The amount of bluff recession Increased steadily from the beginning of the 

 study In August 1970 through 1973. Though this coincided with the Increase In 

 lake levels, the study period was too short to evaluate long-term lake level 

 effects. Peak amounts of recession as shown In Figure 13 occurred during 

 periods of intense storm activity. 



During the final period of surveys (October 1973 to December 1974), the 

 lake levels stabilized, and there were few significant storms. Consequently, 

 most of the profile lines began to stabilize. With respect to the other pro- 

 file lines, the dramatic changes at profile lines 4 and 16 were anomalous dur- 

 ing the study. In both cases, nearby shore protection structures appeared to 

 be affecting the profile lines, dramatically increasing the rate of erosion. 



Although there was considerable variation in bluff recession between pro- 

 file lines, a strong seasonal dependence is shown in Figure 12. Bluff reces- 

 sion is high during late fall and early spring and low during the summer and 

 during periods of ice cover. This variation is interesting because it is out 

 of phase with the seasonal lake level variation but in phase with the annual 

 storm cycle. The effect of storms would probably be greater if they occurred 

 in phase with the seasonal fluctuations in lake level. This is an Important 

 consideration in planning lake level regulation, particularly if the current 

 phase relationship between the storm season and seasonal lake levels is 

 changed. 



The importance of storms was demonstrated 17-18 March 1973 when a major 

 storm caused the bluffs at 11 of the 17 profile lines (65 percent) to erode an 

 average 1.6 meters. This was not only the highest total amount of recession, 

 but also the highest number of profile lines retreating between any two con- 

 secutive surveys. 



In terms of volumetric losses, relative to the Davis (1976) datum (Table 

 4), average change per profile from August 1970 to December 1974 was -35.0 

 cubic meters per meter. Losses varied from -91.7 cubic meters per meter at 

 profile line 4 to only -7.8 cubic meters per meter at profile line 11. Aver- 

 age volumetric losses followed the same trend as bluff recession, increasing 

 from August 1970 to July 1973 and then decreasing between October 1973 and 

 September 1974. No clear relationship between bluff composition and volume 

 change was identified. 



From representative surface sediment samples of the foreshore and back- 

 shore it was found that relative to the backshore, the foreshore had a coarser 

 sand fraction and higher gravel concentrations. The foreshore also displayed 

 greater variability in content both between surveys and between profile lines. 

 Average median grain size for the sand fraction of the foreshore samples was 

 0.32 millimeter versus 0.29 millimeter for the backshore. Other useful sedi- 

 ment statistics are given in Table 5. 



2. Discussion. 



An important aspect of this and almost every study of Great Lakes erosion 

 is the complexity of the problem and the variability of both the lakeshore and 

 the processes. It appears that for every rule, there is an exception like the 

 sand bluff at profile line 11 which eroded the least of all profile lines, 

 including the seemingly more resistant till or mixed sand and till bluffs, or 

 the six profile lines that did not erode during the 17-18 March 1973 storm. 

 Though some of the anomalous results may be caused by poor profile line selec- 

 tion or by surveying errors, for the most part they are probably real. 



37 



