are small compared to the short-term variability so the direction of the 

 trend, if any, is not clearly shown by this data set, A longer record 

 may be necessary to establish a convincing trend. 



The seasonal nature of the above MSL volume was shown in Figure 16. The 

 material removed from the beach during the autumn and winter is apparently 

 replaced (or nearly so) during spring and summer. Extending the profile lines 

 below MSL would allow determination of the offshore changes which have been 

 shown to be important in the beach process. It is possible that the volume 

 change of an extended profile line measured relative to some below MSL datum 

 would be zero if material removed from the beach is stored in the offshore 

 within the range of profile line measurement. Empirical eigenf unction anal- 

 ysis is very useful in showing the regions where changes in beach shape take 

 place. Aubrey (1979) demonstrated that the second temporal eigenfunction 

 showed removal of beach material from onshore and storage in the offshore 

 zone at Torrey Pines Beach. Unless profile line measurements are taken with 

 the method of analysis in mind, it is only fortuitous that a "higher order" 

 analysis technique, which is more powerful and sophisticated, will provide 

 additional insights. The application of empirical eigenfunctions to the 

 Torrey Pines data was fruitful because the study was designed, in part, to 

 develop and test the method. Empirical eigenfunctions did not provide in- 

 sights into Holden Beach processes that were not available through more 

 traditional and straightforward analysis methods, but the reasons may be due 

 more to the limitations of the data than to the technique. Thougli not use- 

 ful for the Holden Beach data, there are indications that empirical eigen- 

 functions will be helpful in the interpretation of temporal and spatial vari- 

 ability of other data in this series. 



The results of this study suggest that Holden Beach has at least three 

 separate systems to be investigated and interrelated in order to understand 

 processes, such as differences in response to environmental forces (erosion 

 rates, variability of profile changes, and mean slopes) along the beach. 

 These are the Lockwoods Folly reach (profile lines 1 through 3) the central 

 reach (profile lines 4 through 18) and the Shallotte reach (profile lines 19 

 through 21). Refraction of waves around shoal area, strong tidal currents, 

 and shifting channels near inlet reaches require special, localized 

 observations . 



Changes observed along the central reach were visually correlated with 

 wind and water level records taken at Wilmington. These correlations were 

 not altogether satisfactory because of the location of Wilmington relative 

 to Holden Beach; direct wind and water level observations at the site would 

 have shown a more reliable correlation with beach changes. 



The identified erosion events were fairly well correlated with high 

 water levels and strong winds during an observation period. The conditions 

 which cause accretion, however, are not easily identified since high water 

 levels and strong onshore winds occurred during these intervals as well. 

 Accretion events seemed to be correlated with gentle onshore winds occur- 

 ring for several days before the survey (Figs. 23 to 27). Profile line 

 measurements must be taken more frequently in order to isolate the effects 

 of individual events. 



54 



