variety of scientific disciplines for many years (Lorenz, 1959), but it has 

 only recently been applied to examining variability within the coastal zone. 

 When applied to analysis of a profile line resurveyed over a period of time, 

 the method is useful in determining the topographic variability in the onshore- 

 offshore direction and in time. A comparison of the eigenf unctions of a series 

 of profiles is useful in determining the longshore variability. The technique 

 has been applied to studies on beaches, islands, and other coastal and bathy- 

 metric features on both the Atlantic and Pacific coasts (Winant, Inman, and 

 Nordstrom, 1975; Vincent, et al., 1976; Resio, et al., 1977; Aubrey, 1979). 



Ihe objective of eigenfunction analysis is to separate the temporal and 

 spatial dependence of the data set so that it can be represented as a linear 

 combination of corresponding functions of time and space (Winant, Inman, and 

 Nordstrom, 1975). This helps identify processes responsible for profile line 

 changes, assists in evaluation of their relative importance, and aids the 

 identification of specific events. 



The shape of a single profile line changes between measurements in response 

 to the many processes (e.g., waves, wind, water level, etc.) active on the beach. 

 A careful evaluation of the profile line measured frequently over time may 

 reveal systematic changes in its shape. Regular seasonal changes in profile 

 area, for instance, were obvious on west coast beaches before being 

 quantitatively confirmed by empirical eigenfunction analysis (Shepard, 1963; 

 Aubrey, 1979). Along a single profile line, zones of maximum variation are 

 to be expected in the region of maximum wave energy dissipation. This has 

 also been confirmed by empirical eigenfunctions on west coast beaches (Aubrey, 

 1979) . However, the technique does not explain the physical reason for the 

 variability. In the case of beach profiles, the sand is moved in response 

 to wave forcing in a manner which is assumed to be deterministic, or at least 

 statistically predictable. It is hoped that since the wave forcing provides 

 most of the variability, the eigenfunctions will reflect this mechanism. By 

 examining the temporal structure of the beach eigenfunctions along with spatial 

 structure, the decision can be made as to whether, in fact, the eigenfunctions 

 represent some physically meaningful process. This has been shown to be the 

 case in nearshore profile studies (Aubrey, 19 79). 



Profiles obtained during the BEP do not extend beyond about the -0.61-meter 

 MSL shoreline. For that reason, beach variability associated with sediment 

 motion and seasonal sand storage in the offshore zone, below MSL, are not 

 included in the study and impose a limitation in the method of analysis. It 

 is known that the breaker zone and nearshore are regions of active transport 

 both onshore-offshore and alongshore. Offshore bars act as periodic storage 

 areas for sand that is later supplied to the beach under favorable wave condi- 

 tions. The time periods and detailed response of these regions cannot be 

 determined from the available data. 



IV. RESULTS 



1 . Temporal Variability . 



a. Long-Term Changes . The long-term erosion rates along Holden Beach 

 have been studied by several investigators who compared the shoreline posi- 

 tions on historical maps and sequences of aerial photos. The net erosion 

 along the east end of the island (beginning between profile lines 3 and 4) 



