The reoccurring region of seasonal hypoxia in the central basin of Lake Erie (“the dead zone”) has been of significant interest to researchers over the past several years. Surprisingly however, no efforts to characterize the endemic microbial community, responsible for the consumption of oxygen in this system, have been published. To understand how the microbial community may be interacting with this event, we have begun to characterize microbial members by using molecular tools. Phycoerythrin-rich cyanobacteria appear abundant and active in a narrow region (∼ 1.5 m) below the thermocline during hypoxic conditions, reaching abundances of greater than 10⁵mL⁻¹and being the primary agent releasing 1.5 mg O₂ L⁻¹above the daytime demands in this region. Sequencing of 16S rDNA amplicons, generated with universal eubacterial primer sets, from the Lake Erie's hypolimnion during seasonal oxygen depletion demonstrated that cyanobacteria, most closely related to phycoerythrin-rich Synechococcus spp., dominate during rapid drawdown of oxygen (0.083 mg L⁻¹d⁻¹in 2004) in this region. Analyses of another conserved marker of phylogeny (RuBisCO) has been used to confirm the presence of these cell types. Numerous distinct taxa of heterotrophic bacteria are also represented in the 16S library. The results of this study suggest that novel groups of cyanobacteria may persist within the Lake Erie dead zone during hypoxic conditions and, along with the heterotrophic community, strongly influence system geochemistry.

We wanted to confirm the presence of round gobies (Neogobius melanostomus) in the upper Niagara River and determine if there was a negative association between the catches of round gobies and smallmouth bass (Micropterus dolomieu). Our approach was to compare the catch rates of round gobies and smallmouth bass by anglers based on their method of fishing (from boat or from shore) and the fish they sought (anything or black bass). During 2003, both boat anglers and shore anglers caught round gobies in the upper Niagara River. The catch rates of round gobies and smallmouth bass for shore anglers who sought anything were positively and significantly correlated (r = 0 .081, P < 0.001). The catch rates of round gobies and smallmouth bass for shore anglers who sought black bass were negatively and significantly correlated (r = 0.141, P < 0.005). Correlations were not done using the catch rates of round gobies and smallmouth bass by boat anglers because they caught so few round gobies. Our results suggest that round gobies did not disrupt fishing for smallmouth bass on the upper Niagara River during 2003 if disruption is defined exclusively by the association between catch rates of smallmouth bass and round gobies. However, catching round gobies did lead to angler perceptions of poor fishing quality.

Hydroacoustics can be used to assess zooplankton populations, however, backscatter must be scaled to be biologically meaningful. In this study, we used a general model to correlate site-specific hydroacoustic backscatter with zooplankton dry weight biomass estimated from net tows. The relationship between zooplankton dry weight and backscatter was significant (p < 0.001) and explained 76% of the variability in the dry weight data. We applied this regression to hydroacoustic data collected monthly in 2003 and 2004 at two shoals in the Apostle Island Region of Lake Superior. After applying the regression model to convert hydroacoustic backscatter to zooplankton dry weight biomass, we used geo-statistics to analyze the mean and variance, and ordinary kriging to create spatial zooplankton distribution maps. The mean zooplankton dry weight biomass estimates from plankton net tows and hydroa-coustics were not significantly different (p = 0.19) but the hydroacoustic data had a significantly lower coefficient of variation (p < 0.001). The maps of zooplankton distribution illustrated spatial trends in zooplankton dry weight biomass that were not discernable from the overall means.

Mud-dominated sediments in Presque Isle Bay are contaminated with metals and hydrocarbons derived from developed watershed and atmospheric sources. Prior to this study, the quantities, rates, and spatial distribution of long-term sedimentation and erosion in the bay were largely unknown. As a result, the fate of contaminated bay-floor sediments and possible rates of natural recovery for this Area of Concern (AOC) could not be determined. To provide baseline data useful to state and federal agencies monitoring recovery of the bay, this paper identifies: (1) the quantities, rates and patterns of 20^th Century sedimentation and erosion, (2) the major sediment inputs and outputs for the bay, and (3) the implications of the sedimentary regime on possible future rates of bay recovery. Bathymetric and sedimentological data show that 20^th Century net accumulation totaled approximately 3.94 ×10⁶ m³ which is equivalent to a dry sediment loading of 5.92 ×10⁹ kg (5.92 ×10⁶ t), or 6.29 kg/m²/yr (1.28 lb/ft²/yr) when averaged over the accretional 70% of the bay. This external loading represents approximately 50% of total accretion because externally derived sediments are augmented with resuspended sediments from shallow-water parts of the bay. The principal sediment inputs were littoral drift from ephemeral and permanent inlets (∼42%), artificial infilling along the shoreline (∼28%), streams (∼16%), bank/bluff erosion (∼12%), and biological production (∼2%). Dredging was the principal output. Based on long-term average sedimentation rates and patterns, recovery of the AOC through natural sediment capping will take at least several decades if source contaminants are removed.

The bioaccumulation of polychlorinated biphenyls (PCBs) is examined in the food web of Lake Winnipeg using measured contaminant concentrations, stable isotopes of nitrogen, and a food web model. Measured concentrations of the sum of 103 PCB congeners are higher in south basin water, sediment, and biota compared with the north. The trophic positions of the top predators as well as the extent of biomagnification of PCBs per unit trophic level do not differ significantly between the north and the south basins. We therefore conclude that the higher PCB concentrations in the south basin are due primarily to higher PCB loadings via riverine sources to the south rather than food web processes. In contrast, the data from the north basin suggest lower total loadings of which a higher fraction is from atmospheric deposition. We find that rainbow smelt (Osmerus mordax) are not associated with elevated exposure of contaminants to top predators of the north basin. This surprising result is attributed to their reduced fitness in this relatively shallow and warm system, which may prevent them from feeding at an elevated trophic level compared with other forage fish. Finally, we hypothesize that high nutrient associated DOC in water decreases PCB bioavailability to lower trophic level organisms and hence the entire food web.

Water mass movement within the Great Lakes may rapidly transport fish larvae from favorable nursery areas to less favorable habitats, thereby affecting recruitment success. During 2001 and 2002, we released satellite-tracked drifting buoys in eastern Lake Michigan to follow discrete water masses, and used ichthyoplankton nets to repeatedly sample larval fish within these water masses. Observed nearshore water currents were highly variable in both direction and velocity. Current velocities far exceeded potential larval fish swimming speeds, suggesting that currents can potentially rapidly advect fish larvae throughout the lake. Evidence suggests that while paired drifters released during 2002 were able to track relatively small alewife (Alosa pseudoharengus) and yellow perch (Perca flavescens) larvae within an alongshore coastal current, paired drifters released during 2001 failed to track larger alewife larvae when flow was more offshore and highly variable. These results are consistent with the decorrelation scales associated with alongshore and offshore transport.

The St. Clair River delta, the largest in the Laurentian Great Lakes system, is located in Lake St. Clair at the mouth of the St. Clair River. It straddles the border between the state of Michigan in the USA and the province of Ontario in Canada. The current study aims at characterizing and explaining the formation and evolution of the distributary channels in the St. Clair River delta and the delta itself. The delta is classified as a river dominated feature with classic “bird's foot” structure similar to the Mississippi delta model. The delta is composed of two surfaces. The older surface at the delta apex was deposited at a higher lake level some 3,500 to 5,000 years B.P. The current delta started to form following a drop in lake level some 3,500 years B.P. and continues to build to the present day. The delta consists of seven active deep channels averaging 11 m in depth entering a lake with a mean depth of 3 m with much shallower water in the delta front region. The channels are stable and are actively eroding into the sediments of the lake creating both sub-aqueous and sub-aerial levee deposits with crevasses. The inter-distributary bays are being filled with sandy deposits created by wave energy and by crevasse deposits. At the erosional front of each distributary a narrow erosional notch “leading channel” is being formed which appears to control the direction of lakeward erosion of each deep distributary channel. The emplacement of the delta body in the shallow receiving water body has been termed “burrowing” delta formation and is the mechanism controlling the formation of this sedimentary feature.

The Mid-Lake Reef Complex (MLRC), a large area of deep (> 40 m) reefs, was a major site where indigenous lake trout (Salvelinus namaycush) in Lake Michigan aggregated during spawning. As part of an effort to restore Lake Michigan's lake trout, which were extirpated in the 1950s, yearling lake trout have been released over the MLRC since the mid-1980s and fall gill net censuses began to show large numbers of lake trout in spawning condition beginning about 1999. We report the first evidence of viable egg deposition and successful lake trout fry production at these deep reefs. Because the area's existing bathymetry and habitat were too poorly known for a priori selection of sampling sites, we used hydroacoustics to locate concentrations of large fish in the fall; fish were congregating around slopes and ridges. Subsequent observations via unmanned submersible confirmed the large fish to be lake trout. Our technological objectives were driven by biological objectives of locating where lake trout spawn, where lake trout fry were produced, and what fishes ate lake trout eggs and fry. The unmanned submersibles were equipped with a suction sampler and electroshocker to sample eggs deposited on the reef, draw out and occasionally catch emergent fry, and collect egg predators (slimy sculpin Cottus cognatus). We observed slimy sculpin to eat unusually high numbers of lake trout eggs. Our qualitative approaches are a first step toward quantitative assessments of the importance of lake trout spawning on the MLRC.

Gyres and seiches are two prominent features of lakes. Gyres largely transport sediments, nutrients, and algae in the horizontal direction. Seiches, on the other hand, can contribute to the vertical mixing in lakes. Theoretical analysis, statistical methods, and numerical models are used to investigate gyres and seiches in Lake Okeechobee, the largest subtropical/tropical lake in North America. The lake has a 1,730-km² surface water area, a typical length of more than 50 km, and a mean depth of 3.2 m. Both the Empirical Orthogonal Function (EOF) method and the numerical model results indicated that lake circulation is typically dominated by a two-gyre pattern, especially in the winter. The northwest wind or southeast wind leads to a cyclone (a counterclockwise rotation gyre) in the southwest and an anticyclone (a clockwise rotation gyre) in the northeast. Because the mean velocity field in the lake is very weak, the first two EOF modes play an important role in lake transport. The mechanism of gyre formation in the lake is clearly explained in a theoretical analysis. Power spectra analysis on measured and modeled water elevations at four stations revealed that Lake Okeechobee has a seiche signal of 5 hours or so. The seiche range is typically around 10 cm. Results from the theoretical analysis, power spectral analysis, and numerical modeling all agree with each other very well. The findings in this study should be useful to understand the lake processes, to guide field data collection programs, and to assist decision making on lake management.

Sea lamprey (Petromyzon marinus) are a nuisance aquatic species in the Great Lakes and Lake Champlain that have devastated native fish populations and hampered the restoration of sport fisheries. This study examined inter-basin movement of sea lamprey in Lake Champlain to identify tributaries that contribute parasitic-phase sea lamprey and provide information for prioritizing those tributaries for sea lamprey control. A total of 4,125 recently metamorphosed sea lamprey was captured in tributaries to Lake Champlain and marked using coded wire tags between the fall of 2001 and winter 2003. These sea lamprey migrated to the lake to prey on salmonids and other fishes and returned to tributaries to spawn about 12–18 months after migration. We recaptured 6 tagged sea lamprey from the lake from spring 2002 through winter 2004, and 35 from tributaries in spring 2003 and 2004. We noted no apparent trends in movement among basins. Sea lamprey were collected at distances up to 64 km from their natal tributaries. Tributary contributions of parasites were significantly different from expectations in the 2002 parasitic-phase cohort (χ² = 9.668, p < 0.011, 3 df), suggesting differential survival rates among out-migrating transformers from different tributaries. Estimates of the lake-wide out-migrating transformer population for the 2002 and 2003 parasitic-phase cohorts were 269,139 ± 55,610 (SD) and 111,807 ± 23,511 (SD). Results from this study suggest that sea lamprey movement is not inhibited by causeways dividing sub-basins, but movement among sub-basins is somewhat constrained. This indicates that management efforts to control sea lamprey should continue to treat the lake as a single system.

Resurgences in toxic cyanobacterial blooms, especially Microcystis, have been observed in the lower Great Lakes over the last decade in areas where total phosphorus (TP) levels are below historically high levels. Compliance with regulatory standards for TP in municipal wastewater effluent has likely depended on increased use of iron chloride as a flocculating agent for phosphorus removal in some wastewater treatment plants (WWTP). Since some ecological research suggests that increased supply of biologically available forms of iron may stimulate cyanobacterial growth, TP and iron loadings from WWTPs in the Hamilton Harbour and Toronto areas were examined. TP loadings from the Hamilton and Toronto WWTPs over the last 10–15 years have, in general, not increased. Effluent sampling frequency for iron is much lower than for a regulated parameter such as TP, consequently there is greater uncertainty in loading estimates. Moreover, iron loading estimates are available for only one plant from 1996 and three plants from 1999 so long-term loading trends cannot be established. Iron loading from the Humber WWTP in Toronto increased 5-fold between 2001–2004 while iron loading from the Woodward WWTP in Hamilton is quite high for its size. Iron consumption (i.e., dosing) data are available for three plants from the early 1990s. There was a clear increase in iron consumption by the Burlington-Skyway WWTP beginning in 1997 which was accompanied by a clear decrease in TP loading. However, there were no clear long-term trends in iron consumption in the Toronto plants even when normalized for flow. Increased monitoring and reporting of iron in WWTP effluent by all WWTP operators is recommended along with a review of methods for iron retention.

Burbot (Lota lota), are native Lake Superior piscivores that share similar habitat and food resources with other predators including lean and siscowet lake trout (Salvelinus namaycush). To better understand their ecological role in the fish community, we combined fisheries assessment information from 1970 to 2002 with a bioenergetics model to estimate their predatory impact in the Apostle Islands region of Lake Superior. Relative abundance declined in the Apostle Islands region from 3.41 fish/1,000 m in 1978 to 0.27 fish/1,000 m in 1998. Fishing mortality was minimal based on creel estimates and observed bycatch in the commercial fishery. Burbot < 400 mm consumed a higher fraction of small prey items such as Mysis relicta, fish eggs and sculpins (Cottidae) while larger burbot were almost exclusively piscivorous. Overall diet composition (by mass) was represented primarily by Coregonus spp. (64%) and rainbow smelt (Osmerus mordax) (17%). We estimated the burbot population size in the Apostle Islands between 1970 and 2000 at 56,541 to 1,585,035 age 1 fish. This population consumed between 56 and 1,584 metric tonnes (0.13 to 3.54 kg/ha) of prey. Increases in both lean and siscowet lake trout abundance have likely contributed to the decline in burbot abundance through predation and/or competition for food resources. Given the current burbot population in the Apostle Islands, this species is unlikely to control production of prey fish or invertebrate taxa.

This study evaluated yellow perch (Perca flavescens) diet in southern Lake Michigan to determine whether prey consumed fluctuated with abundance of zooplankton, benthic invertebrates, and fish species during the period 1984 to 2002. Some change in benthic community abundance was evident from samples collected in the region during the period, including the naturalization of the round goby and the zebra mussel between 1993 and 2002. In addition, changes in fish abundance were evident from 1984 to 2002, when spottail shiner (Notropis hudsonius) and alewife (Alosa pseudoharengus) increased, while yellow perch, and rainbow smelt (Osmerus mordax) declined. Non-indigenous species eaten by yellow perch in 2002 included spiny water fleas (Bythotrephes longimanus), round gobies, and alewives with the latter two species dominating the diet by volume. Yellow perch did exhibit prey preferences, although they consumed a variety of different organisms over the period of study. This euryphagous characteristic of yellow perch is expected to promote its persistence in southern Lake Michigan, despite a changing prey base.

Planktonic community metabolism (photosynthesis and respiration) was assessed in the oligotrophic east basin of Lake Erie, from November 2001 to October 2002 using O₂ and ¹⁴C methods. Areal gross production (AGP; mmol O₂ m⁻² d⁻¹) exceeded areal respiration (AR; mmol O₂ m⁻² d⁻¹) in the surface mixed layer for 69% of the observations during the study period. The median AGP:AR for the entire study period was 1.32. A significant positive relationship between AGP and AR existed, but AGP explained only 25% of the variation in AR. AGP:AR varied seasonally, being below 1.0 in the fall/winter of 2001 and in early spring 2002. High (>> 1.0) AGP:AR was observed in late spring (May) and AGP:AR remained > 1.0 for most of the summer stratified period (July–September). AGP:AR was > 1.0 in the fall of 2002, but the magnitude was less than observed during spring. The results supported traditional concepts of the seasonal production and consumption cycles in planktonic communities of large oligotrophic lakes. Parallel incubations of ¹⁴C uptake and gross O₂ production determined with the light and dark bottle method revealed a mean experimental gross photosynthetic quotient (PQ_G) of 1.29 ± 0.48, indicating that the ¹⁴C method used here had a variable but, on average, close relationship to gross production as it is commonly measured.

Pheromonally-mediated trapping is currently being developed for use in sea lamprey control in the Laurentian Great Lakes. To identify and test lamprey pheromones a practical procedure was needed to isolate relatively large quantities of pheromone from lamprey holding water. The present study developed such a technique. It employs Amberlite XAD7HP, an adsorbent resin which we found can extract over 80% of the sea lamprey migratory pheromone from larval holding water at low cost and with relative ease. This technique allowed us to collect tens of milligrams of all three components of the sea lamprey migratory pheromone, eventually permitting both identification and successful field testing. This technique might also be used to collect pheromones released by other species of fish.

The land use of the Great Lakes region has changed significantly during historical times, and continues to change. As a preliminary step in investigating the overall effect that this might have on climate, attention is focused here on one forcing factor and one effect—land surface roughness length and lake effect precipitation, respectively—that are anticipated to be particularly sensitive pieces of the land use-climate interaction. On both a monthly basis and in an individual case of lake effect precipitation, a reduction of land surface roughness reduces the total amount of lake effect precipitation. It also reduces the degree to which the precipitation is focused on the area closest to the lakeshore. The largest reductions occur immediately adjacent to the lakeshore in an area smaller than the overall lake effect zone. In the individual lake effect event that is investigated here, precipitation increases in some places farther inland when surface roughness is reduced. Because this increase in precipitation farther inland appears to be associated with significant topography, this result is most valid for lake effect zones where there is a high topographic relief, such as near southeastern Lake Erie (the main focus of this study), and to the south and east of Lake Ontario. This displacement in location of precipitation is particularly crucial where the boundary of the drainage basin is near the shoreline, and can indicate a flux of moisture out of the Great Lakes drainage basin and into another basin.

A recent empirical model of glacial-isostatic uplift showed that the Huron and Michigan lake level fell tens of meters below the lowest possible outlet about 7,900 ¹⁴C years BP when the upper Great Lakes became dependent for water supply on precipitation alone, as at present. The upper Great Lakes thus appear to have been impacted by severe dry climate that may have also affected the lower Great Lakes. While continuing paleoclimate studies are corroborating and quantifying this impacting climate and other evidence of terminal lakes, the Great Lakes Environmental Research Laboratory applied their Advanced Hydrologic Prediction System, modified to use dynamic lake areas, to explore the deviations from present temperatures and precipitation that would force the Great Lakes to become terminal (closed), i.e., for water levels to fall below outlet sills. We modeled the present lakes with pre-development natural outlet and water flow conditions, but considered the upper and lower Great Lakes separately with no river connection, as in the early Holocene basin configuration. By using systematic shifts in precipitation, temperature, and humidity relative to the present base climate, we identified candidate climates that result in terminal lakes. The lakes would close in the order: Erie, Superior, Michigan-Huron, and Ontario for increasingly drier and warmer climates. For a temperature rise of T°C and a precipitation drop of P% relative to the present base climate, conditions for complete lake closure range from 4.7T P > 51 for Erie to 3.5T P > 71 for Ontario.