27 



to those of the undistorted openings. In all cases^ the 



adjustments required consisted only of rounding the corners 



of the sills and abutment walls of the distorted openings 

 to reduce contraction effects. 



b. The proposed barriers were then installed in the models 

 and navigation openings (if any) were adjusted as found 

 necessary during the flume tests so that their discharge 

 coefficients were correct throughout the full range of 

 head differentials. 



c. Both the astronomical tide and hiirricane tide generators 

 were then readjusted to reproduce the same tides in the 

 ocean portion of the model as occurred during the base 

 test condition to be used for evaluating the effects of 

 the plan in question. The readjustment procedure was 

 necessary because installation of different plans in the 

 model caused various changes in the tidal prism of the bay 

 system^ both for conditions of astronomical tides and hur- 

 ricane tides^ which in turn would have affected the range 

 of the ocean tides in the model had not the generators 

 been readjusted for each barrier plan. 



d. The plan in question was then subjected to all astronom- 

 ~ ical tide and hurricane tide tests for which information 



was desired, and all necessary measurements of resulting 

 tidal and current phenomena throughout the bay system 

 were obtained. In the case of a few plans, it was de- 

 sired to determine in detail the distribution of current 

 velocities in one or more of the navigation openings for 

 conditions of maximvim hurricane tide head differential and 

 maximum astronomical tide head differential across the 

 structures. Because of the small size of the navigation 

 openings in the distorted model, these data were obtained 

 by observing the maximum head differentials in the model, 

 establishing these head differentials in the flume con- 

 taining the undistorted model of the opening in question, 

 and obtaining the necessary measurements of velocity dis- 

 tribution therein. 



By following this sequence of steps, all errors in model results that 

 would have been caused by the distorted model scales were eliminated, and 

 the test results presented herein may be considered quantitative with 

 respect to the effects of the various structures in the prototype. 



Middle Bay Barriers 



39. The elements of the Middle Bay barrier plan are shown on fig. 15- 



