L’étude intitulée «Reproducibility of Thermal History Reconstruction From Apatite Fission‐Track and (U‐Th)/He Data» par Richard A. Ketcham, Peter van der Beek, Jocelyn Barbarand, Matthias Bernet et Cécile Gautheron vient d’être publiée dans la revue G3 « Geochemistry, Geophysics, Geosystems ».
We report a new interlaboratory exercise to evaluate the reproducibility of apatite fission‐track (AFT) and (U‐Th)/He (AHe) data and thermal history analysis. Twelve laboratory groups participated, analyzing apatite separates from two previously studied localities. Ten groups returned AFT data from 13 analysts, five groups returned AHe data, one contributed apatite U/Pb data, and nine contributed thermal history models. Submitted AFT age data were generally consistent with the original studies and each other to within uncertainties, although there were departures, particularly among results obtained using laser ablation inductively coupled plasma mass spectrometry for uranium determination. AFT‐confined track‐length data showed more variation, which correlated between samples, suggesting that they may reflect analyst‐specific factors. Accounting for anisotropy using c axis projection reduced the variation and correlation. AHe ages showed more dispersion than observed in the previous studies, with one sample showing many ages consistent with prior work but also many significantly older ages, and the other roughly matching prior work but showing symmetrical scatter suggesting a 1SE uncertainty of 17–21%. Thermal history models generally agreed in their major features but showed considerable variation in detail, due to differences in data and data entry, model setup, and modeling software approach. Addressing pitfalls in data entry and model setup improved congruence, as would greater emphasis on AFT length and etch figure calibration. Because of scatter, AHe data had to be entered selectively into the models to achieve reasonable results. Although cautionary in several respects, study results point to promising avenues for improving the reproducibility of thermal history modeling.