The California current system—hypotheses and facts☆

The primary purpose of this paper is to describe the seasonal variation of the various currents which comprise the California Current System—the California Current, the California Undercurrent, the Davidson Current and the Southern California Countercurrent—and to investigate qualitatively the dynamical relationships among these currents. Although the majority of information was derived from existing literature, previously unpublished data are introduced to provide direct evidence for the existence of a jet-like Undercurrent over the continental slope off Washington, to illustrate ‘event’-scale fluctuations in the Undercurrent and to investigate the existence of the Undercurrent during the winter season. The existing literature is thoroughly reviewed and synthesized. In addition, and more important, geostrophic velocities are computed along several sections from the Columbia River to Cape San Lazaro from dynamic heights given by Wyllie (1966), Budinger, Coachman and Barnes (1964), and Reed and Halpern (1976). From these data and from long-term monthly wind stress data and vertical component of wind stress curl data (denoted curl τ) given by Nelson (1977), interesting new conclusions are made. 1. The flow that has been denoted the California Current generally has both an offshore and a nearshore maximum in its alongshore coponent. 2. The seasonal variation of the nearshore region of strong flow appears to be related to the seasonal variation of the alongshore component of wind stress at the coast, τyN, at all latitudes. Curl τ near the coast may also contribute to the seasonal signal, accounting for the lead of maximum current over maximum wind stress from about 40°N northward. Large-scale flow separation and fall countercurrents that of headlands may account for the sudden occurrence of late summer and fall countercurrents that appear as large anomalies from the wind-driven coastal flow south of 40°N. 3. From Cape Mendocino southward a northward mean is imposed on the nearshore current distribution. The mean is largest where curl τ is locally strongest, in particular, off and south of San Francisco and in the California Bight. It may be responsible for the portion of the Davidson Current that occurs off California, for the San Francisco Eddy and for the Southern California Eddy or Countercurrent. When southward wind stress weakens in these regions, the northward mean dominates the flow. Flow separation in the vicinity of headlands may also be responsible for these northward flows. There is some evidence that during periods of northward flow a mean monthly τyN-driven southward current occurs inshore of the mean northward flow. At all latitudes, wind-driven ‘event’-scale fluctuations are expected to be superimposed on the seasonal nearshore flow. 4. The spatial distribution and seasonal variation oftthe offshore region of southward flow appear to be related to the spatial distribution and seasonal variation of curl τ. The seasonal variation of curl τ in these areas, curl τl, is roughly in phase with the seasonal variation of τy near the coast and roughly 180° out of phase with the seasonal variation of curl τ near the coast. Southward flow lags negative curl τ by from two to four months. The offshore region of southward flow is strongest during the summer and early fall. The mean annual location of the maximum flow is at about 250–350 km from shore off Washington and Oregon, and at 430 km off Cape Mendocino, 270 km off Point Conception and 240 km off northern Baja. The offshore branch of the flow bends shoreward near 30°N, which is consistent with the shoreward extension of the region of negative curl τ, so that by Cape San Lazaro (25°N), a single region of strong flow is observed within 200 km of the coast. 5. A third region of strong southward flow occurs at distances exceeding 500 km from the coast. The spatial distribution of this flow appears to be related to the spatial distribution of curl τ. 6. The mean northward flow known as the Davidson Current consists of two regions in which the forcing may be dynamically different—seaward of the continental slope off Washington and Oregon and between Cape Mendocino and Point Conception, the mean monthly northward currents appear to be related to the occurrence of positive curl τ; along the coast of Oregon and Washington the northward currents are not related to the occurrence of positive curl τ but are consistent with forcing by the mean monthly northward wind stress at the coast. 7. A region of southward flow that is continuous with the California Current to the south is generally maintained off Oregon and parts of Washington during the winter. This southward flow appears to separate the northward-flowing Davidson and Alaskan Currents in some time-dependent region south of Vancouver Island. The banded current structure is consistent with the distribution of curl τ, if southward flow is related to negative curl τ. 8. The seasonal progression of the California Undercurrent may be related both to the seasonal variation of the offshore region of strong flow (hence to curl τl) and to the alongshore component of wind stress at the coast. South of Cape Mendocino a northward mean also seems to be superimposed on the flow. This mean may be related to the occurrence of strong positive curl τ near the coast. Velocities at Undercurrent depths have two maxima, one in late summer and one in winter. The slope Undercurrent is indistinguishable, except by location, from the undercurrent that is observed on the Oregon-Washington continental shelf.