Table 9 is similar to Table 8 and gives the distribution functions for higher high waters, 

 all high waters, hourly tides, all low waters, and lower low waters. Figures 28, 29, and 30 are 

 based on data of the type shown in the column for hourUes. Figure 31 and all figures in 

 Appendix B are based on data from column 4 of the respective parameters. 



Zetler and Cummings (1967) have shown that the values of the harmonic constants, 

 based on observations for different years, may vary by 4 to 8 percent. The following dis- 

 cussion shows that differences of the same magnitude can be expected between probabihty 

 estimates based on a primary tide location for a location and that obtained by using primary 

 calculations for a reference station. 



VIU. APPLICATION OF THE TIDE PROBABILITY TABLES 



The NOS publishes tide predictions, using equation (2) for only 50 reference tide 

 stations in the United States. Tide predictions for all other locations (i.e., secondary 

 stations) are normally obtained by modifying the predictions for the particular reference 

 stations designated in Appendix C. For aU U.S. Atlantic coast tide stations, the mean tidal 

 range is given in Appendix C and in NOS tide tables; for all U.S. Gulf of Mexico and Pacific 

 coast tide stations, the mean diurnal range is pubUshed in the tide tables. The tidal height 

 values given in tlie probability tables have been normalized with respect to one-half the 

 mean tidal range for Atlantic coast tide stations and with respect to one-half the mean tidal 

 range or tlie mean diurnal tidal range as indicated in Table 4 and tlie station tables in 

 Appendix B for the Gulf of Mexico. The mean diurnal range was used for all Pacific coast 

 tide stations. Thus, the astronomical tidal height probabilities for any location can be 

 estimated by multiplying the values tabulated in Appendix B for the appropriate reference 

 station by the appropriate tidal range parameters as obtained from Appendix C, from the 

 NOS tide tables, or from the latest bench-mark sheets. It appears from Figure 17 and the 

 discussion of the gulf coast low water datum, that diurnal range might be more suitable than 

 the mean range for some gulf coast locations where the mean range has been indicated and 

 vice versa. If this is the case, the height values can be converted by the ratios of the mean 

 and mean diurnal ranges from Table 7. The computed probabilities will not be affected. 



Actual tide levels near the coast are affected by both meteorological and gravitational 

 forces. The events which cause the gravitational tides, and the meteorological effects on sea 

 level are, in general, uncorrelated, but the effect of wind in generating a sloping water 

 surface is inversely related to the water depth. Consequently, the wind effect on sea level is 

 slightly greater at low tide than at high tide. The effect is greater where tliere is a long 

 shallow-water fetch in an upwind direction from shore, than where the water is relatively 

 deep. This interaction between the tide and the wind effect on water level can be shown by 

 examining the difference between the observed and predicted tides during storms (Fig. 32). 

 The interaction between tide and storm surge at any location is a function of wind velocity 

 and wind duration as well as the stage of the tide. 



74 



