A broad (about 40 km wide) cold filament observed off Point Arena in advanced very high resolution radiometer images at the end of June 1986 evolved into a narrow (10--20 km wide) cold filament embedded in a large-scale (about 200 km alongshore) cool anomaly by mid July. The cool anomaly was part of a much larger-scale anomaly, stretching from north of Cape Mendocino to south of San Francisco, in which several cold filaments were embedded. The filament oof Point Arena, surveyed by ship from July 7 to 19, 1986, extended offshore around the edge of a cyclonic center. The horizontal temperature gradient was sharpest (up to 1.5 ¿C/2.2 km) across the southern edge of the filament, and a salinity front was located within the cold filament; temperature and salinity fronts were not coincident. The surface salinity was high (33.2 psu (pratical salinity unit)) in the cyclonic center and decreased across the cold filament to about 32.6 psu. The entire cool anomaly was denser than surrounding water, and there was a sharp density front (1 sigma-t unit/10 km) as its northern edge.

At the inner (cyclonic) edge of the filament the temperature and salinity variations were density compensating. The cold filament was advected offshore by a jet which was symmetric about its axis, located about 10 km north of the cold filament core. The surface manifestation of the cold filament was highly variable in space and time: the location of the filament core changed by about 5 km in 14 hours along one transect; the velocity field was less variable. Several narrow (10 km wide) water mass anomalies were evident, including a salinity subduction zone in the northern segment of the cold filament. Thermohaline intrusions 50--100 m thick may have been caused by complex frontal interactions upstream of the anomaly and offshore in a region of jet confluence. Fron a second survey (July 27 to August 5, 1986) the offshore jet had reoriented to alongshore, possibly related to an observed increase in the poleward flow over the continental slope. Prior to the second survey there were changes in the nearshore wind stress and the large-scale wind stress curl. ¿ American Geophysical Union 1989