The northward subduction of the Paleo-Tethys oceanic crust in the Paleozoic to Mesozoic is critical for the tectonic evolution of the Qinling-Tongbai-Hong’an-Dabie-Sulu-Imjingang-Gyeonggi orogenic belt. However, the Paleozoic geological fingerprint of the Paleo-Tethys oceanic crust subduction in the Dabie-Sulu orogenic belt remains obscure. In the present study, apatite from the Muniushan monzogranite in the Houkuang area was analyzed to constrain the age of metamorphism in the Jiaobei Terrane and is regarded as the response to the Paleo-Tethys oceanic crust subduction in the Early Permian. Muniushan apatite with obvious negative Eu anomaly is enriched in LREE and depleted in HREE. The chondrite-normalized REE patterns of apatite correspond with I-type granitoids and mafic igneous rocks, implying a magmatic origin. Igneous apatite grains have reset compositional zonation in the cathodoluminescence image. Apatites plotted on a support vector machine apatite classification biplot and Eu/Y-Ce discrimination diagram shows a tendency from the region of “mafic igneous rocks and I-type granitoids” to “low and middle metamorphic”. This evidence consistently suggests that the Muniushan apatite suffered metamorphism at a later stage. The twenty-six apatite grains from the Muniushan monzogranite yield a metamorphic age of 297 ± 8 Ma (by LA-ICP-MS U-Pb), which is different from the Muniushan zircon SHRIMP U-Pb results of 2110 ± 4 Ma, indicating the metamorphism occurred in the Early Permian and reset the U-Pb system of apatite. The Early Permian metamorphism that occurred in the Jiaobei Terrane is synchronous to the subduction of the Paleo-Tethys oceanic crust and is the response to the Paleo-Tethys oceanic crust subduction.

Machine learning provides solutions to a diverse range of problems in high-dimensional datasets in geosciences. However, machine learning is generally criticized for being an enigmatic black box as it focusses on results but ignores the processes. To address this issue, we used supervised decision boundary maps (SDBM) to visually illustrate and interpret the machine learning process. We constructed a SDBM to classify the ore genetics from 1551 trace element data of apatite in various types of deposits. Attribute-based visual explanation of multidimensional projections (A-MPs) was introduced to SDBM to further demonstrate the correlation between features and machine learning process. Our results show that SDBM explores the interpretability of machine learning process and the A-MPs approach reveals the role of trace elements in machine learning classification. Combining SDBM and A-MPs methods, we propose intuitive and accurate discrimination diagrams and the most indicative elements for ore genetic types. Our work provides novel insights for the visualization application of geo-machine learning, which is expected to be a powerful tool for high-dimensional geochemical data analysis and mineral deposit exploration.

The alkaline complex in the southwest region of Luxi Terrane of the North China Craton is spatially correlated with the newly discovered Longbaoshan REE deposit. Its petrogenesis, however, remains ambiguous. In this study, we present an integrated petrology, whole-rock geochemistry, sphene U-Pb and rare earth element data from the Longbaoshan alkaline complex to investigate the petrogenesis, magma source and tectonic evolution. The Longbaoshan alkaline complex consists of mafic to intermediate rocks of hornblende diorite and alkaline hornblende syenite porphyry, biotite monzonite porphyry and aegirine diorite porphyrite. The hornblende diorites show a composition of low SiO2, high MgO, Fe2O3 and moderate Na2O, CaO and are metaluminous and medium-to-high-K calc-alkaline. The hornblende syenite porphyries, biotite monzonites and argirine diorite porphyrites display a relatively higher content of SiO2, Na2O, K2O and Al2O3 and lower contents of MgO, Fe2O3 and CaO and are metaluminous, peralkaline, high-K calcic-alkaline and shoshonite. The sphene U-Pb data shows that the parent magma of the hornblende diorite was emplaced at ca. 120 Ma. All these samples show a common depletion in Th, Nb-Ta and Zr-Hf and enrichment in large ion lithophile elements (e.g., Pb, Ba, Sr) and Light Rare Earth Elements. The magma may have experienced fractionation of pyroxene, amphibole, sphene, apatite and zircon during its evolution. The variable La content, La/Sm, Rb/Sr and (Ta/Th) N ratios indicate that the parent magma may produce by partial melting of a mantle source that was interacted with sediment-derived melts in a subduction setting. Therefore, we propose that the parent magma of the Longbaoshan alkaline complex was derived from a lithospheric mantle which was metasomatized by sediment-derived melt in a prior subduction process. The enriched magma was emplaced through an extension process and experienced subsequent fractionation and assimilation with the continental crust during the rollback of the Paleo Pacific Ocean plate.

No abstract