June 1974. The average bluff recession rate for 27 points north of the plant 

 and 25 points south of the plant, independent of other factors, was 3.86 meters 

 per year, a value equal to that found by this study. Individual recession 

 values varied from a low of 0.8 meter per year to a high of 7.0 meters per year. 



This average rate was also similar to the 3.9 meters per year measured by 

 Tanner (1975) for the period 1970 to 1973 using the same aerial photos (spe- 

 cific measurement locations were not given). A higher rate of recession (4.2 

 meters per year) was determined for the same area between 1964 and 1970. Tanner 

 presented an exponential relationship between bluff retreat and lake level, wave 

 characteristics, and other unspecified parameters. 



Long- and short-term bluff recession rates reported in various sources are 

 summarized in Table 11. To better illustrate the different time periods con- 

 sidered, they are shown in Figure 34 along with variations in annual lake level 

 since 1860. A major decrease in the long-term average lake level occurred 

 around 1890 due primarily to changes in the outflow conditions of the Lake 

 Huron Basin (Brunk, 1968). Though the effect of this change on bluff reces- 

 sion rates is difficult to assess, the data of Powers (1958) and Beach Erosion 

 Board (1956) should be affected. The high levels before 1890 may also account 

 for the peak recession rate of 3.21 meters per year measured near reach D by 

 the Beach Erosion Board for the period 1830 to 1872. The overall bluff reces- 

 sion rate of 3.8 meters per year determined by this study (Table 9) and by both 

 Tanner (1975) and the Michigan Department of Natural Resources (1974) is higher 

 than the 2.65 meters per year reported by Seibel (1972) for a shorter period 

 between 1970 and 1972. 



All of the annual rates determined for the five reaches are higher than the 

 long-term rates of about 0.55 meter per year reported by Powers (1958) and Beach 

 Erosion Board (1956). It is interesting that the long-term recession rates in- 

 creased from 0.55 to about 1.1 meters per year when the period changes from 1830 

 to 1954 or 1956 (125 years) to 1938 to 1970 or 1974 (34 years). This doubling 

 of the recession rate may be an indication of a general increase in bluff reces- 

 sion in recent years. 



From the data in Table 11 an engineer, developer, or land manager could le- 

 gitimately estimate bluff recession using a long-term rate between 0.5 and 1.2 

 meters per year. Over a 30-year period, the implied recession would be between 

 15 and 36 meters. For comparison, station 99 in reach A lost 30 meters of bluff 

 during this study alone. Consequently, even the selection of the higher rate 

 may not provide a suitable buffer zone. This situation is further proof of a 

 lack of understanding of the bluff recession phenomena and the usefulness of a 

 particular recession rate value. The lack of predictable lake level and storm 

 cycles is another complicating factor. Since either lake levels or severe storm 

 frequency cannot be predicted with any confidence for a long enough period, it 

 is impossible to determine a priori whether to use a high or low recession rate. 



Cohn and Robinson (1976) attempted to predict lake levels by Fourier anal- 

 ysis of historic lake level records between 1860 and 1970. They were able to 

 determine prominent cycles of 1, 8, 11, 22, and 36 years. The model correctly 

 predicted the rise in lake level between 1970 and 1975 and forecasted a general 

 decrease in lake levels between 1975 and 1980. Peak lake levels are expected 

 in 1985 and 1993. 



52 



