Past the First Date: Ensemble Concordant–Discordant Comparison for Zircon Pb‐Loss Analysis

Abstract Zircon uranium–lead (U–Pb) geochronology is a cornerstone of Earth science for resolving the timing of deep‐time processes, yet the U–Pb system does not always behave as a closed chronometer. Selective loss of radiogenic Pb during fluid alteration or thermal disturbance can shift isotopic ages and generate discordant analyses. Some zircon data sets may preserve more than one Pb‐loss episode, yet many Pb‐loss modeling approaches return a single “best” time of Pb‐loss. The concordant–discordant comparison (CDC) test estimates Pb‐loss timing by finding the modeled disturbance age at which reconstructed discordant analyses best match the concordant population. Here, we extend the CDC framework to retain supported local optima across Monte Carlo realizations rather than reducing each realisation to a single optimum, and report 95% stability bounds that describe run‐to‐run repeatability in peak position rather than geological age uncertainty. This redesign can resolve the timing of multiple Pb‐loss episodes, recovering overprinting metamorphic, hydrothermal, or magmatic events. Testing the method against simulated zircon U–Pb data sets with prescribed Pb‐loss histories, we find that the ensemble CDC yields the lowest overall median absolute error of the benchmarked modeling approaches (19 Myr) and assigns age estimates to 27 of 36 true events. Errors are lowest for six single‐event benchmark examples, with a 90th‐percentile absolute error of 11 Myr, increasing to 83 Myr in overlapping multi‐event cases. Applied to an Archean Yilgarn granite, the ensemble resolves two Pb‐loss peaks at 454 and 1113 Ma, consistent with regional Pan‐African cooling and the Mesoproterozoic Warakurna Large Igneous Province respectively.