Unmixing detrital geochronology age distributions

摘要: Despite recent advances in quantitative methods of detrital provenance analysis, there is currently no widely accepted method unmixing geochronology data. We developed a mixing model that determines proportions for source samples through inverse Monte Carlo modeling, wherein mixed are compared to randomly generated combinations distributions, and range best retained. Results may then be used constrain forward optimization routine find single best-fit mixture. Quantitative comparison based on the Kolmogorov-Smirnov (KS) test D statistic Kuiper V cumulative distribution functions, Cross-correlation coefficient finite mixture distributions (probability density plots or kernel estimates). demonstrate capacity this series tests synthetic data sets published empirical set from North America known proportions; proof-of-concept testing shows capable accurately highly complex distributions. apply two unknown Colombia central China. Neither yields perfect fits, which provides cautionary note potentially inadequate characterization and/or samples, highlights importance such accurate interpretation sediment provenance. Data size appears major control results; small (n < 100) lead misinterpretation The available as MATLAB-based stand-alone executable (.exe file) graphical user interface.