Three sediment cores from Lake Ontario, Lake Erie and Indiana Harbor Ship Canal were collected, segmented and analyzed for Aroclor and non-Aroclor polychlorinated biphenyl congeners (PCBs). PCBs associated with the commercially produced Aroclor mixtures 1248 and 1254 dominate the sediment signal and the sum of all congeners (ΣPCB) peaks in concentration and accumulation around 1970 in the Great Lakes. This trend is very similar to Aroclor production history. In the Indiana Harbor Ship Canal, PCBs appear around 1935 and remain at very high levels between 1940 and 1980, probably reflecting the history of use at the nearby steel mill. In contrast, the non-Aroclor PCBs in the Lake Ontario and IHSC sediment cores, including PCB11 and heavily chlorinated congeners PCB206, 207, 208 and 209 reach a peak in the 1950s, decline and peak again in the 1970s or in the early 1980s. All five congeners have been previously measured in commercial paint pigment. PCB11 was found to peak about 5 years later than ΣPCBs, and is probably associated with the production or use history of diarylide yellow pigments. The temporal distribution profiles of these non-Aroclor PCBs are well correlated with the production history of paint pigments and dyes. Although it is well known that the production of Aroclor PCBs is preserved in Great Lakes sediments, this study is the first to show that production of non-Aroclors are also preserved in the sediments as a record of long term trends in environmental exposure.
Oneida Lake, New York is a eutrophic freshwater lake known for its abundant manganese nodules and a dynamic manganese cycle. Temporal and spatial distribution of soluble and particulate manganese in the water column of the lake were analyzed over a 3-year period and correlated with other variables such as oxygen, pH, and temperature. Only data from 1988 are shown. Manganese is removed from the water column in the spring via conversion to particulate form and deposited in the bottom sediments. This removal is due to biological factors, as the lake Eh/pH conditions alone can not account for the oxidation of the soluble manganese Mn(II). During the summer months the manganese from microbial reduction moves from the sediments to the water column. In periods of stratification the soluble Mn(II) builds up to concentrations of 20 micromoles or more in the bottom waters. When mixing occurs, the soluble Mn(II) is rapidly removed via oxidation. This cycle occurs more than once during the summer, with each manganese atom probably being used several times for the oxidation of organic carbon. At the end of the fall, whole lake concentrations of manganese stabilize, and remain at about 1 micromole until the following summer, when the cycle begins again. Inputs and outflows from the lake indicate that the active Mn cycle is primarily internal, with a small accumulation each year into ferromanganese nodules located in the oxic zones of the lake.
Collection and analysis of an extensive set of seismic-reflection profiles and cores from southern Lake Michigan have provided new data that document the history of the lake basin for the past 12,000 years. Analyses of the seismic data, together with radiocarbon dating, magnetic, sedimentologic, isotopic, and paleontologic studies of core samples, have allowed us to reconstruct lake-level changes during this recent part of the lake's history.The post-glacial history of lake-level changes in the Lake Michigan basin begins about 11.2 ka with the fall from the high Calumet level, caused by the retreat of the Two Rivers glacier, which had blocked the northern outlet of the lake. This lake-level fall was temporarily reversed by a major influx of water from glacial Lake Agassiz (about 10.6 ka), during which deposition of the distinctive gray Wilmette Bed of the Lake Michigan Formation interrupted deposition of red glaciolacustrine sediment. Lake level then continued to fall, culminating in the opening of the North Bay outlet at about 10.3 ka. During the resulting Chippewa low phase, lake level was about 80 m lower than it is today in the southern basin of Lake Michigan.The rise of the early Holocene lake level, controlled primarily by isostatic rebound of the North Bay outlet, resulted in a prominent, planar, transgressive unconformity that eroded most of the shoreline features below present lake level. Superimposed on this overall rise in lake level, a second influx of water from Lake Agassiz temporarily raised lake levels an unknown amount about 9.1 ka. At about 7 ka, lake level may have fallen below the level of the outlet because of sharply drier climate. Sometime between 6 and 5 ka, the character of the lake changed dramatically, probably due mostly to climatic causes, becoming highly undersaturated with respect to calcium carbonate and returning primary control of lake level to the isostatically rising North Bay outlet. Post-Nipissing (about 5 ka) lake level has fallen about 6 m due to erosion of the Port Huron outlet, a trend around which occurred relatively small (± ∼2 m), short-term fluctuations controlled mainly by climatic changes. These cyclic fluctuations are reflected in the sed-imentological and sediment-magnetic properties of the sediments.
Composite, nearshore, whole fish samples of selected species, collected in fall 1983 from 13 Lake Michigan tributaries and Grand Traverse Bay, were analyzed for a wide range of pesticides and priority pollutants using gas chromatography-mass spectrometry. This study was carried out to identify existing source areas for known and previously unrecognized toxic substances. Our strategy was to analyze those resident fish with the highest likely levels of contaminants. All fish analyzed (eight species from southern Michigan to the upper peninsula) exceeded the 2 mg/kg FDA action levels for PCBs, while 50% of the samples exceeded the DDTr IJC objective of 1 mg/kg. St. Joseph River common carp (Cyprinus carpio) carried the heaviest contaminant burden of all fish examined for PCBs (27.6 mg/kg), DDTr (10.2 kg/mg), and toxaphene (3.3 mg/kg); chlordane levels (0.85 mg/kg) were second highest to those in Kalamazoo River common carp (0.87 mg/kg). Concentrations of PCBs, toxaphene, DDT, DDE, and other pesticides were higher in bottom-feeding fish, such as common carp, than in top predators, e.g., northern pike (Esox lucius). Bottom feeders are relatively fatty fish, and live and feed near contaminated sediments, which increases their potential to bioaccumulate fat-soluble contaminants. Pesticides were also present in elevated concentrations in fish from sites with higher industrial and agricultural development.
A multi-segment water quality model based on the QWASI (Quantitative Water Air Sediment Interaction) fugacity/aquivalence approach is developed for the Bay of Quinte. The model of the bay consists of seven water column compartments, five sediment compartments, the atmosphere, and two compartments in Lake Ontario. Chemicals enter the bay from tributaries and runoff, industrial and municipal point sources, the atmosphere, and Lake Ontario. Over 50 rate expressions are used to describe the transport and transformation of chemicals through and between segments under steady-state conditions. A food chain model applicable to organic chemicals and consisting of six trophic levels is also described. The bay model was calibrated with concentration data for arsenic and pentachlorophenol and tested with copper and zinc data sets. Since varying amounts of information were available for the remaining chemicals, the bay and food chain models were used to “back-calculate” the loadings of 10 chemicals based on known in-bay water, sediment, or fish concentrations. For all chemicals (four inorganics and 13 organics), the model gives order-of-magnitude estimates of loadings, concentrations in water, sediment and biota, amounts, rates of movement, and estimates of persistence from which appropriate remedial measures are discussed.
The predaceous cladoceran, Bythotrephes cederstroemi (Schoedler 1877), was first noted in the Great Lakes in late 1984, and had spread to Lakes Huron, Erie, Ontario, and Michigan by late summer of 1986. Zooplankton samples collected from Batchawana Bay, Lake Superior on 15 September 1988 contained large numbers of B. cederstroemi (122.8 individuals/m2). Geographic variation in Great Lakes B. cederstroemi was observed when this sample was compared to samples from Lake Erie. There was a greater percentage of males present in the Lake Superior sample as well as a greater number of females carrying sexual resting eggs. B. cederstroemi from Lake Superior weighed more than those from Lake Erie. These differences are probably the result of physiological responses to different environmental conditions, and not genetic differentiation of B. cederstroemi populations among the Great Lakes.
The St. Clair River is the outlet channel for Lakes Michigan and Huron to the lower Great Lakes. The river's hydraulic characteristics naturally regulate Lakes Michigan-Huron's water levels by controlling the amount of water that flows out of the lakes. Accurate determinations of the outflows are necessary, in conjunction with the St. Marys River flows and lake level data, to determine the water supplies to Lakes Michigan-Huron. The hydraulic regime of the river has been changed many times since the mid-1800s primarily due to dredging for deeper draft navigation and sand and gravel mining. In addition there have also been minor effects due to shipwrecks at the head of the river. For water resource studies of the Great Lakes system it would be highly desirable to determine Lakes Michigan-Huron's water supplies for the period 1860-1900. Additional water supply values would extend the available period of study by 40 years or approximately 45 percent. This period is also particularly important as it contains extreme high water supplies that led to record Lakes Michigan-Huron water levels. This study comprised an analysis of the existing discharge and dredging data for the 1860-1902 period to determine if discharge equations could be developed with sufficient accuracy to compute monthly St. Clair River flows. The analysis demonstrated that it is not possible to quantify St. Clair River flows prior to 1900, thus limiting the period available for determining Lakes Michigan-Huron water supplies to this century. The study also confirmed the previously determined 0.18 m lowering of Lakes Michigan-Huron between 1900 and present due to channel dredging in the upper river.
An examination of g ridded data developed from quality controlled land station temperature data for the 1901-1987 period for the Great Lakes basin reveals temperature trends not previously reported. On an annual basis, from early 1900 through the mid 1950s, a 5-year moving average shows that overall basin temperatures increased. A cooler regime prevailed for the remainder of the period of record. Seasonal 5-year moving averages show that spring temperatures increased throughout the period. Summer temperatures were highly variable, but with an early upward trend through the mid 1940s and a recent downward trend for the remainder of the period. Fall temperatures slowly warmed from the early 1900s until the mid 1960s, but have recently shown a cooling trend. Winter temperatures show an upward trend with intermittent wide swings from the early 1900s through the mid 1950s after which a lower temperature regime has prevailed. Temperature differences between a warm and a cool year and the long term mean show markedly varying patterns.
To set the scene for the papers which follow, this review presents the principal findings of hydrodynamic and related chemical and sedimentological studies in Lake Erie beginning with the 1928/29 cooperative surveys of the central and eastern basins (Fish 1929, 1960) and ending with the (also cooperative) “Project Hypo” (Burns and Ross 1972) and the papers assembled in the 1976 special number, “Lake Erie in the Early Seventies” (J. Fish. Res. Board Canada, Vol. 33). Attention is also drawn to the far-reaching and interweaving influence of the physical phenomena on biological and sedimentological processes operating in the lake. Where needed to place the half-century's findings in context or to lead into the papers which follow in this issue, some references to work published before 1928 or after 1977 are also included. After an introduction, the review examines: surface surges and seiches; thermal stratification and its coupling with the distribution of oxygen and phosphorus in bottom water; long-term and short-term circulation patterns; sedimentation and sediment transport.
These data, collected during 1942–1944 by Dr. David C. Chandler, describe the density, biomass, and growth of a now extinct population of burrowing mayfly nymphs (primarily Hexagenia limbata) that lived in the sediments of western Lake Erie near South Bass Island. The growth dynamics of this population have not previously been documented. Female nymphs grew faster than males and were about 4 mm longer than males at emergence each year. Significantly fewer nymphs were collected in 1943 than in 1942 or 1944. Before they were extinguished by low dissolved oxygen in 1953, mayfly nymphs were abundant (about 350 weighing 10 wet g m−2) near this island and throughout most of western Lake Erie. The western basin once supported a biomass of 9.6 t · km−2 or at least 17,600 metric tonnes of mayfly nymphs. If burrowing mayflies recolonize the sediments of western Lake Erie, these data could be used to assess the extent of their recovery.
Rediscovered material from an extensive sediment survey in the Straits of Mackinac and adjacent parts of Lakes Michigan and Huron provides one of the earliest quantitative distributional records of mayflies in the main basins of the Great Lakes. About 300 sites were sampled once each during two summers for macrobenthos, as well as for sediment textural characters. Mayflies from the survey were either Ephemera (probably E. simulans) or Hexagenia (at least partly H. limbata), and occurred primarily at depths less than 20 but up to 66 m. Both taxa were restricted to sand, silty sand, or sandy silt bottoms, and occurred more frequently in the Lake Michigan and southern Lake Huron parts of the study area than in the northern Lake Huron part of the Straits. Limited data on body length frequencies indicate a life cycle longer than 2 years for at least part of the Hexagenia population. Areas of low densities were related to lower temperatures and indicators of low pelagic primary production. Absence of ephemerids from several subsequent surveys of parts of the study area raises the possibility that mayflies have disappeared from the Straits since 1956.
Relationships between large-scale environmental factors and the incidence of type E avian botulism outbreaks in Lake Michigan were examined from 1963 to 2008. Avian botulism outbreaks most frequently occurred in years with low mean annual water levels, and lake levels were significantly lower in outbreak years than in non-outbreak years. Mean surface water temperatures in northern Lake Michigan during the period when type E outbreaks tend to occur (July through September) were significantly higher in outbreak years than in non-outbreak years. Trends in fish populations did not strongly correlate with botulism outbreaks, although botulism outbreaks in the 1960s coincided with high alewife abundance, and recent botulism outbreaks coincided with rapidly increasing round goby abundance. Botulism outbreaks occurred cyclically, and the frequency of outbreaks did not increase over the period of record. Climate change scenarios for the Great Lakes predict lower water levels and warmer water temperatures. As a consequence, the frequency and magnitude of type E botulism outbreaks in the Great Lakes may increase.Research Highlights►Type E avian botulism outbreaks in Lake Michigan were examined from 1963 to 2008. ►Avian botulism outbreaks occurred most frequently in years with low water levels. ►Surface water temperature in Lake Michigan was higher in outbreak years. ►The frequency of botulism outbreaks did not increase over the period of record.
Recruitment of young-of-the-year (YOY)rainbow smelt (Osmerus mordax) per parent was nqt dependent upon water temperature (rate of warming and average temperature) during spawning, Glugea infestation, or parental stock-size. Only in the eastern basin of Lake Erie was recruitment to age 1 + years-old positively correlated with the YOY index in the preceding summer. Recruitment of age 1 + year-old smelt was not related to parental stock-size. Recruitment (age 1 + years-old) per spawner was inversely related to the abundance of age 1 year-old smelt present during the year of hatch in both the eastern and central basins of Lake Erie. This could be taken as evidence to support the hypothesis that the alternating dominance of year-classes (1963–74) was caused by cannibalism of YOY by age 1 year-old smelt. After 1974, alternating year-class-dominance was not evident; we considered three possible explanations for the absence of strong year-classes. First, the size at age declined in both basins during the study so that more age classes could have fed optimally on young-of-year (YOY), resulting in a poorer survival of YOY smelt. Second, the harvest of smelt increased greatly after 1974, particularly for yearling smelt, so that year-class strength measured by the abundance of yearlings was reduced by fishing. Third, predation on smelt by stocked salmonids probably increased after 1974 as the stocking rates increased. Adult total mortality rates (ages 2 and 3 years-old) were positively related to stock density providing some evidence for density dependent mortality of smelt. There was no evidence that the variations in adult mortality were associated with the incidence of Glugea.
Lakewide surface heat flux and components are computed for Lake Erie from detailed measurements conducted during the stratified season from 1967 to 1982. These results are formed into monthly means and compared to other long-term estimates for the period 1952 to 1968, thus providing a review of the climatological estimate of the range and variability of these parameters. Details of the lakewide monthly means of pertinent hydrometeorological and limnological observations are briefly outlined. Lake Erie heat storage is derived from intensive lakewide temperature surveys from 1967 to 1982. A verification of the computed surface heat flux estimates is accomplished by comparison between measured heat storage and computed heat storage which is based on the summation of daily heat flux calculations. A positive residual of 2,557 cal/cm2 is estimated for the long-term heat budget. Comparisons of the measured and computed heat storage for individual years over the 15-year period shows good agreement for most years.
A one-dimensional thermocline model is presented which is used to estimate daily vertical temperature distributions, thermal layer thicknesses, thermal interface depths, and vertical diffusion. Comparison of computed temperature with observed vertical profile data (1967–1982) for the three basins shows good agreement. The median relative error between observed and computed mean temperature for the central basin hypolimnion is approximately 5 percent. These computed temperature profiles are used to derive statistical distributions which can be useful for designing sampling frequencies in the Great Lakes Surveillance Program. In particular, analyses are presented on the frequency of occurrence of certain thermal structures particularly favorable for the development of anoxia. It is found that, for a hypolimnion thickness of less than 4-m depth and a turbulent diffusivity less than 1 cm2/s, there is a high likelihood of anoxia development in the central basin of Lake Erie.
Surficial sediments were collected in 1968 and core sediments in 1976 from Lake Ontario. These were analysed for organochlorine insecticides and PCB. Residues of organochlorine compounds were higher in the three Ontario depositional basins than in sediment in the non-depositional zones. PCB was present at the highest concentration, with mean levels of 57 ppb for the whole lake. Residues of PCB in the three basins exhibited only minimal differences.ΣDDT was the second most frequently found contaminant. Both parent DDT and its two metabolites were present in sediment, giving a mean residue of 42.8 ppb for the whole lake. Differences in residues for the three basins were again minimal. DDT was present in sediments estimated to be deposited between 1958 and 1976. HEOD was present in only 40% of sediments and the mean residue for the whole lake was 0.6 ppb. The Niagara basin contained mean residues (1.4 ppm) of HEOD much above the other basins.Chlordane was virtually absent from the lake in the sediments collected in 1968 but appeared in cores between 1964-1976. Endosulfan appeared in lake sediments as a spill of deck cargo at a location in the Niagara basin. This spill dispersed with the current down the south shore to deposit in the Rochester basin.
Residues of DDT, PCBs, and dieldrin in Lake Michigan bloaters (Coregonus hoyi) decreased between 1969 and 1986. All values for contaminant residues were measured and reported as μg/g of whole fish, wet weight; the error limits represented the 95% confidence interval half widths. Total DDT content declined 87%, from 9.9 in 1970, when its use was banned, to 1.3 in 1974, and then averaged 0.88 in 1978–1986 with a low of 0.67 in 1986. Between 1976–1986, p,p’DDD averaged 0.08 and p,p’DDT averaged 0.1; p,p’DDE reached a low of 0.48 in 1980, then averaged 0.53 in 1980–1986. Dieldrin averaged 0.25 until 1974, when it was banned, peaked at 0.55 in 1978, and decreased to 0.20 by 1986. PCBs, first analyzed in 1972, declined 64% (from 5.7 to 2.2) between 1972 and 1980 and were at 1.64 in 1986. The great decline in PCBs occurred following the 1976 ban. A 49% decrease in the lipid content of bloaters from 1980 to 1986 suggested a dietary change possibly due to changing food source makeup.
The Buffalo District, Corps of Engineers’ Lake Erie Wastewater Management Study and Heidelberg College's Water Quality Laboratory supported a tributary water quality monitoring program from 1974 to 1980 of the major United States tributaries to Lake Erie. This program was designed to measure nutrient loads by monitoring concentration changes occurring in association with increased streamflow. Soluble orthophosphate loads, chloride loads, and silica loads decreased from 1970 to 1980. Nitrogen species were highly variable and increased over the period. Total phosphorus loads to Lake Erie have decreased during the period as a result of phosphorus removal at wastewater treatment plants. The effect of the phosphorus reductions can be seen in the lake concentrations and were predicted by a three-basin phosphorus budget model developed in the early 1970s. The results show that phosphorus removal programs are having the predicted effect on Lake Erie water quality.
Five yearly estimators of total phosphorus (TP) concentrations in the central basin of Lake Erie from 1970 to 1986 were calculated and compared to evaluate their utility in the assessment of long term trends. Data were selected from stations within a defined polygon for those years in which at least four surveys were conducted, including spring and fall isothermal conditions and at least two surveys during stratification. Included were an arithmetic average of all survey means during each year, a time-weighted average of all survey means, a spring (isothermal) average, a summer (stratified) average, and an autumn (isothermal) average. The rate of decline in TP concentrations from the spring-only estimator (−0.276 μg TP L-1 yr-1) was similar to that of the time-weighted multiple survey estimator (−0.214 μg TP L-1 yr-1). Additional spring survey data from 1987 through 1991 confirmed that the negative trend continued. The stratified season and the autumn-only estimators exhibited no significant trends, but were biased low and high, respectively, relative to the spring and multiple survey estimators. Despite the reductions in TP concentrations since 1970, the annual normalized rate of oxygen depletion in the hypolimnion of the central basin showed only a slight downward trend (-0.03 mg L-1 mo-1). In 1988 and 1989, however, the depletion rates were lower than any reported during the previous 20 years.
Patterns of PCB congener bioaccumulation were examined in archived herring gull (Larus argentatus) eggs collected from Big Sister Island in Green Bay, Lake Michigan, and Scotch Bonnet Island in Lake Ontario from 1971 to 1982 as part of the Canadian Wildlife Service's Great Lakes Herring Gull Monitoring Program. Concentrations of 97 PCB congeners were measured. From 1971 to 1982, ecological half-lives of most congeners, particularly the tri- through hexachlorobiphenyls, were greater in eggs from Green Bay than Lake Ontario. Comparing sum PCB levels in eggs collected in 1971 and 1982, concentrations declined 80% at Scotch Bonnet Island and 74% at Big Sister Island. PCB congener patterns were different in eggs from the two colonies. Principal components analysis showed that inter-site differences in congener patterns became more apparent after 1976. This indicated that regional PCB sources were the most influential in determining patterns of biologically-available PCBs during the 1971 to 1982 period in these two lakes, via recycling of historical PCBs from sediments or gradually decreasing loading. Trend analysis of selected congeners specific to Aroclors 1242, 1254, and 1260 revealed that the rapid decline of less chlorinated congeners, observed from 1971 to 1976 in Lake Ontario, was explained by a decrease in loading of Aroclor 1242 to the lake. At both colonies, ecological half life of the congeners was significantly (p < 0.001) correlated with log Kow and with –log HLC. Changes in PCB composition, after 1976 in Lake Ontario and from 1971 to 1982 in Green Bay, could be explained by differences in the physical behavior of individual congeners affecting removal by volatilization and sedimentation.
Larval fish were sampled in Long Point Bay each summer in 1971–78 by towing ½– and 1-m diameter plankton nets at about 1 m/s. We report on the analysis of 2,576 tows, giving detailed consideration to sampling problems and statistical methods. The most common larvae were rainbow smelt (Osmerus mordax), caught in about 50% of the tows, and yellow perch (perca flavescens), clupeids, and cyprinids, caught in about 10 to 20% of the tows. Most of the data contained too high proportions of zero catches to be transformed to approximately normal distribution, and were therefore analyzed using the log-linear model rather than by analysis of variance. Except for cyprinids, larval fish seemed to be scarce in surface waters during the day. Small (4 to 10 mm long) smelt and perch larvae were caught more frequently in 526– than in 782–μm mesh nets. Very few perch and cyprinid larvae longer than 10 mm were ever caught. This, and a more rapid decline of daytime than nighttime catches of larval smelt through the mid-summer, suggests that the nets were avoided by at least the larger larvae. If avoidance of sampling gear is substantial, larval fish abundances may be overestimated in years of slow growth when the larvae remain catchable longer. We saw no evidence that the recent industrial development at Nanticoke, Ontario, has affected larval fish abundances so far. As the annual mean densities of larval fish varied 15-fold or more, it would require several years of observation to detect any but drastic future change from the current levels of abundance.
Surficial sediment samples (214) were collected in 1972-73 on a systematic grid throughout the Bay of Quinte, Lake Ontario. PCB analysis was carried out on all samples and the data were collated for depositional basins and non-depositional zones. The mean residue for the whole bay was 48 ± 43 ng/g. Mean residues in the basins and non-depositional zones were not significantly different, being 50 and 47 ng/g respectively. The Trenton basin had the highest mean residue of 74 ng/g and Adolphous basin had the lowest at 33 ng/g. Variations in concentrations were found within basins and within the non-depositional zone. The over-all distribution was patchy but high concentrations off the Trent River and east of the Moira River indicated these two rivers as the major sources of PCB to the bay.
Dissolved oxygen profiles taken in Lake Ontario in 1972 indicate the presence of a distinct and persistent metalimnetic oxygen minimum during the stratified season. Evidence indicates the phenomenon occurred in previous years as well. The depth and magnitude of the minimum were closely related to the thermocline depth and strength of stratification. Lowest minimum values in 1972 occurred in early to mid September and were 8.6 mg/l dissolved oxygen and 82% saturation. Offshore the minimum decreased from west to east across the lake and was lesser in magnitude nearshore and in the northeast. During the nonstratified period oxygen concentrations remained relatively constant with depth at approximately saturated values.
A high-volume sampler, located on the City of Chicago water intake crib (Dunne Crib) 3 km offshore in Lake Michigan, collected atmospheric particles over each 24-hour period from 12-30 September, 1973. Samples were analyzed via instrumental neutron activation analysis for 20 elements: Al, Br, Ca, Ce, Cl, Co, Cr, Cu, Fe, Hg, K, La, Mg, Mn, Na, Sb, Se, Th, V, Zn. Concentrations of many elements show order of magnitude variations with wind direction. For most elements highest levels occur when southern winds transport airborne material from local sources. Lowest levels, occurring for cross-lake wind trajectories, are still above concentrations observed in remote areas by others. Sources of atmospheric trace elements are identified on the basis of meteorological data, and comparison is made of their concentration in atmospheric paniculate matter with the composition of soils. The dependence of Br concentrations on wind direction indicates Chicago as a primary local source. In contrast, certain metals (Cr, Fe, Mn, and Zn) have peak atmospheric concentrations for winds from the Gary area. Residence times for background elements (Ce, K, Na, Th), calculated using a simple model in which concentrations decrease exponentially during transit of air parcels over the lake, are consistent with previous estimates of over-water residence times (5-20 hrs). Average concentrations in offshore air are generally compatible with particle-size corrected inventories of emissions from local sources.
Measurements relating to water transparency were made at roughly monthly intervals during the period May through November 1973 on Lake Superior. Spectrometer data augmented with extensive transmissometer and Secchi disc observations at a grid of stations over the lake were used to compute mean vertical extinction coefficients and photic depth for photosynthetically active radiation in the wavelength band 400-500 nm. These permitted description of the temporal and spatial variability of photic conditions throughout the major photosynthetic production period.The data show spatial and temporal variability in water transparency between areas and, for some zones, considerable departures from the lakewide mean are evident. The mean vertical extinction ranges from approximately 0.15 to 0.25 m−1 for the majority of zones and relatively good correspondence is observed with seasonal changes in chlorophyll a concentration. Considerably lower rates of light attenuation occur in the deeper midlake zones. Photic zone depth in these areas characteristically range from 20 to 30 meters. Reduced water transparency generally predominates along coastal areas in the proximity of Duluth, Thunder Bay, the southwest red clay portions of the lake, Apostle Is., Nipigon-Black Bay, Marathon, and Whitefish Bay. Photic zone depth in these areas are primarily less than 20 metres throughout the photosynthetic production period. In the vicinity of Duluth, mean photic depth values of less than 10 metres are observed with extreme minimums approaching 2 metres.
As a part of the cooperative scientific efforts between the US and the USSR on aquatic ecosystems, a comparison of analytical methodologies and laboratory capabilities was undertaken between 1973 and 1977. The samples exchanged between participating laboratories in the two countries included water samples from Lake Baikal, and standard reference samples from the U.S. Environmental Protection Agency. In general, the results from the analysis of exchanged materials show a remarkable comparability for parameters of environmental significance.
The horizontal and vertical distribution of Lake Superior phytoplankton was studied from May to November/December 1973 at approximately monthly intervals from 34 stations distributed over the entire lake. Taxonomic identification and enumeration were carried out using the inverted microscope H. Utermohl technique. The mean phytoplankton biomass showed a homogeneous distribution with very low biomass concentration across the lake. It did not show any inshore/offshore differentiation, although relatively high values were observed in the western end. On a lakewide basis the seasonal composition was dominated by the phytoflagellates and diatoms, with the bluegreens and greens contributing the least. Lake Superior is characterized by a large number of species. Approximately 285 taxa were identified, the majority of which have not been reported before in Lake Superior or the other Laurentian Great Lakes. Refs.
A variety of chemical and physical parameters were measured on fifteen cruises conducted on Lake Ontario from April to November, 1974. Analyses of chlorophyll a, including pheopigments, indicated that chlorophyll concentrations followed a biomodal seasonal pattern. Significant correlations were found between chlorophyll a and the other parameters measured. First order autoregressive equations were established for all measured parameters. Multiple regression analyses indicated that 74% of the spring, 49% of the summer and 78% of the fall variability in chlorophyll a concentration could be explained with the physico-chemical parameters measured in the present study.
A statistical model was developed to evaluate the temporal and spatial variability inherent in limnological sampling of large lakes. The model was applied to Secchi disc, transmissometer readings, temperature, integrated total particulate carbon, integrated total paniculate nitrogen, and integrated chlorophyll a. The data were obtained during the fifteen Lake Ontario surveillance cruises conducted between April - December, 1974. Based on the model, Lake Ontario was classified into a set of seven statistical zones for each parameter. An estimate of the seasonal cycle for each parameter was also obtained. It was found that the seasonal variability was more pronounced than the spatial variability for all parameters analyzed. However, a significant horizontal variability did exist, and this variability must be recognized in future modelling efforts.
Forty-five suspended solids collected during 1974 and 1976 from 12 streams on the Canadian side of the Great Lakes contained no detectable residues of triazines to a limit of 0.05 μg/g. Of ninety-two stream waters sampled in July, 1977, 77% contained atrazine, 52% desethylatrazine, and 28% simazine above a detectable level of 0.03 μg/L. Mean concentrations of these three triazines in the 92 streams were 1.6, 0.3, and 0.2 μg/L respectively. The highest mean triazine residue occurred in those stream waters entering Lake Erie (4.0 μg/L); there was little difference in the mean concentrations of waters entering Lakes Ontario (1.1 μg/L), St. Clair (1.3 (μg/L), and Huron (1.4 μg/L). While triazine herbicides were found entering the Great Lakes there was no threat to water quality as defined by the objectives of the International Joint Commission.
An intercomparison experiment was carried out at two sampling stations on Lake Erie on June 10, 1975 by three laboratories that are conducting environmental monitoring programs on the lake. The laboratories simultaneously collected water samples from three sampling depths at each station. The samples were divided into thirds, one third to be analyzed by each of the laboratories using their standard procedures. A number of physical, chemical, and biological properties were determined from each of the water samples. One of the principal results of the statistical analyses is that significant differences exist for most of the properties under study due to which laboratory analyzed the samples. However, the effect due to which laboratory obtained the samples is much less important. These results are in direct agreement with results reported in 1974 by Robertson, Elder, and Davies, IFYGL Chemical Intercomparisons, Proc. 17th Conf. Great Lakes Research.
During the last twenty years, intensive efforts have been aimed at reducing the eutrophication of Lake Erie. Although point source inputs have been greatly reduced, models indicated that non-point source inputs would need reduction as well, to meet phosphorus management goals for Lake Erie. Since non-point inputs enter the lake via tributary inflow, it is important to examine tributary records for evidence of trends in nutrient concentrations which might reflect success in reducing non-point inputs. Data series for the Maumee, Sandusky, and Cuyahoga Rivers, and for Honey Creek, spanning 9 to 16 years and up to 6,500 observations, were examined for trends in nutrients and suspended solids. Mean daily flows and two forms of weighted mean concentrations were compiled at monthly intervals, and studied using parametric and non-parametric techniques of trend detection. Total and soluble reactive phosphorus, suspended sediment, and nitrate-plus-nitrite were examined. Flow and suspended sediment generally showed statistically non-significant minor trends. Total and soluble phosphorus both showed downward trends, statistically significant for most data series, of 5 to 40 μg/L per year. Nitrate-plus-nitrite showed usually statistically significant increases of 10 to 140 μg/L per year, except for the Cuyahoga data, which showed a statistically significant downward trend of about 70 μg/L per year. These results are important both because they reflect important progress in the remediation of Lake Erie, and because they demonstrate the possibility of detecting trends in tributaries, given sufficient data and appropriate statistical approaches.
This article reports on organochlorine contaminant bioavailability and temporal trends in nearshore waters of the Great Lakes. Collections of young-of-the-year spottail shiners (Notropis hudsonius) were used as biomonitors of contaminants. Guidelines and criteria adopted for the protection of wildlife were used to assess the importance of contaminant accumulations in spottail shiners. Of the 10 compounds analyzed, only PCB and mirex concentrations in spottail shiners exceeded Aquatic Life Guidelines. Fifteen of 37 (41%) sites sampled in 1990 had shiners with PCB concentrations over the guideline. Mirex was only found in fish from the Niagara River, the Credit River in western Lake Ontario, and the St. Lawrence River at Cornwall. Total PCB and DDT residues in fish collected in 1990 were significantly (p<0.05) lower than residues in fish from the 1970s. While PCB and DDT residues continued to decline at most sites, downward trends at some sites have stabilized.
An intensive survey of water quality parameters was conducted on Lake Michigan during 1976 and 1977. A dynamic phytoplankton simulation model (MICH1) was developed to investigate the observed field data and to use in forecasting lake responses to various phosphorus loading scenarios. The 1977 data indicated that the southern basin of Lake Michigan lost up to 3 μg P/ L compared to concentrations observed in 1976. In an attempt to simulate this rapid depletion of phosphorus, MICH1 required an implicit representation of suspected effects of an extensive ice cover observed during the winter of 1976–77. This included increasing the net apparent settling rate eight fold during ice cover. A total phosphorus model (TPM) was used in conjunction with MICH1 ;for forecasting. These forecasts indicate a steady-state total phosphorus concentration of 7 μg P/ L, given a target load recommended by the 1978 Water Quality Agreement. The projected time to obtain 95% of steady-state response to a load change was 7–14 years.
Data collected on lake-wide cruises in 1976 were used to study seasonal and vertical variations in water temperature, transparency, chlorophyll a, and nutrients in Lake Michigan. Data were analyzed according to subsets corresponding to the northern and southern open lake. Comparisons (t-tests) of data from the open lake indicated that the average water temperature was cooler and average water transparency was greater in the northern lake than in the southern, but with the exception of total phosphorus, average nutrient concentrations did not differ between the northern and southern parts. It was found that physical-chemical characteristics of nearshore and Straits of Mackinac stations differed significantly from open lake stations. Seasonal phytoplankton dynamics in the open lake were related to seasonal and vertical changes in silica and nitrate nitrogen. The spring phytoplankton bloom occurred before the lake was strongly stratified thermally. After thermal stratification was well developed, epilimnetic concentrations of chlorophyll a decreased, probably due to some combination of nutrient limitation and zooplankton grazing, and maximum chlorophyll a concentrations were found below the thermocline. Epilimnetic silica concentrations decreased after thermal stratification and diatoms were replaced in the phytoplankton assemblage by green and blue-green algae in late summer. Total phosphorus averaged only 8 μg/L on a lake-wide basis and thus only small but significant reductions in absolute concentration can be expected from phosphorus control programs. However, over a period of several years, these small reductions in concentration may be difficult to verify from total phosphorus measurements which have relatively large sampling errors and variances.
Successful development of a commercial, small-mesh trap net fishery for round whitefish (Prosopium cylindraceum) in Saginaw Bay, Lake Huron, prompted analysis of round whitefish fillets for selected contaminants [dieldrin, dichlorodiphenyl trichloroethane and metabolites (DDT), polychlorinated biphenyls (PCBs), polybrominated biphenyls (PBBs), chlordane, methyl mercury, and total mercury] to determine the fish's suitability for human consumption. Concentrations of all contaminants measured in 1977–1978 were below U. S. Food and Drug Administration action guidelines for fish; PBBs were below detection limits (0.001 mg/kg) for all fillets analyzed. Only PCBs and total DDT residues displayed a strong direct relationship with fish length. Fillets from males consistently exhibited higher mean concentrations of contaminants than did females. No changes in contaminant levels with season were observed.
Water samples from the central and eastern basins of Lake Erie were analyzed for volatile chloro- and chlorofluorocarbons for a one-week period in each of 1977 and 1978. The following contaminants were observed with mean concentrations and standard deviations in 1978 with the corresponding values for 1977 in parentheses: Dichlorodifluoromethane (Freon 12) 76±38 (73±36) ng°L−1; trichlorofluoromethane (Freon 11) 34±26 (46±40) ng·L−1; chloroform 15±4 ng·L−1; carbon tetrachloride 19±11 (37±20) ng·LT−1; trichloroethylene 20±13 (11±9) ng·L−1.Except for carbon tetrachloride and trichloroethylene, the distribution patterns appear to show little correlation with each other. Chloroform concentrations were relatively uniform, however the carbon tetrachloride and trichloroethylene concentrations appear to be point-source related with discharges along the south shore of Lake Erie. Freon 12 concentrations were generally higher in the eastern basin than in the central basin, while Freon 11 levels were high and relatively uniform throughout the study area.
A regression model was used to determine the trend in time and the effects of sampling site, age, and weight on DDE, PCB, mirex, dieldrin, chlordane, and mercury concentrations in lake trout from Lake Ontario from 1977 to 1988. The model allows the use of the entire data set collected by the Canadian Federal Department of Fisheries and Oceans monitoring program for lake trout in this lake up to 1988, and provides estimates of logarithmic mean concentrations in each age group independent of the overall (across all ages) contaminant: body size relationship. Mercury concentrations decreased steadily during the study period with a half life of 11 years. DDE concentrations decreased rapidly from 1977 to 1980 but then remained relatively constant. Mirex concentrations generally decreased, but with much year-to-year variation. Dieldrin concentrations decreased from 1978 to 1988, but chlordane concentrations remained relatively constant. PCB concentrations decreased from 1977 to 1981, increased in 1982 until 1984, and decreased again in 1985. The overall trend in PCB concentrations was a gradual decline with a half life of roughly 10 years. Deviations about this trend were closely related to the rate of growth of the alewife population, a principal food of lake trout, suggesting that food web interactions play an important role in regulating PCB concentrations in lake trout. Lipid concentrations in lake trout also decreased from 1977 to 1988, complicating interpretation of the time trends in the lipid soluble organic contaminants. PCB, DDE, and mirex concentrations in trout increased from east to west in Lake Ontario. All contaminant concentrations increased with increasing age and body size, but the body size effect was less within an age class than across age classes.
Submersed macrophytes were surveyed at 595 stations located throughout the St. Clair-Detroit River system between Lakes Huron and Erie, 23 August to 13 October 1978. Eurasian watermilfoil (Myriophyllum spicatum), first recorded in the system in 1974, became the fourth most common submersed macrophyte in the system by 1978. However, it has not been reported as a widespread nuisance in this system as it has in many other large water bodies in the United States. Observations made during the present study, and interpretation of an aerial photograph, suggest that M. spicatum was a minor nuisance to small boat navigation in portions of the system. Information presented in this study provides a baseline against which future changes in the occurrence of M. spicatum in the St. Clair-Detroit River system can be measured.
The benthic macroinvertebrates of the nearshore zone of Lake Erie in the vicinity of Cleveland, Ohio, were sampled at 12 stations in October 1988 and May 1989 to determine the present pollutional status of the benthic community and to record changes over the past decade. The most abundant macroinvertebrates were, respectively, oligochaete worms, sphaeriid clams, and chironomid larvae. Of 75 taxa identified, 22 were oligochaetes, 16 were chironomids, and 11 were mollusks. Community composition in most of the harbor was suggestive of highly organically enriched sediments and was dominated by five tubificid worm species and pill clams (Pisidium spp.). The open water community, representative of less enriched conditions, was dominated by the tubificid Limnodrilus hoffmeisteri as well as Pisidium spp. and midges of the Chironomus anthracinus? group.Improvement in the habitat quality of the study area, including the harbor, since the late 1970s was evident from a dramatic increase in the number of taxa at all sites, a reduction in the proportion of oligochaete worms, a large increase in the abundance of sphaeriid clams and midge larvae, and the widespread distribution though low abundance in open water sediments of caddisfiy larvae (Oecetis sp.) which were absent in the late 1970s. Midges of the genus Chironomus remained largely absent from the harbor, and the abundance of the genus Procladius remained low around the mouth of the Cuyahoga River. The abundance of oligochaete worms has increased substantially around the river mouth over the decade. This increase may indicate an actual improvement of conditions overall, as might be expected with declining concentrations of toxic substances but continued organic enrichment of the sediments.
In situ hypolimnetic oxygen depletion measurements were conducted during four summer cruises in 1979 at two central basin stations in Lake Erie to evaluate the relative contribution of the sediments to the oxygen demand. Sediment oxygen demand (SOD) rates were determined by measuring the rate of oxygen decrease within a triangular benthic chamber; water column oxygen demand (WOD) rates were determined using 24-hour light and dark bottles placed in situ. Results indicated that the SOD contribution to the hypolimnetic oxygen depletion rate decreased throughout the summer from about 81% to only 30% with an initially high rate due to spring algal biomass sedimentation and lower rates in late summer due to depressed oxygen levels. The WOD rate contribution increased from 19% to 70% throughout the stratified period due to the decomposition of settling algal cells. Comparing the overall volumetric summer in situ rates (0.126 mg O2/L/day) with cruise-interval depletion rates (0.365 mg O2/L/day), the in situ rates were about 300% higher. This is attributed to unaccounted oxygen sources to the hypolimnion and because the in situ rates measure the gross WOD and SOD rather than measuring the net effects they exert.
Congenital malformations are relatively uncommon in most wild bird populations. Here we document the occurrence of bill malformations in double-crested cormorant (Phalacrocorax auritus) chicks from colonies in Green Bay and elsewhere in the Great Lakes and in reference areas off the Great Lakes, in the years 1979 through 1987. In the Great Lakes, 31,168 cormorant chicks were examined during 147 visits to 42 colonies. Seventy of these chicks had crossed or deflected bills or bills in which the mandibles differed in length. Only two of the 20,962 chicks examined during 82 visits to 35 colonies in reference areas located in the prairies and northwestern Ontario had bill defects. The probability of observing a malformed chick on a visit to a colony in Green Bay was 10 to 32 times greater than on a visit to a colony in the reference areas. Bill defects were observed at only two (6%) of the colonies in the reference areas. This proportion was exceeded in six of eight geographic regions within the Great Lakes and was highest (73%) in Green Bay colonies. The prevalence of malformed chicks in Green Bay (52.1 per 10,000)was markedly greater than in all other regions during this period. These bill defects are an example of developmental asymmetry and are an indicator of developmental stability within local populations. Other investigations in Lake Michigan suggest that chemicals such as polyhalogenated aromatic hydrocarbons (e.g., PCBs) that induce aryl hydrocarbon hydroxylase are responsible for the defects observed defects.
Storm surges were observed at a number of locations in the Great Lakes on the morning of April 6, 1979, in response to a severe storm. Winds gusting up to speeds of 28 m/s and aligned along the axis of the lake generated extreme water level excursions and set-up at a number of locations on Lake Erie. A set-up of 4.5 m between Buffalo and Toledo may be the largest recorded water level set-up on Lake Erie. Simulation of the storm surge on Lake Erie by a dynamical method yielded water level hindcasts within 15% of the observed water levels.
Currents and water temperatures were recorded at a large-scale grid of fixed moorings in Lake Erie from May 1979 through June 1980. Currents measured in the lower half of the central basin water column were mostly return flows (beneath the surface wind drift) driven by the surface pressure gradient. Often observed was a complex system of Lake Erie circulation gyres as predicted by models. Another common occurrence was for one of the central basin gyres to become dominant and envelop the whole basin in either uniform clockwise or anticlockwise flow. It is not fully certain why one of the circulation cells grows as opposed to the others, but the curl of the wind stress had influence. The currents were more barotropic than predicted by full Ekman layer current models. Tidal-like currents driven by the longitudinal seiches of Lake Erie dominate the island-filled passages between the western and central Lake Erie basins, with currents across the whole island chain very closely in phase. Processes of hypolimnion volume entrainment are suggested from the central basin temperature recordings. Large volume water exchanges between the central and eastern basins occurred after the water mass in the vicinity of the shallow ridge that separates them had become unstratified. These and other topics are discussed as the large data set generated from the experiment is explored.
Analyses of coho salmon from each of the Great Lakes by a single laboratory produced residue data on the accumulation of environmental contaminants which have been banned, severely restricted, or are currently permitted in the basin. Coho salmon from Lake Superior contained only trace amounts or low levels of most toxic substances quantified; Lake Erie fish were contaminated with low levels of a number of pesticides and industrial compounds; relatively higher residues were detected in coho from Lake Huron and Lake Michigan; and the highest concentrations for a number of compounds were found in fillets from coho from Lake Ontario. Contaminant concentrations in migratory coho salmon indicate open lake contaminant problems rather than point source or nearshore conditions. Tissue residues were less than USFDA action levels, used by many agencies in assessing the severity of fish contaminant problems. Only mirex concentrations in fish collected from Lake Ontario exceeded a USFDA action level. The data reported in this study generally agree with recent findings from individual state contaminant monitoring programs. Problems with varying analytical and sampling techniques preclude direct comparisons with previously published data of other studies.
To test the hypothesis that suspended fine-grained particles moving downslope within the nepheloid layer in Lake Michigan are periodically reintroduced into the nearshore and euphotic zones during upwelling events, temperature and transparency profiles were recorded and water samples analyzed for total suspended materials (TSM) during a strong upwelling event. The resultant data confirmed that there is periodic reintroduction of suspended materials into the nearshore and epilimnion during such events, and provided insight into the importance of the general resuspension process, especially in regard to differences between known sedimentation rates and the rate indicated by trap collections. Both upwelling and downwelling currents are disruptive processes that tend to keep the suspended particulates in motion and prevent them from rapidly becoming a permanent part of the bottom sediment. These currents redistribute suspended particulates and the associated chemical load, and may resuspend surficial sediments, especially from the slope and shelf regions. The reintroduction of fine-grained materials into the euphotic zone through upwelling events can play a large role in the long-term behavior and fate of persistent contaminants.
The concentrations of DDE, PCBs (1:1 mix of Aroclor 1254:1260), mirex, HCB, QCB, and tetrachlorobenzenes in eggs of herring gulls are presented for nine colonies in Lake Huron in 1980, as part of intensive investigations of herring gull numbers, biology, and contaminant burdens in that year. Additional data are given for seven other organochlorine contaminants at two colonies. These data are analyzed in relation to eggshell thickness, clutch size and reproductive output, and population trends at sample colonies. Levels of all contaminants measured were markedly lower than during the 1970s, and eggshells were only 6% thinner on average than during the pre-DDT era. Levels of most organochlorine (OC) residues were highest in eggs from Saginaw Bay, which remains one of the most heavily contaminated sites on the Great Lakes. Discriminant function analyses provided reasonably good regional, but not colony-specific, separation of eggs based on suites of OCs. Reproductive output was relatively high, and we found little evidence of impaired reproduction attributable to OC contamination or burden in 1980. Lake-wide, herring gull numbers were increasing slightly, but decreasing at colonies on the main body of Lake Huron.
Coho (Oncorhynchus kisutch) and chinook (O. tshawytscha) salmon were collected from seven Michigan tributaries to the Great Lakes and analyzed for chlorinated hydrocarbon pesticides, polychlorinated biphenyls (PCB), and metals. Analyses of standard skin-on fillets of these salmon revealed the presence of PCBs, DDT, dieldrin, zinc (found in one chinook), and mercury, but no detectable residues of aldrin, chlordane, heptachlor, heptachlor epoxide, toxaphene, arsenic, cadmium, chromium, copper, or lead. Chlorinated hydrocarbon contaminants were present at significantly lower levels than were found in salmon samples collected in 1971. No sample of either species exceeded FDA action levels for DDT, dieldrin, or mercury; however, 6 of 63 coho and 29 of 50 chinook salmon exceeded the FDA action level for PCBs. In general, chinook salmon had higher contaminant levels than did coho salmon collected at the same time and location. Salmon collected from the southernmost Lake Michigan tributary had highest contaminant levels (especially PCBs) while salmon from the Lake Erie tributaries had lowest contaminant levels. Samples analyzed as skinless fillets showed dramatically reduced levels of PCBs and DDT analogs compared to standard, skin-fillets from the same fish. A strong linear correlation was observed between residues of DDT and PCBs in samples analyzed
Quantitative analysis of siliceous microfossils in a finely sectioned (0.5 cm) core of near-surface sediments from Lake Ontario indicates the lake is responding to reductions in phosphorus loading. The magnitude of response, however, is very small. Increased diatom accumulation and modified species composition indicates some relaxation of silica limitation, particularly in early spring. Abundance of some diatom species associated with grossly polluted areas of the Great Lakes has been reduced, but most recently deposited assemblages are composed of species tolerant of eutrophic conditions. No reappearance of species assemblages abundant in Lake Ontario prior to 1935 and presently associated with oligotrophic conditions in the upper Great Lakes were noted. Changes in species composition are consistent with observed recent changes in N:P ratio.
Common snapping turtle eggs were collected at nesting sites from two locations in 1981 and eight locations in 1984 in Ontario, Canada, and analyzed for chlorinated hydrocarbons. Nine locations were within the Great Lakes-St. Lawrence River basin and one location, Algonquin Provincial Park, served as a control site outside the basin. Total PCBs ranged from 0.057 to 4.76 mg/kg (wet wt.) among the Great Lakes-St. Lawrence River samples. Mean total PCB concentration at Algonquin Park was 0.187 mg/kg. Eggs from Hamilton Harbour, Port Franks, Bay of Quinte/Murray Canal, and Lake St. Clair were the most contaminated among the ten sample locations. There was statistically significant variation in concentrations of all organochlorine compounds among sites. In some locations, there was high variation in contamination among clutches.A pool of eggs from Hamilton Harbour contained 67 ng/kg of 2378-tetrachlorodibenzo-p-dioxin and 14.0 ng/kg of23478-pentachlorodibenzofuran. Some dioxin congeners were present in turtle eggs at concentrations higher or equal to that in herring gull eggs from Hamilton Harbour. Comprehensive GC/MS analysis of the Hamilton Harbour eggs also revealed the presence of trace amounts of o,p-dicofol, octachlorostyrene, and toxaphene.Geographic variation in contaminant levels in snapping turtle eggs from wetlands is similar to that in spottail shiners and herring gull eggs collected in the pelagic zone of the Great Lakes. This may be due to the consumption of migrant fish by snapping turtles in nearshore wetlands.