NOTES ON THE BEHAVIOUR OF FRESH WATER ENTERING 



THE SEA 



By John P. Tully, Pacific Oceanographic Group, Nanaimo, B.C., Canada 



It has been observed(l, 2, 3, 4) that fresh water flowing into the sea from 

 any river forms a layer of brackish water flowing on the normal sea-water 

 of the region (Fig. 1). 



In the upper zone of fresh-water influence the salinity increases to 

 seaward, and with depth to an inflexion in the salinity-depth relation. 

 The upper zone of fresh-water influence is characterized by horizontal 

 and vertical salinity gradients that are large compared to those in the 

 underlying normal sea-water in the lower zone or beyond the influence 

 of the river. 



In general the fresh water from the land drainage first overruns the 

 sea-water in the river estuary where the divergence and course of the 

 outflow is determined by the shores. Such a situation is oceanographic- 

 aUy simple because the flow is either seaward or shoreward, depending 

 on the tide, and there is only one source of sea-water and one of fresh 

 water. Such systems are exemplified by estuaries, sunken river valleys, 

 fiords, and narrow inlets(l). 



When the outflow* reaches the ocean, or a large seaway where it is 

 not confined by the shore-line, the situation becomes complex because 

 the direction of flow may vary with time, the tidal movements are not 

 simply reversing, and there may be several sources of fresh and sea-water 

 in any region. Complex structures and tide rips are usuaUy found in 

 such places. 



In any case fresh water is continuously supplied from land drainage, 

 and does not accumulate indefinitely. Therefore it is persistently 

 displaced seaward at a rate equal to its supply. The water near the river- 

 mouth is less dense than that to seaward. Consequently, according to 

 Bjerknes, the gradient (solenoidal) effect tends to increase the circula- 

 tion such that the surface water is accelerated seaward, and the deeper 

 water shoreward. Since circulation of this type is present, but does not 

 increase indefinitely, the gradient effect is balanced by the inertial (Cor- 

 iolian) effect and friction. This results in the so-caUed gradient flow 

 along the isobars, but does not define the displacement of water across the 

 isobars from higher to lower pressure. However, such a flow does occur, 

 as evidenced by the mixing of fresh and sea-water. For convenience the 

 component of the pressure gradient which is not balanced by the 

 Coriolian force, and results in trans-isobaric flow, may be termed the 

 •displacement head. 



Evidently the effect of the Coriolian and the displacement forces must 

 exist in any instance of flow involving fresh and sea water. In an open 

 seaway or the ocean, where the boundaries are not restrictive, the grad- 

 ient flow is large compared to displacement across the isobars, but in a 

 simple channel the latter is the only mechanism by which stationary 

 conditions may be maintained. 



The isostatic forces could be evaluated from suitable data by simple 

 calculations. However, the definition of a horizontal isobaric surface 

 required for reference presents considerable difficulty, since the mag- 

 nitude of the frictional forces in coastal regions cannot be accurately 

 assessed at the present time. Furthermore, the presumption of a con- 

 stant circulation over a short period of time cannot be assumed, because 

 the continuaUy varying tidal and run-off conditions make it doubtful if 



267 



