How Fast is Fast? Speed of Abrupt Climate Change from Subannual-Resolution Nitrogen Isotope Measurements in Ice Cores Investigator: Jeffrey P. Severinghaus jseveringhaus@ucsd.edu (Principal Investigator current) Abstract The proposed research will address the speed and magnitude of the abrupt temperature changes that are now recognized to have occurred during the Holocene and the preceding glacial period, using sub annual measurements of nitrogen and argon isotopes in ice cores. Previous work has shown that nitrogen and argon isotopes record a signal of rapid temperature change due to isotope fractionation by thermal diffusion in air in the porous snow overlying the polar ice sheets. This work will increase the sample resolution by two orders of magnitude, to monthly sampling, to address the question of exactly how fast were the abrupt changes in climate. This approach is proposed based on the recent recognition that gas ratios in ice cores are annually-layered much like the better-known annual layers in the ice matrix. Because of the details of the air enclosure process, summer layers should contain air that is much less affected by smoothing than the seasonally-averaged bulk measurements made to date on ice cores. Thus very high-frequency information (order of 1-3 years) on climate change is present in the gases, if sampled at sufficient resolution.. The results will constrain hypothesized mechanisms for the abrupt changes involving the atmospheric circulation and ocean reorganization.