UNMIXING DETRITAL ZIRCON U-PB AGE DISTRIBUTIONS

Detrital zircon U-Pb geochronology has become an invaluable technique in addressing source-to-sink questions of sediment provenance. Despite recent advances in quantitative methods of data comparison, at present there is no widely accepted method of identifying source groups and quantifying their contributions from detrital zircon age distributions. We developed an inverse Monte Carlo mixture model that determines relative contributions of source samples by comparing randomly weighted source sample age distributions to mixed samples using the Kolmogorov-Smirnov (KS) test D statistic, Kuiper test V statistic, and Cross-correlation coefficient. We demonstrate the capacity of this model through a series of tests on synthetic data and published empirical data from two source-to-sink systems: Colombian modern river sediments and catchment sources, and loess samples from central China. Proof-of-concept testing shows the model is capable of reproducing known proportions of highly-complex age distributions when both source and mixed samples are well-characterized. Results from modern-river sand and loess samples cannot be perfectly matched, which provides a cautionary note of inadequate characterization of sediment sources and/or mixed samples. Further, this point highlights the importance of such characterization for accurate interpretation of sediment provenance, while also providing a tool to identify incomplete source data sets. Sample size appears to be a major control on mixture model results; small (n < 100) detrital data sets may lead to misinterpretation of sediment provenance. This method has been developed into a MATLAB-based graphical user interface (GUI) as stand-alone executable (.exe file) and application (.app) programs. The GUI and source codes are openly available to the scientific community.