o 



BOSTON. MASS. 



1920 1930 1940 1950 1960 



20r 



15 



10 



5 







1970 



20 

 O 10 



e 







1.0 





:;f\M 



TEMPERATlfRE 

 I 



B- 



26 



27 



28 , 



29*' 



30 



31 



5 - 



Figure 7. Annual variation in mean 

 sea level at Boston, Massachusetts, 

 compared with mean annual water 

 density (inverted scale), temperature, 

 and salinity at the tide gauge 

 station, plus annual discharge of the 

 Charles River including urban waste 

 water (Emery and Uchupi 1972). 



the summer and winter solstices 

 (Figure 10). As Emery and Uchupi 

 (1972) pointed out, there is also a 

 "fortunate circumstance" such that the 

 times of lowest low tide come during 

 the night or early morning, so that 

 animals and plants accustomed to life 

 below the tide line are not exposed to 

 excessive heating and desiccation 

 (Figure 10). 



While the National Ocean Survey 

 (NOS), National Oceanic and Atmos- 

 pheric Administration, 

 annual predictions of the 

 heights and times for many 

 secondary locations along 

 the tidal pattern found in 

 often be quite different 

 observed or predicted at 



publishes 

 daily tide 



primary and 

 the coast, 



a marsh may 

 from that 



the nearest 



reference station. Generally, the 

 tidal signature found inside a narrow 

 opening, behind a barrier spit, or up 

 a winding tidal creek will show a 

 reduced tidal range and a delayed time 

 of high water; flood tide will be 

 shorter than ebb with a faster mean 

 current speed. 



The tide heights published in the 

 NOS tables are given with reference to 

 mean sea level, an elusive datum we 

 have spent some time discussing in 

 earlier sections. To eliminate or at 

 least reduce much of the short-term 

 variability in sea level measurements, 

 the NOS uses a 19-year average of 

 hourly tide gauge records for 

 most of its tidal work. The choice 

 of this averaging interval has 

 some astronomical significance and 

 represents a practical choice, given 

 the lengths of records available for 

 most stations and the level of 

 variability in the yearly data (Hicks 

 1968). Sea level, however, is often 

 estimated by NOS from a tidal record 

 that is only 1 month to 1 year 

 long by comparing it to a nearby 

 station with a 

 and similar tidal 

 Similarly, it is 

 to develop yearly 



19-year record 



characteristics. 



relatively easy 



tidal predictions 



12 



