Answers are grounded in this paper and cite their sources.
Continental crustal differentiation in subduction zones transforms primitive mafic material into intermediate to felsic compositions through partial melting and magma interaction. The Neoproterozoic Jinping Complex in the Ailaoshan region, southwestern China, preserves an exceptional record of transcrustal magma evolution spanning ca. 783–697 Ma that provides quantitative constraints on crustal melting within the MASH (melting, assimilation, storage and homogenization) framework. Using binary melt mixing tools, this paper explains how a range of distinct magmatic rocks formed within crust exposed to prolonged mantle magma intrusion and heat transfer. Mantle sources metasomatized by slab-derived melts and fluids are inferred to have generated a compositional range of juvenile mafic lower crust. Melting of these rocks with residual garnet at ≥1.2 GPa, with best-fitting endmembers at 800–950 °C and H2O amounts of 1.0–1.3 wt.%, produced high Ba–Sr porphyritic granites with juvenile isotopic signatures. In contrast, muscovite-poor Group B monzogranites are best explained by the mixing of melts generated at 0.6 and 1.4 GPa, 875–950 °C and H2O amounts of 1.0–1.3 wt.% from mafic crust derived from slab-melt-modified mantle. For both suites, mixing melts from different protoliths improves the fit more than varying the melting conditions of one protolith. Gneissic granites record magma mixing, with their isotopic signatures indicating a contribution from evolved crust. Their felsic members are best reproduced by mixing melts generated at 0.6 and 1.2 GPa and 950 °C, with H2O amounts of 1.3 and 3 wt.%, respectively. Muscovite-rich Group A monzogranites, by contrast to the other rocks, reflect reworking of intermediate to felsic crust. Combining the constraints on the source composition and melting conditions for these different rock types reveals how repeated mantle magma intrusion and heating drove crustal melting of a variety of rocks and magma interaction.
No takes yet. Share an insight, caveat, or question.
Huang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: