A 



90 

 80 

 70 

 60 

 50 



- 40 



- 30 



■ 20 



■ 10 

 



E 

 E 



>- 



-D 



1940 



1950 



1960 



1970 



Figure 4. A. Average sea-level time series for all U. S. tide gauge 

 stations with the exception of Alaska and Hawaii (Hicks 1978). B. Global 

 mean sea-level trend from tide gauge data (modified from Gornitz et al. 1982) 



Mean sea-level fluctuates seasonally (Pattullo \966; Nummedal and Humphries 

 1978). Along the U. S. gulf coast the annual amplitude is about 25 cm (Marmer 1952). 

 Sea level is maximum in early fall due to the steric effect (thermal expansion of 

 seawater above the thermocline). Other factors affecting seasonal sea level include 

 freshv^ater runoff from the continent (Meade and Emery 1971) and persistent winds 

 (Behrens et al. 1977). 



To test whether the thermal expansion of water also could have a long-term effect 

 on rising sea level, Gornitz et al.(l982) correlated the global mean sea-level trend for the 

 last century and the global mean temperature curve for the same time period derived by 

 Hansen et al. (1981). Using 5-year running means of both parameters they obtained a 

 correlation coefficient of 0.8. Best regression fit was obtained for a time lag of 18 years 

 between the temperature and sea-level rise curves. This lag time is of the same order as 

 the thermal relaxation time of the upper layer of the ocean. The findings suggest that at 

 least part of the observed global sea-level rise is attributable to the thermal seawater 

 expansion. A simple one-dimensional model of the heat flux into the ocean and the 

 attendant thermal expansion suggests that only about half of the observed rate of global 

 sea-level rise can be attributed to steric expansion; the balance may reflect a slow, but 

 steady, melting of polar ice sheets as well as lowering of global groundwater levels. 



PREDICTION OF FUTURE CHANGES IN GLOBAL SEA LEVEL 



Sea-level studies have traditionally been historical and empirical. The derived sea- 

 level curves have been so variable (Bloom 1977) as to make trend extraction and future 

 predictions all but impossible. Yet scientifically based estimates of future sea levels 

 should be a key component in decisions regarding the use and protection of low-lying 

 coastal lands. The findings reviewed above now permit such a prediction. 



168 



