A record of temperature and monsoon intensity over the past 40 kyr from groundwater in the North China Plain

The well-known temperature dependence of noble gas solubilities in water enables the determination of absolute paleotemperature records from radiocarbon dated groundwater. An advantage of the noble gas paleothermometer is that noble gas temperatures (NGTs) can directly be related to stable isotope ratios from the same groundwater archive in order to calibrate the stable isotope thermometer over long time scales. Our record from heavily used aquifers of the North China Plain constitutes a unique combined NCT and stable isotope data set from the region affected by the East Asian monsoon. In addition to providing a reliable estimate for the glacial-interglacial temperature change, the data enable the separation of the effects of temperature and precipitation (monsoon strength) on the stable isotopes. Although the record extends from modern waters to near the limit of the radiocarbon dating range at about 40 kyr BP, the last glacial maximum is not represented, probably because and conditions limited groundwater recharge during that period. The coldest observed period at about 30 kyr BP was 4.6 +/- 1.2 degrees C cooler than the Holocene. Whereas NGTs show no systematic variation during the Holocene, the delta O-18-values increased by about 1 parts per thousand between the early and late Holocene. A similar increase of delta O-18 has been observed before in speleothems and ascribed to decreasing monsoon strength in response to decreasing insolation during the Holocene. Our results confirm that the Holocene trend in stable isotopes is not related to temperature. In contrast, delta O-18 and NGT are correlated within the Pleistocene group of samples, indicating a delta O-18-NGT slope of 0.18 +/- 0.05 parts per thousand degrees C-1, similar to the temperature effect in modern precipitation of the region. Thus, of the overall increase of delta O-18 by about 2 parts per thousand between 30 kyr BP and the late Holocene, about half is due to a temperature increase, whereas the other half is due to a decrease of monsoon strength. (C) 2008 Elsevier B.V. All rights reserved.