Deep Subduction in Earth history: Seeking for traces in the sedimentary record
by Jan Schönig
Date of Examination:2021-12-14
Date of issue:2022-02-24
Advisor:Hilmar von Eynatten
Referee:Hilmar von Eynatten
Files in this item
This file will be freely accessible after 12.12.2022.
EnglishThe global-scale horizontal movement of tectonic plates driven by the sinking of cold and dense lithosphere, known as plate tectonics, is a major process linking the Earth’s surface with mantle. Thus, plate tectonics strongly affect geochemical cycles, mantle convection, crustal growth rates, as well as thermal and tectonic regimes. Numerous hints, particularly from the detrital zircon record, indicate that plate tectonics operate since at least the Archean–Proterozoic transition, although opposing views persist. How plate tectonics evolved on a global-scale from the onset to the modern-style regime, which includes cold and deep subduction defined by low-temperature/high-pressure and ultrahigh-pressure (UHP) metamorphism, is a highly controversial and hotly debated topic in Earth Sciences. Particularly the lack of blueschists and UHP rocks from the pre-Neoproterozoic record represents a key argument for a late onset of modern-style plate tectonics. A main limitation in the search for modern-style plate tectonic regimes in deep time is the fragmentarily preserved crystalline rock record that may or may not be representative for the respective time interval, and the virtual absence of techniques to make use of the sedimentary record that would enable to more representatively investigate subduction regimes through time on a global scale. Recently, a novel method was introduced to trace the erosion of UHP metamorphic rocks by screening mineral inclusion assemblages of detrital garnet for the presence of coesite, and thus potentially opening new avenues to seek for the operation of deep subduction in deep time. However, these findings are restricted to a small catchment in the Western Gneiss Region of Norway, raising some fundamental question to be addressed before applying the concept to ancient sediments. This thesis applies the novel detrital approach to two Phanerozoic orogens, demonstrating that: (i) mineral inclusions in detrital garnet are capable to record UHP rock occurrences, also for regional river catchments; (ii) besides coesite, diamond-grade rocks effectively transfer UHP signatures to the sedimentary record in the form of diamond inclusions; and (iii) combining the information from mineral inclusions and garnet chemistry provides new insights regarding the UHP rock cycle of the study areas. For the central Saxonian Erzgebirge of Germany, this includes evidence for a much wider extent of UHP metamorphism than previously assumed and the involvement of mafic as well as felsic rocks in the UHP cycle. This includes felsic country-rock gneisses that underwent partial melting and re-equilibration during exhumation and have previously been supposed to have reached peak conditions below the coesite stability field. It is concluded that previously described UHP lenses and the surrounding country rocks were subducted as a largely coherent slab, which has important implications to understand the buoyancy development of the subducting/exhuming continental crust of UHP terranes. For the D’Entrecasteaux metamorphic complex of Papua New Guinea, detrital garnet chemistry, coesite co-existing with graphite inclusions, melt inclusions, and elastic geothermobarometry reveal, for the first time, a complete UHP rock cycle starting with a metasedimentary protolith that originated from the Earth surface, deep subduction to UHP conditions, exhumation under increasing temperature conditions, and erosion to form the studied beach placer. In addition, for the Erzgebirge, the large number of monomineralic coesite inclusions and coesite inclusions that partially transformed to quartz enabled a detailed investigation of preservation factors, showing that: (iv) a small size <9 µm and a low frequency of coesite inclusions favors the garnet host to stay intact in spite of inclusion overpressures developing during exhumation, and thus coesite inclusions are shielded from external conditions and fluids, which enables their monomineralic preservation; (v) bimineralic coesite/quartz inclusions ruptured their host garnet at a late stage during exhumation at temperatures of ~330 °C; and (vi) the heterogeneous grain-size distribution of detrital coesite-bearing garnet can be explained by inclusion frequency. Thereby, mafic and felsic UHP garnets are initially large, but mafic garnet contains a low number of inclusions resulting in minor disintegration and enrichment in the coarse fraction, while felsic garnet contains variable amounts of inclusions, whereby coesite-poor grains are enriched in the coarse fraction and coesite-rich grains extensively disintegrated into smaller fragments resulting in an enrichment in the fine fraction. Furthermore, the thesis presents several technical advancements, which include that: (vii) the 250–500 µm grain-size fraction is most efficient in terms of analytical time to invest compared to information value to gain; (viii) based on a newly developed discrimination scheme using a large database and a machine-learning algorithm, garnet chemistry represents an efficient tool to pre-screen and pre-select grains ahead of the time-consuming inclusion analysis; and (ix) hyperspectral Raman imaging provides an alternative to reduce the user-assisted analytical time. In main conclusion, analyzing mineral inclusions in detrital garnet represents a robust and efficient approach to capture the distribution and characteristics of UHP rocks exposed at the surface at the time of sediment generation and deposition. The methodological framework is mature and has a high potential to tackle the issue whether modern-style plate tectonics operated on a global scale in pre-Neoproterozoic times.
Keywords: Raman spectroscopy; Provenance; Garnet; Ultrahigh-pressure metamorphism; Mineral inclusions; Plate tectonics