we propose a 2D ANSI workflow as an intermediate step prior to a full-scale 3D ANSI survey and an affordable tool in brownfield exploration, e.g., when trying to update current geological models beyond the drilled area. We test the applicability of this approach by analysing selected receiver lines from a 3D passive dataset acquired over the Kylylahti mine in Finland. Our methodology aims at choosing the optimal processing strategy at possibly lowest acquisition (2D geometry) and computational (small amount of data) cost. We address the fundamental questions in ANSI, i.e., (i) how much AN should one record and (ii) which SI processing approach should one choose. Therefore, we test different processing steps necessary to produce virtual shot gathers (VSG): preprocessing, selection of the ambient-noise portion, and selection of the method for retrieving the impulse responses between the receivers (crosscorrelation - CC, crosscoherence - CCh, multidimensional deconvolution - MDD). We conclude that trace energy normalization and high-pass filtering are the preferred preprocessing steps, while the best imaging is obtained when VSGs are retrieved using MDD applied in the noise-volume approach or CC in the event-driven approach. An event-driven approach may significantly reduce the acquisition time: for the Kylylahti dataset, using 10 events with energetic body-wave arrivals, extracted from one hour of data, was enough to provide results comparable to the results from the noise-volume approach using the complete one hour of noise.

We present a method for automatic detection and classification of seismic events from continuous ambient-noise (AN) recordings using an unsupervised machine-learning (ML) approach. We combine classic and recently developed array-processing techniques with ML enabling the use of unsupervised techniques in the routine processing of continuous data. We test our method on a dataset from a large-number (large-N) array, which was deployed over the Kylylahti underground mine (Finland), and show the potential to automatically process and cluster the volumes of AN data. Automatic sorting of detected events into different classes allows faster data analysis and facilitates the selection of desired parts of the wavefield for imaging (e.g., using seismic interferometry) and monitoring. First, using array-processing techniques, we obtain directivity, location, velocity, and frequency representations of AN data. Next, we transform these representations into vector-shaped matrices. The transformed data are input into a clustering algorithm (called k-means) to define groups of similar events, and optimization methods are used to obtain the optimal number of clusters (called elbow and silhouette tests). We use these techniques to obtain the optimal number of classes that characterize the AN recordings and consequently assign the proper class membership (cluster) to each data sample. For the Kylylahti AN, the unsupervised clustering produced 40 clusters. After visual inspection of events belonging to different clusters that were quality controlled by the silhouette method, we confirm the reliability of 10 clusters with a prediction accuracy higher than 90%. The obtained division into separate seismic-event classes proves the feasibility of the unsupervised ML approach to advance the automation of processing and the utilization of array AN data. Our workflow is very flexible and can be easily adapted for other input features and classification algorithms.

The main issues related to passive-source reflection imaging with seismic interferometry (SI) are inadequate acquisition parameters for sufficient spatial wavefield sampling and vulnerability of surface arrays to the dominant influence of the omnipresent surface-wave sources. Additionally, long recordings provide large data volumes that require robust and efficient processing methods. We address these problems by developing a two-step wavefield evaluation and event detection (TWEED) method of body waves in recorded ambient noise. TWEED evaluates the spatiotemporal characteristics of noise recordings by simultaneous analysis of adjacent receiver lines. We test our method on synthetic data representing transient ambient-noise sources at the surface and in the deeper subsurface. We discriminate between basic types of seismic events by using three adjacent receiver lines. Subsequently, we apply TWEED to 600 h of ambient noise acquired with an approximately 1000-receiver array deployed over an active underground mine in Eastern Finland. We develop the detection of body-wave events related to mine blasts and other routine mining activities using a representative 1 h noise panel. Using TWEED, we successfully detect 1093 body-wave events in the full data set. To increase the computational efficiency, we use slowness parameters derived from the first step of TWEED as input to a support vector machine (SVM) algorithm. Using this approach, we detect 94% of the TWEED-evaluated body-wave events indicating the possibility to limit the illumination analysis to only one step, and therefore increase the time efficiency at the price of lower detection rate. However, TWEED on a small volume of the recorded data followed by SVM on the rest of the data could be efficiently used for a quick and robust (real-time) scanning for body-wave energy in large data volumes for subsequent application of SI for retrieval of reflections.