HR: 0800h
AN: PP41C-0696 [Abstracts]
TI: Objectively Derived Age Models for Marine Isotope Stage 3 Paleoclimate Records From Ice, Land and Sea
AU: * Pisias, N G
EM: pisias@coas.oregonstate.edu
AF: College of Oceanic & Atmospheric Sciences, Oregon State University
104 COAS Admin. Bldg, Corvallis, OR 97331, United States
AU: Clark, P U
EM: clarkp@onid.oregonstate.edu
AF: Department of Geosciences, Oregon State University, Corvallis, OR 97331, United States
AU: Brook, E
EM: brooke@@geo.oregonstate.edu
AF: Department of Geosciences, Oregon State University, Corvallis, OR 97331, United States
AB:
A primary goal of paleoclimate studies is to determine the mechanisms and dynamics that are responsible for
climate change on long time scales and during times of very different Earth boundary conditions. To achieve this
goal, we must be able to determine the spatial and temporal relationships between all elements of the climate
system. Recently Clark et al. (2007) combined the spatial and temporal modes of variability defined from a set of
38 paleoclimate time series with ocean and atmospheric models to develop a scenario for the sequence of
system responses to climate change observed during MIS-3. Clark et al. (2007) used the published time scales
for these records in their data analysis. Empirical Orthogonal Function (EOF) analysis showed that there were
two dominant spatial/temporal modes in these diverse time series. Clark et al. (2007) referred to the first EOF as
the Northern Signal, which is most important in paleoclimate records from much of the Northern Hemisphere,
and the second EOF as the Southern Signal, which is most important in most Antarctic and Southern Ocean
records. We will show that the EOF analysis can also provide a framework for refining the chronologies of the
paleoclimate records used in these types of studies. Numerical experiments show that the time series of the
modes extracted by EOF analysis average out random chronologic errors in the individual time series used in the
analysis. Thus it is possible, by mapping the original time series into simple weighted combinations of the
extracted EOF modes, to remove these random errors from the chronology of each time series. Applying this
strategy to 38 time series from MIS-3 shows that: 1) average errors in chronologies range from 200 to 1800 years
with a mean of 620 years; 2) that the modes of variability extracted by EOF analysis are reasonably robust to these
chronologic uncertainties; and 3) most of the time series analyzed are not simply one mode or the other but are a
complex pattern of combination of the Southern and Northern modes of climate variability.
This work was supported by an NSF grant, PALEOVAR to Oregon State University, University of Oregon and
University of Minnesota.
Clark, P.U., S.W. Hostetler, N.G. Pisias, A. Schmittner, and K.J. Meissner, 2007. Mechanisms for a ~7-kyr Climate
and Sea-Level Oscillation During Marine Isotope Stage 3, AGU Monograph, in press
DE: 4938 Interhemispheric phasing
SC: Paleoceanography and Paleoclimatology [PP]
MN: 2007 Fall Meeting