Strong acid mine drainage (AMD) processes in the flooded, formerly open pits in the Lusatia area present an enormous environmental challenge for the rehabilitation of the post-mining landscape. Extensive and costly monitoring is required for optimal AMD management and remediation planning and control. Because of the large size of the area and the dimension of the problem, the regular sampling can only provide limited point data, which needs to be extrapolated to the entire area. Consequently, the search for effective approaches for extrapolating the point data to the area of all water bodies is essential for rehabilitation success monitoring and for understanding the dependencies between AMD and environmental factors such as land use, weather conditions, geology, and hydrogeology. The main aim of this study was to investigate the suitability of Sentinel-2 multispectral imagery and artificial neural networks (ANNs) for the quantitative mapping of acid mine drainage (AMD) constituents, such as dissolved iron, pH value, and sulfate in large water bodies, for an area of approximately 7220 km2 (the area of the pit lakes is about 185 km2). Correlations between different chemical water parameters were also investigated. An extensive water monitoring dataset was used to train artificial neural networks for the identification of dependencies between the multispectral remote sensing data and the water quality ground measurements. Respective relationships have been identified, especially for dissolved iron and pH. These trained ANNs have been used to produce water quality maps with high spatial (10 × 10 m) and temporal (any cloud-free period) resolution, which show the wide variability of water quality in the different parts of the mining region. Concrete sources of AMD can be identified using the water quality maps of single lakes, and the success of sanitation measures such as liming was visualized. The approach opens many doors for the optimization of both the monitoring program and sanitation technology.

To ensure Europe increases its domestic production of high quality and responsibly produced raw materials, the development of innovative technologies for 3D geological modeling in mineral exploration is paramount. The Erzgebirge in Germany provides an excellent framework to showcase the application of artificial intelligence and in particular Artificial Neural Networks (ANN) for 3D mineral prospectivity mapping. The Erzgebirge belongs to the Variscan Belt, withholding 800 years of mining history and it is also famous for Ag, Sn, W, Fe, Cu, Li mineralizations among others. The Bockau deposit is located at the western section of the Erzgebirge. The target area is a Paleozoic metasediment body that was formed during the Variscan orogeny. The metasediment body consists primarily of alternating micaschist, phyllite and quartzite and dips mostly 25° to 240° SW. The metasediment is surrounded by Late Variscan plutons which partly led to contact metamorphic zones. In addition there is a large Quartzite body which was mined near to the surface in the 17th century for Sn, following a stratiform tin anomaly which can reach up to 4000 ppm Sn. Thanks to the long mining history, the Bockau deposit condenses a large amount of geological, geochemical, geophysical and mineral data. To increase mineralogical knowledge of the deposit and to help identify drilling targets, a hybrid approach for 3D mineral predictivity mapping is implemented. Potentially mineralisation-controlling factors are identified in knowledge-driven genetic exploration models, taking into account the borehole data, major faults, electromagnetic data, intrusive bodies, contact metamorphic zones and lithological borders, followed by data-driven weighted ANN predictive modelling implemented in the in-house developed advangeo® 3D Prediction Software. The predictive model is guided by structural variables such as the euclidean distance to fault planes, lithological surfaces and to metamorphic contact zones. The model is also constrained by geophysical data by a magnetic susceptibility model obtained from an airborne magnetic data inversion. Finally, Sn anomaly data from boreholes is implemented as training data for the prediction. The results show the probability distribution of Sn mineralisation occurrence in 3D over a voxel model formed by blocks of approximately 684 m3 13(x), 13.5(y) and 4(z), increasing the mineralogical knowledge of the deposit and guiding exploration efforts complementing the decision making process for drilling new targets. The results are validated by iteratively implementing the jackknife method, splitting the training data into validation and training subsets. The first prediction iteration is performed with a subset containing 77 % of the Sn content data from boreholes as training data, followed by 50 and 30 % subsets. Thus, allowing at each iteration to perform a quantitative evaluation of the prediction by comparing the validation subset with the Sn content of the borehole that was not used for the prediction. The paper has been prepared in the frame of the Horizon 2020 co-funded project GOLDENEYE, which has received funds through the Grant Agreement 869398.

Lithium (Li) was recently added to the list of critical raw materials by the European Union due to its significance for the green energy transition. Thus, the development of new toolchains to make Li exploration more economic and more effective is needed. Stream sediment analysis can play an important part in these new tool chains. In this work, two historical stream sediment datasets covering parts of the Fregeneda-Almendra pegmatite field in the Douro region (Portugal) were reprocessed considering two distinct approaches: spatial interpolation through inverse distance weighting (IDW) and the catchment basin approach using the concentration area (C-A) fractal analysis. The following objectives were delineated: (i) determine pathfinder elements for Li, considering relevant associations in the mineralization sources; (ii) compare the performance of both approaches; (iii) identify new target areas for Li. In the case of spatial interpolation, the highest Li values were associated to granites although the use of key elements allowed lithological discrimination and the delineation of target areas. In the catchment basin approach, fractal analysis proved to be effective in decreasing the number of areas of interest with high accuracy (>75%) when considering the previously mapped Li-pegmatites. One of the limitations identified was the number of anomalous basins related to the granites, despite the use of pathfinder elements allowing discriminating granite- from pegmatite-related Li anomalies. Comparing the two approaches, the spatial interpolation method is more adequate for the early stages of exploration (reconnaissance), while the catchment basin approach is more suited for prospect-scale exploration. Field validation of the results identified one pegmatite containing Li mineralization and three others with favorable signs for Li mineralization in the Douro region.

The growing importance and demand of lithium (Li) for industrial applications, in particular rechargeable Li-ion batteries, have led to a significant increase in exploration efforts for Li-bearing minerals. To ensure and expand a stable Li supply to the global economy, extensive research and exploration are necessary. Artificial neural networks (ANNs) provide powerful tools for exploration target identification. They can be cost-effectively applied in various geological settings. This article presents an integrated approach of Li exploration targeting using ANNs for data interpretation. Based on medium resolution geological maps (1:50,000) and stream sediment geochemical data (1 sample per 0.25 km2), the Li potential was calculated for an area of approximately 1200 km2 in the surroundings of Bajoca Mine (Northeast Portugal). Extensive knowledge about geological processes leading to Li mineralisation (such as weathering conditions and diverse Li minerals) proved to be a determining factor in the exploration model. Furthermore, Sentinel-2 satellite imagery was used in a separate ANN model to identify potential Li mine sites exposed on the ground surface by analysing the spectral signature of surface reflectance in well-known Li locations. Finally, the results were combined to design a final map of predicted Li mineralisation occurrences in the study area. The proposed approach reveals how remote sensing data in combination with geological and geochemical data can be used for delineating and ranking exploration targets of almost any deposit type.

Self-organizing maps (SOM) are a useful tool to analyze and interpret gridded datasets like potential field or stream sediment geochemistry data. The data are transformed from geographic space to SOM space where they can be clustered according to overall similarity. By transforming the clusters back to geographic space, geological interpretation of the clusters is facilitated. We present the application of a multilayer perceptron (MLP) in SOM space to produce mineral predictive maps. The reduced number of grid cells in SOM space greatly enhances the performance of the MLP and the tolerance to noise in the input data, compared to an application of the MLP to the original data. The method is applied to tin skarn deposits in the German part of the Erzgebirge. The training and validation data required for the MLP are compiled from mining and exploration records. The input data for the SOM are reprocessed gravimetric, magnetic, stream sediment geochemistry, geologic and tectonic data sets. Potentially ore-controlling spatial relationships, such as the distance to different types of partly covered granite intrusions, are derived from a regional scale 3D geological model. The resulting mineral prediction map allows the definition of exploration zones for detailed studies. The paper has been compiled in the frame of "NEXT - New EXploration Technologies" project. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 776804.

Artificial neural networks (ANN) are a comprehensive data-driven modelling approach for creation of mineral predictive maps. Based on a “self-learning” process, this technology can be used to interpret almost any geo-scientific data for generation of both qualitative (prediction of locations) and quantitative (prediction of grades, tonnages) predictive maps. By analysing the footprints of known mineralisations in the framework of available geo-scientific data, the approach generates trained ANN`s that are further used to generate predictive maps.

Artificial Neural Networks (ANN) are used for statistical modeling of spatial events in geosciences. The advantage of this method is the ability of neural networks to represent complex interrelations and to be “able to learn” from known (spatial) events. The software advangeo® was developed to enable GIS users to apply neural network methods on raster geodata. The statistic modeling results can be developed and displayed in a user-friendly way within the Esri ArcGIS environment. The complete workflow is documented by the software. This chapter presents five case studies to illustrate the current possibilities and limitations of spatial predictions with the use of artificial neural networks, which describe influencing factors and the selection of known events of the phenomenon to be modeled. These applications include: (1) the prognosis of soil erosion patterns, (2) the country-wide prediction of mineral resources, (3) the vulnerability analysis for forest pests, (4) the spatial distribution of bird species, and (5) the spatial prediction of manganese nodules on the sea bottom.

Artificial neural networks (ANN) are used for statistical modeling of spatial events in geosciences. The advantage of this method is the ability of neural networks to represent complex interrelations and to be “able to learn” from known (spatial) events. The software advangeo® was developed to enable GIS users to apply neural network methods on raster geodata. This statistic modeling can be displayed in a user-friendly way within the ESRI ArcGIS environment. The complete workflow is documented by the software. This paper presents three pilot studies conducted to illustrate the possibilities of spatial predictions with the use of existing raster datasets, which described influencing factors and the selection of known events of the phenomenon to be modeled. These applications included (1) the prognosis of soil erosion patterns, (2) the prediction of mineral resources, and (3) vulnerability analysis for forest pests.

We have studied the regolith landform distribution in the area of Gaoua, western Burkina Faso, using an integration of geophysical and remote sensing data. Concentration maps of K, Th, U, as well as their ratios, were computed from airborne gamma-ray spectrometry data to assess the geochemical composition of the regolith. The mineralogy of the surfaces was mapped via the analysis of multispectral ASTER and Landsat scenes. Pauli-decomposition data retrieved from polarimetric ALOS PALSAR and Radarsat-2 images were included to characterize the surface properties of the regolith material. Morphometric variables such as slope, curvature, and relative relief were derived from the SRTM digital elevation model to quantify the topographic parameters of the different regolith landforms. An artificial neural network implementation, ADVANGEO, was then employed to extract four basic regolith landform units from the satellite and airborne data. Relic ferruginous duricrusts rich in hematite and goethite belonging to the High glacis, erosional surfaces represented by rock outcrops and suboutcrops, alluvial sediments, and soft pediment materials of the Middle and Low glacis were mapped successfully in the region. The results were compared with the existing geomorphological maps, an independent visual classification, and field observations. We found that the distribution and shape of the iron-rich duricrusts are more accurate than portrayed in the current maps. The best results, with an overall accuracy of 94.21% and a kappa value of 0.92, were obtained for a dataset consisting of gamma-ray spectrometry data combined with derivatives of the SRTM digital elevation model augmented by Landsat, and polarimetric radar data. The approach demonstrates for the first time the potential of machine learning in regolith landform mapping. The proposed combined analysis of airborne geophysics and remote sensing data can be adopted easily in other regions with similar long-term lateritic weathering histories worldwide.