MAXIMUM DEPOSITIONAL AGES IN HIGH-PRECISION CHRONOSTRATIGRAPHY: LA-ICPMS–CA-ID-TIMS TANDEM DATING OF ZIRCON AND BAYESIAN AGE–DEPTH MODELING, CENOMANIAN–CAMPANIAN COLVILLE FORELAND BASIN STRATA, CENTRAL NORTH SLOPE, ALASKA

U–Pb geochronology is refining many long-standing, relative-age frameworks of sedimentary basins. Here, we use laser ablation-inductively coupled plasma mass spectrometry (ICPMS) and chemical abrasion-isotope dilution-thermal ionization mass spectrometry (TIMS) to establish high-precision detrital zircon maximum depositional ages (MDAs; n=6) and tephra zircon ages (n=15) for Upper Cretaceous strata that crop out in northern Alaska at or near Ninuluk, Tuluvak, Shale Wall, Schrader, and Uluksrak bluffs. TIMS results are projected onto a 1656-m-thick composite section, revealing relations and trends that are not resolved by the biostratigraphy and requiring chronostratigraphic revisions up to ~5.5 Ma. Bayesian age–depth modeling of the section in modifiedBchron bears many benefits, including superposition and geologically sensible sedimentation rate priors, as well as realistic posterior uncertainties relative to conditioning data. Furthermore, detrital likelihoods are input as uniform probabilities extending between an MDA and the superjacent tephra age, circumventing the common yet problematic treatment of MDAs as ages. We also derive posterior ages (base) and stratigraphic accumulation rates for each unit: Nanushuk (Ninuluk sandstone: ~96.8 Ma, ~95 m/Ma), Seabee (Shale Wall facies: ~95.0 Ma, ~120 m/Ma; clinoform facies: ~93.4 Ma, ~305 m/Ma), Tuluvak (marine: ~92.7 Ma, ~438 m/Ma; non-marine: ~92.4 Ma, ~122 m/Ma), Rogers Creek (~91.2 Ma, ~41 m/Ma), Barrow Trail (~85.6 Ma, ~102 m/Ma), Sentinel Hill (~83.7 Ma, ~166 m/Ma), and lowermost Prince Creek (~82.1 Ma, ~25 m/Ma). The mid-Campanian unconformity is <~82.1 Ma and >~79.3 Ma and reflected in Prince Creek’s ~25 m/Ma rate. Tandem dating of individual grains delineates difficulties in establishing accurate ICPMS ages, with average offsets of –2.7% and –2.3% for detrital (n=36) and tephra (n=93) zircon, respectively; similar ICPMS offsets occur in published tandem datasets. Existing ICPMS-based MDA algorithms do not explicitly address systematic or geologic errors, and youngest single grain and youngest statistical population results for our samples are too young by ~6 Ma and ~3 Ma (averages), respectively. Tandem dating and Bayesian age–depth modeling are compelling approaches for absolute-age chronostratigraphic research that employs MDAs.