A two-dimensional gravest empirical mode determined from hydrographic observations in the Subantarctic Front
Document Type
Article
Date of Original Version
1-1-2001
Abstract
South of Australia, where the baroclinicity in the Subantarctic Front extends almost to the seafloor, the geopotential height of the sea surface (Ø) and the vertical acoustic travel time (T) exhibit a tight empirical relationship to each other and to the entire vertical structure of temperature T(p), salinity S(p), and specific volume anomaly δ(p). Measurements of T provide proxy estimates of the profiles TG(T, p), SG(T, p), and δ(T, p), based on a two-dimensional "gravest empirical mode" (GEM) representation of the vertical structure fitted to 212 hydrographic stations. A seasonal model was fitted in the near-surface layers using additional historical data. At each depth in the range 150-3000 dbar, more than 96% of the variance in both the T and δ fields is captured by the GEM representation. During the Subantarctic Flux and Dynamics Experiment (Mar 1995-Mar 1997), inverted echo sounders (IES) and current meters were moored in a 2D array along the WOCE SR3 transect south of Australia. At each IES site, proxy-estimated T(p) time series were compared with independent moored temperature records, confirming agreement within 0.29°, 0.26°, 0.10°, and 0.04°C rms differences at depths 300, 600, 1000, and 2000 dbar. Between laterally separated IES sites, geostrophic velocity profiles estimated from horizontal gradients in Ø(p) agree with moored current records within 0.07, 0.05, and 0.03 m s-1 at 300, 600, and 1000 dbar relative to 2000 dbar. A simple argument based on conservation of potential vorticity is suggested to account for GEM dominance of the spatiotemporal variability.
Publication Title, e.g., Journal
Journal of Physical Oceanography
Volume
31
Issue
8 PART 1
Citation/Publisher Attribution
Watts, D. R., C. Sun, and S. Rintoul. "A two-dimensional gravest empirical mode determined from hydrographic observations in the Subantarctic Front." Journal of Physical Oceanography 31, 8 PART 1 (2001). doi: 10.1175/1520-0485(2001)031<2186:atdgem>2.0.co;2.