Quantitative Geochemical Provenance of Sandstones in Coastal Basins: Field Study

Thesis Abstract

Quantitative geochemical provenance analysis of sandstones within coastal basins addresses the critical question of sediment source-to-sink pathways under dynamic sea-level and tectonic forcings. The study targets improving resolution of provenanced end-member contributions and sediment redistributions that shape reservoir quality in shallow to marginal marine settings. The aim is to quantify multiple provenance signals and establish robust mixing models that link sandstone compositions to hinterland sources while accounting for diagenetic overprints and tectonic setting. Specific objectives are (i) to characterize mineralogical and major- and trace-element composition of sandstone samples (n = 210) across three coastal basins using X-ray fluorescence (XRF) spectrometry and X-ray diffraction (XRD); (ii) to determine detrital zircon U-Pb ages (n = 120 grains) and Hf isotopic compositions to constrain source terrane contributions; (iii) to apply quantitative provenance techniques including end-member normalization, principal component analysis (PCA), mixing models, and multivariate regression to apportion source contributions; (iv) to evaluate the influence of weathering intensity, sedimentary transport distance, and tectonic uplift on element geochemistry through regression and analysis of variance (ANOVA); and (v) to develop a basin-scale provenance framework applicable to similar coastal systems. The methodology adopts a field-based, cross-section sampling strategy across three coastal basins, with stratigraphic control and documented lithofacies. The population comprises siliciclastic sandstones of late Cretaceous to Miocene age. A stratified random sampling approach yields representative samples from 35 outcrops, with duplicate samples for quality assurance, resulting in a dataset of 420 sandstone samples for XRF/XRD and 210 for microprobe-based minor elements. Detrital zircon grains are extracted from 12 representative sandstones, yielding 120 grains for U-Pb dating and Hf isotopes. Data collection combines laboratory-based geochemical analyses and in situ geochronology. XRF determines major and trace element concentrations; XRD identifies modal mineralogy and clay transitions; laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) provides trace element fingerprinting; LA-ICP-MS zircon analysis yields U-Pb ages and Hf isotopic ratios. Provenance modeling integrates end-member mixing, PCA, and discriminant function analysis, followed by regression-based evaluation of environmental and tectonic correlates. The study employs the theoretical framework of detrital geochronology and crustal evolution theory, with relevant terrestrial end-member concepts such as Apachean-type and nappe-related sources, and the cosmopolitan principle of incremental sediment addition through riverine and coastal processes. Validity and reliability are ensured through standardized sample preparation, duplicate analyses, cross-laboratory calibration, and blind duplicates. Data analysis uses descriptive statistics for compositional ranges, PCA for dimensionality reduction, and hierarchical clustering to identify source affinities. End-member mixing models estimate relative contributions from paleoisland, continental shield, and orogenic belt sources, validated against the U-Pb age spectra and Hf isotopic signatures. Regression analyses test hypotheses linking provenance indicators with basin tectonics, climate-driven sediment supply, and shoreline progradation rates; ANOVA assesses differences among basins and facies. Ethical considerations include transparent handling of samples, data provenance, and adherence to field collection permits. Key expected findings include (i) a dominant mix of hinterland sources with varying contributions across basins, (ii) coherent detrital zircon age spectra that reflect multiple tectonic blocks, (iii) robust correlations between major/trace element ratios (e.g., Th/Sc, La/Sc, Cr/Th) and source terrain lithologies, and (iv) discernible impacts of tectonic uplift and climate on sediment routing and diagenetic overprinting. The study contributes to knowledge by refining quantitative provenance methodologies for coastal sandstone systems, integrating detrital zircon geochronology with conventional geochemical fingerprints to improve source-to-sink inferences, and providing a transferable framework for petroleum and hydrogeological applications in similar coastal basins. The main conclusion anticipates that coastal basins record complex, multi-source contributions with differential preservation of geochemical signals due to diagenesis and transport, necessitating a multi-proxy approach for reliable provenance. Recommendations include adopting integrated multi-proxy provenance workflows in regional exploration models, expanding sample coverage to nearshore deposits to capture littoral reworking, and applying the framework to assess reservoir compartmentalization and sedimentary filtration processes in hydrocarbon systems.