level readings at Father Point (Pointe Au Pere) Quebec, for 11 years of available records 

 between 1941 to 1956. Although additional records were available, this period was selected 

 as representing the most reliable data. First-order level Knes were run from Father Point to 

 Kinston to estabUsh the elevation of Lake Ontario. 



It was assumed that the mean water level of Lake Ontario from June to September 

 (1952-58) defined a level surface. First-order leveling was run from western Lake Ontario to 

 eastern Lake Erie to define the elevation of Lake Erie. The mean water level of Lake Erie 

 from June to September (1952-58) was assumed to define a level surf ace. First-order levels 

 were run from western Lake Erie to southern Lake Huron to estabUsh the level of Lakes 

 Michigan and Huron. Lakes Michigan and Huron are assumed to have the same level because 

 of tlie wide and deep connection of both lakes at the Straits of Mackinac. First-order 

 leveling was run from northern Lake Huron to eastern Lake Superior to estabhsh the level of 

 Lake Superior. As with Lakes Ontario and Erie, the mean water level from June to 

 September (1952-58) was used to establish the datum for tlie entire shoreline of each lake. 

 All calculations were made to a resolution of 0.001 foot. Elevations assigned in the IGLD 

 (1955) do not, in general, agree exactly with elevations assigned to the same bench marks in 

 the NGVD system based on orthometric leveling. The differences, however, have never 

 exceeded 2 feet. 



VI. LONG-TERM VARIATION IN MEAN SEA L^VEL 



Although MSL with respect to the land is a relatively stable reference surface, it varies 

 irregularly with time (see Fig. 7) and location. In general, the sea level is either rising with 

 respect to tlie land or shows no discernible trend at low latitudes, and is faUing with respect 

 to the land at northern latitudes. The variability in sea level during this century, as revealed 

 by many tide gage records in the United States, is clearly shown in Figures 21 to 25 and 

 Table 5. In Figure 21, the data points for New York are connected by straight lines to 

 facilitate recognition of both the general trend and the variation about this trend. Records 

 for other stations in the figure are shown as point values only to demonstrate that, although 

 the general trend and longer perturbations are similar at all locations, there is also a great 

 deal of scatter with the shorter perturbations. The southernmost station (Charleston) is 

 repeated in Figure 22 and three additional stations are added. Again, a general trend toward 

 rising sea levels is clearly apparent. AU records are highly correlated with respect to some 

 perturbations, particularly the high sea level of the late 1940's, but not with aU perturbations. 

 The message is repeated for the four stations in the Gulf of Mexico (Fig. 23). 



Records for three stations on the west coast of the conterminous United States and for 

 Honolulu are shown in Figure 24. The same upward trend and general simUarity of the 

 perturbations about the trend are revealed. 



A change in this general trend is seen in Figure 25 for Alaska. The record from the 

 northernmost station (Skagway) shows a clear trend toward decUning sea levels relative to 

 the land; the record from the southernmost station (Ketchikan) shows a great deal of scatter 

 but no clear trend. 



53 



