Application of grouped detrital zircon analyses to determine provenance and closely approximate true depositional age:Early Cretaceous McMurray-Clearwater succession,Canada

In the Lower CretaceousMcMurray-Clearwater succession of the intracontinental Alberta Foreland Basin,Canada,detrital zircon U-Pb geochronology samples(referred to herein as DZ samples)have been used to interpret the strata as representing a paleo-continental-scale drainage system.However,the majority of DZ samples are relatively small(n≈90–100),and syndepositional DZ(i.e.,crystallization age<5 Ma older than depositional age)are rare.This has forced a reliance on dinocysts with long stratigraphic ranges to chronostratigraphically subdivide the McMurray-Clearwater succession rather than employing maximum depositional ages(MDAs)derived from DZ samples.Herein,43 DZ samples(taken from20 subsurface cores)are assigned to 1 of 5 stratigraphic intervals,and in each stratigraphic interval all associated DZ samples are combined to produce a grouped DZ sample.Analysis and comparison of individual and grouped DZ samples are used to(1)assess variability in provenance through time and space,and(2)assess the accuracy of chronostratigraphically subdividing the succession using MDAs.Along the main paleo-drainage axis,a comparison of dissimilarity between DZ samples from the same stratigraphic interval,as well as between stratigraphic intervals,reveals increasing average dissimilarity between individual DZ samples and their respective grouped DZ samplewith increasing spatial separation of samples.These data indicate that in the McMurray Depocenter some sediment is sourced from local tributaries,leading to geographical provenance variability.CalculatedMDAs for all 43 DZ samples and groupedMDAs(gMDAs)for the 5 grouped DZ samples are compared to an ash-derived absolute age and existing biostratigraphy.In theMcMurray Formation,comparison of MDAs to gMDAs shows that in basins with rare syndepositional DZ,the gMDA method improved depositional age estimates by transforming low-confidence MDAs(e.g.,youngest single grains)into high-confidence(multi-grain)gMDAs.In the Clearwater Formationwhere syndepositional DZ are plentiful(i.e.,>5%of the total DZ population),calculating maximumlikelihood ages fromgrouped DZ samples avoids negatively biased(i.e.,too young)MDAs.We suggest grouped DZ samples and the gMDA method be used in systems with multiple DZ samples from a well-defined stratigraphic interval as a means of assessing variability in provenancewithin a depositional system and for improving estimates of depositional ages using DZ.