A combination of TIN meshing, geometric extrapolation and linear interpolation for deep ore bodies modeling: Case of Sulcis Coal Basin (Sardinia, Italy)

The realization of any kind of environmental and/or mining project is not possible without a strong knowledge of the main features of the geological body involved in. Usually the evaluation of these characteristics is based on reality models built on a limited set of data deriving from different survey campaigns. Consequently, the decision-making and the planning processes are based on models that are, obviously, affected by errors. The seriousness of these errors is closely related to the modeling technique, to the data accuracy, to the approximations that have been taken to build them and, especially, to the real variability of the ore body inhomogeneity inside the study domain. Therefore from an engineering point of view the consequences can be very dangerous especially where, because of the presence of geological noises elements (i.e. tectonic faults and folds), it is not possible to apply directly a statistical and/or a geostatistical interpolation technique (because of the limitations of these techniques) without producing high errors of evaluation (like could happen in the target area of this study). For this reason, in these cases, the only possible solution is proceeding with a preliminary analysis of the available data and of the environmental/geological features in order to define the most appropriate modeling technique to hold the evaluation errors down. In this paper, exploiting GIS analysis tools, an innovative 3D modeling technique based on a combined use of TIN meshing, geometric extrapolation and linear interpolation will be presented and its results on the Sulcis Coal Basin (Sardinia, Italy) will be introduced and discussed.