141 



head in a short section of a tidal channel can be closely reproduced by 

 a simple harmonic fluctuation, and the proper coefficient of friction 

 selected, the computed primary current affords a fair representation 

 of the actual currents, but a closer approximation would be secured 

 by applying- the corrections developed in the preceding paragraphs. 



Flood 



o 



2 J 



I I I I I I I I I LJ I I I I I I I I I I I I 



O IS 



Solar hours 



24 



Figure 40.— Current in Seebonk River, R. I., showing effect of short-period constituents. 



LIMITATIONS OX THE COMPUTATION OF CURRENT VELOCITIES FROM THE 

 OBSERVED HEADS IN A SHORT SECTION OF TIDAL CHANNEL 



284. As is well recognized, the velocity in a natural channel cannot 

 be reliably determined from the observed head even when the flow is 

 steady. In a short section of a tidal channel, the computation of tlie 

 velocities from the observed heads presents further complications. 

 These heads are the relatively small differences between the changing 

 tidal heights at gaging stations at the two ends of the selected section 

 of channel. Considerable accidental errors are inevitable in taking 

 off the tidal heights from the somewhat irregular curve produced by a 

 recording tide gage, and even greater errors in the timed readings of a 

 staff gage. When these departures happen to be in opposite directions 

 they produce errors which are large in proportion to the head. The 

 heads derived from the dift'erences of observed hourly readings are apt, 

 therefore, to vary so erratically as to afford little basis for a determi- 

 nation of the velocities. The most workable procedure is to find the 

 harmonic fluctuations which most nearly represent the actual fluctua- 

 tions of the tides at the ends of the section, and to derive therefrom 

 the corresponding harmonic fluctuation of the head. Obviously, more 

 consistent results may be secured from average tide curves than from 

 observations made during 1 day. 



285. In a long tidal channel, the heads between the entrances 

 usually are so large that accidental errors in the observed tidal heights 

 at these entrances become of minor importance ; but in such a channel 

 the currents may be due more to the storage and release of water in 

 the tidal prism than to the head between the entrances. 



286. In short, a direct measurement of the actual velocities in a 

 channel, however crude, is more reliable than the most refined calcu- 

 lation from the varying head and an assumed coeffici(Mit of roughness. 



