Radiocarbon calibration/comparison records based on marine sediments from the Pakistan and Iberian Margins
Résumé
We present a new record of radiocarbon ages measured by accelerator mass spectrometry (AMS) on a deep-sea core collected off the Pakistan Margin. The C-14 ages measured on the planktonic foraminifera Globigerinoides ruber from core MD04-2876 define a high and stable sedimentation rate on the order of 50 cm/kyr over the last 50 kyr. The site is distant from the main upwelling zone of the western Arabian Sea where C-14 reservoir age is large and may be variable. Many independent proxies based on elemental analyses, mineralogy, biomarkers, isotopic proxies, and foraminiferal abundances show abrupt changes correlative with Dansgaard-Oeschger and Heinrich events. It is now common knowledge that these climatic events also affected the Arabian Sea during the last glacial period through changes in the Indian monsoon and in ventilation at intermediate depths. The stratigraphic agreement between all proxies, from fine-to coarse-size fractions, indicates that the foraminiferal C-14 ages are representative of the different sediment fractions. To build a calendar age scale for core MD04-2876, we matched its climate record to the oxygen isotopic (delta O-18) profile of Hulu Cave stalagmites that have been accurately dated by U-Th (Wang et al. 2001; Southon et al. 2012; Edwards et al., submitted). Both archives exhibit very similar signatures, even for century-long events linked to monsoonal variations. For comparison, we have also updated our previous work on core MD95-2042 from the Iberian Margin (Bard et al. 2004a, b, c), whose climate record has likewise been tuned to the high-resolution delta O-18 Hulu Cave profile. Sophisticated and novel statistical techniques were used to interpolate ages and calculate uncertainties between chronological tie-points (Heaton et al. 2013, this issue). The data from the Pakistan and Iberian margins compare well even if they come from distant sites characterized by different oceanic conditions. Collectively, the data also compare well with the IntCal09 curve, except for specific intervals around 16 cal kyr BP and from 28 to 31 cal kyr BP. During these intervals, the data indicate that C-14 is somewhat older than indicated by the IntCal09 curve. Agreement between the data from both oceanic sites suggests that the discrepancy is not due to local changes of sea-surface C-14 reservoir ages, but rather that the IntCal09 curve needed to be updated in these intervals as has been done in the framework of IntCal13 (Reimer et al. 2013a, this issue).