The study of Pacific circulation is as long as the history of Pacific navigation. In the 18th century scientists began to theorize about the causes of the currents and water distributions. Peterson et al. [1994] offer an excellent summary of the early days of charting the ocean's surface currents; they include an interesting account of some of the pitfalls encountered by overeager theoreticians. Prestwich [1876] summarized, on the eve of the Challenger expedition, the state of knowledge regarding the distribution of temperature with depth in the world's oceans, a principal scientific issue being that of the continuous increase of seawater density to the freezing point. Illustrated by meridional sections of temperature through each of the oceans he summarized prior conclusions that the deeper waters are supplied from higher latitudes, that there must be upwelling of this water and that there is upwelling at the equator. He concluded that deep water was formed in the North Atlantic and Antarctic and not in the North Pacific.
Wüst's [1929] treatise on the deep circulation of the Pacific Ocean has been neglected compared with his longer and much more thorough work on the Atlantic Ocean [ Wüst, 1933]. The latter remains a benchmark of traditional oceanography and important background for anyone seeking to understand the water properties and circulation of the Atlantic. The Pacific treatment suffered from a much poorer database; even so, Wüst was able to outline some basics of the abyssal circulation, showing the northward path of bottom waters through Samoan Passage with spreading eastward south of Hawaii. An incorrect inference was of deep water formation in the Okhotsk, which resulted from assigning minimum temperatures from maximum-minimum thermometers to the bottom [Mantyla, personal communication].
The surface currents of the Pacific
have been known for many decades and the
basics are easy to discern in surface
dynamic topography relative to 1000 dbar
(1 dbar = 10
Pa) [ Wyrtki, 1975;
Reid and Arthur, 1975]. The surface
circulation consists of the cyclonic
subpolar gyre in the north, the
anticyclonic North Pacific subtropical
gyre, the cyclonic and very narrow
northern tropical cell including the
North Equatorial Countercurrent, the
westward South Equatorial Current at the
equator and to the south, the
anticyclonic South Pacific subtropical
gyre, and the Antarctic Circumpolar
Current. Reid showed that the
subtropical circulations shrink poleward
with increasing depth in the North
Pacific [1965] and in the South Pacific
[ Reid, 1986]. He also pointed out a ``C-shape''
associated with the western part of the
subtropical gyre circulations, in which
the western boundary current has a
westward and equatorward recirculation
just equatorward of and east of the
boundary current and its eastward-flowing separated extension. The
recirculation connects back into the
eastward flow of the subtropical gyre at
a lower latitude.
The ocean's variability has been studied for many years. Wyrtki et al. [1976] presented maps of the eddy kinetic energy of the global ocean based on ship drifts. Mizuno and White [1983] showed the variable paths of the Kuroshio. Many syntheses of upper ocean thermal data acquired using expendable bathythermographs in the 1970's in the North Pacific showed the prevalence of westward propagation [ e.g. White and Saur, 1981; White, 1982]. The importance of eddies in transporting heat was pointed out for the Antarctic Circumpolar Current by deSzoeke and Levine [1981]. One of the first results from satellite altimetry was mapping global sea surface height variability, showing high energy in the western boundary current regions and Antarctic Circumpolar Current, with relatively high energy through the western boundary region and across the Pacific in the tropics just north of the equator, and low energy in the eastern Pacific [ Cheney et al., 1983].
The layered water mass structure of
the global ocean is apparent in
treatments such as Reid and Lynn [1971].
In the Pacific, the dominant layers
consist of: the upper ocean with
alternating fresh and saline bands
directly influenced by surface exchange;
the relatively fresh intermediate water
layers (Antarctic and North Pacific);
the low oxygen, high silica Pacific Deep
Water formed in the north through
upwelling and diffusion and intruding
southward; the high salinity, higher
oxygen Circumpolar Water intruding
northward; and the cold Antarctic Bottom
Water intruding northward. The
northward spreading water masses are
separated from the southward spreading
Pacific Deep Water by a jump in
temperature at about 1-2
C [ Craig
et al., 1972], associated with a
vertical stability maximum [ Reid and
Lynn, 1971].
The horizontal circulation which conveys these layers meridionally is tortuous, with large zonal excursions around gyres and topography. Because of the unknown reference velocity, maps of properties on isopycnals, showing the contrasting water masses, are often used to aid in determining flow directions. An update of Wüst's [1929] bottom mapping, including many other properties, is by Mantyla and Reid [1983] who used high quality deep data to show sources and pathways of flow. Reid [1965] used properties on the isopycnals which characterize the northern and southern intermediate waters in the Pacific to illustrate their circulation. Reid [1981] used properties on a deeper isopycnal, lying at about 2000 m, to show the global sources of water, and aid in interpreting the dynamic heights relative to a deeper reference level. Reid and Lynn [1971] showed salinity on a deep isopycnal characteristic of the North Atlantic Deep Water, showing the spread of its influence via the Antarctic to the Indian and Pacific Oceans.