An evaluation of inoceramid single-prism sclerochronology

Abstract

Inoceramid prisms are among the most common microfossils that can be found in Late Cretaceous strata. Since these biomineral units were formed by sequential accretion in the outer shell layer of the bivalves, they potentially serve as archives of short-term (circa-annual) paleowater temperature fluctuations. In the present study, we tested to what extent intra-prismatic stable isotope variations of individual inoceramid prisms reflect sub-annual δ13C and the δ18O patterns. We obtained multiple carbonate samples from prisms recovered from the sediment and from a cross-sectioned inoceramid shell fragment and compared data from these samples to data from classical sclerochronological sampling (i.e., sequential drilling). The analyzed shell material belonged to the species Platyceramus platinus from the Smoky Hill Chalk Member (Niobrara Formation, Kansas, USA). Scanning electron microscopy (SEM) was used to evaluate the diagenetic alteration of the studied prisms and allowed to identify suitable material for stable isotope analyses. The δ18O values of drilled carbonate samples (δ18ODrill) and of prisms (δ18OIP) compare very well to each other when temporally aligned (R2 = 0.98). The stable carbon isotope values (δ13CDrill; δ13CIP), however, show less agreement (R2 = 0.28), which is likely attributable to different time-averaging of the samples. We discuss how to identify the most suitable prisms for sub-annual paleotemperature reconstructions from disarticulated material using microgrowth patterns and geometrical features of the prisms as proxies for changing growth rates. Finally, we apply the novel sampling strategy to naturally occurring disarticulated prisms from the sediment and compare their intra-prismatic isotope values (δ13CIPS and δ18OIPS) to those of bulk-analyzed prism fragments from the sediment (δ13CBPS and δ18OBPS). Sclerochronological analysis of the prisms from the sediment yields δ18OIPS fluctuations of 1.89‰, virtually identical to the δ18OIP chronologies of the prisms extracted from the shell cross-section (1.90‰). The overall ranges of the δ18OIP and δ18OIPS values are identical (2.60‰) and capture very well the expected seasonal stable oxygen isotope oscillation derived from the δ18ODrill values (2.61‰). The range of the δ18OIPS data is two-fold larger than that of the δ18OBPS values (1.19‰), suggesting a strong effect of time-averaging on the stable oxygen isotope data of the bulk-analyzed prisms fragments. According to our novel statistical model, the resolution of δ18OBPS data can be increased by preferentially analyzing short (ca. 300 μm) individual prisms fragments (at least n = 20).