U.S. Geological Survey
Open-File Report 03-120
Peter Barnes1, Guy Fleisher2, James V. Gardner1, and Kristen Lee1
1-U.S. Geological Survey
Perspective view of Hog Island Reef in Lake Michigan. The distance across the bottom of the image is about 6700 m with a vertical exaggeration of 10x.
We apply state of the art laser technology and derivative imagery to map the detailed morphology and of principal lake trout spawning sites on reefs in Northern Lake Michigan and to provide a geologic interpretation. We sought to identify the presence of ideal spawning substrate: shallow, "clean" gravel/cobble substrate, adjacent to deeper water. This study is a pilot collaborative effort with the US Army Corps of Engineers SHOALS (Scanning Hydrographic Operational Airborne Lidar Survey) program. The high-definition maps are integrated with known and developing data on fisheries, as well as limited substrate sedimentologic information and underlying Paleozoic carbonate rocks.
Figure 1. - Location map of 6 sites mapped with SHOALS system along with USGS 30-m topography and regional shaded bathymetry developed from Holocombe et. al., 1996. We also show the regional underlying Paleozoic geologic formations mapped on land and inferred offshore. (modified from Michigan DNR, 1987) A- Boulder Reef, B-Gull Island Reef, C-Little Traverse Bay, D-Trout Island Shoals, Trout Island and the west coast of High Island, E-Hog Island Shoal, F- Dahlia Shoal and Ile Aux Gallet. Link to larger view,(JPEG 142 Kb).
Lake Michigan supported the largest and most valued commercial and recreational lake trout fishery in all the Great Lakes before it was driven to collapse in the middle of the last century by over fishing, predation by the exotic sea lamprey, and habitat degradation (Holey et al., 1995). With sea lampreys now under control, a Federal and state sponsored program to rehabilitate lake trout relys on intensive stocking of hatchery-reared fish. The current stocking strategy seeks to re-colonize lake trout in areas of suitable habitat for growth and spawning. The reefs in northern Lake Michigan, one of two historical areas that produced the majority of spawning activity in the early 1930s (Dawson et al., 1997), are prime targets for restocking by Federal and state agencies (Figure 2).
Figure 2. - Location map and spawning and nursery areas for lake trout and yellow perch (modified after Goodyear et al., 1982).
To date, stocked lake trout have become abundant enough to support sport and commercial fisheries and have developed significant populations on some targeted sites on rocky reefs in northern Lake Michigan (Holey et al. 1995). However, many of these fish have been found to stray to varying degrees into other areas during spawning. A study was initiated to document these movements and to understand the habitat attraction of these other areas and to quantify the distribution of spawning lake trout and associated physical habitat. This study includes areas with document historic catches of native lake trout (Figure 2), a representative distribution of rocky reefs that have and have not been recently stocked, and a mixture of near shore and offshore sites (Figure 1). An important part of this study is gaining an understanding of the geologic habitat attributes of successful spawning substrate, the potential for those attributes to change the regional distribution of that habitat.
Dawson (et al., 1997) found evidence of concentrations of historic spawning lake trout to be associated with Silurian rock outcrops along the lake's north and northwest shores and middle Devonian rock outcrops in the northeastern area of the lake (see Fig. 1, and Table 1). Another cluster of offshore spawning reefs occurs at the major geological unconformity occurring between upper Silurian and middle Devonian rocks; an unconformity which often exhibits slumping and irregular fracturing on the lakebed. This process may be partially responsible for some reefs in northern Lake Michigan. The extent and geological details of the unconformity and fracturing are not well known. Studies by Sommers (1968) and others (summarized in Soller, 1998, Figure 3) indicate the presence of glacial outwash, glacial till, and modern lake sediment as offshore surface deposits in addition to the Paleozoic outcrops. The coast and upland areas of the islands and mainland are also draped with glacial features (Farrand et al., 1984) which could extend offshore and be found on bathymetric highs (reefs).
Table 1. - Lithilogic Description of basement Paleosoic Rocks (from Dorr and Eschman, 1984 and Catacosinos, et al, 2000)
Late and Middle Devonian (mapped as one unit in Fig 1)
Middle and Early Devonian
Upper Silurian (mapped as one unit in Fig 1)
Figure 3. - Modified from Quaternary geologic map of Lake Superior region 4° x 6° Quadrangle, United State and Canada, showing areas mapped with high resolution LIDAR morphology and onshore topography, 1984 and 1982 Quaternary Geology Michigan Natural Features Inventory and MI DNR. Link to larger view, JPEG 185 kb.
Recent detailed morphologic assessments of the Great Lakes basins (Holcombe et al.,1996) are based on historic depth measurements, with data densities derived on a 2-kilometer to 90 meter grid spacing. We use this grid as the regional setting for our more detailed study.
Modern techniques can greatly benefit the interpretation of lakebed morphology; of offshore geologic character and geologic processes; and for addressing the relation of habitat character and sites preferred for lake trout spawning. These approaches provide geo-referenced lakebed measures at meter horizontal and decimeter vertical resolution. To achieve this goal the project cooperated with the U.S. Army Corps of Engineers to use a high-resolution, airborne laser system SHOALS (Scanning Hydrographic Operational Airborne Lidar Survey) to map approximately 200 square kilometers of shallow lakebed morphology on a 4 meter grid at decimeter resolution. We (J.V.G.) are also working to develop and apply new techniques to extract spectral reflection and amplitude information from the laser data to allow additional classification of biologic and geologic substrate information.
Elevation data of detailed substrate morphologies was obtained from six coastal and offshore study sites (Figure 1) during 8 aircraft flights over a period of 6 days, August 27 to September 1, 2001. On August 27 and 28 water clarity (Secchi disc 6.5-7.5m) and spectral reflectance were measured at Boulder and Gull Island Reef during overflights by the SHOALS aircraft. The relatively good water clarity allowed the SHOALS system to record bottom reflections from 25 -30 meters. Differential GPS navigation and elevation data were referenced to the International Great Lakes Datum (IGLD1985) - (See 1995 Report). The long term (1900-1990) lake level is close to 176.5m (NOAA/GLERL web site). Lake level at the time of our survey was about 176m (Figure 9.1). This is the elevation (176m) through which we referenced water depth relative to IGLD (1985) during the lidar survey (Lake water level 30 kb gif).
The SHOALS system decimeter elevation/location data were merged with coarser NOAA gridded bathymetry (Holcombe et al. 1996) and a 30m USGS digital elevation model (DEM) topography to generate the six map images that underpin this report (Figures 2 through 7). Three criteria are thought to be important in forming ideal lake trout spawning habitat. 1) Coarse substrate with voids; 2) "Clean" substrate (devoid of biologic growth and fine sediment); and 3) Adjacency to steep slopes with access to deeper water. To assess these criteria illumination, color coding and vertical exaggeration were subjectively selected to best enhance viewing the substrate morphology. Oblique images were selected that best emphasized the habitat morphology. Sparse substrate samples, underwater diver and useful but local (Boulder Reef and Gull Island Reef) video observations supplement the morphologic information.
Three geologic regimes are interpreted from the morphology, bedrock, glacial and modern. These form the basis for substrate, habitat and morphologic classification. Representative examples of these morphologies are shown in Figure 5.1. Southwest-northeast striking Paleozoic carbonate rocks underlie most of the northwestern rim of the Michigan Basin and crop out at the coast, on offshore islands (Michigan DNR, 1987) and at Hog Island Reef (Sommers, 1968; Figure 6). Morphologic scarps and bedding(?) lineations at all sites (See especially Figures 5 & 6) suggest bedrock at or near the surface at all of the mapped areas, but specific location confirmation is lacking. Outcrops are also thought to be present along the nearshore and western portions of the area surveyed in Little Traverse Bay.
Overlying the Paleozoic bedrocks are pervasive glacial deposits in the west of Figure 3 including compacted clay tills and outwash gravel and sand (Sommers, 1968). The morphology and video observations (Thomas Edsall, Gregory Kennedy, personal communication, 2001) suggest NW-SE lineations in basal till and one to three kilometer diameter) lobe-like cobble and boulder moraines with linear outwash (?) ridge-deposits to the east of Boulder, Gull Island and Dahlia Reefs ( Figures 2.1, 3.1, and 7.1). Erosional or depositional ridges and channels radiate outward to the east and south from the shallowest (3m) portion of these features, and include a pronounced 1 km long and straight ridge at the shallowest part of the Gull Island Reef (Figure 3.1 and 3.2). Post-glacial reworking appears minimal in depths greater than 10m as the glacial bedforms do not appear modified in response to post glacial hydraulic regimes. The coarse textured glacial material extends to beyond the depths of the survey as a surficial deposit to the west and south at Boulder and Gull Island Reef's where we have video data to more than 40 m depth (Thomas Edsall, Gregory Kennedy, personal communication, 2001)
Modern lake-bed sand deposits appear as overprints on the glacial and bedrock surfaces and occur as thin down-drift (to the east) bedforms, sand sheets and depositional lobes. Along the coast of Little Traverse Bay, en-echelon nearshore bars with shore normal bedforms are well developed, particularly toward the head of the bay.
Along the west coast of High Island, rough nearshore morphology (Bois Blanc Formation?) changes abruptly to a smooth (sand?) bottom along a 2-3m scarp in about 10 m water depth (Figures 5.1 and 5.2). The bottom of this scarp may represent the location of a previous (glacial) shoreline, although this sharp transition and shoreline-like feature is not obvious in similar water depths at other mapped sites.
Modern hydraulic processes appear to control the morphology of Trout Island Shoal (Figures 5.1 and 5.3). The shoal appears to be a large sand(?) body migrating eastward over underlying bedrock (glacial or units of the Devonian Bois Blanc Formation (Figure1)?). The eastern slopes of the shoal are relatively steep. Three generations of recurved spits are present on the northeast part of the shoal. In the eastern (lee ?) of the shoal northwest-southeast linear ridges (glacial or bedrock?) are partly obscured by patches of smooth sand(?).
Southwest of Ile Aux Gallet on Dahlia Shoal (Figure 7.1) numerous smooth 300-500m wide amphitheater-like features opening to both the southwest and northwest are present in water depths greater than 15m. and indent the shallower, rougher relief (of cobbles and boulders or outcrop?). The smooth morphology of these features suggests they are presently depocenters.
Post-glacial active reworking of the substrate appears minimal in depths greater than 10m based on the absence of hydraulic bedforms and sample and video observations of cobble and boulder substrates or smooth depositional sediment deposits. Comparison of 1989 and 2001 videos of Boulder and Gull Island reefs indicates a change from nearly mussel and algae free cobble and boulder substrate to mussel and algal dominated cobble and boulder substrate in 2001.
Relevance to lake trout spawning habitat - In terms of the three criteria are thought to be important in forming ideal lake trout spawning habitat. 1) Coarse substrate with voids; 2) "Clean" substrate (devoid of biologic growth and fine sediment); and 3) Adjacency to steep slopes with access to deeper water, we believe that the glacial deposits could be important habitat. Deposits such as outwash gravel and boulders would form void filled substrate that if free of algae and zebra mussels and adjacent to deeper water might be ideal. The morphology of ridges and lobate glacial forms at Boulder Reef, Gull Island Reef and Dahlia Reef may be the source of these features and associated substrate habitat, however, their "cleanliness" may be compromised by algae and zebra mussel growth. Laser waveform data is being analyzed for benthic albedo information which we hope will lead to an ability to assess both substrate texture and also substrate "cleanliness" and facilitate remote assessment of benthic habitat.
The trout spawning reefs in northern Lake Michigan show complex morphology and habitat variability related to patchiness of bedrock, glacial and modern geologic deposits. In addition, the spawning criteria of coarse substrate with adjacency to deeper lake water are common in the areas mapped. The presence of "clean" spawning substrate was not determined by mapping, but video form pre and post zebra mussel invasion suggest higher infestation of mussels and algae on cobble and gravel substrate since 1989. The lidar data gathered in this study will serve as a long-term baseline for monitoring the character and status of lake trout habitat. We also expect these results to help define new techniques to remotely identify high quality spawning habitat that can play a significant role in enhancing and developing naturally reproducing lake trout stocks in Lake Michigan.
Catacosinos, P.A. M.S. Wollensack, W.B. Harrison III, R.F Rynolds,
and D.B. Westjohn, 2000, Stratigraphic Nomenclature of Michigan,
MI Dept of Env. Quality, Geological Survey Div. and Mich. Basin
Geological Soc. one sheet.
Dawson, K. A., R. L. Eshenroder, M. E. Holey, and C. Ward. 1997. Quantification of historic lake trout (Salvelinus namaycush) spawning aggregations in Lake Michigan. Can. J. Fish Aquat. Sci. 54(10):2290-2302.
Dorr, J.A. and Eschman, D.F. 1970, Geology of Michigan: The Universtiy of Michigan Press, Ann Arbor.
Edsall, T.A., Poe, T.P., Nester, R.T., and Brown, C.L., 1989, Side-scan sonar mapping of lake trout spawning habitat in Northern Lake Michigan; N. Amer. Jour. of Fisheries Management. v.9, p. 269-279.
Farrand, W.R., Mickelson, D.M., Cowan,W.R. and Goebel, J.E. 1984, Quaternary Geologic Map of the Lake Superious 4º x 8º Quadrangle, United States and Canada. U.S. Geological Survey, Misc. Geologic Investigation Series Map I-1420(NL-16). 1:1,000,000.
Holcombe, T.L., D.F. Reid, W.T. Virden, T.C. Niemeyer, R. De la Sierra, and D.L. Divins (1996). Bathymetry of Lake Michigan. A color poster with descriptive text anddigital data available on CD-ROM. U.S. Dept of Commerce, National Oceanic and Atmospheric Administration, National Geophysical Data Center, Boulder, CO. Report MGG-11.
Holey, M. E., R. R. Rybicki, G. W. Eck, E. H. Brown, Jr., J. E. Marsden, D. S. Lavis, M.L.Toneys, T. N. Trudeau, and R. M. Horrall. 1995. Progress toward lake trout restoration in Lake Michigan. J. Great Lakes Res. 21(Supplement 1):128-151.
Krueger, C.C., B.L. Swanson, and J.H. Selgeby. 1986. Evaluation of hatchery-reared lake trout for establishment of populations in the Apostle Islands region of Lake Superior, 1960-1984. In Fish Culture in Fisheries Management, ed. R.H. Stroud, pp. 93-107. Bethesda, MD: American Fisheries Society.
NOAA/NOS/CO-OPS water level station data (2001). Retrieved from
Soller, David R., 1998, Sheet B of "Map showing the thickness and character of Quaternary sediments in the glaciated United States east of the Rocky Mountains". U.S. Geological Survey Miscellaneous Investigations Series Map I-1970, scale 1:1,000,000.
Sommers, L.H., 1968, Preliminary report on geological studies in northern Lake Michigan using underwater observation techniques; Internat. Assoc. for Great Lakes Research, Proceedings, 11th conference on Great Lakes Research, v. 239-244.