shift in shoreline position (Hands, 1976). Lofgren (1969) discusses many examples of land 

 subsidence due to the application of water. Although subsidence due to this cause can be 

 troublesome in many continental locations, it is not likely to affect the relative elevation of 

 land and sea in the coastal zone. 



Worldwide changes of sea level during the past few thousand years have been summarized 

 by ChappeU (1974), Morner (1976), Pirazzoli (1976), and Pouchet (1976). Each of these 

 references provides improved comprehension of the subject and supplies perspective for 

 modern measurements. 



2. Conclusions and Recommendations. 



This brief examination of the records for sea level variability in this century and over tlie 

 past few thousand years demonstrates that the sea level has been changing for as long as any 

 type of records exists. The general trend toward rising sea level at midlatitudes and faUing 

 sea level at northern latitudes has been continuing for many centuries. Many factors 

 contribute to the changes and, except for subsidence due to the removal of subsurface fluids 

 and other minerals, there appears to be little reason for believing that future predictions can 

 be made with great confidence. The optimum procedure to use for any essential 

 extrapolation depends on the reason for extrapolation. For structure design, extrapolation 

 of observed trends for the next 100 years is perhaps the most conservative policy. Extrapo- 

 lation, however, should always be based on the best available data. 



3. Data Sources for Detailed Studies. 



The NOS maintains a running summary of the montlily MSL, the mean and extreme high 

 and low waters of tlie month, and many other tidal statistics. Photocopies of these records 

 may be obtained from NOS upon request. A sample NOS summary of tide level and sea level 

 is provided in Figure 26. This particular sample was selected because it shows several 

 realistic characteristics of the records. For example, few observations may be missing in a 

 month where enough data are collected to provide a meaningful estimate of the tabulated 

 quantity. In these cases the estimate is shown in parentlieses. Erroneous data may be 

 entered in the record and not immediately detected; e.g., repairs to a gage may cause a small 

 shift in gage zero that is not measured until the next visit of a survey party. In these cases, 

 the incorrect data are ruled out and corrected values entered. Because tide level and sea 

 level are entered on the same sheet in this sample, it is readily seen that although MSL and 

 MTL are highly correlated, they are different. Even where annual means are considered, 

 MTL may be above MSL in some years and below MSL in others. 



The National Geodetic Survey (NGS) (part of the National Oceanic and Atmospheric 

 Administration) continually relevels various survey lines throughout the United States. 

 Elevation changes between surveys may be used to map tlie extent of vertical crustal 

 movement. If the engineer needs information on vertical changes for a particular problem 

 area, the request to NGS should specify uncorrected data to compute vertical change. For 

 many geodetic purposes, it is necessary to combine a mixture of time-scrambled data; 



59 



