Advanced proxies for provenance, erosion and transport mechanisms of modern stream sediments – An application of SEM-based quantitative mineralogical analysis

In order to analyse stream sediments for provenance with respect to erosion and transport mechanisms, several methods are established (e.g. bulk sediment geochemistry, mineralogy (provided by XRD) and indicator mineral analysis). In this study, we make use of automated mineralogy by Mineral Liberation Analyser (MLA) as a potentially tool to advance sediment provenance studies.
The MLA combines backscattered electron (BSE) imaging with energy dispersive X‐ray spectrometry (EDS) generating compositional data each sediment particles of the sample. The provided data include particle as well as mineral grain parameters (i.e. size and shape) as well as the mineralogical composition and properties (e.g. elemental composition, density) of each particle (including individual constituting mineral grains). The aim is to join the provided parameters in a holistic model including statistical automatisms. In order to ensure a valid combination of the heterogenic and voluminous set of data provided by MLA, robust statistical analyses are needed. These statistical analyses unveil trends and dependencies in suites of related samples. Furthermore, in this study bulk geochemistry and XRD measurements are integrated to guaranty the quality of the introduced method and subsequently to critically assess the benefit of the measurement.
The study area is located in the Vogtland region of the Free State of Saxony (Germany). The variscan bedrocks comprise plutonic (i.e. different types of granite) and metamorphic units (mica schists, phyllites and quartzites), which are very well studied. Especially since, the Vogtland and the neighbouring Erzgebirge are well known for the occurrence of granite‐related mineral systems, represented as polymetallic deposits (skarn‐, vein‐, stockwork‐, and greisen‐type). In addition, this area is menial populated, suggesting a restricted anthropogenic contaminations of the stream sediment.
First results of this study, give rise for a clear improvement in the detection of lithological changes of the source rock composition and transport features of the unconsolidated sediments. This can be easily identified, based on the modal mineralogy, geochemical changes and grain‐parameter patterns. In addition, mixing of the material can be calculated with respect lithological changes along the river path. Another issue, is to detect the anthropogenic contamination of the sampled material and to balance the impact to the chemical composition.