LocatOre: pathways to mineral discoveries through computer-based modelling of geochemical data – case studies of chlorite vectoring at the Batu Hijau and Ujina porphyry copper deposits

Using chlorite analyses from the Batu Hijau (Indonesia) and Ujina (Chile) porphyry copper deposits, we illustrate the effectiveness of LocatOre, a new software application, to aid geochemical exploration for porphyry systems. Using mineral chemistry data generated by laser ablation inductively coupled plasma mass spectrometry analysis, LocatOre facilitates knowledge-driven and data-driven analyses that help users to predict deposit locations from broadly distributed samples. Knowledge-driven analysis of spatial variations in chlorite trace element ‘proximitor’ ratios (e.g. Ti/Sr) can be used to initially estimate potential deposit locations and to identify their related alteration domains using 2D and 3D modelling features. Data-driven analysis can then refine predictions of likely deposit locations in each domain. The software simulates multiple potential fluid sources in each region of interest to identify those that represent the most likely location of the causative intrusion. Linear regression modelling is used to predict distances of each sample from the intrusion using spatial variability in chlorite compositions. Potential heat sources are filtered according to modelled R 2 values to highlight strong correlations between chemical and spatial variables. For regression modelling, the locations of both porphyry deposits were predicted correctly using data-driven analysis, with R 2 values ≥0.79 and ≥0.54 for sources generated using datasets from 70 and 34 chlorite-altered samples from Batu Hijau and Ujina, respectively.