The locations of minerals and mineral-forming environments, despite being of great scientific importance and economic interest, are often difficult to predict due to the complex nature of natural systems. In this work, we embrace the complexity and inherent "messiness" of our planet's intertwined geological, chemical, and biological systems by employing machine learning to characterize patterns embedded in the multidimensionality of mineral occurrence and associations. These patterns are a product of, and therefore offer insight into, the Earth's dynamic evolutionary history. Mineral association analysis quantifies high-dimensional multicorrelations in mineral localities across the globe, enabling the identification of previously unknown mineral occurrences, as well as mineral assemblages and their associated paragenetic modes. In this study, we have predicted (i) the previously unknown mineral inventory of the Mars analogue site, Tecopa Basin, (ii) new locations of uranium minerals, particularly those important to understanding the oxidation-hydration history of uraninite, (iii) new deposits of critical minerals, specifically rare earth element (REE)- and Li-bearing phases, and (iv) changes in mineralization and mineral associations through deep time, including a discussion of possible biases in mineralogical data and sampling; furthermore, we have (v) tested and confirmed several of these mineral occurrence predictions in nature, thereby providing ground truth of the predictive method. Mineral association analysis is a predictive method that will enhance our understanding of mineralization and mineralizing environments on Earth, across our solar system, and through deep time.

Breyite is the second most abundant mineral inclusion in super-deep diamonds after ferropericlase. Though breyite stability extends to 300 km along typical mantle geotherm, this phase is often assumed to be the product of retrograde transformation of CaSiO3-perovskite, and thus has the potential to retain informa-tion from as deep as 800-1000 km. In this study, we determined the depth of formation of a breyite inclusion still enclosed in its host diamond from Juina, Brazil, by X-ray diffraction. The measured >5 % smaller unit cell for breyite indicates a stored residual pressure showing that the breyite was entrapped between about 9(1) and 10(1) GPa. These are the highest estimates of formation pressure ever determined for a breyite inclusion. For ambient mantle temperatures higher than 1400-1500 degrees C, these pressures would exceed the maximum P of the breyite stability field. Breyite in this diamond cannot be primary but is rather a back -transformation product from CaSiO3-perovskite formed in the transition zone or the lower mantle. The co-existence magnesite in diamond JU55 and the slab -association of sublithospheric diamonds is evidence of carbon transport to lower mantle depths.

Two karrooite crystals, one with a disorder parameter (X = Ti content in the M1 site) of 0.070(5) and the other with X = 0.485(5), were mounted together in one diamond-anvil cell and studied by single-crystal X-ray diffraction at several pressures up to 7.51 GPa. The most noticeable effect of increasing cation disorder on the high-pressure behavior of the structure is to increase the compressibilities of the mean < M12-O > bond length from 0.00148(2) GPa(-1) in the ordered sample to 0.00163(7) GPa(-1) in the disordered one and decrease those of the mean < M1-O > bond length from 0.00243(5) to 0.00193(12) GPa These changes are responsible for the compressibility difference between the two phases observed by Hazen and Yang (1997). Both compressibilities of the mean < M-O > bond lengths and the octahedral volumes in two phases decrease linearly with increasing the Ti contents in the octahedral sites. All octahedra in two samples become less distorted as pressure increases, but those in the more disordered structure exhibit larger decreases in terms of the octahedral angle variance than the corresponding ones in the more ordered structure. The influence of pressure on the interatomic angles is small compared to the interatomic distances, suggesting that compression of the karrooite structure is controlled primarily by the bond-length shortening, rather than by bend-angle bending. The strong compressional anisotropy of the structure is a consequence of the differential compressibilities of the weaker Mg2+-O and stronger Ti4+-O bonds and the complex edge-sharing linkage involving the M1 and M2 octahedra.