Sonia Bauer, Mark Swenson, Annalisa Griffa, Arthur Mariano
Atlantic Oceanographic and Meteorologic Laboratory/NOAA
bauer@aoml.noaa.gov
(Abstract received 10/07/2000 for session A)
ABSTRACT
Eddy diffusivity is a statistic that provides a quantitative measure of the diffusive transport caused by mesoscale motions. The principal objective of this study is to estimate eddy diffusivities of the surface velocity field in the tropical Pacific Ocean. The observations are satellite-tracked drifting buoys spanning the years 1979 through mid-1996. The data were assembled and distributed by the data acquisition center at the NOAA Atlantic Oceanographic and Meteorological Laboratory (AOML) and were obtained as part of the Equatorial Pacific Ocean Climate Study (EPOCS) and World Ocean Circulation Experiment--Tropical Ocean-Global Atmosphere (WOCE-TOGA) programs. The tropical Pacific surface current system is characterized by: 1) nonstationarity due to rapid response time of equatorial and near-equatorial dynamics; 2) strong meridional shear in the large-scale mean; and 3) an energetic mesoscale velocity component. Taylor (1921) defined eddy diffusivity as the integral of the autocovariance of Lagrangian eddy velocities -- requiring both stationary and homogeneous statistics of the eddy field. Eddy velocities were obtained in this study by removing the splined mean field to eliminate mean shear as described in Part 1 (Bauer, et al 1998) and binned spatially to group data that have similar dispersion characteristics. Temporal partitions were made to create stationary eddy statistics. Zonal diffusivity estimates vary with latitude from about 5x107 to 76x107 cm2/s. Largest estimates are in regions of strong meridional shear and large eddy variance between 4°S and 10°N. However, meridional diffusivity estimates are nearly constant throughout the tropics varying from 2x107 to 9x107 cm2/s. Simple autoregressive models provide the analysis with estimates of Lagrangian integral time scale which is a measure of the turbulence decorrelation time scale. First order (AR(1)) and second order (AR(2)) autoregressive processes adequately describe the eddy transport statistics. All zonal eddy transport statistics and meridional statistics from low variance regions (generally poleward of 4°S and 10°N) can be modeled by AR(1) statistics modified by inertial wave oscillations. Meridional statistics of the near-equatorial regions (between 4°S and 10°N) are characterized by large meridional eddy variance associated in part with tropical instability waves (TIW's), and by low meridional diffusivity: Wave motion increases eddy variance but does not diffuse water parcels with periodic trajectory motion. Therefore, meridional eddy diffusivities are low in these regions and are modeled by AR(2) statistics. An independent confirmation of the diffusivity estimates in the central and eastern Pacific was obtained by comparing tracer flux divergence computed from a parameterization using diffusivity estimates of our analysis and that from direct eddy Reynolds stress flux divergence. Our results show that diffusivity can be estimated for regions that have not been considered before because of lack of data and because of the complexities of the velocity field.