above, has been shown to agree with independent 

 observations of ship drift over large parts of the 

 North Pacific (Reid, 1961). No single level, how- 

 ever, is strictly appropriate for use as a "level of 

 no net motion" in the subarctic. Drogue measure- 

 ments in the study area (Budinger et al., 1964) 

 indicated a weak current at 1,000 m. depth, so 

 that slight error from this source is expected m 

 charts of geopotential referred either to surfaces 

 of 500 dbar. or of 1,000 dbar. 



Current direction denoted by dynamic topog- 

 raphy varies from year to year, particularly in 

 the offshore and plume provinces (fig. 11). Geo- 

 strophic speeds, proportional to geopotential 

 gradients, are slow and generally uniform beyond 

 the nearshore province; values of 3 to 8 cm. 

 sec."' are typical. 



Redistribution of mass due to nearshore up- 

 weUing results in a dynamic topography that 

 consistently suggests intensified southerly flow 

 (30-50 cm. sec."') near the coast, and an eddy 

 or loop with even greater velocities at about lat. 

 43° N. It is unlikely, however, that geopotential 

 gradients reflect current velocities in nearshore 

 areas as accurately as in the offshore and plume 

 provinces. Time lags in the response of distri- 

 bution of mass to changes in ■wind-driven transport 

 probably negate the assumption of a steady state 

 in the nearshore provinces. 



The ^\and-induced component of flow, the 

 Ekman transport, is superimposed on geostrophic 

 flow in the upper layer of the sea. Ekman transport 

 by blocks of 1° latitude, averaged zonally from 

 longs. 130.5° W. to 124.5° W. in June and July, 

 was calculated from the source and by the method 

 given later in this paper. These averages demon- 

 strate that wind-induced flow is generally to the 

 west or southwest and that the flow — in particular 

 its zonal component — intensifies with decreasing 

 latitude (fig. 12). No such marked tendencies are 

 apparent from analogous meridional averages, 

 except that of direction. 



Average current speeds computed from Ekman 

 transport, presumed to extend to 30 m. depth, 

 generally are lower, often by a factor of 10, than 

 geostrophic speeds. The effect of Ekman transport 

 on distribution of variables is probably very 

 important, however, where Ekman transport is 

 not parallel to geostrophic transport — in the 

 nearshore province, parts of the offshore province, 

 and near the plume boundaries. 



Tongue Structures 



Recurved, tonguelike isopleths are apparent in 

 distributions of temperature, saUnity, density, 

 and oxygen concentration over the upper-zone 

 waters. Along any approach to the coast from 

 the offshore province, one thus encounters a 

 maximum in temperature (fig. 3), minimums in 

 sahnity and density (figs. 4 and 5), and both 

 a minimum and a maximum in oxygen con- 

 centration (fig. 6). These large-scale extremes are 

 mainly confined to the surface layers but occur 

 occasionally at greater depths. Each large-scale 

 extreme, except the oxygen maximum, occurs 

 within the area here defined as the plume province. 



The sahnity minimum is due simply to the 

 presence of the low-salinity plume. In the absence 

 of runoff, saUnity at the sea surface would pre- 

 sumably decrease monotonically from about 

 33.5°/oo in the upweUing areas to about 32.5°/oo 

 with increasing distance offshore. 



The tonguelike ridge of higher temperatures is 

 ascribable to two effects of the plume itself. 

 Because the secondary halocline at the plume-sea 

 water interface partly coincides with the ther- 

 mocline, stability is augmented in this interface 

 layer; consequently, downward heat flux through 

 the thermocline is less than that across the un- 

 augmented thermocline of the offshore province. 

 Retention above the thermocline of heat gained 

 in siu-face layers thus tends to be greater in the 

 plume province than beyond. Second, this greater 

 stability in the plume may be considered to 

 dictate a preferred site for development of the 

 summer thermocline at depths generally less than 

 20 m.' Wind-mixing in the offshore province, in 

 the absence of the effect of the plume on stability, 

 is effective to greater depths. The offshore ther- 

 mocline accordingly develops at greater depths in 

 the manner described by Tully and Giovando 

 (1963). By this difference in mixed-layer depth 

 (depth to the top of the summer thermocline), 

 heat gained in the surface layer is constrained 

 to a smaller volume in the plume than offshore, 

 and hence the plume province has higher summer 

 temperatiu-es. The shoreward decrease of temper- 

 ature in the nearshore province results simply 

 from upwelling of cold water. 



A measure of the extent to which the plume 



' With the partial exception of July 1963, the offshore limit ol the plume 

 province is in fact approximated by the 20-m. isopleth of thermal mixed layer 

 depth (fig. 13). 



OOEANOGRAPHIC CONDITIONS IN NORTHEAST PACIFIC OCEAN 



613 



