THE PACIFIC OCEAN 



107 



2. The convergence in latitude about 40° north off 

 the coast of Japan in the figure corresponds to the west- 

 ern part of the northern polar front as shown by Merz. 



3. The westerly current in the inner part of the 

 Gulf of Alaska is seen in both maps. 



4. The Equatorial Countercurrent runs in nearly the 

 same regions on the two maps. 



5. The line separating the westerly and easterly 

 currents in the eastern part of the South Pacific practi- 

 cally coincides with the corresponding line of subtropic 

 convergence as shown by Merz. 



It is hardly a coincidence that the surface currents, 

 which are derived from dynamic computations, agree 

 with the observed surface currents, in spite of the fact 

 that the latter result from the combined effect of the 

 wind and the primary distribution of density. This 

 agreement must be interpreted as indicating that the ef- 

 fect of the wind is to maintain a certain distribution of 

 density, and the computation of the currents on the basis 

 of the density distribution actually includes part of the 

 effect of the prevailing winds. 



Palmen (1930) has recently discussed a number of 

 observations from the Gulf of Bothnia which demonstrate 

 in a striking way the effect of the wind on the distribu- 

 tion of density. When the wind blows in the direction of 

 the Gulf the light water is accumulated along the right- 

 hand shore and the heavy water along the left-hand shore 

 The current, which is computed from this distribution of 

 density (the convection current according to Ekman's 

 terminology) has a velocity corresponding to the velocity 

 of the wind current which would be produced under the 

 given circumstances. This example deals with conditions 

 in a narrow bay, but it is probable that the results are of 

 general importance and that even in the open ocean we 

 may find that the wind changes the distribution of density 

 in a corresponding manner. This would mean that a 

 prevailing wind maintains an abnormal distribution of 

 density. If the wind should stop blowing, the normal dis- 

 tribution of density would be re-established, and the 

 dynamic computation would give the current which would 

 be present if the tangential force exerted by the wind on 

 the surface were absent. Supposing these considerations 

 to be correct, we may regard our dynamic charts as 

 representing the total currents resulting from the dif- 

 ferences in density which would occur in the absence of 

 wind, and from the abnormal distribution of density 

 which is established and maintained by the action of the 

 wind. 



As to the character of the wind current we remind 

 the reader of Ekman's theory. According to this the 

 total transport of water is directed 90° to the right of 

 the direction of the wind in the Northern Hemisphere, 

 and 90° to the left in the Southern Hemisphere. The 

 depth to which the wind current reaches depends on the 

 latitude and on the eddy viscosity, which again is a func- 

 tion of the stratification of the water. 



In general, it is assumed that at some distance from 

 the equator the wind currents reach to less than 100 me- 

 ters in depth, but the effect on the distribution of density 

 must reach much deeper. Since the surface water is 

 light, a transport of surface water to the right of the di- 

 rection of the wind leads to an accumulation of light 

 water on the right-hand side of the wind, and on the left- 

 hand side the light surface water must be replaced by 

 heavier water from greater depths. On the right-hand 

 side of the wind the surfaces of equal density are de- 

 pressed, and on the left-hand side they are raised. The 



effect may reach to considerable depths and, owing to 

 this "abnormal" distribution of density, a current in the 

 direction of the wind is created. 



In the open ocean the maintenance of an abnormal 

 distribution of density represents only part of the effect 

 of the wind. If it represented the total effect, the condi- 

 tion 



dvi, 

 (^d^)o 



dv- 



would have to be fulfilled. Here v is the coefficient of 

 eddy viscosity, v^ and Vy, the components of the convec- 

 tion current, and Tx and Ty, the components of the tan- 

 gential streas of the wind. This condition, which may 

 be satisfied in a narrow channel, is never fulfilled in 

 the open ocean, since a decrease of the required magni- 

 tudes of the velocity near the surface does not occur. 

 Pure drift currents will, therefore, be present beside 

 the convection current, but under stationary conditions 

 the climatological factors may balance their effect on 

 the distribution of density. We shall not enter any fur- 

 ther on this subject but shall, in the following, consider 

 only the currents which are associated with the distri- 

 bution of density. 



We shall first examine the currents in the tropo- 

 sphere, which extend to a depth of about 500 meters. In 

 the North Pacific the dominant feature is represented by 

 the anticyclonic current system which has its center in 

 latitudes 25° to 30° north, and in longitude about 180°. 

 It is perhaps not correct, however, to use the term 

 "center" because, apparently, we find an axis of maxi- 

 mum elevation of the isobaric surfaces stretching from 

 the region to the south of Japan toward the Hawaiian Is- 

 lands. On the northern side of this axis we find currents 

 toward the east, and on the southern side currents to- 

 ward the west or southwest. 



A similar current system is probably present on the 

 Southern Hemisphere, but our observations are not ex- 

 tended over a sufficiently wide area to disclose the dif- 

 ferent branches of this system. In our charts we find 

 the westerly current represented between latitudes 0° 

 and 20° south, although it appears to have a less stable 

 character than the corresponding current in the North- 

 ern Hemisphere. The easterly current is seen to the 

 south of latitude 30° south between longitudes 80° and 

 120° west. Between the tropical westerly currents we 

 find the Equatorial Countercurrent which is in longitude 

 140° north and latitude 11° north where it runs as a very 

 strong and narrow current, and in longitude 175° west 

 appears as a rather broad and weak current extending to 

 both sides of the equator, but to the greatest distance on 

 the northern side. 



It is of advantage to discuss separately the different 

 branches of the current systems in the two hemispheres 

 and we shall, as previously, begin with the most south- 

 ern part of the South Pacific. 



In the southeastern part of the Pacific our charts 

 show the northern branch of the South Pacific east drift. 

 The current runs toward the east between latitudes 30° 

 and 40° south, and can be followed from the surface to a 

 depth of 500 meters, but the velocity decreases down- 

 ward and is very small below 300 meters. Above a 

 depth of 400 meters the current appears to turn toward 

 the west in latitude 30° south, but below 400 meters a 

 closed circulation appears to be present between longi- 

 tudes 80° and 120° west. 



