139 



conditions. Distortion, however, essentially requires additional flow 

 resistance to be provided at the bottom by means of artificial roughness 

 elements. Calibration of such a model, particularly one involving two 

 fluids, can be a tedious process. It took several years to make the San 

 Francisco Bay model fully operational. Other major physical models of this 

 nature are those of the Mississippi River and Chesapeake Bay. 



Extensive field investigations were carried out in the fifties for 

 understanding salinity intrusion through direct (prototype) evidence, and 

 to calibrate physical models such as the ones noted above. Reference may 

 be made to the investigation by Sir Claude Inglis (Inglis and Allen, 1957) 

 on the Thames in England, and by Prof. Pritchard (Pritchard, 1952) on the 

 Chesapeake Bay. In both estuaries, as elsewhere, studies which began in 

 the fifties (and earlier) have been of an ongoing nature, and have 

 continued until the present time. The reason for this is both a continued 

 interest in the basic aspects of the mechanism of mixing between salt water 

 and fresh water in the real environment (see, for example, Dronkers and van 

 de Kreeke , 1986), and also because new engineering problems continually 

 arise and, therefore, must be examined afresh. Physical models of large 

 estuarine systems such as the San Francisco Bay, Chesapeake Bay, the 

 Mississippi and New York Harbor have been retained, and are used as needed 

 by the Corps of Engineers and other agencies. As fresh input for 

 modification and calibration of such models, prototype studies are 

 cpnducted, although many now tend to be highly site-specif ic , given the 

 costs involved in field work. 



Computer technology has made it possible to develop sophisticated 

 numerical models for handling estuarine hydrodynamics including salinity 

 intrusion. The early models, in the sixties and early seventies, were 

 typically one -dimensional, simulating cross-sectional average processes in 

 the longitudinal direction (e.g., Harleman et al . , 1974; Miles, 1977). 

 These were followed by two-dimensional, depth- averaged models, e.g., such 

 as the one incorporated in the TABS-2 system of estuarine numerical models 

 used by the Waterways Experiment Station (Thomas and McAnally, 1985). More 



