A NEW CONFIDENCE-LIMITS BASED METHOD FOR COMPARING DETRITAL ZIRCON U-PB AGE DISTRIBUTIONS

Detrital zircon U-Pb ages distributions have become one of the most widely used tools for tectonic and paleogeographic analysis in the geosciences. Yet, to date, no viable method has emerged for assessing similarities and differences between age distributions. We present a new method, similar to that of Stock et al. (2006; Geology), where the age distribution in a single sample (measured PDF, n ≈100) is compared to a parent distribution (Predictor PDF, n >>100) generated by Monte Carlo simulation of a possible sediment source. We sample 100 ages, at random and with replacement, from a “predictor” PDF and repeat the process 1000 times. We then average the 1000 PDFs from the resampled distributions, and calculate two standard deviations to represent the variability in the resampled results. The PDF from a measured sample can then be compared to the 2-sigma range of the modeled predictor PDF. A simple difference calculation highlights areas of mismatch, which allows rapid recognition of missing or additional age components in the sample distribution. Importantly, this method assumes that we have sampled the a priori probability distribution of the source area; thus, parent distributions with low n (i.e., <500) may not accurately reflect either the range or probability of age components in the source area. Despite generating a large n using resampling, the predictor PDF will only be as representative as the original parent distribution. Therefore, we suggest using only very large datasets (i.e., ≈ 1000 ages), which are statistically most likely to represent the true parent distribution. We compare a measured distribution from Middle Jurassic sandstone of the Blue Mountains Province, eastern Oregon, to two geologically plausible source regions (i.e., parent distribution): (1) Jurassic eolian sandstone of the Colorado Plateau, and (2) uplifted Triassic backarc basin deposits of western Nevada. Our results suggest that the age distribution in Middle Jurassic sandstone of the Blue Mountains province most closely resembles that of the uplifted Triassic backarc basin.