U.S. GEOLOGICAL SURVEY OPEN FILE REPORT 94-255 The Southern Lake Michigan Coastal Erosion Study CD-ROM By C.F. Polloni1, C.L. Brown2, D.W. Folger1, D.S. Foster1, and A.L. Brill3 Display software by Russell A. Ambroziak4 Documentation by Christine A. Cook4 1U.S. Geological Survey, Woods Hole, MA 02543 2U.S. Forest Service, Jacksonville, OR 97530 3Duke University, Durham, NC 27708 4U.S. Geological Survey, Reston, VA 22092 OVERVIEW the following is reprinted with permission from J. Great Lakes Res. 20(1):2-8, Internat. Assoc. Great Lakes Res., 1994 OVERVIEW OF THE SOUTHERN LAKE MICHIGAN COASTAL EROSION STUDY David W. Folger1, Steven M. Colman1, and Peter W. Barnes2 1U. S. Geological Survey, Woods Hole, MA 02543 2U. S. Geological Survey, Menlo Park, CA 94025 Chicago's waterfront was flooded repeatedly between 1985 and 1987 [figure_1] when Lake Michigan rose to its highest level of this century. Damage was extensive and the need for a comprehensive study to evaluate causes, effects, and the potential frequency and magnitude of future coastal erosion and lake-level changes was apparent. In 1987, the U.S. Geological Survey (USGS) and Illinois State Geological Survey (ISGS) conducted a reconnaissance survey with sidescan sonar of the failing shore defense structures along the Chicago shoreline (Chrzastowski and Schlee 1988). This effort produced information that was immediately useful to the U. S. Army Corps of Engineers and various state and city planners. It showed that state-of-the-art marine techniques could be applied successfully to the study of shoreline erosion in southern Lake Michigan and resulted in funding for a more comprehensive study. Congressman Sidney Yates, of the U. S. House of Representatives, whose District includes the Chicago waterfront, sponsored a bill funding a five-year (1988-1992) study to be carried out by the USGS in cooperation with the ISGS to gain a better understanding of the magnitude and frequency of processes associated with lake level fluctuations. The urgency for conducting such a study was highlighted by the extensive damage that occurred during the 1985- 87 high water levels and the fact that, historically, each major episode of high water level has exceeded the previous one since the turn of the century [figure_2] (Bishop 1990). The study was divided into three parts: Framework, Lake Level, and Processes. The first, Framework, was designed to build upon earlier geologic studies of the lake such as those of Hough (1938, 1958), the ISGS, summarized by Wickham et al. (1978), and Lineback et al. (1971, 1979), and Cahill (1981). The Framework studies provide a more complete picture of the geologic structure, stratigraphy, and surficial sediment distribution especially of the nearshore zone. The second part of the study, Lake Level, was designed to provide new information on the magnitude and frequency of pre-historic fluctuations. Although considerable work has been done in this area [see recent reviews by Hansel et al. (1985) and Larsen (1985,1987)], relatively little has focussed on late Holocene shoreline features or the record from deep-water sediments. The third part of the study, Processes, was designed to evaluate some of the mechanisms responsible for changes in the shoreline and lake-bed, particularly those that are poorly understood or aspects of them that have not been addressed by such past important process studies as those by Hands (1983), Fox and Davis (1973), Wood et al. (1988), Seibel (1986). To carry out these studies, participants, in addition to those from the USGS and ISGS, included researchers from the Indiana Geological Survey, University of Indiana, Purdue University, University of Michigan, Northeastern Illinois University, Oregon State University, Woods Hole Oceanographic Institution, University of Rhode Island, and the University of Washington. The 14 papers included in this volume represent much of our 5-year research effort. Four papers in the Framework section include: Late Wisconsinan and Holocene geologic history of the Illinois- Indiana coast of Lake Michigan (Chrzastowski and Thompson), Coastal geomorphology and littoral cell divisions along the Illinois- Indiana coast of Lake Michigan (Chrzastowski et al.), The geologic framework of southern Lake Michigan (Foster and Folger), and Survey of littoral drift sand deposits along the Illinois and Indiana shores of Lake Michigan (Shabica and Pranschke). Three papers in the Lake Level section include: Lake level history of Lake Michigan for the past 12,000 years: the record from deep lacustrine sediments (Colman et al.), Lake Michigan's late Quaternary limnological and climate history from ostracode, oxygen isotope, and magnetic susceptibility records (Forester et al.), and Isostatic uplift history and Holocene landforms (Larsen). Seven papers in the Processes section include: Rates and processes of bluff recession along the Lake Michigan shoreline in Illinois (Jibson et al.), Wave climate and nearshore lakebed response, Illinois Beach State Park (Booth), Modeling beach and nearshore profile response to lake level change (Wood et al.), Contemporary and historical rates of eolian sand transport in the Indiana dunes area of southern Lake Michigan (Olyphant and Bennett), The influence of ice on southern Lake Michigan coastal erosion (Barnes et al.), Video monitoring of nearshore ice in southern Lake Michigan (Kempema and Holman), and A sediment budget for southern Lake Michigan: source and sink models for different time intervals (Colman and Foster). Summaries of these papers are included below. FRAMEWORK The physiography or geomorphology of the Illinois-Indiana coastline developed in response to glacial processes that formed the lake margin and, subsequently, to coastal processes that modified it by erosion and accretion. Glacial landforms significantly influenced coastal evolution and paleogeography. Prior to 2.5 ka separate spit systems terminated on the eastern and western shores of the lake. During the last 2.5 ka the systems coalesced forming the present coastline that continues to be modified mainly by erosion (Chrzastowski and Thompson). The 156-km coast can be divided into three provinces. The northernmost is the Zion beach-ridge plain that extends from the Wisconsin border southward into Illinois as far as the city of North Chicago. This province is 19 km long, has a maximum width of 1 km and, stratigraphically, contains sediments as thick as 11 m. The second province, the Lake Border morainal-bluff coast, extends for 25 km from North Chicago to Winnetka. The shoreline has eroded into end moraines of the Lake Border Morainic Complex, producing bluffs that are as high as 30 m. The bluffs decrease in height southward in the third province, the Chicago/Calumet lacustrine plain, which rims the southern margin of the lake. This stretch of coastline extends 112 km from Winnetka, IL to the Indiana-Michigan state line. The relief ranges from low lying areas near lake level to the Indiana Dunes, which are as much as 50 m high. Fifty-eight percent of all uplands along the Illinois-Indiana coast range from 0- 10 m in height. All of the Zion beach-ridge plain and most of the Chicago/Calumet lacustrine plain are no more than 5 m above the mean lake level (Chrzastowski, Thompson, and Trask). Offshore from these coastal features, compiled and interpreted maps of lake bottom sediment texture and stratigraphy show that much of the lakefloor is a dynamic environment; within about 20 km of shore currents, induced by storm waves, transport fine sand and silt, resulting in a patchy, continually changing distribution of lacustrine sediment overlying a till-gravel pavement. Only north of Waukegan, Ill. and east of Michigan City, Ind. does silty sand completely cover the surface of Wadsworth Till, which is 09a glacial deposit composed mostly of silty clay. A nearshore sand wedge is thickest north of Waukegan and thins southward to Chicago. Large volumes of sand offshore are limited to complex northeast- trending ridges formed at the outer margin of the nearshore sand wedge between Waukegan and Lake Forest. South of Lake Forest, the nearshore sand wedge is limited to within 300 m of the shoreline. Though several meters of sand are present in some areas of the bottom in Indiana Shoals, net erosion of the lake floor has taken place there over the last 20 years. West of Michigan City, the nearshore sand wedge is thin or absent. Thus, over much of the survey area, erosion or non-deposition has been taking place exposing the 10-40 m thick Wadsworth till or gravel-boulder lag deposits, and in some places the underlying Devonian shale or Silurian and Devonian carbonates (Foster and Folger). Sand has been diverted and trapped by harbors and lakefills along the Illinois and Indiana shore, depleting the supply of littoral sand for longshore drift. Most sand (38 X 106 m3) on the western side of the lake is concentrated along a 13.7-km stretch of coast north of Waukegan Harbor but much less is present in all the remaining profiles surveyed as far south as Evanston, Ill. Comparisons of sand thickness in several areas south of Waukegan reveal that from 1975 to 1991 sand has been greatly depleted, exposing the underlying clay-rich till to erosion (Shabica and Pranschke). LAKE LEVEL Analyses of the stratigraphy of the late Quaternary sediments deposited in Lake Michigan provide a good record of lake-level changes during the last 12,000 years. High lake levels were associated with the last major glacial advance with at least two incursions of flood waters from glacial Lake Agassiz. Drastically lowered lake levels occurred when ice retreat opened the North Bay outlet, draining Lake Michigan to the Chippewa low phase, at least 80 m below its present level. The subsequent rise in water level from the Chippewa low to the Nipissing high is well documented by the distinctive, planar, trangressive Chippewa unconformity that left little evidence of intermediate lake-level positions. A dramatic change in conditions in the lake occurred after about 5 ka; among other changes, ostracodes and mollusks are poorly preserved after this time, eliminating many tools for reconstructing paleolimnologic conditions. However, short-term variations in grain size and magnetic properties of the sediments, over time periods of 200 to 500 years, may be due to lake level changes (Colman et al.). Before 5 ka, isostatic changes in the elevation of outlets and climate change controlled lake-levels. Ostracode species statigraphy, which is related to lake volume and chemistry, provides a way to reconstruct the lake's early history. Stable isotope values from both ostracodes and mollusks also provide information about lake volume history and a test of the ostracode reconstructions. The ostracode species stratigraphy subdivides the lacustrine sedimentary record into five episodes. During the oldest episode, the lake contained cold water with very low solute content owing to lake interaction with glacial ice. Then, in the next episode, lake level fell, probably as glacial ice retreated from the basin leaving the North Bay outlet exposed. The glacial retreat may have been hastened by a generally dry climate including warm, dry summers. The third episode shows the lake expanding due to both isostatic rebound and to a wetter, colder climate. In the fourth episode, lake level retreats due to another period of ariditiy, but those retreats are punctuated by Agassiz flood events temporarily negating the impact of climate. Finally, in the last episode, the lake's solute content is elevated and lake level may have fallen below or not kept pace with the rising outlet as the lake responded to the full impact of mid-Holocene climate aridity. Following 5 ka, the loss of ostracodes from the sediment record eliminates this technique for reconstructing more recent lake level history (Forester et al.). Study of beach ridge complexes at Sturgeon Bay, Wisc., Whitefish Point, Mich., and Gary, Ind. provide a reconstruction of differential postglacial uplift in the upper Great Lakes. The landforms in the three complexes were correlated and dated, in a manner similar to tree rings, supported by 14C dating. The rebound rates calculated were not exponential, as expected, but linear over the past 4000 years. For Whitefish Bay the rate was 0.34 m/century and for Sturgeon Bay the rate was 0.09 m/century relative to the south shore (Gary, Ind.) of Lake Michigan. These are similar to rates calculated from historic lake level measurements. Lake Superior was uplifted above Lake Huron about 2100 years ago separating the mean water surfaces by 6.92 m. Lake level fluctuations during the past few thousand years appear to have exceeded those measured historically by as much as a factor of two. (Larsen). PROCESSES Between 1872 and 1987, rates of bluff retreat from Wilmette to Waukegan vary from 10 to 75 cm/yr between discrete segments of bluffs [figure_3]. The average rate of retreat for the entire area, however, does not vary significantly between 1872-1937 and 1937- 1987 and ranges from 20-25 cm/yr. No obvious correlation appears to exist between lake levels, rainfall, abundance of groins, and retreat rate. Local variations in retreat rate do, however, correlate closely with lithologic variations. Bluffs that contain lake-plain sand and silt have higher retreat rates than clay-till bluffs. However, the bluffs have little curvature across these boundaries indicating that the variations average out over time, producing long- term parallel bluff retreat. New data from cone penetrometer tests are being combined with the recession data to generate a model that will predict the annual sediment contribution to the lake from the eroding bluffs (Jibson and Staude). In contrast to the bluff erosion north of Chicago, sand is being added to the shoreline and blown landward into the Indiana Dunes area. Most eolian sand movement there occurs in late fall and spring when wind speeds are still high but marram grass and snowcover are not abundant. One dune has grown vertically by 75 cm since studies began in the fall of 1990. Storm winds from the north transport sand most effectively and produce blow-out areas. Moisture is the second most important factor relative to wind speed in controlling the volume of sand moved. Highest transport rate (>126 g/cm/hr) were observed at wind speeds of 7 m/s from the north under dry conditions. In contrast, transport was low (26 g/cm/hr) during almost equally strong southerly winds (6 m/s) because the sand surface was damp from precipitation during the storm, which inhibited grain movement (Olyphant and Bennett). In constrast to these areas of particularly dynamic sediment movement, the lakebed of the outer nearshore zone at Illinois Beach State Park shows apparent significant sediment movement only during major storms which typically occur with a frequency of years to decades. When grain size and morphologic data observed in this study were compared to similar data sets from 1946, 1973, and 1978, no conspicuous changes were noted. This implies that the area is nondepositional; that is, evidence for significant erosion or deposition in the area over the time represented by the studies was lacking. A non-depositional interpretation is generally consistent with the non-depositional or erosional nature of the southern Lake Michigan lake floor not far to the south. However, the Illinois Beach lakebed, which is characterized by a substantial sheet of loose sand over much of its area may be more dynamic in terms of sediment movement (more sediment transported in and out) than the area to the south. Alternatively, the area is unique: the presence of relict sands coupled with the continual flux of of sand from the north have created an offshore environment that is buffered against, rapid, large-scale changes, at least in the short term. The extent and magnitude of cyclic, seasonal changes within this outer nearshore lakebed are unknown (Booth). Closer to shore, a nearshore ice complex (NIC) develops and decays repeatedly along the coast of southern lake Michigan between December and March acting as an ephemeral sea wall causing cut and fill along its lakeward margin [figure_4]. Sediment, entrained in ice as it forms is rafted alongshore and, to a lesser extent, offshore in amounts similar to those supplied by erosion of the northern Illinois bluffs. Factors that impact the transport of littoral sediment by ice include: 1) the multiple break-up and refreezing of a NIC during a single winter; 2) littoral sand, suspended in water nearshore by anchor ice, is formed into the NIC by waves; 3) after sediment is entrained in ice, long range transport is possible; 4) wave energy, rather than melting, breaks up the NIC releasing sand- laden ice to longshore and offshore ice rafting; and 5) jetties and promontories enhance sediment loss to deep water by deflecting ice streams offshore. In short, ice does not protect the coast from erosion (Barnes et al.). A time-lapse video camera system set up on the southwestern shore of Lake Michigan showed that longshore ice drift was most often to the southeast at speeds between 0.047 and 0.377 ms-1 and cross- shore drift was offshore at speeds between 0.009 and 0.126 ms-1. Based on these data and published data, estimates for the longshore sand flux ranges from 0.008 to 0.760 tonsh-1 and the cross-shore flux from 0.0016 to 0.096 tonsh-1. These observations clearly demonstrate the feasibility of using video to monitor the movement of ice during daylight hours and their application for assessing sediment transport by ice (Kempema and Holman). With waves, currents, and ice as obvious major contributors to coastal erosion, an effort was made to assess the validity of various coastal evolution models. For examples, a large set of beach and nearshore profile data along the Indiana shoreline were used to evaluate the widely applied equilibrium profile concept. Results show that the profile form h=Axm ( where h=depth, A is a coefficient related to wave energy dissipation, x is distance offshore, and m is an arbitrary exponent) is appropriate to describe the average profile shape in southern Lake Michigan. Values of m are even more closely grouped than those determined on ocean coasts, which may, in part, be due to the small geographic area encompassed by the study. Statistically, the relation between A values and mean grain size showed weak, but not significant, agreement in the inner part of the profiles from the water's edge to the inner bar trough. Little or no support, thus, was found for the use of Moore's curve relating the coefficient A to mean sediment size. There appears to be a phase lag between lake level change and profile response that must be resolved before accurate modeling can be successfully carried out (Wood et al.). Efforts to quantify and balance the sediment budget for the southern Lake Michigan basin have been largely successful. Sediment sources are bluffs, rivers, aerosols, and basin import; sediment sinks include the lake basin, basin export, and nearshore sand, beaches, and dunes. During the last 100 years, when historical measurements are available, the primary source of sediment is bluff erosion and the primary sink is deposition in the deep basin. The budget nearly balances at 3 X 106 t/yr. About half the sand derived from bluff erosion is deposited in deep water modern sediments. The other half appears to be deposited nearshore, and on beaches and dunes. For periods farther back in time the sediment sinks can be estimated fairly well but the sources are more difficult to evaluate but bluff erosion has probably been the dominant source (Colman and Foster). DATA AVAILABILITY Raw data and as much interpreted data as possible will be assembled in a Geographic Information System (GIS) and subsequently made available in a CD-ROM to the user community. To acquire the CD-ROM, address requests to: Map Sales U. S. Geological survey Box 25286 Denver, CO 80225 or call: 1-800 USA MAPS FUTURE RESEARCH Of the work that remains to done, especially important are more intensive studies of beaches and nearshore sand depletion because of their importance to bluff and substrate erosion and impact on recreation for this populous area. Hydrologic (wave climate, currents) and suspended sediment studies, largely unaddressed in this project, are of equal importance not only for quantifying rates of sediment transport but for addressing such important problems as pollutant mobilization and distribution. Bibliography Bishop, C. T. 1990. Historical variation of water levels in Lakes Erie and Michigan-Huron. J. Great Lakes Res.16: 406-425. Cahill, R. A. 1981. Geochemisty of recent Lake Michigan sediments. Illinois State Geological Survey.Circular 517, Champaign, IL . Crzastowski, M. J. and Schlee, J. S. 1988. Preliminary sidescan sonar investigation of shore-defense structures along Chicago's northside lake front: Wilson Avenue groin to Ohio Street Beach. Illinois State Geological Survey Environmental Note 128. Fox, W. T., and Davis, R. A., Jr. 1973. Simulation model for storm cycles and beach erosion on Lake Michigan. Geol. Soc. Am. Bull. 84:1769-1790. Hands, E. B. 1983. The Great Lakes as a test model for profile responses to sea level changes. In CRC Handbook of Coastal Processes and Erosion, ed. P. D. Komar, pp 167-189, Boca Raton, Florida. Hansel, A. K., Michelson, D. M., Larsen, C. E., and Schneider, A. F. 1985. Late Wisconsinan and early Holocene history of the Lake Michigan basin. In Quaternary geological evolution of the Great Lakes, pp. 39- 54, ed.. P. F. Karrow, and P. E. Calkin, Geological Association of Canada Special Paper No. 30. . Hough, J. L. 1935. The bottom deposits of southern Lake Michigan. J. Sedimentary Petrology 5: 57-80. Hough, J. L., 1958. Geology of the Great Lakes. Urbana, IL: University of Illinois Press. Larsen, C. E. 1985. Lake level, uplift, and outlet incision, the Nipissing and Algoma Great Lakes. In Quaternary Evolution of the Great Lakes, ed. P. F. Karrow and P. E. Calkin, pp. 63-78. Geol. Assoc. Canada Spec. Paper No. 30. Larsen, C. E. 1987. Geologic history of glacial Lake Algonquin and the upper Great Lakes. U. S. Geological Survey Bulletin 1801, 36 p. Lineback, J. A., Gross, D. L., Meyer, R. P., and Unger, W. L. 1971. High resolution seismic profiles and gravity cores of sediments in southern Lake Michigan. Illinois State Geological Survey, Environmental Geology Notes no. 47, Champaign, IL, Lineback, J. A., Dell, C. I., Gross, D. L. 1979. Glacial and postglacial sediments in Lakes Superior and Michigan. Geol. Soc. Am. Bull 90: 781-791. Seibel, E.,1986. Lake and shore ice conditions on southeastern lake Michigan. In Southeastern nearshore Lake Michigan: impact of the Donald C. Cook Nuclear Plant, ed. R. Rossmann, pp. 401-432, University of Michigan, Great Lakes Res. Div., Pub. No 22, Ann Arbor, Michigan. Wickham, J. T., Gross, D. L., Lineback, J. A., Thomas, R. L. 1978. Late Quaternary sediments of Lake Michigan. Ill. State Geol. Surv. Environ. Geol. Notes 48. Studies of Lake Michigan Bottom Sediments Number 13. Champaign, Illinois. Wood, W. L., Hoover, J. A., Stockberger, M. T., and Zhang, Y. 1988.. Coastal stiuation report for the State of Indiana. Great Lakes Coastal Research Laboratory, School of Engineering, Purdue University, West Lafayette, Indiana. FIGURE CAPTIONS Figure 1. Large wave hitting the seawall at Rogers Park, Chicago, Ill. Photo courtesy of M. Chrzastowski. Figure 2. Estimated prehistoric (A.-Colman et al. this volume; B.-Larsen 1985) and historic (Bishop 1990) lake levels. Figure 3. Bluff erosion at Lake Bluff, Ill. Photo courtesy of R. Jibson. Figure 4. Nearshore Ice Complex (NIC) at Wilmette, Ill. January, 1991 (Photo courtesy of D. Folger) End of text; press the Escape key to continue.