Search

Amidst the rapid advancement of artificial intelligence and information technology, the emergence of big data and machine learning provides a new research paradigm for mineral exploration. Focusing on the Tethyan metallogenic domain, this paper conducted a series of research works based on machine learning methods to explore the critical geochemical element signals that affect the metallogenic potential of porphyry deposits and reveal the metallogenic regularity. Binary classifiers based on random forest, XGBoost, and deep neural network are established to distinguish zircon fertility, and these machine learning methods achieve higher accuracy, exceeding 90%, compared with the traditional geochemical methods. Based on the random forest and SHapley Additive exPlanations (SHAP) algorithms, key chemical element characteristics conducive to magmatic mineralization are revealed. In addition, a deposit classification model was constructed, and the t-SNE method was used to visualize the differences in zircon trace element characteristics between porphyry deposits of different mineralization types. The study highlights the promise of machine learning algorithms in metallogenic potential assessment and mineral exploration by comparing them with traditional chemical methods, providing insights into future mineral classification models utilizing sub-mineral geochemical data. [ABSTRACT FROM AUTHOR]

Mantle metasomatism refers to the interaction between mantle melt, fluid, and mantle rock. It not only affects the physical and chemical properties of the lithospheric mantle but also plays an important role in the process of metal and gem mineralization. In order to explore the nature and evolution of metasomatism in the lithospheric mantle of the Mengyin area in the eastern part of the North China Craton, this paper combines the previous data of garnet inclusions in diamonds and analyzes the major and trace elements of garnet xenocrysts in the Shengli No. 1 kimberlite pipe from the EPMA and LA-ICP-MS experiments. The experiments show that the garnet xenocrysts of the Shengli No. 1 kimberlite pipe are mainly lherzolitic and harzburgitic garnets. The content of Zr and TiO2 in some garnets are low, which are the characteristics of depleted garnets. Conversely, another group of garnets display high Zr and TiO2 contents, indicative of high-temperature melt metasomatism. When comparing the Ti/Eu ratio of the depleted garnets to that of the primary mantle, a significantly lower value is observed. Additionally, the (Sm/Er)N value undergoes minimal changes, while the Zr/Hf value exceeds that of the primary mantle. These characteristics are indicators of carbonatite melt metasomatism. Garnets that are affected by high-temperature melt metasomatism exhibit low (Sm/Er)N content, a significant variation in the Ti/Eu ratio, and a Zr/Hf value greater than that of the primary mantle. These characteristics indicate the influence of kimberlite melt metasomatism. Garnets impacted by carbonatite melt metasomatism display a strong sinusoidal distribution pattern of rare earth elements (REE) and are often found as lherzolitic garnet xenocrysts and garnet inclusions in diamond. On the other hand, garnets influenced by kimberlite melt metasomatism exhibit a slight sinusoidal REE distribution pattern in harzburgitic garnets and a slight sinusoidal REE distribution or a flat pattern from medium rare earth elements (MREEs) to heavy rare earth elements (HREEs) in lherzolitic garnet xenocrysts. Based on these findings, it is evident that there are at least two types of metasomatism occurring in the lithospheric mantle of the Mengyin area in the eastern part of the North China Craton. The first type involves the metasomatism of early carbonatite melt to the mantle peridotite. Garnets formed under this condition exhibit high Sr and LREE contents, as well as low Zr, Hf, Ti, Y, and HREE contents, indicating depletion characteristics. The second type entails the metasomatism of late kimberlite melts affecting the mantle peridotite. Garnets formed under this process display high Zr, Hf, Ti, Y, and HREE contents. [ABSTRACT FROM AUTHOR]

The Daocheng batholith consists of granite, granodiorite and K-feldspar megacrystic granite, which is located in the north Yidun Arc. It is a barren batholith in contrast to plutons of the same age that contain major copper deposits, such as Pulang to the south. In the Daocheng, abundant mafic microgranular enclaves (MMEs) mainly developed within granodiorite and K-feldspar megacrystic granite, which are characterized by quenched apatite, quartz eyes and plagioclase phenocrysts. LA-ICP-MS zircon U–Pb dating of host granodiorite yielded ages ranging from 223 Ma to 210 Ma, with a weighted mean of 215.3 ± 1.8 Ma. Zircons from MMEs yielded ages ranging from 218 Ma to 209 Ma, with a weighted mean of 214.2 ± 1.4 Ma. Geochemical analyses show that granodiorite is high-K, calc-alkaline and I-type, with SiO2 contents ranging from 67.90% to 70.54%. These rocks are metaluminous to marginally peraluminous (A/CNK = 0.98–1.00) and moderately rich in alkalis with K2O ranging from 3.28% to 4.59% and Na2O ranging from 3.18% to 3.20%, with low MgO (1.08%–1.29%), Cr (12.7 ppm–16.8 ppm), Ni (5.19 ppm–6.16 ppm) and Mg# (35–49). The MMEs have relatively low SiO2 contents (56.34%–60.91%), higher Al2O3 contents (16.06%–17.98%), higher MgO and FeO abundances and are metaluminous (A/CNK = 0.82–0.83). The MMEs and host granodiorite are enriched in light rare-earth elements (LREEs) relative to heavy rare-earth elements (HREEs), with slightly negative Eu anomalies, and enriched in Th, U and large ion lithophile elements (LILEs; e.g., K, Rb and Pb), and depleted in high field strength elements (HFSEs; e.g., Nb, Ta, P and Ti), showing affinities typical of arc magmas. The zircon εHf(t) values (−6.28 to −2.33) and ancient two-stage Hf model ages of 1.92 to 1.25 Ga, indicating that the magmas are generally melts that incorporated significant portions of Precambrian crust. The relatively low silica contents and high Mg# values of the MMEs, and the linear patterns of MgO, Al2O3 and Fe2O3 with SiO2 between the MMEs and host granodiorite, showing the formation of MMEs are genetically related to magma mixing. The Daocheng granodiorite is characterized by much lower zircon Ce4+/Ce3+ (average of 3.53) and low fO2 value (average of ∆FMQ = –10.84), whereas the ore-bearing quartz monzonite porphyries in the Pulang copper deposit are characterized by much higher zircon Ce4+/Ce3+ (average of 52.10) and high fO2 value (average of ∆FMQ = 2.8), indicating the ore-bearing porphyry intrusions had much higher fO2 of magma than the ore-barren intrusions considering that the high oxygen fugacity of the magma is conducive to mineralization. [ABSTRACT FROM AUTHOR]