Logo: to the web site of Uppsala University

Reflection seismic data were acquired within two field campaigns in the Blötberget, Ludvika mining area of central Sweden, for deep imaging of iron-oxide mineralization that were known to extend down to 800-850 m depth. The two surveys conducted in years 2015 and 2016, one employing a seismic landstreamer and geophones connected to wireless recorders, and another one using cabled geophones and wireless recorders, aimed to delineate the geometry and depth extent of the iron-oxide mineralization for when mining commences in the area. Even with minimal and conventional processing approaches, the merged datasets provide encouraging information about the depth continuation of the mineralized horizons and the geological setting of the study area. Multiple sets of strong reflections represent a possible continuation of the known deposits that extend approximately 300 m further down-dip than the known 850 m depth obtained from historical drilling. They show excellent correlation in shape and strength with those of the Blötberget deposits. Furthermore, several reflections in the footwall of the known mineralization can potentially be additional resources underlying the known ones. The results from these seismic surveys are encouraging for mineral exploration purposes given the good quality of the final section and fast seismic surveys employing a simple cost-effective and easily available impact-type seismic source.

Many metallic mineral deposits have sufficient physical property contrasts, particularly density, to be detectable using seismic methods. These deposits are sometimes significant for our society and economic growth and can help to accelerate the energy transition towards decarbonization. However, their exploration at depth requires high-resolution and sensitive methods. Following a series of 2D seismic trials, a sparse, narrow source-receiver azimuth, 3D seismic survey was conducted in the Blotberget mine, in central Sweden, covering an area of approximately 6 km(2) for deep-targeting iron oxide deposits and their host rock structures. The survey benefited from a collaborative work by putting together 1266 seismic recorders and a 32 t vibrator, generating 1056 shot points in a fixed geometry setup. Shots were fired at every 10 m where possible, and receivers were placed at every 10-20 m. Notable quality data were acquired despite the area being dominated by swampy places as well as by built-up roads and historical tailings The data processing had to overcome these challenges for the static corrections and strong surface waves in particular. A tailored for hardrock setting and processing workflow was developed for handling such a dataset, where the use of mixed 2D and 3D refraction static corrections was relevant. The resulting seismic volume is rich in terms of reflectivity, with clear southeast-dipping reflections originating from the iron oxide deposits extending vertically and laterally at least 300 m beyond what was known from available boreholes. As a result, we estimate potential additional resources from the 3D reflection seismic experiment on the order of 10 Mt to be worth drilling for detailed assessments. The mineralization is crosscut by at least two major sets of northwest-dipping reflections interpreted to dominantly be normal faults and to be responsible for much of the lowland in the Blotberget area. Moreover, these post-mineralization faults likely control the current 3D geometry of the deposits. Curved and submerged reflections interpreted from folds or later intrusions are also observed, showing the geological complexity of the study area. The seismic survey also delineates the near-surface expression of a historical tailing as a by-product of refraction static corrections, demonstrating why 3D seismic data are so valuable for both mineral exploration and mine planning applications.

Diffraction imaging can help better understand small-scale geological structures. Due to their often-weak signal, in order to image them, it is necessary to separate diffraction signals from the rest of the wavefield. Many different methods have been developed for diffraction wavefield separation, and the newest trend includes the application of artificial neural networks and deep learning. Available case studies with a deep-learning approach for diffraction separation show good results when applied to synthetic and sedimentary setting datasets where diffraction signals are either strong or have pronounced characteristics. Examples, however, are missing from crystalline or hardrock geological settings where the signal-to-noise ratio is by far lower and diffraction signals are usually within a complex reflectivity medium, have steep tails and are usually incomplete. In this study, we showcase the application of a deep semantic segmentation model on synthetic seismic, real ground-penetrating radar, and hardrock seismic datasets. Synthetic seismic sections were generated using different random noise levels and coherent noise resembling a complex reflectivity pattern interfering with diffraction tails. For the real GPR dataset, diffraction signals were successfully delineated, although in some locations reflections were picked up because of their similar pixel values as the apex of the diffractions. As for the real seismic dataset, through a number of approaches, we were able to completely delineate a single diffraction within several inlines that was generated from a massive sulphide body. The algorithm also enabled us to recognize an incomplete diffraction, at the edge of the seismic cube, which was never labelled. This diffraction originated from outside of the seismic volume and may be a target for future mineral exploration programmes, thanks to the deep semantic segmentation algorithm providing this possibility.