HR: 1340h
AN: A23B-0795    [Abstracts]
TI: Stationarity of the ocean component of the climate signal
AU: * Tokmakian, R
EM: rtt@nps.edu
AF: Dept. of Oceanography, Naval Postgraduate School, 833 Dyer Rd, Bldg 232, Rm 328, Monterey, CA 93943-5122 United States
AB: Empirical orthogal functions (EOFs) have been used extensively in recent years to explore the primary spatial and temporal distribution of the ocean's variability on both basin and global scales. Such empirical segmentation of time series are sometimes used as a way to extend time series back in time. This paper using both EOFs and Cyclic EOFs [Kim, et al. 1999] explores the spatial distribution of the ocean's variability and its relationship to the dynamic climate signal and how the primary signal patterns are modified in time. Using a relatively high resolution ocean model (0.2 degrees) forced with a 40+ year time series of realistic NCEP atmospheric fluxes, the stationarity of the interannual and longer sea surface height and upper ocean temperature signals are quantified and related to the forces applied at the ocean's surface. (Stationarity: A stationary process has the property that the mean, variance and autocorrelation structure do not change over time. For this purpose, any long term trend is removed.) With the model as an example time series, the stationarity of the ocean's upper ocean quantities is further examined through the use of satellite measurements of the ocean. While a percentage of the ocean's variability is relatively stationary, a significant proportion is not. Regional causes of the non-stationarity in the signal are explored.
DE: 3337 Numerical modeling and data assimilation
DE: 3339 Ocean/atmosphere interactions (0312, 4504)
DE: 3360 Remote sensing
DE: 1620 Climate dynamics (3309)
SC: Atmospheric Sciences [A]
MN: 2004 AGU Fall Meeting