Understanding how new felsic crust is formed and subsequently evolves through time is critical to identifying the geodynamic regimes that have dominated various parts of Earth history, and have important implications for feedbacks between the lithosphere and biosphere, such as controlling the influx of continental detritus into the oceans. In recent years, several trace element-based geochemical proxies have been proposed to allow determination of paleo-crustal thicknesses, which have been calibrated primarily using data collected from modern-day arcs. The application of these proxies through deep time has revealed surprising results, including the suggestion that the mid-Proterozoic continents were atypically thin compared to those in the Archean and the Phanerozoic; however, a range of factors may influence commonly cited trace element ratios (e.g. Sr/Y) rather than just crustal depth, leading to additional and unexpected magnitudes of uncertainty. Here we perform geochemical modelling to deduce the effect of variable bulk-rock composition and geothermal gradient on the trace element signature of felsic melts generated in arc systems. Using a range of protoliths representative of deep arc crust, the results show that considerable care must be taken when analysing simple trace element ratios of granitoid melts and making direct interpretations of the pressure of crystallisation. In particular, changes in geothermal gradients and differences in arc basalt composition impart strong controls on the relative stability of garnet and plagioclase during metamorphism and partial melting, and wide ranges of Sr/Y and La/Yb may be produced in derivative felsic melts produced at the same crustal depth. The interpretation of mid-Proterozoic continental arcs being atypically thin may instead be an artefact of underestimation of the active geothermal gradient at the time of magma formation, which acts to reduce Sr/Y and La/Yb ratios, even in normal thickness (∼35–40 km) crust. Furthermore, we argue that the potentially garnet-free residua during the formation of mid-Proterozoic felsic magmas points to crust formation without lower crustal foundering, and thus, that this commonly invoked paradigm for formation of the continental crust may only be applicable to certain periods of Earth history.

中文翻译：

---

地壳熔化过程中石榴石的稳定性：对弧岩浆作用和大陆形成的化学测压和长期变化的影响


了解新的长英质地壳是如何形成以及随后随时间演化的，对于确定主导地球历史各个部分的地球动力学状况至关重要，并且对岩石圈和生物圈之间的反馈具有重要影响，例如控制大陆碎屑流入地球。海洋。近年来，已经提出了几种基于微量元素的地球化学代理来确定古地壳厚度，这些厚度主要使用从现代弧收集的数据进行校准。这些代理在深层时间中的应用揭示了令人惊讶的结果，包括认为中元古代大陆与太古宙和显生宙的大陆相比异常薄；然而，一系列因素可能会影响通常引用的微量元素比率（例如 Sr/Y），而不仅仅是地壳深度，从而导致额外的和意想不到的不确定性。在这里，我们进行地球化学建模，以推断可变的块体岩石成分和地温梯度对电弧系统中产生的长英质熔体的微量元素特征的影响。使用代表深弧地壳的一系列原岩，结果表明，在分析花岗岩熔体的简单微量元素比率和直接解释结晶压力时必须非常小心。特别是，地温梯度的变化和弧形玄武岩成分的差异对石榴石和斜长石在变质作用和部分熔融过程中的相对稳定性产生了强有力的控制，并且在衍生的长英质熔体中可能会产生大范围的Sr/Y和La/Yb。相同的地壳深度。 对中元古代大陆弧异常薄的解释可能是低估了岩浆形成时活跃地温梯度的人为因素，即使在正常厚度下（∼35 –40 公里）地壳。此外，我们认为中元古代长英质岩浆形成过程中可能不含石榴石的残留物表明地壳形成没有下地壳沉没，因此，这种普遍引用的大陆地壳形成范式可能只适用于某些时期地球历史。