MeO 6S 517 20 OE teal | MR 80-2 (Ad -AOST 262) The Effect of Structures and Lake Level on Bluff and Shore Erosion in Berrien County, Michigan, 1970-74 by William A. Birkemeier MISCELLANEOUS REPORT NO. 80-2 APRIL 1980- Approved for public release; distribution unlimited. U.S. ARMY, CORPS OF ENGINEERS COASTAL ENGINEERING RESEARCH CENTER TC. Kingmen Building 203 Fort Belvoir, Va. 22060 NSS Me So-2- Reprint or republication of any of this material shall give appropriate credit to the U.S. Army Coastal Engineering Research Center. Limited free distribution within the United States of single copies of this publication has been made by this Center. Additional copies are available from: National Technical Information Service ATTN: Operations Division 5285 Port Royal Road Springfield, Virginia 22161 The findings in this report are not to be construed as an official Department of the Army position unless so designated by other authorized documents. NN TAN 0 0301 0089849 i} UNCLASSIFIE SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered) READ INSTRUCTIONS 1. REPORT NUMBER 2. GOVT ACCESSION NO.| 3. RECIPIENT'S CATALOG NUMBER MR 80-2 4. TITLE (and Subtitle) 5. TYPE OF REPORT & PERIOD COVERED THE EFFECT OF STRUCTURES AND LAKE LEVEL ON BLUFF AND SHORE EROSION IN BERRIEN COUNTY, MICHIGAN, 1970-74 Miscellaneous Report 6. PERFORMING ORG. REPORT NUMBER 8. CONTRACT OR GRANT NUMBER(s) 7. AUTHOR(s) William A. Birkemeier 10. PROGRAM ELEMENT, PROJECT, TASK EB GANIZAT AME AND ADDRESS Sac OF MING ORGANI oe AREA & WORK UNIT NUMBERS Department of the Army Coastal Engineering Research Center (CERRE-CP) Kingman Building, Fort Belvoir, Virginia 22060 11. CONTROLLING OFFICE NAME AND ADDRESS Department of the Army Coastal Engineering Research Center Kingman Building, Fort Belvoir, Virgi 14. MONITORING AGENCY NAME & ADDRESS(if different from Controlling Office) D31194 ee ae April 1980 13. NUMBER-OF PAGES ATX: 15. SECURITY CLASS. (of thia report) UNCLASSIFIED DECL ASSIFICATION/ DOWNGRADING SCHEDULE 15a. 16. DISTRIBUTION STATEMENT (of this Report) Approved for public release; distribution unlimited. 17. DISTRIBUTION STATEMENT (of the abstract entered in Block 20, if different from Report) 18. SUPPLEMENTARY NOTES 19. KEY WORDS (Continue on reverse side if necessary and identify by block number) Berrin County, Michigan Coastal structures Lake Michigan Bluff recession Great Lakes Shore erosion ABSTRACT (Continue on reverse sids if necessary and identify by block number) Rates of bluff recession and shoreline change along five 1.6-kilometer reaches located within Berrien County, Michigan, between 1970 and 1974 were measured by use of aerial photos. Annual measurements were made at 30.5- meter intervals, except for two adjacent reaches where biannual measurements were made. The overall average rate of recession for the five reaches was 3.8 meters per year. Average recession rate varied from 2.4 meters per year 20. Continued FORM DD . jan 73 1473 EDITION OF NOV 65 1S OBSOLETE UNCLASSIFIED SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered) ene UNCUASS TELE Ee aaa SECURITY CLASSIFICATION OF THIS PAGE(When Data Entered) for a reach with low foredunes to 4.5 meters per year along a reach with a high sandy bluff. The greatest amount of recession resulted from a signifi- cant storm occurring 16 to 18 March 1973. Simple regression analysis of the data from both lake level and storm parameters identified storms as the primary cause of recession. However, the data set was too small and at such a unique point in the long-term lake level cycle (the crest of a rising peak) to quantify the effect of lake level. The effect of a 579-meter-long seawall constructed during the study is discussed; the volume of material eroded downdrift of the wall nearly equaled the amount of material removed from the sediment supply by the seawall. The procedures used in analyzing the air photos and their accuracy are described in an Appendix. Guidance is also given for determining the number of measurement points needed per distance along the shore depending on the desired accuracy of the bluff recession rates. 2 UNCLASSIFIED SECURITY CLASSIFICATION OF THIS PAGE(When Data Entered) PREFACE This report is published to improve the understanding of Great Lakes bluff recession and the factors controlling it. The report is the result of a study of a series of air photos taken between 1970 and 1974 of 8 kilometers of shore- line in Berrien County, Michigan. The work was carried out under the coastal processes program of the U.S. Army Coastal Engineering Research Center (CERC). The report was prepared by William A. Birkemeier, Hydraulic Engineer, under the supervision of Dr. C. Galvin, Jr. and C. Mason, Coastal Processes Branch, Research Division. The author acknowledges the assistance of many CERC staff members, includ- ing S. Hildenbrandt for his painstaking collection of the data and review of the report; T.J. Lawler for developing some of the computer programs; K. Jacobs for aid in some of the data analysis; and Dr. D.L. Harris, C. Mason, A.E. DeWall, and E.B. Hands for their helpful reviews and comments which greatly benefited the report. Reviews by Dr. C.J. Galvin, Jr. (formerly of CERC), C. Johnson, U.S. Army Engineer Division, North Central (who originally suggested the study), RovELlkin’, Urs. Army Enigineer District, Detroit, and C. Kureth, The Traverse Group, Ann Arbor, Michigan, contributed greatly to improving and consolidating the final report. Comments on this publication are invited. Approved for publication in accordance with Public Law 166, 79th Congress, approved 31 July 1945, as supplemented by Public Law 172, 88th Congress, approved 7 November 1963. ae id TED E. BISHOP Colonel, Corps of Engineers Commander and Director II IV CONTENTS CONVERSION FACTORS, U.S. CUSTOMARY TO METRIC (SI). INTRODUCTION . 1. Study Area. Site 2. Available Data. . . ENVIRONMENTAL CHARACTERISTICS OF THE STUDY AREA. Lake Levels . Waves . Storms. 6 3 Littoral Material and Transport 6 Ice . PHOTO MEASUREMENTS . Data Collection . Reach A . Reach B . Reach C . Reach D . Reach E . > fH DUuBRWNe DW OPWN COMPARISON OF RESULTS. 1. Comparison Between RACH. z 2. Previous Berrien County Erosion Santos ‘ 3. Prediction of Bluff Recession . 4, Explanatidn of Seawall's Effect . SUMMARY. 1. Results . 2. Further Resonradhn. LITERATURE CITED . APPENDIX ANALYSIS PROCEDURE . i 2 TABLES General characteristics of study reaches . Summary of visual breaking wave data at Warren Dunes State Park, 26 October 1971 to 4 December 1974. Winter deepwater design waves for a 50-year storm. List of aerial photos. Summary of bluff, shore, and beach data for reach A. Summary of bluff, shore, and beach data for reach B. Summary of bluff, shore, and beach data for reach C. Summary of bluff, shore, and beach data for reach D. 13 14 14 15 CONTENTS TABLES--Continued Summary of bluff, shore, and beach data for reach E. Comparison of average bluff recession and shoreline changes for all reaches . Summary of bluff recession rates reported in various sources Reach A recession and process data used in linear regression model Longshore current measurements in reach B. .........-.+-2+.-. Comparison of volumeteric losses behind and adjacent to seawall. FIGURES Location of study reaches. Annual average of Lake Michigan water level as recorded at Ludington, Michigan, from 1951 to 1974. Monthly mean Lake Michigan water levels at Ludington, Michigan . Aerial view of lakeshore ice along reach B....-.. ....+245-s Composite aerial photo of reach A taken 23 November 1974 . Severe bluff erosion at the northern end of reach A. Rates of bluff recession and shoreline change along reach A. Rates of bluff recession and shoreline change along reach A from 19 November 1970 to 23 November 1974. Composite aerial photo taken 23 November 1974 showing reach B with main sections identified. View southward from station 13 . View southward from station 28 along the seawall View northward from station 42 showing the dune section north of lie SCMwA 5 6 66 016150 0 6 Composite aerial photo of the northern section of reach B taken 19 November 1970. Rates of bluff recession and shoreline changes along reach B . Aerial photos showing development of the downdrift cut south of the seawall from 16 November 1972 to 20 November 1973 . Page 10 12 12 16 19 20 22 23 25 26 Di, 27 28 29 Sill 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 CONTENTS FIGURES--Continued Variations in recession rate for different sections in reaches A and B compared to reach A . View of the downdrift cut from the southern end of the seawall . Bluff at CERC profile line 16 recently stabilized by precast- concrete seawall following a period of severe erosion . Rates of bluff recession and shoreline change along reach B from 19 November 1970 to 23 November 1974. Composite aerial photo of reach C taken 23 November 1974 . Low foredune topography typical of reach C . Dune buggy trails through the reach C dunes between stations 30 Erne “OS 65,95 OC Comparison between the rate of bluff recession and the average beach width along reach C . Rates of bluff recession and shoreline change along reach C from 15 April 1971 to 23 November 1974 . Composite aerial photo of reach D taken 23 November 1974 . View of bluff and beach northward from station 27 in reach D . Rates of bluff recession and shoreline change along reach D. Rates of bluff recession and shoreline change along reach D from 19 November 1970 to 23 November 1974. Composite aerial photo of reach E, 23 November 1974. View southward showing beach and foredune morphology in reach E. Rates of bluff recession and shoreline change along reach E. Rates of bluff recession and shoreline change along reach E from 15 April 1971 to 23 November 1974 . Cumulative bluff recession and shoreline change for each reach . Periods of previous study and historic lake level variations . Variations in the variables used in the regression analysis. Effect of lake level on the bluff recession rate . Page 32 33 33 35 36 37 38 38 59 a ST 42 42 43 44 45 46 46 48 53 55 SY/ CONVERSION FACTORS, U.S. CUSTOMARY TO METRIC (SI) UNITS OF MEASUREMENT U.S. customary units of measurement used in this report can be converted to metric (SI) units as follows: eee Multiply inches square inches cubic inches feet square feet cubic feet yards square yards cubic yards miles Square miles knots acres foot-pounds millibars ounces pounds ton, long ton, short degrees (angle) by To obtain eee eee 254 millimeters PD oy sya centimeters 6.452 square centimeters AKG) SY) cubic centimeters 30.48 centimeters 0.3048 meters 0.0929 square meters 0.0283 cubic meters 0.9144 meters 0.836 square meters 0.7646 cubic meters 1.6093 kilometers 259/20 hectares 1.852 kilometers per hour 0.4047 hectares Ike SHS newton meters ONS ex Om kilograms per square centimeter 28.35 grams 453.6 grams 0.4536 kilograms 1.0160 metric tons 0.9072 metric tons 0.01745 radians 5/9 Celsius degrees or Kelvins! Fahrenheit degrees Ifo obtain Celsius (C) temperature readings from Fahrenheit (F) readings, use formula: C = (5/9) (F -32). To obtain Kelvin (K) readings, use formula: Ke G/9) @ 82) * 278 is. sta AK. : eR a ot) bod ania mie Jinagp? ‘oping awe 26 + Sari Ons ea TS y0s eee, ae Bog Lae: Ae b Myith AA STO Rey OG) EO) ba ts in A Ca oS Oe ARR cohen Le cP th AM DET tow of WARS Gy ilages 2k Got Hem Gia * eambpptie Re fe ehh nied i Rp the NEN TIRE ANN HE LTR eI ONE MAE NET PRM rm UN Oe pe Ta " ow te = Ta Bate at. Chap cs atowe eminem Re tLe 0 ae bat ae bes hy were f Chndrete Seay] keegekieret @ perial af 4 2 Ng GEL w 4 - Q Bc bets ann eee phe. i9 ALE S etow] i ¥ cH ehey y Cie) ELAS £Q Novemhe2 Sy Zs , Movennoy ar 7 . ae a oct Pw oe c Peers ' mi hs ( My : 20 Composite nerve RCO O Pah OC. cave Does 1974545 ‘ STE TOR, f Bie ah i 3 es bevearchiiareideseneer Any ps a etn eo es ISOM otdua pel £0.60 ze Byrn t ey m Eteyoeh (Eo ogee E ane he te rnaih it ae AU. : ; Btoefow. oS BPIO.O. at a eras oe) Giape RO nN — By Ms = rs 25 Leg eos beta ne mitigate cf bl uit rapes ay gard he AB oe atte PEDO > » ' ei _ ‘ r ihe res oat bith PEs VE BUG I @ac 24 Kered orohivd?’, eeeheinias ghorehi ne: c6 ge CeO hy: BO ch , 4E hor a7 thas Neva SNFaG oie tere Ph OA “a'dt{ eS Fargai ts gee it Ch. GCrpsite Gerled pitti oF fete & Cahmwe sy Nev ens (al @, ing eee PST Vee 3) VhOR SD ih Ls 26 View ay heer Mtines) Eraw tap aH 2? Dae’ 1 Areiou Roe See wt i, 27 iste ect Mnee Caoree tin ane shiureri! ithe Se a1 ie Pb) ae ea tater Ines ovadpe T4q easegadta EGE « Tere,t- Co Ryren Of Dv fechePl om ahd Sorel iis, chenke wlan reach 10 NS veut DUT! “Baoley’ min Be ae HH ERE VN és J OMNI ha WES RL LS ay dery cao i a Bt wrk) OLE api by i ‘eiergotda athe. Ge : 30. View sorelword ‘Sipwide CVeCW Aas. Foreciuney PATE Mo egy Pati ¥ aos Sao. e), aro ft (a : ten Mee Bir e eg » bol) (BRL LL eee en ae i} oni af Roegty, isi AP oberon: ; ~ RTUE, iv} ats ’ Seton, aul shows Piles wrong, 7 on ah taAWer Lb IST) te re Markee ey PAR Ly. vee bits oe a) hk a“ ned is veo ddegin eh (aatamgant. # Ree hing change’ ir m a sbegh ee ee =v nen ee ttal areas aa aniicyy« Strownlanensiensapepastalabirn ii pager. fet Poe hates are inh “ets SENT MF id | viggt soem, Ds MONE) 89 be vb ote ES aT OR Lphy + Nite fe, pala as aay byte, Whats t M7 Uittec* of san vel wih Gh Hi eaeoiea yr pata yee THE EFFECT OF STRUCTURES AND LAKE LEVEL ON BLUFF AND SHORE EROSION IN BERRIEN COUNTY, MICHIGAN, 1970-74 by William A. Birkemeter I. INTRODUCTION The staggering loss of public and private property along the Great Lakes during the period of high lake levels, which peaked in 1973 and 1974, focused renewed interest on understanding the lakeshore erosion problem and on develop- ing methods to minimize it. This study examines, by use of aerial photos and other available data, the shoreline and bluff-line erosion which occurred along Berrien County, Michigan, between November 1970 and November 1974, a period of rising lake levels. A major emphasis of the study is the investigation of the spatial and temporal variation in bluff recession along both protected and unprotected shorelines. The effect of lake level is also discussed but the period covered by the data set is too short to adequately cover this phenomena. 1. Study Area. The study area is located in the southeastern section of Lake Michigan near Stevensville, Michigan (Fig. 1). Shoreline use includes summer and permanent residences, undeveloped parkland, and the Donald C. Cook Nuclear Plant which was under construction during the study period. Five 1.6- kilometer reaches of shoreline were selected for study; three of the reaches (A, B, and C) are north and two (D and E) are south of the nuclear plant (see Fig. 1 for locations of each reach). Selection was based on bluff type, height, and local structures. The reaches were selected away from the power- plant to minimize the influence of the construction of a temporary harbor and an associated sand-bypassing project at the plant. The temporary harbor and its effect are discussed in Johnson and Hiipakka (1976). General character- istics of the study reaches are given in Table 1. Reaches A and B form a continuous 3.3-kilometer stretch of shoreline com- posed of predominantly sand bluffs ranging from 3 to 15 meters in height. The Table 1. General characteristics of study reaches. Reach | Orientation Length Beach Bluff Bluff Offshore Shore Residences width height type slope! protection \ structures (km) (m) (m) (No. Distance from nuclear Powerplant A N. 27° EB. 1,71 4 to 8 10 to 15 Sand-till bluff Minor 11 N. 26° E. 1.62 6 to 11 3 to 14 Sand-till bluff Major N. 26° E. 1.52 5 to 13 3 to 7 Sand foredune None N. 19° E. 1.62 2etond 3 to 7 Sand foredune Minor None N. 30° E. 1.62 | 6 to 17 Sand foredune 42°02'30" + + 42°02'30" STEVENSVILLE pAKe MICHIGAN Z REACH C / 2 © LAKE MICHIGAN 41°57'°30 + Donald C. Cook Nuclear Plant rene vi = | rite BRIDGMAN km | 2 3 Depth Contour Interval: oft + Figure 1. Location of study reaches. primary difference between reaches A and B is a 579-meter-long seawall con- structed in reach B during the study period. The two reaches were compared to determine the effect of the seawall on the surrounding shoreline. Reaches, C, D, and E differ from A and B in that the active bluff line is composed primarily of low foredunes fronting a ridge which reaches heights of 36 meters above the mean lake level. Reaches C and E (on opposite sides of the powerplant) differ slightly in orientation and both are undeveloped; reach D is developed. None of these three reaches include any major shore protection structures. 2. Available Data. The primary data used in this study were aerial photos taken monthly of the shoreline from July 1970 to December 1974. These photos were originally used to monitor the effects on the adjacent shoreline of the temporary harbor and the sand-bypassing project at the powerplant. Each photo set covered about 18 kilometers of shoreline centered around the nuclear powerplant. Nominal scale was 1:3,600 with 40 to 60 percent overlap for stereo viewing. The air photo analysis procedure and its accuracy are discussed in the Appendix. Other data collected during the study period include (a) hourly wind meas- urements at the powerplant, (b) visual observations of daily wave and wind characteristics from Warren Dunes State Park (within reach E), and (c) monthly ground surveys of 17 eastern Lake Michigan profile lines (including profile line 16 in reach B collected by Davis, Fingleton, and Pritchett (1975), Davis (1976), and the U.S. Army Engineer District, Detroit). Background data for the study area are presented in Section II. Section III discusses the data for each reach and the changes that occurred. Section IV compares the results, both between reaches and to the results of other Berrien County studies; speculation about the effects of lake level changes and storms on the rate of bluff recession and about the effects of seawalls on adjacent shorelines is also presented. A summary and recommendations for future studies are given in section V. IIT. ENVIRONMENTAL CHARACTERISTICS OF THE STUDY AREA Many important factors influence the rate of bluff and shoreline change. These factors can be divided into ''shore factors" and "process factors."' Shore factors include the shape, composition, and orientation of the beach and bluff, which are relatively easy to determine for any particular area. Process fac- tors include the wind and wave climate, water level variations, and storm type and frequency which are not so easily determined. Secondary factors, such as ice cover and runoff, are also important. 1. Lake Levels. The most widely discussed process factor affecting bluff recession is the fluctuation in lake level. Although a high proportion of bluff recession prob- ably results from individual storms, lake level appears to be a controlling factor (Hough, 1958; Seibel, 1972; Maresca, 1975; Berg and Collinson, 1976). This study covers the final 3 years of a steady 9-year period of increas- ing lake level from an annual mean of 175.49 meters, International Great Lakes Datum (IGLD), in 1964 to 176.92 meters in 1973. During 1974, the final year of study, the lake stabilized at a level just slightly less than the 1973 level. The variation in lake level from 1951 to 1974 as recorded by the National Oceanic and Atmospheric Administration (NOAA) (1971; 1972; 1973; 1974; 1975) is shown in Figure 2. 177.0 176.8 580 176.6 a 579 e€ 176.4 @ a > 176.2 5/8 a = 2 176.0 i=) _ 175.8 ot 175.4 1950 1955 1960 1965 1970 1975 Yeor Figure 2. Annual average of Lake Michigan water level as recorded at Ludington, Michigan, from 1951 to 1974 (IGLD). Figure 3 shows the mean monthly lake level from 1970 to 1974 and the maxi- mum and minimum mean daily water levels for each month. The average seasonal lake level variation from lows in the beginning of the year to highs in summer is 0.34 meter which equals the long-term average given by Seibel (1972). A major increase in the average lake level occurred in 1972 when there was little seasonal decrease in lake level following the summer peaks. 177.2 58! 177.0 i r | if u i ee 176.8 Moximum Daily Mean 4" ali > 0e- 09--002- 001— 40¢- S30UOND! SUiBIOUS IE pes ae See |e a ee ee ape sabuoy) auljasoys voissaday yjnig : 3 OP- Ss 3 Ov- volssaday J4ynig = > 00}- S\- = < ool- te OS - SI-S =072= ~ 02- Is e = [\ if ‘$ \ p \ in ‘ 0 0) 0 Oo} — -—- > — HY -— D+ -- + - FO 0 ' - $261 idy Si—0261 AON Op 1 vt / 4 / / NZ » np My; @ vA) Ma wi Ve = u y c a 7 y o e OO! ! €26140NW 02-2261 AON O1 OSie= S001 os & S os) 2. = } © =s D . \ oO 2 A A ae %, on MK aN = awh 2 Op I A a = 0 7 palo sae 2a | OS. > OoF-= = —-—-— Y= YY SHH =a FIO GV 2 j Aue © WA hel Pees 3 . PbL6FAON 91-1261 3QV SIV ip 3 5 wd \ = SP yd ae N

- = ® BS \ ! \ 1 ~ \ i SS an Se wo a ‘ ~ @ fr \W H \ ast Komp WV bs A\ = = ATS Z2xX Y aan et < oO M \ J ENN = as oj —-— eee AINA Ee EN =—=10 oo = VD ke \ N i S| ae ae | ae Seer ee es eee ee 1 ie Y = 2261390 81-1261 AON 91 £61) AOW OZ—€ 26) AON OZ (ore) Os rexel} os ROS ay) . AY 1 A= Pra) Re 2 ay , Sf ! Wo vA) n Wr ye 1 JBN (VOSS Masri vs i io ‘ pL6LAON E2—-b26) AOW OZ oo} os ‘ ‘ Ol—- ee are er ore ee oY \ oy \ U u Se) \ t = iy U 2261 AON 91-2261 490 8h oo! os 22 Rate of Change (m/yr) 60 Figure 8. Shoreline Change Bluff Recession 70 80 Station 90 100 110 (ft/yr ) Rates of bluff recession and shoreline change along reach A from 19 November 1970 to 23 November 1974. \gpeach width adjusted to lake level (176.79 meters, IGLD). 28) Table 5. Summary of bluff, shore, and beach data for reach A (57 stations Bluff recession Shoreline change Beach width! Period Date Rate (oj Max. Rate 0 Max. (oj Change j length (m) | (a/yr) | Qr/yr)_ | Qi/yr) || @ (m/yr) | Gi/yr) | @/yr) (m) (mo. ) 19 Nov. 1970 Fie 2.2 oF 359) 14.6 0.9 2. 13. =25). 15 Apr. 1971 1.4 2.5 Boal 7.8 5.0 Uo 16 Nov. 1971 2.8 6.7 88 Bad) -8.0 18 Apr. 1972 Sera; So) 4.9 16.7 736 U 16 Nov. 1972 Boe 7.6 Yoal 32.0 -2.6 20 Mar. 1973 2.6 3.9 3.5 12.8 0.8 20 Nov. 1973 SZ SoH Bos 10.4 S673 20 May 1974 1.9 3.8 2.8 12. 23 Nov. 1974 19 Nov. 1970 I 3.6 3.6 2.2 16 Nov. 1971 6.0 6.0 3.4 16 Nov. 1972 5.2 5.2 3.8 20 Nov. 1973 3.6 3.6 1.9 23 Nov. 1974 19 Nov. 1970 to 18.4 4.6 Ito's} 23 Nov. 1974 Sa Reach) Be Reach B (Fig. 9) extends 1.62 kilometers from the southern edge of reach A to the Chalet on the Lake housing development. The sand bluff decreases in elevation from about 15 meters at the northern end to less than 3 meters at the southern end. Except for reach D, reach B is the most heavily developed reach with 28 houses, one-half of which are between stations 22 and 42 where a 579-meter-long seawall was constructed during the study. The sequence of development of the seawall is important in understanding the changes that occurred along reach B. To facilitate analysis, reach B is divided into five areas. Station 13 (Fig. 10) is approximately the same location as CERC profile line 16 discussed by Davis, Fingleton, and Pritchett (1975) and Davis (1976). The section be- tween stations 14 and 22, referred to as the ''downdrift cut," is located imme- diately downdrift (south) of the long seawall (Fig. 11) which protects the shoreline between stations 23 and 41. North of the seawall, between stations 42 and 46, is a high, unprotected and lightly vegetated sand dune (Fig. 12). Two smaller seawalls, one 91 meters long between stations 47 and 50 and one 30 meters long at station 54, are in the northern end of the reach. Both of these seawalls were constructed before the beginning of this study. Figure 13 shows reach B as it appeared in November 1970. Note the absence of a beach in front of the sand dune area. A beach averaging 11 meters wide fronts the seawall area; a similarly wide beach also fronts the downdrift sta- tions. The shoreline is straight from the 91-meter seawall southward. The existence of two seawalls at the bluff toe in the area where the long seawall will be built is an indication of previous erosion. Figure 14 shows the rates of bluff recession and shoreline change along the reach for the same time periods as reach A. Vertical lines separate the areas shown in Figure 12. During the first period, November 1970 to April 1971, the average bluff recession rate for the full reach was 4.3 meters per year. However, most of the recession occurred in two areas of the reach--the dune area between stations 40 and 47 and the area between stations 21 and 27. By April 1971, construction of a concrete seawall had started in the vicinity of station 32. The width of the beach gradually increased from zero at the northern end of the reach to about 14 meters at the southern end. Construction of the full length of the seawall was completed by November 1971, though not in the final steel sheet-pile form. Bluff recession was moderate from April to November 1971, averaging only 1.1 meters with the dune section retreating the most. Beaches had narrowed in front of the seawall while a beach up to 30 meters wide appeared in front of the dune. Major changes occurred between November 1971 and April 1972. The beach in front of the dune disappeared along with 8 meters of the dune bluff. The bluff behind the seawall retreated less than the dune area and the downdrift cut be- gan to form. The bluff near station 13, south of the cut, was unchanged prob- ably due to the relatively wide beach between stations 1 and 16. The average recession rate for the reach was 6.8 meters per year. 24 *semnqonijs uot}9930I1d sLoys AFTIUSPT *POTFIWUSPT SUOTIIES UTeW YIM g YOVoL smozie {4nd 4JTIpumop pue ounp Jo S[Teiep MoYs S9TUT SUTMOYS PpL6T LEquOAON ¢z usyxez OoYd [TeTIee 93 TsSOduo) LpUMOg © 6 9iIn3sTy 25 Figure 10. View southward from station 13 (also CERC profile line 16 in Davis, Fingleton, and Pritchett, 1975, and Davis, 1976). Photo taken 8 May 1976. 26 Figure 11. View southward from station 28 along the seawall (16 October 1976). Figure 12. View northward from station 42 showing the dune section north of the seawall (16 October 1976). All “STTWMBOS BSUTISTXO O9JeITPUT SMOIIy “OL6T IOqUOAON 6T USyXe q YOROT FO uOTIDeS uTEYyILOU ey} FO OR0Yyd TetTr9e 93TsOdUOD ft “ST oansty 28 *SOTBOS [POTIIOA JUSLTOFFIP 9I0ON °g YOROT SuoTe soesueyd suTTeLOYS pue UOTSS9d0L JFgn[Tq FO soqeYy ‘pT andy 09 OS Ov ees 02 OL (0) eS = 09 os Ov o¢e 02 ol 0 OS1—7S¢ 09-- 002— — ———---- sabuoy) aurjasoys -- sabuoy) auijasoys o8- Wolsseoey 43018 uolssavay 4yynig = 0b- > 3 09- 002- $9njQ uMOp 00l- O€- 5 < OOo!- Os- SI- = ~ = — a pijs esnoy > = 02- ool- -— 4SI- 0 (0) 0 0 02- 0 0 0 ool os oo1 1261 AON 91-1261 24V SI oot (4A/4}) @40Y UoISsadaYy 4jNIE joe) (44744) 840y uoIssaday 4yynig ZL6b JOY 81-1261 AON 91 (44743) sabuoyd auijasoys $0 a40y tote} Leye}} 00! {Jomoas aung 261 AON EZ—bLELAOW OZ ool 409 45!4puaog |Jomoas aung ||]DMDaS 9} aul] 21!304g JY39D $99 4) 14puMog 9b Ul] 211304g DYID 29 The rate of bluff recession decreased between April and November 1972, to only 4.1 meters per year. The recession rate increased south to north with the southern end and some sections behind the seawall experiencing the least change. Figure 15 shows the reach between stations 35 and 13 in November 1972. The seawall is composed of steel sheet pile from station 28 to the northern end. From station 23 to 28 the beach is protected by the older and lower concrete wall behind which is evidence of recent bluff erosion. No beach exists lake- ward of the seawall. The first evidence of a beach appears at station 24; a narrow beach also fronts the dune. The most serious erosion occurred from November 1972 to March 1973 due to the intensity of the early spring storm (see Section II,3). The average rate of bluff recession during the period reached 10.1 meters per year for. reach B. The bluff at station 42 retreated 14 meters. The seawall, completed during this period, did not fully protect the bluff behind it as evidenced by one small building which toppled down the bluff causing considerable recession at Station 27. During this period the downdrift cut became better defined, ex- tending from station 22 to about station 15. The bluff at station 22 retreated 11 meters. No beach was within the downdrift cut, though south of it the beach widened quickly to a maximum width of 23 meters at station 13. Bluff recession continued at a reduced rate during the final three periods of study. The rate of bluff recession was the lowest behind the seawall and along the dune. However, the bluff at the downdrift cut was actively retreat- ing, and the cut appeared to be lengthening (see the November 1973 photo in Fig. 15). The exact effects of the seawall on the bluff downdrift are diffi- cult to fully assess because of seawall construction within the cut. This resulted in the formation of a second cut, to the south of the first (see the November 1974 photos in Fig. 9). The sequence of events described above is illustrated in Figure 16. Aver- age recession rates for each area are plotted in the figure, and compared with the rate for reach A since it represents the unprotected bluff recession rate. Variations in lake level are also shown. The reach A recession rate increased during the period of rising lake levels and then stabilized at a lower rate when the lake levels stabilized in 1974 (not including the usual seasonal variations). This stability may be attributed to other factors, particularly to the absence of severe storms in 1974. The dune section experienced a dra- matic reduction in bluff recession rate after November 1973 which followed an equally dramatic period of erosion. The unstable areas were the downdrift cut and CERC profile line 16 (station 13). The downdrift cut shows a decrease in recession rate in November 1974, probably due to efforts to stabilize the area. These same measures probably accentuated the problem at CERC profile line 16 and farther south. The increase in recession was verified by ground surveys which measured an increase in bluff recession from 1.8 meters (Davis, 1976) between August 1970 and July 1973 to 9.4 meters between October 1973 and November 1974 (Birkemeier, in preparation 1980). The downdrift erosion has continued; however, an October 1976 field visit found the bluff slope in the cut area well vegetated and stabilized (Fig. 17) and the bluff at CERC profile line 16 stabilized by the installation of a pre- cast concrete seawall (Fig. 18). Measurements at CERC profile line 16 indi- cated an additional 9 meters of recession since November 1974. The bluff 30 “UOTITPUOD PL6T TOqUWeAON TOF 6 9ANBTY 9eg. *(wWo}I0G) F/6T TEqueAON QZ 03 (dod) ZLET IOGUOAON OT WOLF TTeMeES OY} FO YANOS 3nd 3FTAIpuMOp 9y}z FO JUowdoToAep Burmoys sojoyd Tetsey S T oan3sty Sil §81¢ = 177.2 Lake Michigan S00 + 580} 2176.8 Fae a] oe 176.6 : 579" 51764 Stations 6 Reach B To CERC Profile Line 16 1-13 15 co Downdrift Cut 2 (0) — A RT RT TC ER ee EES SE Nee E S55 2 Sean ee Jee meee 2 15 Gs 1,900-ft Seawall 23-41 S 0 = erage ie een 315 SS coe a Sand -Duneracr | ee eee 42-46 O15 * Oita ESCO AtaSeawall S/n eee 47-50 50 15 i SG i Reach A: (ij: Cea pe — 51-111 ae Wel is 1971 1972 1973 1974 1975 Time Figure 16. Variations in recession rate for different sections in reaches A and B compared to reach A (dashline). 32 Figure 17. View of the downdrift cut from the southern end of the seawall. Heavy vegetation on the bluff slope indicates successful stabilization (16 October 1976). Figure 18. Bluff at CERC profile line 16 recently stabilized by precast-concrete seawall following a period of severe erosion (16 October 1976). SS) between stations 1 and 13 was also retreating despite a series of new sandbag groins. Probable explanations for the dramatic changes of the dune and for the downdrift cut are given in Section IV,4. Overall rates of bluff and shoreline change for the full period of study are shown in Figure 19. Data measured from the air photos are summarized in Table 6. The table is based on simple averaging of all the stations in reach B and does not separate the protected and unprotected sections of shoreline. Table 6. Summary of bluff, shore, and beach data for reach B (54 stations). | Bluff recession Shoreline change Beach width? | Period Date Rate 0 Rate to Max. 0 Change || length (m)_[ Gi/yr) | Ga/yr) | 6 (m) | (m/yr) | Gi/yr) | (a/yr) (m) | (@) (m)__|__(mo) 1.8 4.3 4.9 Max. m/yr) i) 19 Nov. 1970 6.2 | 5.7 20. 8.6 19.5 | 15 Apr. 1971 7.6 | 4.2 | 1.4 1.1 10.5 10.1 38.1 16 Nov. 1971 10.6 | 5.8 | 3.0 2.8 15.2 | -70.0 18 Apr. 1972 6.7 | 5.0 | -3.9 2.4 7.9 | -29.0° 16 Nov. 1972 55 |) 4.0 3.4 | 20 Mar. 1973 6.1 | 3.4 i 1.5 | 20 Nov. 1973 6.9 | -1.4 | 2.0 20 May 1974 4.7 | -1.0 1.9 25 Nov. 1974 4.4 |-1.5 19 Nov. 1970 Soy 2.9 12 16 Nov. 1971 5.8 | 4.4 5.2 12 16 Nov. 1972 5.5 | -1.9 4.9 17.1 || -0.7 | -0.7 12 20 Nov. 1973 6.9 | 2.0 3.9 : : 14.6 || -5.6 | -5.6 6.2 | -18.9 12 | 23 Nov. 1974 || | 8.2 | 4.4 | -2.5 | ; 19 Nov. 1970 to 16.9 4.2 2.5 11.9 -2.2 2.2 -7.1 48 | 23 Nov. 1974 lgeach width adjusted to lake level (176.79 meters, IGLD). ae) Reachwes Reach C extends for 1.52 kilometers southward from the last cluster of homes near the Grand Mere Lakes (Fig. 20). No houses are within the reach. The geomorphology of reach C differs from reaches A and B because of a low foredune (Fig. 21) which fronts and protects a high, well-vegetated dune maldigey. To keep definitions consistent, the term bluff line used in this discus- sion refers to the active edge of the foredune system. The bluff line could be determined by stereoscopic viewing but identification was more difficult than the bluff line in reaches A and B. Because of the lack of cultural fea- tures, it was also difficult to establish reference points and to match suc- cessive air photos. Only reach C of the five reaches showed any lakeward movement of the bluff line due to foredune accretion. All of the accretion occurred south of the bend in the orientation of the shoreline between stations 20 and 30 (see Fig. 20). St, = ae © oO ry ee a ‘= = = wo So m1 2 ae (iu) Oo @ b— J C3) (S) (am) wh (am) (7p) 20 6 4 =z eae ~ Bluff Recession if + L = imam iT t —— ~ oO -4£ = @ -6 = So ox -8 -10 -12 10 20 30 40 50 Station Figure 19. Rates of bluff recession and shoreline change along reach B from 19 November 1970 to 23 November 1974. Se) ‘Op pue OF sUOTIeIS Uus9MI0q STTeIZ ABB8nNq suNp 930N *plL6T LOqUOAON ¢Z, Udye DJ YSeet Fo OOYd TetI9e 931sSoOduUOD "0% e4nsTy 36 Figure 21. Low foredune topography typical of reach C (17 October 1976). During the 4-year study period, the most noticeable change in reach C oc- curred to the inland dunes in the northern part of the reach. Because of the easy access and the rolling topography, the area has become popular with four- wheel drive enthusiasts. The air photos clearly document a widening of the trails and an increase in the number of roads across the dunes. A ground photo of the area is shown in Figure 22. No attempt was made to determine if the in- creased use of the area had an effect on the rate of bluff recession. The area has recently been closed to vehicular traffic (C.L. Kureth, The Traverse Group, Ann Arbor, Michigan, personal communication, 1979). Data were taken from the air photos at about l-year intervals between 15 April 1971 and 23 November 1974. The results of the bluff recession rate and average beach width computations are shown in Figure 25. The shift from accre- tion to recession occurs between stations 15 and 16 with the bluff or foredune being stable or accreting south of station 15. The average rate of accretion for stations 0 to 16 was 2.0 meters per year; the remaining stations retreated 3.2 meters per year, the lowest rate for any of the reaches. The rate of shore- line change over the 4 years was an almost constant -3.7 meters per year, though the rate was lowest at the ends of the reach. Although measurements of beach width are of questionable accuracy due to the difficulty in establishing a repeatable landward bound, the beach width data (Fig. 23) correlate well with the bluff recession rates. In general, where the beach was wide, the dune or bluff line either stabilized or accreted; where the beach was narrow, the dunes retreated. During October 1973 to Novem- ber 1974, the beach width averaged 5.6 meters and was fairly constant along the SU Figure 22. Dune buggy trails thro stations 30 and 40 (17 ugh the reach C dunes between October 1976). 18 Oct. 1973-23 Nov.1974 16 Nov. 1972- Bluff Recession Rate (ft/yr) 18 Oct.1973 Average Beach Width (ft) =) =10— = Bluff Recession E £10 Avg. Beach Width poe -15"-50 -100--30 0 10 20 30 40 Station Figure 23. Comparison between the rate of bluff recession and the average beach width along reach C. Note different vertical scales. 38 reach. During the same period, the bluff recession was also a fairly constant 2.0 meters per year. A simple regression analysis of the average beach width and the recession rate at each station during each period resulted in correla- tion coefficients of 0.6 or higher except for the last period which had a cor- relation coefficient of 0.26. This is an indication of the importance of beach width on bluff or, in this case, dune movement. Average rates of bluff reces- sion and shoreline change for the full 4 years are shown in Figure 24. Reach C data are summarized in Table 7. 20 = = o O 2 oS ‘So 2-20} o [on Bluff Recession Shoreline Changes —40 0 10 20 30 40 50 60 Station Figure 24, Rates of bluff recession and shoreline change along reach C from 15 April 1971 to 23 November 1974. Table 7. Summary of bluff, shore, and beach data for reach C (51 stations). Blurf recession Shoreline change ie |G Max. Rate (oj Max. (of (m/yr) | sin (m/yr) (m/yr) | Gn/yr) [ee m 2.2 15 Apr. 1971 22 Dec. 1971 16 Nov. i972 18 Oct. 1973 23 Nov. 15 Apr. 1971 to 23 Nov. 1974 lBeach width adjusted to le _ ~ Ss eS ZT oek—--—--—- -—-— —--=—-—-—=---—|o = = bh o 34 OT SO o NL —o> CS oe o a Se oS i e 3S . ’ = a 50 y 1006 s 15 Apr. 1971—16 Nov. 1972 e a o S Oe ee ee |e a Sose SAO =< a . wo = 19 Nov.1970—15 Apr. 1971 ‘s @ 50 ea hie ae OO. | \ ! . oe \ 5 / 3 B UX i \ ASS, Va # 0 Of oS ae /— — +0 0 \ \) ‘ Sy / = 8 -20 2-15 —5O} Bluff Recession 100 = £é Shoreline Changes —— ——— — e -40-£ - —100 —200 4-60 3 {OMMNDNZ OM MANIZON INGO MNSOMMINNTCO Station Figure 27. Rates of bluff recession and shoreline change along reach D. Note different vertical scales. 42 Rate of Change (ft/yr) Bluff Recession Shoreline Changes -—-— — peat 0 10 20 30 40 Station Figure 28. Rates of bluff recession and shoreline 90 60 change along reach D from 19 November 1970 to 23 November 1974. Table 8. Summary of bluff, shore, and beach data for reach D (54 stations). Bluff recession Shoreline change Rate (m/yr) Max. (m/yr) Date Beach width! Period 19 Nov. 1970 15 Apr. 1971 16 Nov. 1972 ° al o | Change || length (m) | @) (mo) 3.5 5 ASH) 0) i} 19 Barth) Os Niet 0 | -5.7 Iain Be ToBI elena | LASTER, ——————— : 1 48 lBeach width adjusted to lake level (176.79 meters IGLD). 43 *pL6T equeAoN ¢z ‘gq yoeer Fo ozoyd [etree 93TsSoduo5 "67 omn3Ty 44 Figure 30. View southward showing beach and foredune morphology in reach E. Photo taken near station 6 (17 October 1976). The bluff recession and shoreline change rates for each period are shown in Figure 31. One obvious feature is the regularity in bluff recession rate along the reach, particularly in the first and last period. The only area of high bluff recession was between stations 17 and 26. The recession rate be- tween 15 April and 18 November 1971 was 2.5 meters per year, with a slight in- crease to 2.8 meters per year during the second period with high localized erosion to both the shoreline and bluff line between stations 18 and 23. In reach E (like the other four reaches) the bluff recession rate increased Significantly during 15 November 1972 to 20 November 1973 but the longshore pattern was similar to the preceding period. The bluff retreated 6.3 meters, an amount greater than the 5.1 meters of recession measured for reach A during the same period. The shoreline accreted 4.7 meters during the period, possibly by the buildup of eroded bluff material on the beach. The rate of bluff re- cession decreased during the final period to 2 meters per year and was uniform along the reach. Shoreline changes generally correlated well with the bluff recession during all but the third period. Bluff and shoreline change rates for the full period are shown in Figure 32. Interestingly, the shoreline and bluff-line peaks appear out of phase by six stations. Beach widths during the study varied from 0 to 25 meters and averaged 10.5 meters. The peak beach widths occurred on the beach pads. The widest beaches occurred 20 November 1973 and averaged 17.0 meters in width. Reach E data are summarized in Table 9. 45 50 100 20 Nov. 1973—23 Nov.1974 50 100 ° | I | | | | | | to) SSF 100 Bluff Recession Rate (ft/yr) . t 50 i 100 Rate of Shoreline Changes (ft/yr) 15 Apr 1971—16 Nov.1971 OO ee » =O ie -20 — 2-15 -SOF Bluff Recession 100 = & -40£ -304-100 — -60 0 10 20 30 40 50 60.) Station Figure 31. Rates of bluff recession and shoreline change along reach E. Note different vertical scales. Rate of Change (ft/yr) Bluff Recession ae Shoreline Changes -———— BS) 0 10 20 30 40 50 60 Station Figure 32. Rates of bluff recession and shoreline change along reach E from 15 April 1971 to 23 November 1974. 46 “(A191 sxe.O 6L°OLT) TeAST exeT 02 poasn(pe syiptM Yyoeeg, ys ust oP) ik Soe pottod ,UIPTM yore esueyud SUuTTeLOYS uotsssse0l FFnNTG *(suotze4s S$) J yoeel LOF eJep YoVeq pue “ax0ys ‘z3n~Tq Jo Azewuns °6 STIPL pL6l TL61 ¢L6L cLO6L TL6L TL6L “AON €Z 02 “ady ST | rZ6r "AON £2 “AON 02 “AON OT “AON OT “aidy ST 47 IV. COMPARISON OF RESULTS The changes in each reach have been discussed in the preceding section. More detail was given to reaches A and B because of the greater number of photo sets examined. In this section, the results are compared between reaches and to the results of other investigators. In addition, the relationships between the causa- tive factors of bluff recession and the measured rates are examined along with a possible explanation of the effect of the reach B seawall. 1. Comparison Between Reaches. Cumulative amounts of bluff recession and shoreline change for each reach, as given in Tables 5 to 9, are plotted in Figure 33. The data for four common periods are tabulated in Tabie 10 for a better comparison of the five reaches. Although reach A had the highest overall bluff recession and bluff recession rate, reach D had the highest rate for any single period, losing 7.1 meters between November 1972 and October 1973. Volumetric losses were definitely larger in reaches A and B due to the greater bluff heights. 20 REACH A REACH B REACH C Cumulative Change (m) ° REACH D INERES se Si Ha CS BEE 1974 Yeor Tigure 33. Cumulative bluff recéssion and shoreline change for each reach (data from Tables 6 to 9). Shaded areas indicate periods of ice cover. 48 62 °92T 68°9ZT 1 (80°T)T (an 7 Sine, o°¢ a 7ey 33ey a 7ey a7ey J5OTE *3yd euT{Teroys|| uoTssede1 F35nTq || “8yd9 outfetoys |] uotssaza. VL6T “AON SZ 03 pL6T ‘AON €Z 02 TL6h ‘1dy ST (fL6T *390 8 10) ELér 9°¢ Ra “AON 02 1(26°0)I (w) *8yd autTTaroys £L6T (£261 °320 8T Potiad *“soyoevel [Te LOF sosueyd suT[TeLoys pue uoTSSedeL FInTq esereae Fo uostxredwuoy) “EZ6T 1990290 BT YITM Butpue 10 Buyuuz8eq potsed 03 s1azey, T6°9LT L9°9LT (a) Teast ayet “say (24) yq3uaty pottad z° s CES o2ey 97ey uotssacer F3nTg |] “B8yd eutperoys || uotssadez zznTg “AON 07 03 10) 2261 “AON OT 7Z6T “AON 9T 03 TL6T “adv ST “OT TIP L 49 The lowest overall recession occurred in reach C--the result of the lake- ward movement of the foredune in the second period. If the stations where the accretion occurred are not included, then the overall retreat rate increases from 2.4 to 3.2 meters per year, very close to the overall rate for reach E of 3.5 meters per year. Even though the bluff recession rate for reach C was low, the rate of shoreline change was the third highest, averaging -3.7 meters per year. Most of this shoreline movement occurred during the first period. The changes during the second period are difficult to interpret. In gen- eral, the recession rate peaked during this period and then decreased in the final year. This did not, however, occur in reach C where the bluff at the southern end accreted, thereby reducing the average recession rate: The lowest rates of both shoreline change and bluff recession occurred between the fall of 1973 and November 1974 with similar recession rates along reaches A, B, and D and in reaches C and E. Interestingly, the average reces- Sion rate along reach B, even with the long seawall in place, slightly exceeded the reach A rate. Average recession for all reaches over the full period was 13.6 meters for a rate of 3.8 meters per year. The associated shoreline change was Slightly less. Comparing the recession rates north and south of the powerplant, the rates were higher (to the north) in the first period (April 1971 to November 1972), lower in the second period (November 1972 to October or November 1973) and about equal in the third period (October or November 1973 to November 1974). In view of the different characteristics of the reaches, these differences (positive and negative) are insignificant and it is unlikely that they can be attributed to the construction of the temporary harbor. This agrees with the findings reported by Johnson and Hiipakka (1976). 2. Previous Berrien County Erosion Studies. Numerous studies have been conducted on Great Lakes shorelines. Primary topics include studies of geomorphology, sediment characteristics, and bluff or shoreline changes. A number of these studies have dealt with bluff reces- sion in Berrien County. A comprehensive study by Powers (1958) classified the entire Lake Michigan shoreline according to geomorphology (bluff type, composition, and height). He also relocated section lines where old bluff-line measurements had been made and determined the rate of bluff retreat. Of 134 stations around the lake, Powers reported that 124 eroded an average of 0.45 meter per year, 4 had no change, and the remaining 6 accreted an average of 0.48 meter per year. Periods of coverage varied from 20 to 127 years. Powers also recognized lake level fluctuations, severe storms, and manmade structures as primary factors affecting the recession rate. However, he noted the paucity of measurements needed to quantify the relationship between lake level and bluff recession. Powers (1958) reported that the shoreline in Berrien County consisted pri- marily of 3- to 12-meter-high bluffs and 6- to 38-meter-high dunes, that the beaches averaged 9 to 49 meters wide, and that the average bluff recession 50 rate for points in Berrien County between 1830 and 1956 was 0.60 meter per year, which was higher than the overall lake average. None of his points were within the five study reaches. A report on a proposed beach nourishment project for St. Joseph, Michigan, included an analysis of the bluff recession within the five reaches and a study of the bluff, beach, and nearshore sediment characteristics (Beach Erosion Board, 1956). The report concluded that only 20 to 40 percent of the bluff material was suitable beach-fill material. A peak bluff recession rate of 3.21 meters per year was found in reach D for the period 1830 to 1872. The average rate between 1830 and 1954 for all five reaches was only 0.50 meter per year, a value similar to but lower than that derived by Powers (1958) for about the same period. Seibel (1972) examined the bluff recession since 1938 at four Lake Michigan and two Lake Huron locations, and the relationship between lake level and pre- cipitation, and between lake level, storm frequency, and bluff recession. He determined linear relationships between average lake level and bluff recession for each of the six sites. One of the six sites was at Bridgeman in Berrien County where measurements were made at 27 profile lines, including six within the five study reaches. Data were obtained from aerial photos dated 1938, 1950, 1955, 1960, 1967, 1970, and 1972. Average bluff recession rate was 1.2 meters per year between 1938 and 1970, although individual profiles retreated as much as 9 meters per year. In addition to the long-term rates, Seibel also computed the rate of bluff recession between 1970 and 1972 at 14 points near the powerplant from the same photos used in this study. An average rate of 2.8 meters per year was deter- mined, an increase over the preceding period (1967 to 1970). Most of the in- crease occurred south of the powerplant. Seibel indicates that much of the increase can be explained by the increase in average lake level. An important conclusion reached by Fox and Davis (1970), Seibel (1972), and Johnson and Hiipakka (1976) was the significance of infrequent, severe storms in control- ling the rate and amount of bluff recession. Because the problem of lakeshore property insurance is directly linked to the recession rate in an area, there is considerable interest in predicting future bluff lines for at least the mortgage life of a structure (generally 30 years). Jannereth (1974) described the State of Michigan's effort to pre- dict bluff lines from 1938 and 1974 photos. The results (Michigan Department of Natural Resources, 1975) indicate that except for a small part of reach E, all five reaches are in high risk erosion areas. A bluff recession rate of 1.1 meters per year was determined for reaches A, B, and C, and a rate of 0.5 meter per year for reaches D and E. These values were used to compute a mini- mum setback line equal to 30 times the recession rate. A recommended setback line was also determined by adding another 9 meters to the minimum setback value. The Michigan Department of Natural Resources (1974) participated with the U.S. Army, Corps of Engineers in monitoring the effect of the temporary harbor at the powerplant. They found similar rates of recession north and south of the plant at areas with similar bluff topography for the period July 1970 to 5| 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 Lake Level (m) Table 11. Summary of bluff recession rates reported in various sources. Location Time Average Recession interval lake level rate (m) (m/yr) Powers (1958) Berrien County 1830-1956 0.60 Lake Michigan (all) 1830-1956 0.45 Beach Erosion Board (1956) Reach A 1830-1954 Reach B 1830-1954 Reach C2 1830-1954 Reach D? 1830-1954 Peak 1830-1872 Reach E? 1830-1954 Avg. (all reaches) 1830-1954 Seibel (1972) Bridgeman, Mich. 1938-70 176.11 1.17 1938-50 176.08 0.75 1950-55 176.49 1.48 1955-60 176.05 1.41 1960-67 175.88 0.73 1967-70 176.30 2.78 Near nuclear powerplant 1967-70 176.30 1.95 1970-72 176.54 2.78 Michigan Department of Natural Resources (1974; 1975) Reaches A, B, and C (1975) 1938-74 176.25 1.1 Reaches D and E (1975) 1938-74 176.25 0.5 Near nuclear powerplant (1974) 1970-74 176.71 3.86 Peak 1970-74 7.0 Minimum 1970-74 0.8 Tanner (1975) Near nuclear powerplant 1927-56 176.15 6 1956-64 175.95 5 1964-70 176.05 2 1970-73 176.67 9 linsufficient data; no lake level data collected between 1830 and 1860. 2actual stations near but not in these reaches. 177.0 This Study 38! 176.5 176.0 175.5 1830 575 isles BETS FANE SB : , Powers (1958) 1830 —— EAP SEAT Cease Beach Erosion REA MEMG GD Sate APTA Nee BY kes splat SET Board (1956) | oa ve oles weve Seibel (1972) Michigan Department of Natural Resources (1974, 1975) Tonner (1975) 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 Year Figure 34. Periods of previous study and historic lake level variations. OS) 582 580 Soe 578 577 576 (ft 3. Prediction of Bluff Recession. As discussed previously, Seibel (1972) determined a linear relationship between average lake level and bluff recession. His bluff recession rate meas- urements (Table 11) correlated well with the average lake level for each period (correlation coefficient of 0.73), explaining about 50 percent of the variance. Seibel also considered the average number of storms in each period which was fairly constant and did not correlate well. Since this study includes more detailed measurements over shorter time periods, these relationships were reexamined using simple and multivariant regression analysis. With bluff recession rate, B, as the dependent vari- able, linear relationships with the following independent variables for each period were examined: AL = average of daily lake levels during each period. RL = average rate of lake level change computed from monthly average lake levels. HL = average of the highest 1/4 daily lake levels. W = percent of time that winds were onshore (220° < 6 < 20°) and greater than 26 kilometers per hour as measured at the Muskegon, Michigan, airport (the nearest weather station 137 kilometers to the north). The selection of some variables was arbitrary and the results may have possibly been improved, for example, by increasing or decreasing the cutoff windspeed. However, this was not done since the intent was only to identify the important variables, not to develop the best possible prediction equation. The data were refined by assuming that all bluff recession occurred during the ice-free periods. Therefore, storms and lake levels during the ice-covered periods were not considered and the value of each variable was computed based on the length of the ice-free periods using the estimated periods of ice cover given in Section II,5. (Note: all rates of bluff recession discussed previ- ously have considered the entire period regardless of ice cover.) Only data from reach A were considered because of its lack of major structures and the fact that eight recession rate measurements were made in the reach. These data in final form are given in Table 12. The reduction in period length due to ice cover caused a significant in- crease in the winter recession rates which accentuated the already high winter values compared to the lower summer rates. This is the inverse of the lake levels which are high in the summer and low in the winter. Figure 35 shows the relationship between the variables, including the actual variations in the mean monthly lake level, the mean monthly rate of lake level change, and the number of days per month that onshore windspeeds 54 Table 12. Reach A recession and process data used in linear regression model. Bluff recession Average Avg. rate Avg. high 1/4 Time period Estimated Amount Rate lake level | lake level lake levels onshore winds ice cover change >26 km/hr (m) (a/yr) (m) (m/yr) (m) (pct.) (d) B AL RL HL W \correlation coefficient resulting from linear regression with the bluff recessian rate as dependent variable and independent variables. Period Onshore Winds>26 km/hr AL, HL (m) RL (m/yr) Loke Level (m) Rate of Loke Level Change (m/mo) Nov. 1970 Nov, 1971 Nov. 1972 Nov. 1973 Nov.1974 Apr. 1971 Apr. 1972 Mor, 1973 Moy 1974 Figure 35. Variations in the variables (defined in Table 12) used in the regression analysis. Actual monthly variations in period of onshore winds, lake level, and rate of lake change are also shown. (Note: all lake level values are plotted around the mean value from November 1970 to November 1974.) Vertical lines denote photo dates and ice periods. 55 exceeded 26 kilometers per hour. Clearly, only W has the right phase and amplitude variations as the bluff recession rate, B. Simple regressions be- tween B and the other variables yielded the correlation coefficients given in Table 12. Interestingly, although the rate of bluff recession correlated well with W (correlation coefficient = 0.87), explaining 76 percent of the variation, it negatively correlated with all of the lake level variables. Attempts were made to obtain better correlation by combining variables and by multivariate analysis, but no significant increase was found in the correlation coefficient above 0.87. More meaningful conclusions might be possible if the data were further refined and the data set expanded. One of the weakest variables is W, which estimates storm wave activity during a period. Muskegon, Michigan, data were used because of the uniform quality, but hourly data taken at the powerplant indicate different and gen- erally higher values. The powerplant wind data were not used because of gaps in the data and because of problems in resolving which of two anemometers (at different elevations) were used. Since the ultimate interest is in the wave action and energy reaching the beach, wave data (either actual or hindcasted) should be included. Unfortu- nately, the Warren Dunes State Park LEO data did not cover a long enough period to be useful. Quigley (1976) examined the relationship between the bluff re- cession rate and wave power on Lake Erie and found a strong linear correlation (r = 0.79). However, he proposed a more realistic, nonlinear, relationship in- volving the combined effects of low and high waves and varying lake levels. Other problems with the data in Table 12 which may affect the correlations include the different period lengths, the incomplete knowledge of ice periods, and the imperfect split of the storm periods since the effect of September and October storms fall into the longer summer periods. Another study of the aerial photos (or a similar set) should make more frequent measurements (every month or every other month) and should include a monitoring program of waves, ice, and lake levels. Though the lake level variables did not correlate well with the bluff reces- sion rates in Table 12, its importance on bluff recession is well known. Berg and Collinson (1976) showed that there is a phase lag between bluff recession and lake levels. Part of the reason for the lag is the time required to denude bluffs of protective vegetation as the levels rise and the time needed to re- vegetate the bluffs after the levels start falling. In this study with its unique point in the lake level cycle (a rising peak), the bluffs within reach A were already actively eroding at the beginning of the study and the lag effect may not be significant. The average lake level steadily increased to its peak and then stabilized at a high level while the recession generally in- creased until 1974 when it dramatically decreased. The computations of the lake level variables in Table 12 were based on average values for each ice-free period. If the average lake level, AL, is shifted one period forward to improve the phase relationship with B, the corre- lation coefficient between B and AL changes from -0.41 to 0.01. No increase occurred in the correlation coefficient between B and W when the shifted AL was included as a third variable. 56 The actual monthly variation in the rate of lake level change was periodic during the study; therefore, the true effect of the average rate of lake level change during each study period is difficult to identify using photos spaced at regular intervals each year. The value of the regression exercise is to identify the importance of short- term storm effects (as indicated by W) on the bluff recession rate. To test Seibel's (1972) linear relationship between average lake level and long-term bluff recession rates, the average changes for all reaches over the four periods (Table 10; ice days included) were combined with data from other sources (Table 11). The results (Fig. 36) further support a strong lake level dependence. (ft) 6.5 at 6.0 This Study Seibel (1972) Tanner (1975) Michigan Department of National Resources (1974) on on g ° ae A S (2) on (oo) (S) ine) on Bluff Recession Rate ( m/yr ) 0 175.8 176.0 176.2 176.4 176.6 176.8 177.0 Avg. Lake Level (m) Figure 36. Effect of lake level on the bluff recession rate (data are from Tables 10 and 11). Sif A simple regression analysis of average lake level and bluff recession rate for the 19 data values resulted in a correlation coefficient of 0.69, explaining about 50 percent of the variance. The intercept of the trend line shifts to the right and the slope is steeper than that found by Seibel (1972). The shift is due primarily to the high rates of recession reported by Tanner (1976) for two relatively low lake levels. These two points are important to the overall lake level bluff recession relationship and may be explained by the lake levels shown in Figure 2. A rate of 2.5 meters per year was recorded between 1956 and 1964, a period of generally falling lake levels except for a sharp rise of 0.46 meter between 1959 and 1960. It is speculated that much of the recession occurred during this period. The other point was a 4,5-meter per year recession rate between 1964 and 1970, a period of low average but rapidly rising lake levels. The two data points are important because they indicate that high rates of recession can occur during low lake levels and that other factors than average lake level need to be considered. 4. Explanation of Seawall's Effect. A seawall protects the shoreline by separating land and water areas with a fixed boundary. Because of high wave reflectivity off vertical and sloping seawalls, the rate of erosion tends to increase in front of the seawalls and it is difficult to maintain a fronting beach. According to the Shore Protec- tion Manual (SPM) (U.S. Army, Corps of Engineers, Coastal Engineering Research Center, 1977) the ground at the toe of a seawall, bulkhead, or revetment can be expected to scour below the natural bed to a depth equal to the height of the maximum unbroken wave which can be supported in the original water depth. Similar guidelines are not available for the effects on the shore adjacent to the seawall, although the SPM cautions that when a seawall is built on a receding shoreline, the recession on adjacent shores will continue and may be accelerated. The three-dimensional aspect of seawalls has been discussed by Silvester (1972; 1974; 1977). Silvester (1977) describes a seawall as a means of protecting a shoreline which is receding due to an imbalance between the supply of sediment and the sediment-carrying capacity of the incoming waves. While the construction of a seawall will protect the land behind it, it only accentuates the original problem by further reducing the supply of littoral material to the unprotected region of the beach. This is actually what is occurring downdrift of the long seawall in reach B. Silvester (1977) also theorizes that the interference of incident and re- flected waves produces a short-crested wave system which increases the trans- port of material in front of and immediately downdrift of a seawall over what would have normally occurred without the wall. Farther downcoast, this excess sediment can no longer be carried and it settles out as a shoal. In the specific case of the reach B seawall, the shoal did not appear on any of the air photos, but the lack of a beach within the downdrift cut and the occurrence of a beach farther downcoast offer some support to the Silvester theory. 58 During a visit to the area in October 1976, longshore currents were meas- ured using dye as a tracer. A minor storm was then occurring from the north- west with wave heights of 1.2 meters and periods of 7 seconds. The water depth along the seawall increased from 1.4 meters at the northern end to 1.7 meters near the southern end of the wall. Because of this depth, waves were not break- ing before striking the seawall. Longshore current measurements taken about 2 meters from shore at stations 14, 28, and 44 (see Fig. 9) are given in Table 13. The southward-moving current was about twice as fast near the southern end of the seawall as north of it, and more than three times faster than the current just south of the downdrift cut where the beach begins. Therefore, the current is capable of moving more sediment at the downdrift end of the seawall than it is either updrift of the seawall or south of the downdrift cut. Although the amount of material moved depends on the width of the current, its effectiveness is clearly evident in the complete absence of a beach in front of the seawall. Table 13. Longshore current measurements in reach B, 16 October 1976. Station 44 (north of seawall) 28 (along seawall) 14 (south of seawall) According to Silvester (1977), the effect of the seawall should be localized between the shoal and the downdrift end of the wall. In this specific case, the affected beach appears to be lengthening, partly because of the measures taken to stabilize the cut and partly because of the reduced sediment supply caused by the wall. Using the aerial photo data, the volume of material lost in the downdrift cut was estimated and compared to the volume of material removed from the sedi- ment supply by the seawall. The computations (Table 14) are based on the period between 20 March 1973 and 23 November 1974 when the seawall was completed, the backing bluff had stabilized, and the beach in front had disappeared. Expected rates of recession for the seawall and downdrift cut were computed from the re- cession of reach A during the same period but were adjusted by a factor based on the relative recession rate between each section and reach A in 1971, before the seawall was constructed. Therefore, the expected rate for the downdrift cut is low relative to reach A. Average elevations for each section were sub- jectively determined from topographic maps and a few ground measurements. Table 14. Comparison of volumetric lusses behind and adjacent to seawall. Section Station | Elevation! | Length | 1971 bluff Factor Bluff recession rate Total volume change recession 20 Mar. 1973 to 23 Nov. 1974/20 Mar. 1973 to 23 Nov. 1974 (m) (m) (m) (m/yr) | (m3/yr) Expected Actual Expected | Actual | Difference Reach A 12 1,646 3.6 1.0 4.6 4.6 | 90,859 | 90,859 0 Dune 13 152 8.1 2.3 10.6 etl | 20,946 6,126} -14,820 Seawall 12 579 Best) 0.8 &57 1.5 25,708 | 10,422] -1S,286 Downdrift 9 274 0.8 0.2 0.9 6% 2,219 | 22,441 20,222 cut 396 1.8 0.5 2.3 3.6 | 6,376 9,979 3,603 Total ee) as lied PSG lEstimated from topographic maps and some field data. 2Ratio of the recession rate of each section to the reach A rate for November 1970 to November 1971. NOTE.——Ratio of volume differences: downdrift cut (stations 1 to 22) to dune and seawall (stations 23 to 46) 23,825/30,106 = 0.8. 519 Interestingly, the bluff continued to erode behind the seawall but at a rate 40 percent lower than the estimated rate without the seawall. This fur- ther recession was probably caused by wave and spray overtopping, by slumping of the bluff, and by waves flanking the ends of the seawall. The material eroded in this manner either filled in behind the seawall or contributed to the littoral drift supply. The reduction in loss to the dune area is intriguing since it sustained extensive erosion in the early years of the study. One explanation is that the two seawalls adjacent to the dune acted like artificial headlands while the beach in between evolved into a stable, crescent-shaped embayment (Dean and Maurmeyer, 1977). During the same period, the two sections downdrift of the seawall experi- enced a 380-percent increase in the volume eroded compared to the volume change expected based on the recession rate in this area before the seawall was con- structed. The actual increase in volume equaled about 80 percent of the de- crease in volume of the dune and seawall sections due to the seawall. This one case study does not prove the theory that the additional amount of mate- rial eroded from the shores adjacent to a seawall will approximately equal the amount of material removed from the sediment supply by the seawall, but it does indicate that such a relationship may exist. Because no new material is being added to the system, the downdrift erosion can be expected to continue, though probably at a reduced rate depending on storm frequency, lake level, and the effectiveness of measures to mitigate the erosion. V. SUMMARY 1. Results. This report has dealt specifically with the bluff recession which occurred in Berrien County, Michigan, near the Donald C. Cook Nuclear Plant between 1970 and 1974. Though site specific, some of the findings and the analysis procedures used are applicable to other areas and studies. A major difference between this study and others which have used aerial photos to measure bluff recession is the quality and large scale of the aerial photos. Errors were minimized by the selection of the best sets from the monthly photos in terms of flight path, vegetation, ice cover, waves, offshore bars, and shadows. Measure- ment distances were kept short and interpretation errors were reduced by using stereo images to define the bluff. Measurements to the bluff line, bluff toe, and shoreline were made every 30.5 meters and bluff changes were computed to an accuracy of +1.4 meters (see App.). An adequate measure of the bluff recession rate for a receding reach of shoreline over a 1-year period could be obtained with as few as 20 equally spaced measurements per 1.6 kilometers (App.). The number reduced to 10 for measurements over a 4-year period. Even fewer may be adequate for longer periods. The bluff and shorelines of all of the five reaches eroded significantly during the study. The average rate of bluff recession was 3.8 meters per year 60 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 undeveloped 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, ejeer 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 data in Table 14 indicate that such a relationship may exist. It was deter- mined that the additional volumetric loss of the bluff adjacent to the down- drift end of the seawall approximately equaled the amount of material removed from the sediment supply by the seawall. The length of shore affected in- creased during the study and appeared to be still increasing in 1976 as attempts were made to stabilize the eroding area. 2. Further Research. This study has concentrated primarily on obtaining measurements of bluff and shoreline changes. The aerial photos are, however, limited in providing details on bluff composition, runoff effects, and other features; therefore, the impor- tance of these factors cannot be examined without detailed ground surveys. Most States now have programs to determine long-term recession rates (see Great Lakes Basin Commission, 1974; Michigan Department of Natural Resources, 1975; Berg and Collinson, 1976; Carter, 1976). Further research should be directed toward obtaining a better and more com- plete record of bluff recession and the complex factors controlling it. This requires a well-planned program of ground surveys coupled with aerial photos. Sites should be carefully chosen to include various bluff and shoreline conditions and a reasonable number of cultural features for reference points. Although the aerial photos can easily cover long stretches of shore, only short reaches need to be analyzed. Accurate ground control can be established by ground surveys coupled with contour maps compiled from aerial photos. Measure- ments should be taken at minimum intervals of 1 year and after major storms. Monitoring should continue through one full lake level cycle (20 to 25 years) and preferably longer. A complete and consistent record of the wave energy reaching the beach is also necessary. Wave hindcasting is probably the most cost-effective means for obtaining this information. Accurate observations of ice cover are also required along with lake level measurements. This type of program is necessary to fully understand the relationship between lake level, storms, and bluff recession. It is particularly important if attempts are made to minimize shore erosion by lake level regulation. The systematic long-term research program outlined above would provide con- siderable insight into these long- and short-term relationships. This type of effort is necessary to develop confidence in an ability to predict future Great Lakes bluff and shoreline changes. 62 LITERATURE CITED ASSEL, R.A., ''Great Lakes Ice Cover Winter 1970-1971,"' Technical Memorandum NOS-LSC-D4, Lake Survey Center, National Oceanic and Atmospheric Administra- tion, Detroit Mich., Sept. 1972a. ASSEL, R.A., "Great Lakes Ice Cover Winter 1971-1972,"" Technical Memorandum NOS-LSC-D6, Lake Survey Center, National Oceanic and Atmospheric Administra- tion, Detroit, Mich., Oct. 1972b. ASSEL, R.A., "Great Lakes Ice Cover Winter 1972-1973," Technical Memorandum NOS-LSC-D7, Lake Survey Center, National Oceanic and Atmospheric Administra- tion, Detroit, Mich., Apr. 1974a. ASSEL, R.A., "Great Lakes Ice Cover Winter 1973-1974,'"' Technical Report ERL 325-GLERL 1, Great Lakes Environmental Research Laboratories, National Oceanic and Atmospheric Administration, Ann Arbor, Mich., July 1974b. BEACH EROSION BOARD, "Beach Erosion Control: Berrien County, Michigan," H. Doc. 336, 85th Cong., U.S. Army, Corps of Engineers, Washington, D.C., Nov. 1956. BERG, R.C., and COLLINSON, C., 'Bluff Erosion, Recession Rates and Volumetric Losses on the Lake Michigan Shore of Illinois," Environmental Geology Notes, No. 76, July 1976. BIRKEMEIER, W.A., ''Coastal Changes, Eastern Lake Michigan: 1970-1974," U.S. Army, Corps of Engineers, Coastal Engineering Research Center, Fort Belvoir, Va. (in preparation, 1980). BRUNK, I.W., "Evaluation of Channel Changes in St. Clair and Detroit Rivers," Water Resources Research, Vol. 4, No. 6, Dec. 1968, pp. 1335-1346. BRUNO, R.O., and HIIPAKKA, L.W., "Littoral Environment Observation Program in the State of Michigan," Proceedings of the 16th Conference on Great Lakes Research, International Association of Great Lakes Research, 1973, pp. 492-507 (also Reprint 4-74, U.S. Army, Corps of Engineers, Coastal Engineering Research Center, Fort Belvoir, Va., NTIS 777 706). CARTER, C., ''Lake Erie Shore Erosion, Lake County, Ohio: Setting, Processes, and Recession Rates from 1876 to 1973," Report 99, Ohio Department of Natural Resources, Columbus, Ohio, 1976. COCHRAN, W.G., Sampling Techniques, 2d ed., John Wiley § Sons, Inc., New York, 1963. COHN, B.P., and ROBINSON, J.E., "A Forecast Model for Great Lakes Water Levels," Journal of Geology, Vol. 84, July 1976, pp. 455-465. DAVIS, R.A., "Coastal Changes, Eastern Lake Michigan, 1970-73,"' TP 76-16, U.S. Army, Corps of Engineers, Coastal Engineering Research Center, Fort Belvoir, Weis 5 OGEs UOTE 63 DAVIS, R.A., FINGLETON, W.G., and PRITCHETT, P.C., "Beach Profile Changes: East Coast of Lake Michigan, 1970-72," MP 10-75, U.S. Army, Corps of Engi- neers, Coastal Engineering Research Center, Fort Belvoir, Va., Oct. 1975. DEAN, R.G., and MAURMEYER, E.M., "Predictability of Characteristics of Two Embayments,'' Proceedings of Coastal Sedtments 77, American Society of Civil Engineers, 1977, pp. 848-866. FOX, W.T., and DAVIS, R.A., Jr., "Profile of a Storm-Wind, Waves, and Erosion on the Southeastern Shore of Lake Michigan,'' Proceedings of the 13th Conference on Great Lakes Research, 1970, pp. 233-241. GREAT LAKES BASIN COMMISSION, Proceedings of the Recesston Rate Workshop, 1974. HANDS, E.B., ''Changes in Rates of Shore Retreat, Lake Michigan, 1967-76," TP 79-4, U.S. Army, Corps of Engineers, Coastal Engineering Research Center, Fort Belvoir, Va., Dec. 1979. HOUGH, J.L., Geology of the Great Lakes, University of Illinois Press, Urbana, Til. , 1958. ISTVAN, L., "Shoreline Recession Rate Studies from Aerial Photography," unpublished working paper, Environmental Research Institute of Michigan, Ann Arbor, Mich., Oct. 1974. JANNERETH, M., "State Recession Rate Programs: Michigan," Proceedings of the Recesston Rate Workshop, Great Lakes Basin Commission, 1974. JOHNSON, C.N., and HIIPAKKA, L.W., ''Sand Bypass and Shore Erosion, Bridgeman Michigan," Proceedings of the 15th Coastal Engineering Conference, American Society of Civil Engineers, Vol. II, 1976, pp. 1361-1376. MARESCA, J., "Bluffline Recession, Beach Change and Nearshore Change Related to Storm Passage Along Southeastern Lake Michigan,"' Ph.D. Dissertation, University of Michigan, Ann Arbor, Mich., 1975. MICHIGAN DEPARTMENT OF NATURAL RESOURCES, ''Cook Nuclear Plant, Berrien County Shoreline Monitoring Survey Engineering Division Project No. 34-8422," Interoffice communication, 24 Sept. 1974. MICHIGAN DEPARTMENT OF NATURAL RESOURCES, '1975 Erosion Areas," unpublished, Lansing Mich., 1975. NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION, ''Great Lakes Water Levels: 1860-1970,"' Lake Survey Center, Detroit, Mich., Apr. 1971. NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION, "Great Lakes Water Levels: 1970,'' Lake Survey Center, Detroit, Mich., Aug. 1972. NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION, "Great Lakes Water Levels: 1971,"' Lake Survey Center, Detroit, Mich., Jan. 1973. 64 NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION, ''Great Lakes Water Levels: 1973,'' Lake Survey Center, Detroit, Mich., May 1974. NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION, ''Great Lakes Water Levels: 1974,"' Lake Survey Center, Detroit, Mich., May 1975. POWERS, W., "Geomorphology of the Lake Michigan Shoreline,'' ONR Contract No. NONR-1228 (07), Northwestern University, Evanston, I1l., Mar. 1958. QUIGLEY, R.M., "Processes of Bluff Recession," Proceedings of the Workshop on. Great Lakes Coastal Eroston and Sedimentatton, Burlington, Ontario, Canada, 1976, pp. 101-106. RESTO, D.T., and VINCENT, C.L., "Lake Michigan," Report No. 3, Technical Report H-76-1, Destgn Wave Information for The Great Lakes, U.S. Army Engineer Waterways Experiment Station, Vicksburg, Miss., Nov. 1976. SEIBEL, E., "Shore Erosion at Selected Sites along Lakes Michigan and Huron," Dissertation, University of Michigan, Ann Arbor, Mich., 1972. SEIBEL, E., CARLSON, C.T., and MARESCA, J.W., ‘'Lake and Shore Ice Conditions on Southeastern Lake Michigan in the Vicinity of the Donald C. Cook Nuclear Plant: Winter 1973-1974," Special Report 55, University of Michigan, Ann Arbor, Mich., 1975. SILVESTER, R., ''Wave Reflection at Sea Walls and Breakwaters,"'"' Proceedings of the Instttution of Civil Engineers, Vol. 51, 1972, pp. 123-131. SILVESTER, R., Coastal Engineering, Vols. I and II, Elsevier Publishing Co., New York, 1974. SILVESTER, R., "The Role of Wave Reflection in Coastal Processes,"! Proceedings of Coastal Sediments 77, American Society of Civil Engineers, 1977, pp. 639-655. SNEDECOR, G.W., and COCHRAN, W.G., Stattsttcal Methods, 6th ed., Iowa State University Press, Ames, Iowa, 1967. STAFFORD, D.B., "An Aerial Photographic Technique for Beach Erosion Surveys in North Carolina,'' TM-36, U.S. Army, Corps of Engineers, Coastal Engineering Research Center, Washington, D.C., Oct. 1971. STOKER, A.B., "Measurement of Coastal Bluff Recession from Aerial Photographs: Muskegon County, Michigan,'' Appendix 7 to Summary Report of the Pilot Study Program, Great Lakes Shoreland Damage Study, U.S. Army Engineer Division, North Central, Chicago, I11l., 1976. TANNER, W.F., ''Beach Process, Berrien County, Michigan," Journal of Great Lakes nesiceraelin WoL, it, NOo i, Oces 175, jy, iyo. TANNER, W.F., ed., "Standards for Measuring Shoreline Changes," Report of Workshop Published by Coastal Research and the Department of Geology, Florida State University, Tallahassee, Fla., 1978. 65 TEWINKEL, G.C., "Basic Mathematics of Photogrammetry," Manual of Photogrammetry, Vol. 1, American Society of Photogrammetry, Washington, D.C., 1962, pp. 17-66. THOMPSON, M., ed., Manual of Photogrammetry, 3d ed., American Society of Photo- grammetry, Washington, D.C., 1966, 1,197 pp. U.S. ARMY, CORPS OF ENGINEERS, COASTAL ENGINEERING RESEARCH CENTER, Shore Protectton Manual, 3d ed., Vols. I, II, and III, Stock No. 008-022-00113-1, U.S. Government Printing Office, Washignton, D.C., 1977, 1,262 pp. U.S. ARMY ENGINEER DISTRICT, DETROIT, ''Section III Detailed Project Report on Shore Damage at St. Joseph Harbor, Michigan," Detroit, Mich., May 1973. WILSON, D., and EVERTS, C.H., "Evaluation of the Base Map Technique to Determine Shoreline Changes from Aerial Photography,'' U.S. Army, Corps of Engineers, Coastal Engineering Research Center, Fort Belvoir, Va. (in preparation, 1980). 66 APPENDIX ANALYSIS PROCEDURE Various procedures are currently in use for making measurements from aerial photos. Therefore, it is important to discuss the procedure and to estimate the accuracy before attempting to understand the results. 1. Photo Measurements. One commonly used procedure for taking measurements from aerial photos is to make measurements from carefully selected low elevation reference points to the feature being studied (Stafford, 1971). The reference point must be a clearly defined stable feature which appears on all the sets of photos. After all the measurements have been made and scaled according to each photo, changes can be computed. The primary advantage of this method is that measurements are taken directly from the photo. The disadvantage is that points of measurement cannot be taken at equal intervals along the shoreline and it cannot be used in areas with few cultural features; also, since measurements are made independently, problems in defining the bluff line may cause it to move lakeward, a physical impossibility. This is particularly true when the time intervals are short and the bluff line is relatively stable. A second procedure involves a form of indirect measurement where photo de- tails are optically transferred to a base map by a device like a zoom transfer scope (ZTS). Wilson and Everts (in preparation, 1980) provide an example of this method. The ZTS allows magnification, rotation, and stretching of one image to superimpose the image on another. It is, however, difficult and tedious to use. For this reason, Istvan (1974) recommends that the ZTS should not be used to determine shore and bluff recession because of optical errors, the difficulty of matching photos precisely, and the greater potential for interpreter error and fatigue. A combination of these procedures was used in this study to optimize speed and accuracy. Measurements to the toe of the bluff and the shoreline were made directly on each photo using an appropriately scaled grid of the reference line and the station locations. This procedure was considered accurate enough to identify shoreline and beach width variations. Greater accuracy was desired for the bluff line. Using an acetate overlay, the bluff line on each photo was traced with the aid of a scanning stereoscope. Using the ZTS, the photo with the bluff-line overlay still in place was then superimposed on the corresponding November 1974 photo by matching the center part of the photo. (The ZTS was only used to match the scales; the "stretch" feature was not used. ) After the photos were matched, the bluff line was transferred to an acetate overlay on the November 1974 photos. In this manner, the bluff lines could be compared as they were being drawn and errors due to photo interpretation were reduced. 67 After all the bluff lines were drawn, a grid of the reference line and station locations was placed on the November 1974 overlay and measurements to each bluff line from each station were made. This procedure also eliminated problems in relocating the reference line from photo set to photo set. 2. Sources of Error. The straightforward estimation of distance from aerial photos is based on several convenient assumptions which are not often satisfied practically. Errors can be separated into those which are inherent in the photo and those which are due to the analysis procedure. The various sources of error are well known and frequently discussed (Thompson, 1966; Stafford, 1971; Istvan, 1974; Stoker, 1976; Tanner, 1978). Compensation for some errors may be possi- ble if enough information is available. An estimate of the magnitude of the maximum and most probable error for the procedure used is developed below. Photo errors are due to relief displacement and scale variations within single photos, between photos in a set, and between sets. Since other errors are scale-dependent, their effects have to be combined in any procedure where measurements are made directly on the photos (Istvan, 1974). By using the above procedure, variations in scale among photos in a set or between sets were virtually eliminated (relative to other errors) by optically matching the photos to the November 1974 set. The assumed accuracy of the scale used for the November 1974 photos was verified by the actual ground measurements shown in Table A-1. Objects with well-defined end points were selected without regard for elevation or orienta- tion. The amount of error ranged from 0 to -1.43 meters; the average absolute error was 0.5 meter. Relative errors varied from -3.6 to 4.2 percent, averag- ing 0.03 percent. Table A-1. Comparison between actual and photographic measurements. Error || Relative error (m)_|| (pet) 18.66 || -0.37 Approximate | Distance (m) | Object elevation - 4974 photos | Actual (m above LL) Distance between 1 two groins Seawall 1 Swimming pool 12 SZO2 alt OOz 19.72 || +0.09 Roof 11 26. 82 +0.61 Seawall 1 49.99 || -0.61 Seawall 1 39.53 || -1.43 Seawall 1 34.90 0.15 Tennis court 9 45.72 45.26 0.46 Tennis court 9 36.58 36.42 || 0.16 Between lines in 6 18.29 18.29 0.0 parking lot Seawall 1 33.53 34.14 -0.61 Seawall 1 29.26 Roof 9 21.34 21.49 || -0.15 29.87 | -0.61 Avg. of absolute values 68 Scale variations within a single photo are due to changes in ground eleva- tion and the amount of tilt of the image relative to the ground. For an image parallel to the ground, the scale of the photo can be expressed as (A-1) n iT | where f is the focal length of the camera lens and H is the height of the camera above the ground. Since the scale varies for points of different eleva- tion, the scale for a point at an elevation h above the general ground eleva- tion can be written as (A-2) The problem is compounded when the image is tilted relative to the ground at an angle o. Then equation (A-2) becomes f- y sino SS A-3 S aS (A- 3) where y is the radial distance measured on the photo from the nadir or center of the photo to the point of interest. As seen from equation (A-3), the effect of tilt is to linearly modify the scale in a direction perpendicular to the axis of tilt. The effect of tilt can be minimized through the optimal matching procedure described. With enough information, tilt can be eliminated by recti- fying the photos. According to Tewinkel (1962) about 50 percent of all aerial photos are tilted less than 2° and very few more than 3°. A reasonable estimate of the effect of tilt and topographic relief can be determined from equation (A-3). Starting with equation (A-1) and using the known focal length and nominal scale of f = 0.152 meters S Ss 185,600 a value of H can be determined. H = 547.2 meters Using the data in reach A as an example, the following average and maximum values of the variables in equation (A-3) were determined. Estimated varia- tions in the elevation of the bluff line within a single photo (h, actual) are Haug = 3 meters I ee = 5 meters Distance from station to nadir (y on photo) Yavg = 0.0381 meter Virnaa = 0.0889 meter 69 Then, using a conservative tilt estimate of 2°, the following maximum scale error was determined. S _ Osi52 > O.0880) sim 22. 1 Meee Moe S 3,641 3,600 - 3,641 INS oe SSS Se SS oe = 1.1 percent Similarly, using average values and 2° of tilt, an average scale error can be computed _ Spolil s S,000 — ASavg Ts COO 0.3 percent Because these errors were minimized by the procedure used, AS,,,, is probably the more realistic error. This amount of error can be converted to distance, D, by using the maximum distance measured between the reference line and the bluff line (52 meters). ADpin = 0.03 x 52 = 1.56 meters Therefore, a reasonable assumption is that the amount of error for a single measurement will lie between + 1.56 and - 1.56 meters and be normally distrib- uted about a zero mean. A consequence of this assumption is that 99.7 percent of the errors for all measurements lie within three standard deviations of the mean error. . The standard deviation of the error, 06,4, can then be estimated as 1/6 of the total range, or in this case C= 2 = 0192, meters and the variance of 0.27 square meter Another error inherent in vertical aerial photos is relief displacement. Figure A-1 shows that the location of the top of an object of height h will appear to be farther from the nadir than it actually is. This difference in distance, dR, depends on both the height of the object, h, and its radial distance from the nadir, R. The amount can be calculated as h dR=R i (A-4) Relief displacement is an important error in any area that has considerable relief, e.g., the bluffs. However, the procedure used allowed relief displace- ment errors to be neglected. This was possible because of the optical matching and because the reference line was selected close to the bluff line and at a Similar elevation. Since the quantity of interest is the amount of bluff re- cession, calculated as the difference between two measurements at a single sta- tion, the only source of relief displacement error affecting this quantity is a change in position of a measurement station, relative to the nadir, between photo sets. Even this is minimized by using reference points near the bluff to optically match the photos. 70 Camera Ground Nadir Figure A-1. Effect of relief displacement. An estimate of the probable error in the radial displacement between the reference line and the bluff line at a station was computed for the average quantities given above and was found to equal only 0.23 meter. This was for a single measurement and the actual effect for successive measurements can be assumed to be considerably less. Because the bluff line was usually displaced west from the nadir, relief displacement is always a positive effect and un- like scale variations, it cannot be assumed to follow a normal distribution. A major source of error which is difficult to quantify is that resulting from interpretation and human error. These errors occur from improper inter- pretation of the various beach and bluff features, improper photo matching, and careless measurements. Photo interpretation is a tedious process and should be recognized as such. Errors can be minimized by careful photo selec- tion, by using the simplest possible procedure, and by including frequent checks and remeasurements. Analysis errors resulted from inaccurate bluff-line identification and trans- fer and from the accuracy of the measuring device. Because all bluff measure- ments were made on the November 1974 photos, the errors are independent of scale variation errors. Moreover, since the errors are random, it can be assumed that they have a zero mean and that the amount of error follows a normal distribution. These assumptions allow the standard deviation of the error to be estimated by following the procedure previously used for scale variations due to tilt and Tela ef. (al Measurements were made using an engineer's scale with 24 divisions per centimeter (60 divisions per inch). This allows an accuracy of 1/2 division which at the photo scale equals +0.76 meter. Therefore, the effect of measure- ment error, e¢,, can be determined by estimating the standard deviation 9, Cm = +0.76 meter i182 Gj. So e's 0,25 m 6 o2 = 0.064 m The errors in identifying and marking the bluff line, ep» and in trans- ferring it to the 1974 photo set, e,, are unique to the analysis used in this study. The width of the ink line is 1.5 meters and the line is drawn so that one edge of the line traces the bluff line. To minimize interpretation errors, photos were selected at times when trees were without leaves and when the bluff line was well defined. If a conservative error of +1.5 meters is assumed for each process, then Sh = Saye +1.5 meters S10) Oy SOS ae ee) Dace Oh ye Oe 2 Oe = ORZS Finally, collecting all the error terms Pdi, te Ae 2 2. 2 OTA Ot Oe 07. 7 Ge GE O.27 & Os064 2 O25 2 0525 2 0.86 og = 0.91 meter Therefore, individual distances from the reference line to the bluff line can be measured to an accuracy on the order of +1 meter. This is a reasonable amount of error which has been kept small by (a) a large photo scale (1:3,600), (b) short measuring distances, (c) optical matching of photos, and (d) with all photos at the same nominal scale. The accuracy in determining bluff-line changes can also be determined by OS = Dy = Dy where 6 = change in bluff line D, = measured distance at time t = t, D2 = measured distance at time t = t2 V(6) = o% + 62 = variance of 6 Dy Do Ge Since the nominal scale of all the photos was the same, and since the distances were approximately equal, 2 ee IO oon oT, 0.91 V(6) = 0.91 + 0.91 = 1.82 os = 1.35 meters Therefore, a change in bluff position can be determined to an accuracy of +1.35 meters. This error is quite large and makes it difficult to measure small amounts of change. In bluff recession rate determinations, the accuracy improves for long- period data and decreases for short periods. For example, a bluff recession amount of 20 + 1.4 meters over 4 years reduces to an annual rate of 5 + 0.3 meter per year. Measurements were also made to the toe of the bluff and the shoreline. The accuracy of these measurements is considerably less than for the bluff measure- ments because of increased relief displacement and line definition problems. Changing water levels also affected the accuracy of shoreline measurement and of comparing successive measurements. Stoker (1976) reported on the difficulty of properly identifying the vari- ous beach and bluff features out to the offshore bar and indicated that inter- pretation was the major source of error. 3. Number of Measurement Stations. The errors given above pertain to one station and are too large to detect small changes in bluff recession rates; therefore, measurements were taken every 30.5 meters. This allows mean changes to be specified as small as ta/¥n (de- fined as the standard error of the mean) where n is the number of stations and o is the standard deviation of the rate of bluff change. For example, using the bluff recession along reach A for four l-year intervals (see Table 5), the mean recession rate varied from 3.6 to 6.0 meters per year with the standard deviation varying from 1.9 to 3.8 meters per year. With 57 stations within the reach, those values give a standard error between +0.3 to +0.5 meter per year. An empirical evaluation was made to determine the minimum number of stations needed to obtain a mean recession rate which was within +0.3 meter per year of the mean value determined using all stations. This was done by first removing the linear trend from the 1- and 4-year recession rate data from reach A. Sub- samples of n stations were then obtained by systematically sampling (Cochran, 1963) all the stations at equal increments of k stations such that nk = 57. For each year and for each value of k, k unique subsamples were obtained. Means were calculated for each subsample and the maximum difference between sample mean (x,) selected from the set of 4 x k subsample and the population mean (X57) was plotted versus the number of stations in a sample (see Fig. A-2). The figure also shows a similar line (based on k samples) computed for a rate per year using data over a 4-year interval which has a significantly lower standard deviation than the 1-year data. a) 25 7 6 20 £ Ss Ix u 15 Ix oh Gh is ° = oe = = s 3 10 4-yr Reach A Data E 2 = |-yr Reach A Data 22 ©) | 4 , + 0.3m | 0) (0) ——————— : (0) 10 20 30 40 50 60 n= ra = Sample Size (measurement points per |.6 km) Figure A-2. Comparison of maximum difference between Xg, the mean bluff recession rate of a sample of n equally spaced measurements per 1.6 kilometers, and X57, the mean computed for all 57 stations. Lines are shown for 1- and 4-year periods (curves are hand-fit). From Figure A-2, it appears that for data from 1-year periods, 20 measure- ment stations at equally spaced increments per 1.6 kilometers would result in a mean recession rate within +0.3 meter per year of the value which would be determined using 57 stations. This amounts to a 65-percent reduction in the number of measurements needed. Since Figure A-2 is based on the maximum difference between X, and X57 (instead of the mean difference), it should be conservative. Fewer stations are needed for longer period data as indicated by the shift of the line for the 4-year data to only 10 stations required for the same accu- racy. If an accuracy of +0.3 meter per year is inadequate, Figure A-2 can also be used to select a larger sample size. This rather simplified procedure for estimating measurement station fre- quency is presented for general guidance in setting up a similar study. The actual number of stations needed will vary depending on the quality and scale of the air photos, the time period considered, the accuracy desired, the uni- formity of bluff type, and the analysis procedure. It is suggested that some experimenting be done with measurement density along a short reach of coast before a final station frequency is selected. 74 £09 GO}. HOU awrgcn €0ZOL *Z-08 ‘Ou jaodeX SnosUeTTSeOSTW ‘AequUeD yoIeesey BuTAveuTZuq Te IseoD ‘S'N iSeTzeS ‘II ‘“eTIFL “I ‘“UoTsoze eroys *G ‘“UeSsTYOT_ eyeT “hy *soye] 28019 *¢ ‘*SeANQONAASs sAOYSTJO *Z ‘uesTYOTW ‘AQuNOD useTateg ‘[ *xTpueddy ue ut peqrzosep sie Aoeanooe ATeyq pue sojoyd ate oy} ButzATeue ut pasn saanpsooid ayy *F3nTq Apues y3Ty & YITM yore e BuO0Te sieqeu G*y 0} SauUNpertoyZ MOT YRTM YyoRer e JOJ Siajow 7°Z WOLF peTAeA ae 9yQ faeeh aod siajou g*e seM SaydPaT BATJ BY TOF |JeA uOoTSsade1 aBeteAy ‘Soqoyd TeTAee wory paanseouw arem ‘h/-0/6| UWeenqeq SueSTyoTW ‘AjQuNOD UeTAteg UT Sayoeeat AeJeWOTTY -9°[T eATF BuoTe ssueyd aeuTTeTOYS pue uOTSSa0ea1 F5NTqQ Fo saqey ‘OTT 12809 (7-08 "OU £ TeqUeD YyOARPeSeYy BuTAseU -T8uq TeISeOD *S*n — JAode1 snosueTTeostW) — “wo fz { "TIT : ‘d [rf] “O86. ‘A0TAIeS UOTJeMAOFJUT TeITUYDEeT TeUOTIeN WOAF YVTqGeTTeAe : “eA *pretgsutads £ aequep yoreasey SuTreeuTsuq TeqseoD *S'n : “eA ‘SATOATAg qiog — JeTeweyatg “Vy WeTTTTM Aq / H/-0/61 ‘uesTYOTW ‘AquUNOD uaTateg UT uoTsoie es1oys pue FjJNTq uo TeAeT syeT pue seinjzonz4s Jo 4oIoFjzo Jy ‘y WETTTIM ‘teToweyatTg £e9 CHG: Sou AW gcn” €0720L *Z-08 °OU JtodeA snoOsUeTTSOSTW ‘equa YyOAReSey BuTAseuUTSU_ TeISeOD “S° :Setteg “II “STIFL ‘I ‘uoTsorte etoys *G ‘ueSTYOTW exeT “yh ‘SseyeT qe019 *€ *SeinjdNIAS sioYSssjO *Z ‘uesTYyOTW ‘AQuNOD ueTAteg ‘{ *xTpueddy ue ut peqriosep ere Aoeanooe Atay pue sojoyd ate oy} 3urzATeue ut pesn seinpasoad sayy, ‘F3nTq Apues yszTy & YITA yoRet e BuoTe siaqzewW Gy OF SauNpeOF MOT YITM YORer e AOJ SiejowW 7°Z WOAF poTAeA 97R1 BYR ‘ieak 1ad siajoM g°E seM SoYydPaI @ATJ GY} TOF |}eI UOTSSede1 a8eA1eAy ‘*sojoyd Teter worz paanseow eI0M *h/-O/6[ UeeMqeq SuesTYOTW ‘AQuNOD UsTIzeg UT SoyoeeT ToOqJOWOTTH -9°T PAT] BuoTe ssueyo suTTeIOYS puke UOTSS90eI FJNTq Fo saqey ‘aTITI Ieao0g (Z-08 ‘OU $ TeqUeD YoIeeSey BuTIee8U -T3uqg TeIseoD *S'n — JaodeA snosueTTeosTW) — ‘wo je 8 ARE 8 och [em ‘O86, ‘80TATES UOTIEWIOFUL TeoTUYoe], TeuoTJeN WOAF |TqeTTeae : ‘eA ‘prtetysutads { zaqueg yoieesey BuTiseuT3uq TeqseoD) ‘s'n : ‘BA SATOATOY qiog — Aetoweyatg “Vy WeTTTTM 4q / 4/-0/61 ‘uestyoTW ‘AjuN0D ueTAZeg UT uoTSOZe BOYS pue JJNTq UO TaAeT syeT pue seinqonazqs Jo WoeFjo ou ‘y WETTTIM ‘reTeweyxIzTY Lc9 g-08 “ou AM PES €070L *Z-08 ‘Ou j10de1 snosuURTTeDSTW *AejUeD YyoIeesey BSuTiteeuT3uq Teqseog ‘S*N iseTtesg ‘II ‘“eTATL “I “WoTsoze etoys *G ‘“UesTYOTW eyeT ‘hy ‘sayey qeeig *€ ‘*‘seanjonazqs eT0YyszjO *Z ‘uesTYyOTW ‘AqUNOD ueTAteg ‘f *xTpueddy ue ut peqtiosep erie Aoeanooe AtTeyy pue sojoyd ate ey} SupzATeue ut pesn seanpasorzd ayy ‘zynTq Apues yZtTy & YITM yOeeT e BuoTe srejew C*y OF SeUNpetOF MOT YITM Yoeor e OF SIejowW 4*Z WOLF peTAeA ARPA DY {aeek tod siajzeU g°E SEM SaydeeT BATF BY TOF 97eA uOTSSadeI a8ereAy ‘sojoYd TeTiee wor paanseow aI9M *h/-0/6| UeeMzeq SuesTYOTW ‘AQUNOD UeTAizeg UT Seyorer Ta}SUOT TH -9°T SATJ BuoTe asueyo sUTTeIOYS pue uoTSSaze1 JJnTq jo soqey *8TITI 1aAOD (7-08 ‘ou § zeqUeD YyYoAeeSeYy BuTIseU (-T3uq Teqaseop *S*N — Todea snosueTTeosTW) — “wo /z § *TTE : ‘d [ZL] ‘O86, ‘e0TAIeS UOTJeWAOFUT TeoTuYyoe] TeUOTIeN WorAF aTqeTTeae : “eA ‘pretz3utads { toque ysieesey BuTrseuT3uq TeqseoD *S'n : “eA ‘ATOATAg qAoq — JoToweyIT_ “y wWeTTTTM Aq / ¥/-0261 ‘uesTYyOTW ‘AqunoD ueTazreg UT uotsoie sioys pue JJnNTq uo TaAeT syeT pue seinjonzqs Jo AOeyje sy ‘y WETTTEM ‘2eToweyxrTY £09 ¢=08 “ou awyTgcn- €0720L *Z-08 *Ou 3iode1 snosueTTessTW + ‘1ejUeD YyOIeOSey BSuTAseuTZuy [eqISeOD “S°M iSeTteS “II ‘“eTIFL ‘I ‘uotsote sxz0ys *G -UueSTYOTW PAST “4 ‘SoyeT qeeTD *E ‘*SaANjJONARWS aATOYSFJO *Z ‘“uesTYyOTW ‘AQUNOD UeTAIEg ‘[ *xTpueddy ue ut peqtadsep ore Adeanooe ATSeyW pue sojoyd ate 9yj ButzATeue ut pesn seinpesoid ayy *s3nTq Apues ystTy @ YITM yoeet e BSuoTe siejew G*y 0} SauNpeTOF MOT YIM YoRoI e AOJ Siajew y*z7 WOAF paeTieA ajer dy f1eak ted srajoW g'E SeM soyoPoI PATJ BY AJOF 9ReA uOTSsa9de1 V8etsAy ‘sojoyd Tetiee worlzZ paeinseou e19M 6H/-O/6f Ueemjzeq ‘uestTyoTW ‘AquNOD ueTIteg UT seyoeet T9]OWOTTY -9°T PATJ BuoTe a8ueyo suTTetoys pue uoTSssadet JynTq jo soqey ‘8TIT} TsA09 (Z-08 ‘Ou { JaqUeD YoTeesSey BuTIseUu -T3uq TeISeOD *S*n — 3A0deaT snosueTTeosTW) — -wo 7z § “TTE : -d [Z|] ‘O86[ ‘A9TATeS UOTJEWAOJUT TeEOTUYOeT TeUCTIeEN WoTy eTqeTTeae : ‘eA *prletzysutads § zaquUeD YoTReeSey ButTAveuTsuq TeqseoD *S'n : “eA SATOATAgG qiog — Jateweyztg “vy weTTTIM 4q / 7/-026) ‘uedtyoTW “AquNoD ueTiteg UT uoTSOI9 a10YS pue FJNTq UO TeAdST oyeT pue saanjonazaqzs Jo RZOaFFO 3yL ‘Vy WETTTIM “1eTeweyx1tg £729 CO} Oe awpgcn* €0ZOL "7-08 ‘ou qiodeXt snosueTTecosTW ‘1eqUaD YOTeeSeYy BuTAvauT3uq Teqseog "S'N :seTtes ‘Il “eTIEL ‘I ‘“uoTsoze ezoys *G ‘uesTYOTW eyeT ‘yh *soyeT qeeI9 *E ‘*seanqoONAAs |sA0YSFIO *Z ‘uesTYOTW ‘AQuNOD uetaAreg “7 *xTpueddy ue ut peqtzosep ore Adeanooe ITeYW pue sojoyd ate ayq ButTzATeue uT pesn sainpsooid sayy ‘zyntTq Apues y3TYy e YITM yoeer eB BuOTe siejeul ¢*y 02 SeUNpetOJ MOT YITM YoReT & AO} Sioejow 4°Z WorAF peTaea ajer 9yQ freak az9d srajzow g*E seM SoYyoeeT BATF O42 TOF 3ReAT uoTSsaver aBerteAy ‘sojoyd TeTiee worz peanseou 2am Sh/-O/6| UseMjeq SuesTYOTW ‘AQUNOD UsTAIEg UT SayoeeT IaqOWOTTY -9°T OATJF BuoTe asueyo suTTeOYs pue uoTSsedeA1 FJnTq Fo saqey “O8TIF2 teA0p (Z-08 ‘OU $ ZeqUeD YOTeaSey BuTIseU -T8ug TeqIseoD *S*n — qAodex snosueTTeostH) — ‘wo fz $ "TTT : ‘d [77] “O86, S20TAIES UOTIeMAOZJUT TeodTUYoe] TeuoTJeN WOAF sTqeTTeae : “eA ‘pretysutaids £ tequeD yoTeesey BuTAeeuTZuq TeqIseoD *S'n : “eA ‘SATOATOg qi0q — AeToweyATG *y WETTTEM 4q / 2-0/6, SuesTYyoTW ‘SAqunoD useTAiteg ut uoTSOIa aOYS pue FJNTq UO TeAST syeT pue saanjonajys Jo AoezzZo su ‘VY WETTTIM ‘teTeweyxrtg £79 C=08 sou aurecn* £072OL ‘7-08 “OU J40de1 snosUeTTeOSTW “1aqUeD YyOITeeSoy BuTAseuTBUq Te ISeOD “S°M ‘tSetzes ‘Il ‘“@TIFL “I ‘“uotsoze ot0ys *G “UeBTYOTW exe] “YH “SoyeT 2e01D *€ *SeAinj}ONAAS sAOYSFFZO *Z ‘“uesTYOTW ‘AQuUNOD useTIteg */ *xTpueddy ue ut peqraosep eae Aoeanooe ITsyW pue sojoyd ite 94} ButzATeue ut pesn seinpasoad syy *F3nTq Apues ys8Ty & YIM yore e BuOoTe Ssiajzew G*y 0} SoeuNpeZOFZ MOT YITM YOeeT e LOF SiazoW H°7 WOLF paTAeA 9}eA DYW SaeaKk 19d sieqoW gE seM SoyoeeI BATF FYI AOF 97eA UOTSSS9De1 BBetTeAY ‘*sojoyd TeTiee wory poanseou e1eM h/-0/6[ Weenqeq SuesTYyoTW ‘Ajunop ustTizeg ut seyoeed TeqoWoTTy -9'°T PATS BuoTe as8ueyo sUTTeIOYS pue UuOTSSedeI FjnTq Jo saqey “8TITF 139A00 (Z-08 ‘ou § tajque9 yoIeesey 3uTIeeu -T3uqg TeIseoD *S*n — JAodez snosaueTTedSTW) — ‘wo fz § “TTE : ‘d [ZL] ‘0861 S80TAIGS UOTIeEWIOFUT TeoTUYOeT, TeuOTJeN WOAF STQeTTeAe : “eA *‘ptetysurads £ taquej Yyorteasey BuTisauTZuq TeqseoD *S*p : “eA SATOATAG qio0g — AepToweyzTg “Vy WETTTTM Aq / 4/-0/61 ‘uestTyoTW ‘AjQuNOD ueTiteg ut uoTsoOta siOYys pue FjnNTq UO TeAeT ByeT pue seinjonazqZS Jo Joeszo syL ‘y WETTTIM “tepTowexaTYg £09 C=08) 20u awrscn’ €0ZOL *Z-08 ‘OU jA1odeX snosUPTTeOSTW ‘1ejUeD YoIvesey BuTeeuT3uq Teqseog "S*M :seTteg ‘II “eTITL “I “uoTsoze eto0ysg *G¢ ‘ueSsTYOTW oNeT “hy “seyeT 38019 *€ ‘*‘seAnjonA}s eTOYSTIO *Z ‘uesTYOTW ‘AQuNOD ueTAIEg ‘Tf *xTpueddy ue ut peqtiosep eae Aoeanooe atayy pue sojoyd ate ey ButTzATeue uf pesn seaanpasord syy *F3nTq Apues y3ty & YIM YOReT e BuoTe sizejeW C*h 07 SeUNpetTOF MOT YATM Yoeol e JOJ siajew 4*Z WOAF paTAeA |jeI |YyR faeeh ted siajoW gE SEM sSayoeaT BATJF BY TOF ae uoTssade1 asereay ‘sojoyd TeTazse worZ peinsesu ateM ‘h/-0/6| UeeMqeq ‘uesTYyTW ‘AjuUNOD UseTateg UT sayoeet TaJeWOTTH -9°T BATS BuoTe asueyo suTTeIOYS pue uoTSsave1 JJnTq jo saqey *8TITI 1e8A0D (7-08 ‘ou £ ataqUaD YOIReSeYy BuTIseU -T3uq Teqyseop “Sn — 310deAr snosaueTTedsTW) — “wo /z $ *TTE : *d [7/1] “O86. ‘e0TAZeS UOTJeEWAOFUT [TePOTUYDE], [TPUOTIEN WOAF STqeTTeAe : “eA ‘prety2utads { zequeg ysieasey SuTrseuTsug TeqyseoD *S'n : ‘ea SATOATAg qaoq — AepoMeyITG *y WETTTIM 4q / ¥/-0/61 ‘uestTYyoTW ‘AquN0D UeTiteg UT uoTSOle aioys pue JJNTq UO TOAST aYeT pue seinjzonaqs Jo Adezjo su “y WETTTIM ‘toToweyxrtTg £09 €-08 “ou AW ssn” €07Z0L *Z-08 ‘OU j10deX snosueTTeDSTW ‘oqUeD YyoTeesoy BuTIeeuTSU_ TeISeOD "S'h tS8eTI9§ “II ‘“eTITL ‘I ‘“WoTsors ezoyg *G ‘UeSTYTW eye] “y *"SONeT JB01D “Eg *So9AN}JIONAAS aAT0YSFJO *Z ‘“uesTYoOTW ‘AQuNOD ueTarzeg ‘[ *xTpueddy ue ut peqtiosep ere Adeanooe AITayW pue sojoyd ite 9y} BurzATeue UT pesn saanpssord ayy *FgnTq Apues y3TYy e YITM yoeerl e BuoTe siaqew ¢G*y 02 SeuNpadojZ MOT YITM YoReI e AOJZ SAtajoW 4*Z WOAF peTazeA 9jRe1 9YQ tae9h Aad siajow g'E Sem sayoeo1 PATJ BY LOZ JRE uoTSsade1 e8etsay ‘*sojoyd Tetiee worlzZ poeinseow e1emM Sh/-O/6[ UeeMjeq SuesTYyOTW ‘AQuUNOD UsaTIieg UT soeyoeet IToq,OWOTTY -9°T ATF BuoTe asueyd suTTetoys pue uoTsseder FynTq Fo saq\ey *9TITI aac (7-08 ‘Ou § Jaqueg yoIeasey 3uTiseUu -T3sug TeISeOD *S*n — JaoOdea snosueTTeosTW) — ‘wo ¢z $ “TIE : cd [7/1] "086, SS80TAIeS uoTJeWAOFUT TeoTUYe] TeUCTIeN Wory eTqeTTeae : ‘ep *‘prletgsutads § z9}Uae9 YyoTeesay BuTAsauTSuq Teqseop ‘S'n : ‘eA SATOATOgG JAOg — AsToweyATg ‘Vy weTTTIM 4q / 74-046, SuesTYyoTW ‘AQUNOD ueTIteg UT uoTsoia siloys pue FJNTq uo TeAeT syeT pue sainjzonaqjs Jo JoeFje sy] ‘y WETTTIM ‘1eTewexatg” TET