Potential vorticity in the Pacific and Indian Oceans

WOCE Conference, May, 1998

Lynne D. Talley and Mary Cait McCarthy
Scripps Institution of Oceanography, University of California San Diego
La Jolla CA 92093-0230

Abstract

Potential vorticity and especially its horizontal gradient is essential to propagation of information in the ocean. Even though it is a dynamical quantity, it is a useful tracer of flow since it has identifiable sources and sinks, and complements other properties such as salinity, nutrients and chemical tracers. Potential vorticity distributions on neutral surfaces in the Pacific and Indian Oceans have been constructed using the WOCE data sets supplemented by historical data. Particular attention has been paid to determining which regions are strongly influenced by the mid-laitude gyre circulations compared with the tropical regions, surface convection (mode water/intermediate water/deep water formation areas), and to the vertical extent of topographic influence. Isoneutral maps of other properties, such as oxygen and silicate, supplement interpretation of the potential vorticity fields.

Methods

Isopycnic potential vorticity was calculated as (equation) from CTD and bottle data. Values from both data types are comparable. Relative vorticity is neglected. Q is hand contoured on neutral surfaces (Jackett and McDougall, 1997) in the Indian Ocean and on potential density surfaces in the Pacific Ocean. Surfaces represent mode, intermediate and deep water.

Pacific Ocean

Density Mean Depth Water Mass
sigma0=25.2 150m NPSTMW
sigma0=26.0 200m SPSTMW
sigma0=26.8 400m NPIW
sigma1=31.7 700m AAIW
sigma2=36.96 2700m PDW
sigma4=45.88 4200m LCDW

Salinity section along 150W (WOCE P16). Isopycnals used below are shown as heavy contours.

Absolute value of potential vorticity along 150W (WOCE P16) in units of 10-14 cm-1 sec-1. Isopycnals used below are shown as heavy contours.

Potential vorticity at sigma_theta = 25.2 Beta domination equatorward of 3 degrees. Low Q in the North Pacific is the lightest subtropical mode water (south of Kuroshio Extension). Low Q in South Pacific associated with the subtropical front in the central basin. High Q subducted in eastern part of North and South Pacific subtropical gyres.

Potential vorticity at sigma_theta = 26.0 Beta domination equatorward of 5 degrees. North Pacific subtropical gyre - low in north central region, underlying coldest subtropicl mode water (sigma_theta = 25.4). High Q subducted in east and along Kuroshio Extension. South Pacific subtropical gyre - low north of New Zealand - core of subtropical mode water. Advected eastward, with westward advection of high.

Potential vorticity at sigma_theta = 26.8. Beta domination equatorward of 10-15 degrees. In N. Pacific subtropical gyre - high in Kuroshio and extending eastward across gyre. Slight low along subarctic front to base of pycnocline. Lowest in Okhotsk Sea. In South Pacific - lowest in Tasman Sea ACC region where it is the local Subantarctic Mode Water. Some circulation of this in the Tasman Sea and around New Zealand. Highest in southeast Pacific, with subducted high carried around gyre to low latitudes.

Potential vorticity at sigma_1 = 31.7. Beta domination equatorward of 15-18 degrees. Q nearly uniform in N. Pacific subtropical gyre except for high along Kuroshio, west-east gradient in Okhotsk Sea and south-north gradient in Bering Sea. In South Pacific, high Q along ACC decreasing eastward, lowest in southeastern Pacific, at formation site of AAIW. Low Q spreads norrthward by subduction from this region.

Potential vorticity at sigma_2 = 36.96. Beta domination equatorward of 15-18 degrees. Nearly uniform Q in North Pacific north of 35N. Low Q in eastern South Paciifc, likely created by local hydrothermal heating. Strong Q distortion by flow in South Pacific. Banded high and low Q along ACC - low absolute value of Q on south side and high absolute value of Q on north side. These would partially compensate for relative vorticity differences on either side of the ACC.

Potential vorticity at sigma_4 = 45.88. Beta domination equatorward of 10 degrees. Strong topographic effects: in the northern hemisphere, Samoan Passage and the deep western boundary current distort Q. Island of high Q in central North Pacific. Band of high Q along the ACC.

Oxygen (ml/l) at sigma_4 = 45.88. Highest in south, lowest in north. Long zonally oriented tongues near equator and relatively large northward gradient at 10 S to equator.

[Plot not available (does not convert easily to postscript).]

The tropical beta barrier. As f decreases, potential vorticity must go to zero. At depths greater than 500 to 700 meters, the switch to beta domination occurs around 15 to 18 degrees from the equator. At shallower depths, the vigorous subtropical gyre circulations approach closer to the equator. In the deepest layer, topographic effects also become important.

A simple argument for determining the latitude separating beta domination from gyre circulations: deviation of Q from beta requires strong enough stretching. At low enough f, Q is too difficult to deform through stretching. The two terms in the isopycnic potential vorticity are: (equation beta*y + (f*f/N*N)psi_zz).
Scale values appropriate for the deep ocean yield a latitude of 15-20 degrees for the shift from stretching to beta domination.

Indian Ocean

The Indian Ocean work is currently in press at J. Geophys. Res.:
McCarthy, M.C. and L. D. Talley, 1999. Three-dimensional potential vorticity structure in the Indian Ocean. J. Geophys. Res., accepted.

Conclusions

In the mid to high latitudes, source waters and their circulation set the upper ocean potential vorticity structure outside the tropics. The abyssal PV structure is strongly affected by both the abyssal circulation and topography. In the tropics, PV is dominated by beta. In the transition region between the tropics and subropics, PV has a somewhat higher meridional gradient than in either region. The demarcation between the tropical and subtropical regions likely results from a change in the efficiency of stretching. This change in the nature of the PV field between the tropics and subtropics likely affects planetary wave propagation characteristics.
For more information: ltalley@ucsd.edu