while the shoreline retreated 3.1 meters per year. Individual measurement 

 points lost considerably more with the bluff at one station in reach B reced- 

 ing an average of 11.9 meters per year for a total loss during the study of 

 47.6 meters. 



Bluff recession rates were the greatest along reach A where the sandy bluff 

 was high, unvegetated, and unprotected. Because of the higher bluff, volumetric 

 losses were also greatest along reach A. The lowest bluff recession rates were 

 measured at the two imdeveloped reaches (C and E). Reach C had the lowest rate 

 with some of the dune accreting. The average recession for the points that 

 eroded was similar to the average loss along reach E. The highest average 

 rate of recession occurred during the period that included a severe storm 

 (March 1973). 



By using close measurements points, it was possible to illustrate the high 

 degree of spatial variability in bluff recession rates. Generally, when the 

 amount of bluff recession increased, the standard deviation increased. Over 

 1-year periods the standard deviations of the rate of bluff recession varied 

 from 1.3 meters per year along reach E to 6.7 meters per year at reach C. 



Rates of shoreline change were generally greater than the bluff recession 

 rates with considerably more spatial variation. Because of relief displacement 

 and the difficulty in accurately accounting for the effect of a changing lake 

 level on the shoreline, shoreline measurements were less accurate than bluff 

 recession measurements. 



During the study period, the average recession rate for all the reaches 

 (Table 9) increased along with the lake level, peaking in 1973 (the year of 

 the major storm) and then decreased in 1974 when the lake level stabilized 

 and no major storms occurred. 



Because these data cover a relatively short period at a unique point in 

 the lake level cycle (the rising side of a peak lake level), the effect of 

 lake level cycle could not be definitively determined from the photos exam- 

 ined in this study. Long-term lake level effects were examined by combining 

 the data in Table 10 with the findings reported in various sources. This re- 

 sulted in evidence of a lake level dependency with average lake level explain- 

 ing about 50 percent of the variation in bluff recession rates. 



Although many important variables affect the rate of bluff recession, a 

 very simplistic linear regression approach was used to identify the relative 

 importance of lake level and short-term storm events (indicated by the percent- 

 age of occurrence of high onshore windspeeds) . For the eight available data 

 points from reach A, the only significant correlation was found for the short- 

 term events. This relationship is shown in Figure 35. More data points are 

 needed over a wider range of conditions in order to examine the effects of 

 other variables like wave climate, bluff type, bluff height, orientation, ice, 

 etc. 



An important aspect of this study is the analysis of the effect of the 

 long seawall constructed during the period within reach B. A general expres- 

 sion for the relationship between seawall length and the volume and length 

 of the affected shoreline cannot be made from only one example; however, the 



61 



