2002 Denver Annual Meeting (October 27-30, 2002)

Paper No. 4
Presentation Time: 1:30 PM-5:30 PM

STABLE ISOTOPES OF CYCLIC CARBONATE NODULES DEPOSITED IN CHINESE LOESS: EVIDENCE FOR PRECIPITATION TRIGGERING OF ABRUPT CLIMATIC EVENTS?


LI, Zheng-Hua1, MORA, Claudia I.1 and LU, Huayu2, (1)Dept of Geological Sciences, Univ Tennessee, Knoxville, TN 37996-1410, (2)State Key Laboratory of Loess and Quaternary Geology, Chinese Academy of Sciences, Xi'an, 710054, China, zli@utk.edu

More than 26 carbonate nodule layers are observed in Chinese loess-paleosol sequences formed during the last 1.2Ma. The nodule layers have long been thought to originate by leaching of carbonate from overlying paleosols. We report stable C- and O- isotope compositions of nodule layers from three loess profiles that support an alternative mechanism of formation whereby nodule layers form in the loess proper, at the end of each cold period, due to an abrupt increase in precipitation. The profiles occur near Baoji (34°20’N, 107°39’E), Lintai (35°04’N, 107°02’E), and Luochuan (35°45’N, 109°25’E), located geographically along a trend of decreasing precipitation that is associated with tropical/subtropical Pacific water vapor transport from south to north across the Chinese loess plateau. Disseminated matrix carbonate and nodules from loess layers have similar d13C values (nodule d13C=-5.08±0.82 ‰ V-PDB, n=28), but these values are distinct from compositions of nodules in overlying paleosols (d13C=-8.29±0.45, n=7). Oxygen isotope compositions of loess nodules are more negative (d18O=-9.41±0.23, n=28) than paleosol nodules (d18O=-8.82±0.40, n=7), supporting the hypothesis that the nodule layers formed primarily in situ and during cold periods (loess deposition). Additional observations supporting in situ formation rather than carbonate leaching and translocation from overlying paleosols include an excellent match of carbon isotope variations with magnetic susceptibility and grain size records in these profiles. In situ nodule formation in loess, occurring across the plateau, suggests an abrupt increase in precipitation prior to each interglacial. In the last 500 ka, these wet periods correspond to IRD events in the North Atlantic, early warming in SSTs at low latitudes, and a series of wet events in northern Hemisphere. These data thus support new studies that indicate a possible precipitation trigger for rapid climate change, initiated within tropical regions, then amplified in the North Atlantic area.