Salinity of the Ocean, its Variation in Oceanic Space and in Time 



157 



served values leads to a maximum lateral exchange coefficient of 4-9 x 10^ gcm-^ sec^^ 

 which in view of the intense mixing in the Gulf Stream is of an order of magnitude 

 in good agreement with this coefficient (see p. 105). 



From the extensive data available for the Australian-Asiatic Mediterranean (largely 

 from the "William Snellius" Expedition) Visser (1928) has determined the annual 

 temperature and salinity variations and has discussed them in detail. The rather large 

 annual variations here (more than 2-5%o) are also mainly produced by advection. 

 Table 70a gives, as an example, some values for the eastern Java Sea. While the tem- 

 perature shows the equatorial double wave with maxima in April and December and 

 minima in January and August, the salinity shows only a single main maximum in 

 September and single minimum in May. These phenomena are due to the monsoon 

 change and the associated changes in advection. During the east monsoon cold sahne 

 water flows in from the east (May to August) and the salinity rises ; it remains almost 

 constant during the monsoon change (September to November) and falls from De- 

 cember to February, while the west monsoon carries water of lower sahnity in from 

 western Java Sea. 



Table 70a. Annual temperature and salinity variations in the eastern Java Sea 



In the interval between the monsoons from March to May the changes are only 

 small. It is obvious that here also the advection of water masses of different sahnities 

 is the principal factor involved. 



In the Polar regions the annual salinity and temperature variations may be due not 

 only to the effects of advection but also to ice formation and melting thereby producing 

 large amplitudes. The annual salinity variation may be increased to as much as 25%o 

 or more, but this occurs only in a very thin top layer; the layers underneath show only 

 a small annual variation with a maximum in winter and a minimum in summer. This 

 small annual variation can be regarded as a consequence of ice formation. Table 71 

 shows, as an example, conditions in the homogeneous top layer of the east Siberian 

 Sea from November 1922 to October 1923. 



SvERDRUP (1929) pointed out that between February and the end of May there was 

 an increase of 0-47%o in the salinity of the layer below the top layer. If this is assumed 

 to be due to ice formation, and the ice formed is assumed to have a salinity of 5%o, 

 then the increase observed corresponds to an ice layer 67 cm thick which is in agree- 

 ment with actual measurement of ice thickness. The salinity decrease between May 

 and August is about 0-55%o, corresponding to the melting of 87 cm of ice which is 

 also in agreement with the observed values. 



Footnote continued from opposite page 



Knowing the amplitude of the variation in the region of the flow the quantity Ayj pu can be calcu- 

 lated and knowing p and u a numerical value of the lateral exchange coefficient Ay can be found 

 (see p. 106. et seq.). 



