HR: 13:55h
AN: PP53A-02    [Abstracts]
TI: Glacial/interglacial temperature variations in Soreq cave speleothems as recorded by 'clumped isotope' thermometry
AU: * Affek, H
EM: hagit.affek@yale.edu
AF: Geology and Geophysics, Yale, New Haven, CT 06520,
AU: Bar-Matthews, M
EM: matthews@gsi.gov.il
AF: Geological Survey of Israel, Jerusalem, Israel, 95501,
AU: Ayalon, A
EM: ayalon@gsi.gov.il
AF: Geological Survey of Israel, Jerusalem, Israel, 95501,
AU: Eiler, J
EM: eiler@gps.caltech.edu
AF: Geological and Planetary Sciences, Caltech, Pasadena, CA 91125,
AB: Isotopic composition in speleothem carbonates provide high resolution, well dated climatic records. However, deducing paleotemperature from δ¹⁸O suffers a major disadvantage as it reflects a combination of variations in both temperatures and isotopic composition of the precipitating solution. This is particularly difficult in continental records, due to the complex nature of δ¹⁸O variations in rainwater. In contrast, the carbonate 'clumped isotopes' thermometry (based on abundance of ¹³C-¹⁸O bonds) independently determines carbonate growth temperatures and can constrain the δ¹⁸O values of past precipitation. 'Clumped isotopes' thermometry is based on analyzing mass 47 in CO₂ extracted from carbonates. It is reported using the Δ₄₇ value, which is defined as the deviation of R⁴⁷ from that at a random distribution of isotopologues. At thermodynamic equilibrium, CO₂ approaches the random distribution (Δ₄₇=0) at high temperatures and preferentially clumps ¹³C and ¹⁸O into bonds with each other to produce positive Δ₄₇ values as temperature decreases. This approach was used successfully in paleosol nodules, brachiopods, mollusks and meteoritic carbonates. We applied this approach to Holocene and Last Glacial Maximum (LGM) speleothem samples from Soreq cave, Israel. The observed Δ₄₇ values of LGM carbonates were 0.668±0.004 and 0.673±0.008‰ for samples dated 19 and 20ka, respectively. Late Holocene samples (1-2ka) were 0.641±0.012 and 0.660±0.013‰. 8-11ka BP samples varied between 0.628±0.011 and 0.647±0.009‰. A 56ka BP sample was 0.657±0.011‰. For calibration purposes a sample of a modern speleothem was analyzed, resulting in 0.642±0.006‰. Using the Δ₄₇-temperature calibration of Ghosh et al (2006) results in a growth temperature of 26°C for the modern speleothem, higher than the estimated cave temperature over the growth period (18°C). This difference is consistent with kinetic isotope effects associated with CO₂ degassing, possibly related to surface processes in speleothem production. We therefore corrected all the temperatures obtained by 8°C. This resulted in a temperature of 11-12°C for the LGM samples and 15°C at 56ka BP, similar to the temperatures estimated before for Soreq Cave, using fluid inclusions, and temperature estimates for the Eastern Mediterranean using alkenones. 14-18°C were obtained at 1ka. Temperatures of 17-21°C were obtained for the 8-11ka samples, slightly higher than the modern day temperatures. These temperatures were then combined with the δ¹⁸O values in measured carbonates to estimate the δ¹⁸O of cave water, resulting in LGM water being 1.1‰ more enriched than the modern, and Holocene values being more depleted than the modern (by 0.6‰ at 1ka and 1.0‰ at 8-11ka). Combining this with previously measured δD of fluid inclusions places the 1ka sample within the range of modern rain, close to the Mediterranean meteoric water line. An ~11ka sample is on the global meteoric water line, suggesting higher humidity then currently in the region, consistent with its time correlation with sapropel layers in Mediterranean sediments that are interpreted as reflecting high amounts of rain. The LGM samples are close to the global meteoric water line, and the 56ka sample is between the two lines, consistent with its intermediate temperature.
DE: 1041 Stable isotope geochemistry (0454, 4870)
DE: 1094 Instruments and techniques
DE: 1616 Climate variability (1635, 3305, 3309, 4215, 4513)
DE: 1637 Regional climate change
DE: 4958 Speleothems
SC: Paleoceanography and Paleoclimatology [PP]
MN: 2007 Fall Meeting