General Considerations 



13 



The density of sea water increases with de- 

 creasing temperature and with increasing sahnity 

 and pressure. Except in quite dilute sea water, 

 the temperature of maximum density is lower 

 than the freezing point. The range of density 

 in the open sea is between about 1.02 and 1.06. 

 It may, of course, be lower in inshore waters 

 in the vicinity of river mouths. At constant 

 pressure the major changes in density in the sea 

 are associated with temperature, so that to a 

 first approximation the change of density (com- 

 puted for constant pressure) with depth is in- 

 versely proportional to the change of tempera- 

 ture. 



Many processes in the sea depend on the 

 density distribution. The ocean basins are 

 largely filled with water of relatively high 

 density formed in high latitudes ; overlying this 

 dense water in middle and low latitudes, and 

 separated from it by the pycnocline, is the sub- 

 surface mixed layer, varying from a few meters 

 to several hundred meters in thickness but 

 averaging about 75 meters, of water of high 

 temperature and low density. The relative rate 

 of change of density with depth may be taken 

 as a measurement of the vertical stability of 

 the water (Sverdrup, Johnson and Fleming, 

 1942, p. 417). Stability in the region of the 

 pycnocline is much higher than above or below 

 it, so that exchange of water across it tends to 

 be small. 



All parts of the ocean and its bordering seas 

 are in communication with each other, and are 

 in continuous motion. The rates of movement, 

 however, differ greatly in different areas. Thus, 

 although there is eventual complete interchange 

 of water between all oceans and seas, some parts 

 are partially isolated from others, the exchange 

 between these parts being much slower than 

 within them. 



Near-surface currents and mixing within the 

 upper layer 



Currents in the upper, mixed layer of the 

 sea are primarily generated by winds, and, con- 

 sequently, the major horizontal surface currents 

 of the ocean correspond to the field of wind 

 stress (Munk, 1950). The average locations and 

 velocities of the important surface currents are 

 well known from numerous observations of 

 merchant ships and research vessels, and appear 

 on many charts. 



The velocities and volume transports of the 

 major near-surface currents are large. For ex- 

 ample, the mean speed of the Florida Current 

 is about 193 cm/sec. and of the Kuroshio about 

 89 cm/sec. The volume of water flowing 

 through the Florida Straits in 15 years is equal 

 to the volume of the upper 500 meters of the 

 whole North Atlantic, and the transport of wa- 

 ter by the Kuroshio between the Northern 

 Ryukus and Kyushu in 50 years is equal to the 

 upper 500 meters of the whole North Pacific. 



Because of the large surface currents, intro- 

 duced materials tend to be carried away from 

 the sites of introduction to other parts of the 

 upper mixed layer of the sea. Thus, no area of 

 surface water in the ocean is isolated for long 

 periods from the remaining areas. 



The currents are not steady streams, but have 

 a complicated fine structure, with many eddies, 

 jets, and filaments. In consequence of this 

 turbulence, on both large and small scales, dis- 

 solved materials in seawater are rapidly dis- 

 persed horizontally. The rate of dispersion is 

 about a million times the rate of molecular dif- 

 fusion, and depends on wind speed, current 

 shear, vertical and horizontal density gradients, 

 direction of dispersion, and the dimensions of 

 the area considered. Because of this large num- 

 ber of variables and the lack of knowledge of 

 turbulent processes, it is not possible to predict 

 accurately the horizontal dispersion in particular 

 areas. If even moderately precise values are re- 

 quired, experiments must be conducted in the 

 area of interest. Some of the results of such 

 studies are reported by Wooster and Ketchum 

 (Chapter 4) . 



The rate of vertical diffusion in the upper, 

 mixed layer, although much less than that for 

 horizontal dispersion, is nevertheless about a 

 thousand times greater than molecular diffusion. 

 The extent of vertical stirring in the upper layer 

 depends on the magnitude and uniformity of 

 the wind stress and on the vertical density gra- 

 dient. Convective processes, and, in coastal 

 areas, strong tidal currents, also contribute to 

 vertical mixing. The mixing rate in the upper 

 layer has been measured by changes in the ver- 

 tical distribution of radio isotopes following 

 weapons tests. Revelle, Folsom, Goldberg, and 

 Isaacs (1955) report that in one such test the 

 lower boundary of the radioactive water moved 

 downward at about 10"^ cm/sec. until it reached 

 the thermocline, where it abruptly stopped. 



