Stalagmite Based Reconstruction of Atmospheric Radiocarbon Levels during Deglaciation: Implications for Radiocarbon Calibration

Radiocarbon measurements on dendrochronologically dated tree rings provide the most definitive record of atmospheric changes in radiocarbon concentration (14C). However, the tree ring portion of the current radiocarbon calibration curve (INTCAL04) only extends to 12.4 thousand years Before Present (kyr BP). Efforts to extend radiocarbon calibration timescale beyond 12.4 kyr BP have focused on using 14C measurements in marine varves, lake sediments, corals, and a floating tree ring sequence, but significant disagreements exist among these marine and terrestrial 14C data sets. Speleothems (e.g., stalagmites), which are secondary calcite deposits, form by slow degassing of CO2 enriched cave waters, and can be used to generate records of past atmospheric 14C concentration. Furthermore, stalagmites can be precisely dated by U-Th, and the stable isotopic composition of carbon and oxygen in stalagmites offer an opportunity to assess the atmospheric 14C variations in terms of regional climate change. However, a necessary pre-requisite in using stalagmites for radiocarbon calibration purposes is to quantify the dead carbon fraction or DCF. The DCF is the fraction of `old'14C-free carbon derived from host limestone rocks that is incorporated in a given stalagmite, and must be corrected for in order to extract a true atmospheric 14C record. Here, we present results of Accelerator Mass Spectrometry (AMS) 14C measurements in a stalagmite recovered from Timta cave in the western Himalayas in India. The absolute ages of this stalagmite are constrained by 24 230Th dates, which show that it grew from 14.5 to 11.5 kyr BP. The comparison between stalagmite 14C ages with the atmospheric 14C values from the tree ring portion of INTCAL 04 curve (11.5 to 12.4 kyr B.P.) suggest a small but variable DCF that ranges from 2% to 6%. We observe a surprisingly strong inverse correlation between δ13C, a proxy for vegetation changes overlying the cave, and the DCF, perhaps due to pH changes in the seepage waters, with consequent changes in limestone dissolution in the overburden. For the period covered by the INTCAL04 tree ring data, this correlation constrains the DCF to better than 1%. Assuming that this correlation is valid for the period beyond 12.4 kyr BP, we have used the δ13C data to correct the speleothem radiocarbon results for the DCF and thereby derive a record of atmospheric 14C for the early Younger Dryas and the preceding Allerod interstadial. Our study suggests that caution must be exercised in assuming a constant DCF in any stalagmite based radiocarbon calibration and underscores the importance of the use of environmental proxies as an aid to interpreting 14C values in stalagmites.