2004 Denver Annual Meeting (November 7–10, 2004)

Paper No. 1
Presentation Time: 8:00 AM

THE MISMATCH OF MODEL SIMULATIONS AND OBSERVATIONS FOR OXYGEN ISOTOPIC STAGE 3 IN EUROPE: IS NATURAL VARIABILITY MORE IMPORTANT THAN PREVIOUSLY CONSIDERED IN PALEOCLIMATE STUDIES?


BARRON, Eric J., Department of Geosciences, Penn State Univ, 116 Deike Building, University Park, PA 16802 and POLLARD, Dave, EMS Environment Institute, Pennsylvania State Univ, University Park, PA 16802, barron@ems.psu.edu

Oxygen Isotopic Stage 3 (OIS-3) is a time of remarkable climate interest, given that it is the time of transition between Neanderthal populations and modern humans in Europe and there is substantial evidence for transitions between warm and cold phases of climate during the period. A series of climate model simulations, using the GENESIS GCM, RegCM2 (a mesoscale climate model), and coupled biome models, are utilized to explore the nature of climate change during OIS-3. The Stage 3 Project, organized by Tjeerd van Andel, provided a comprehensive synthesis of climate-related data and characterization of climate forcing factors. Despite a wide variety of well-constrained climate simulations followed by a significant number of sensitivity experiments, the climate models failed to capture the cold phase climate conditions in northern Europe (southern European climates and warm phase conditions appeared to be well simulated). There are three potential solutions to this enigma: (1) adequacy of the model codes, (2) characterization of the boundary conditions and climate forcing factors (orbit, carbon dioxide concentrations, ocean temperatures, ice sheet size), and (3) interpretation of the observations. Careful analysis of the three sources of error leads to one of three hypotheses as an explanation: (1) structural changes in the atmospheric circulation are required that have not been captured because of poor global SST data sets (North Atlantic values are well constrained), (2) model-data discrepancies can be explained as the sum of many small errors in models, boundary conditions, and in the interpretation of observations although these errors must all be of the same sign, and (3) decadal variability in climate, which may not be captured in ocean temperature records or in model simulations, may be more important in governing the record of climate proxies than previously considered.