The objective of this study was to establish a methodology for building a 3D geological model of a mineral vein deposit, encompassing morphology and wolframite content. The available data set includes stope data (vein thickness and wolframite quantity) and borehole data (vein thickness and wolframite modal classes). The data captured from boreholes and stopes differ in terms of their spatial distribution and clustering as well as the size and shape of each sample. Therefore, a specific

more »

... gy was designed to combine these two sources of information. The morphology model mapped vein thickness as a vein proportion variable, by applying a two-phase Direct Sequential Simulation (DSS) locally conditioned to borehole data. Regarding the evaluation of wolframite content, the variables are unable to be related directly to each other and are non-co-located. Therefore, a proximity study was made using DSS to build local conditional cumulative histograms (by borehole class and by mine level). The final model of wolframite quantity was generated using Probability Field Simulation. The proposed approach considers the initially identified problems, allowing the mineral potential of the deposit to be quantified by integrating the results of the two independent methodologies. Minerals 2017, 7, 234 2 of 33 The morphology modelling consists of characterizing the proportion of veins in each grid cell. In contrast, the wolframite content modelling spatially quantifies a variable that represents the wolframite quantity by vein fraction (kg/m 3 ). Both models, morphology and wolframite content, were built independently and then subsequently merged to quantify the mineral resource (tonnes of wolframite). The design of the two methodologies involved the use of a data-driven approach, conditioned to all available information (stopes and boreholes). Data from stopes and boreholes are typically obtained using different protocols and report designations and exhibit different spatial tendencies. The particular characteristics of the data in the present study meant that the use of conventional geostatistical methods was unfeasible, making it necessary to consider and develop alternative modelling approaches. The approach developed for both models, morphology (vein proportion) and wolframite content (wolframite quantity in the vein volume), was a stochastic simulation type. The approach proved to be very suitable for integrating data from boreholes and stopes, considering all the conditions mentioned above. Stochastic simulation-based methodologies generate different scenarios with the same probability of occurrence, thereby allowing estimates to be made of the variable local uncertainty as well as an average scenario [3,4]. Therefore, stochastic-simulation-based methodologies are suitable for quantifying the potential of a mineral deposit. Materials and Methods Geological Background The mine of interest is located in the tin-tungsten metallogenetic province of the central and northern areas of Portugal [5]. The tin and wolframite mineralizations are related to granite intrusions and shearing [6,7]. Locally, the mine is contained within the Schist Greywacke Complex belonging to the Beiras Group of the Central Iberian Zone (Portugal). The mine intersects various units including schist, greisen, and dolerites, with the two main lithologies being clay schist and mottled clay schist. According to Thadeu [8], the upper Pre-Cambrian [6] clay schist and mottled clay schist units have the highest probability of mineralization occurrences. In the area of the mine, many magmatic bodies have been identified, including sub-vertical dolerite dikes, near-horizontal quartz veins, and a non-exposed greisened granite intrusion [9] (Figure 1) . A large number of structures have also been identified, including folds, faults, and joints. Thadeu [9] reports three sets of faults: a N-S system, a NW-SW system, and an ENE-WSW system. The ore deposit occurs within a near-horizontal joint system that hosts the most important wolframite and tin mineralizations. This joint system originated during the intrusion of the granite dome.