Assessing the application of SeaWiFS ocean color algorithms to Lake Erie

The feasibility of satellite-based monitoring of phytoplankton chlorophyll a concentrations in Lake Erie is assessed by applying globally calibrated, ocean-derived color algorithms to spatially and temporally collocated measurements of SeaWiFS remote sensing reflectance. Satellite-based chlorophyll a retrievals were compared with fluorescence-based measurements of chlorophyll a from 68 field samples collected across the lake between 1998 and 2002. Twelve ocean-derived color algorithms, one regional algorithm derived for the Baltic Sea's Case 2 waters, and a set of regional algorithms developed for the western, central and eastern basins of Lake Erie were considered. While none of the ocean-derived algorithms performed adequately, the outlook for the success of regionally calibrated and validated algorithms, with forms similar to the ocean-derived algorithms, is promising over the eastern basin and possibly the central basin of the lake. In the western basin, each of the regional algorithms considered performed poorly, indicating that alternative approaches to algorithm development, or to satellite data screening and analysis procedures will be needed.     

A band-ratio algorithm for retrieving open-lake chlorophyll values from satellite observations of the Great Lakes

The U.S. Environmental Protection Agency's Great Lakes National Program Office (GLNPO) has collected water quality data from the five Great Lakes annually since 1993. We used the GLNPO observations made since 2002 along with coincident measurements made by the Sea-viewing Wide Field-of-View Sensor (SeaWiFS) and the Moderate-resolution Imaging Spectroradiometer (MODIS) to develop a new band-ratio algorithm for estimating chlorophyll concentrations in the Great Lakes from satellite observations. The new algorithm is based on a third-order polynomial model using the same maximum band ratios employed in the standard NASA algorithms (OC4 for SeaWiFS and OC3M for MODIS). The sensor-specific coefficients for the new algorithm were obtained by fitting the relationship to several hundred matched field and satellite observations. Although there are some seasonal variations in some lakes, the relationship between the observed chlorophyll values and those modeled using the new coefficients is fairly stable from lake to lake and across years. The accuracy of the satellite chlorophyll estimates derived from the new algorithm was improved substantially relative both to the standard NASA retrievals and to previously published algorithms tuned to individual lakes. Monte-Carlo fits to randomly selected subsets of the observations allowed us to estimate the uncertainty associated with the retrievals purely as a function of the satellite data. Our results provide, for the first time, a single simple band ratio method for retrieving chlorophyll concentrations in the offshore “open” waters of the Great Lakes from satellite observations.     

Mapping cyanobacterial blooms in the Great Lakes using MODIS

Toxin-producing Cyanobacteria have been documented in Lake Erie and Ontario in the last several years. We developed algorithms to discriminate potentially toxic cyanobacterial blooms from other harmless phytoplankton blooms and to extract relative phycocyanin abundances from Moderate Resolution Imaging Spectrometer (MODIS) satellite data. Lee's quasi-analytical algorithm was used to calculate total absorption and backscatter from the 250 m, 500 m and 1 km bands of MODIS scenes. A non-negative least square algorithm was then utilized to discern relative concentrations of Chlorophyta (green algae), phycocyanin-rich Cyanobacteria (blue-green algae), and colored dissolved organic matter and suspended sediments combined in lake waters using published absorption spectra for these components. MODIS-derived cyanobacterial concentrations and/or bloom distributions from 10 scenes acquired in the summers of 2004 and 2005 were successfully verified against contemporaneous calibrated measurements of pigments that were acquired from measurements made using continuous fluorimetric measurements of surface water (1 m depth) from six cruises, and three additional cyanobacterial blooms reported in the scientific literature between 2002 and 2006. These results demonstrate that this methodology could be used to develop a cost-effective practical screening method for rapid detection and warning of potentially toxic cyanobacterial blooms in the lower Great Lakes.     

A simple water clarity-turbidity index for the Great Lakes

There are a multitude of satellite-derived water clarity and turbidity indicators to support the decision making of environmental managers and policy makers. However, water quality dynamic ranges addressed by these indicators can differ significantly, subjecting non-expert users to potential pitfalls. Here we propose a satellite water clarity-turbidity index (CTI) as a simplified way to capture major changes in water clarity/turbidity across all water types in the Great Lakes. The CTI is defined to merge key information from three prerequisite variables derived from Visible Infrared Imaging Radiometer Suite (VIIRS) measurements, namely, the Secchi disk depth, the particulate backscattering coefficient, and the nephelometric turbidity, which are suitable for clear, intermediate, and turbid waters, respectively. Application to the Great Lakes shows that with one parameter, the CTI can illustrate major spatial and temporal patterns that are not entirely visible with each of the three original indicators alone. Using the CTI, we identified significant decrease in water turbidity in Lakes Michigan and Huron from 2000 to 2005, during which daily variability of CTI in August initially spiked and then gradually decreased most likely owing to diminishing whiting events. The CTI is a convenient and holistic assessment tool for water quality management.     

Variations in chromophoric dissolved organic matter and its influence on the use of pigment-specific fluorimeters in the Great Lakes

The variability of colored material (yellow substances, chromophoric dissolved organic matter [CDOM]) in Great Lakes water on the operation of a submersible pigment-based fluorimeter (FluoroProbe) was assessed. The FluoroProbe is designed to account for the influence of ambient CDOM on fluorimetric analysis of natural waters by using background yellow substance content to correct for fluorescence that is not due to phytoplankton. The objective of this study was to assess FluoroProbe performance as well as determine the variability of CDOM that has a bearing on instrument use. This study focused on Lakes Ontario and Huron in spring and summer and employed up to three FluoroProbes for simultaneous comparisons of performance, as assessed by comparing in situ versus extracted concentrations of chlorophyll-a. Conclusions from this study are: FluoroProbes are not equally accurate; yellow substance correction factors obtained by FluoroProbes are instrument specific; there are seasonal differences in yellow substance correction factors between summer and spring and among thermal strata in summer, in Lakes Huron and Ontario.The observed differences among yellow substance correction factors are attributed to chromophoric material that was assessed independently by measuring CDOM using a separate fluorimetric technique and by measuring in situ UV absorption coefficients. These results suggest that the use of FluoroProbes and related instruments for phytoplankton surveillance and monitoring initiatives must account for differences in ambient CDOM.     

Relationship of road density and marsh condition to turtle assemblage characteristics in the Laurentian Great Lakes

Human-induced degradation of coastal wetlands often leads to altered trophic dynamics and species assemblages. Here we use data from 77 coastal marshes in three Laurentian Great Lakes collected between 2001 and 2007 to examine the relationship between human disturbance (road density and wetland quality) and characteristics of aquatic turtle assemblages, including species richness and abundances. Painted turtles (Chrysemys picta) were encountered disproportionately in degraded wetlands and the probability of occurrence decreased with improved site quality. Abundance of painted turtles peaked, however, at intermediate road density in surrounding 1- and 2-km buffers. Across all sites, species richness was highest and common snapping turtles (Chelydra serpentina) were most abundant in wetlands with intermediate water quality. The common musk turtle (Sternotherus odoratus) was absent from degraded wetlands in the lower lakes (Erie and Ontario) that fell within their historical range, but reached high abundances in marshes of Georgian Bay and the North Channel, a region with relatively low human disturbance. Analysis of sex ratios in painted turtles revealed a significant male bias in an area with high road density, while the sex ratio did not differ significantly from 1:1 in a less developed region, consistent with reports of high female mortality in urban areas.     

Trends in Diporeia populations across the Laurentian Great Lakes, 1997-2009

Benthic communities in the Laurentian Great Lakes have been in a state of flux since the arrival of dreissenid mussels, with the most dramatic changes occurring in population densities of the amphipod Diporeia. In response, the US EPA initiated an annual benthic macroinvertebrate monitoring program on all five Great Lakes in 1997. Although historically the dominant benthic invertebrate in all the lakes, no Diporeia have been found in Lake Erie during the first 13 years of our study, confirming that Diporeia is now effectively absent from that lake. Populations have almost entirely disappeared from our shallow (< 90 m) sites in lakes Ontario, Huron, and Michigan. In Lake Ontario, three of our four deep (> 90 m) sites still supported Diporeia populations in 2009, with densities at those sites ranging between 96 and 198/m². In Lake Michigan, populations were still found at six of our seven deep sites in 2009, with densities ranging from 57 to 1409/m². Densities of Diporeia in 2009 at the four deep sites in Lake Huron were somewhat lower than those in Lake Michigan, ranging from 191 to 720/m². Interannual changes in population size in Lake Huron and Lake Michigan have shown a degree of synchrony across most sites, with periods of rapid decline (1997-2000, 2003-2004) alternating with periods of little change or even increase (2001-2002, 2005-2009). There has been no evidence of directional trends at any sites in Lake Superior, although substantial interannual variability was seen.     

Escherichia coli toxin and attachment genes in sand at Great Lakes recreational beaches

The United States Environmental Protection Agency recommends density thresholds for the fecal indicator organism Escherichia coli in order to ensure the safety of recreational waters. A number of studies published over the past ten years indicate that E. coli is encountered frequently in sand at recreational beaches. While a majority of the sand-associated E. coli may be commensal or environmental strains, the potential for pathogenic strains of E. coli to be present exists. Therefore, the aim of this study was to assess the presence of attachment and virulence genes associated with enteropathogenic and enterohemorrhagic strains of E. coli (EPEC and EHEC) in populations of E. coli recovered from swash zone sand from seven recreational beaches along Lake Huron and Lake St. Clair in eastern Michigan, USA. Genes coding for attachment proteins in EPEC and EHEC were very prevalent in sand E. coli, but genes coding for toxin genes were uncommon. The paucity of genes associated with E. coli toxins suggests that the EPEC and EHEC pathotypes are not common in sand; however, the high prevalence of genes associated with attachment in E. coli pathotypes suggests that these genes are being retained within the beach sand E. coli population.     

First Report of the Asian Fish Tapeworm in the Great Lakes

The Asian fish tapeworm (Bothriocephalus acheilognathi) is an introduced species that has spread throughout much of the United States and into Canada. This cestode was found in a specimen of bluntnose minnow (Pimephales notatus) collected from Grosse Îsle in the Detroit River in September, 2002. This parasite is a known pathogen that can cause weight loss, anemia, and mortality in young fishes. This is the first report of B. acheilognathi in the Great Lakes.     

The Great Lakes Cladophora Model: Development,testing, and application to Lake Michigan

A recent review of the Great Lakes Water Quality Agreement has concluded that while controls on phosphorus inputs to Lake Michigan achieved the desired effect in offshore waters, the nearshore region continues to suffer from elevated phosphorus levels. Failure to achieve trophic state goals in the nearshore is manifested in nuisance growth of Cladophora and attendant impacts on property owners, utilities, and the public health and welfare. This study focuses on a site in Lake Michigan near Milwaukee, Wisconsin, where nuisance growth of Cladophora and associated beach fouling occur regularly. A mechanistic model simulating Cladophora growth, suitable for guiding nutrient management in the Great Lakes nearshore, is presented. The model represents an update of the Canale and Auer framework, reflecting current understandings of Cladophora ecology and offering a user-friendly interface making the software more widely available to decision makers. This Great Lakes Cladophora Model (GLCM) is first validated for the Auer/Canale data set collected in 1979 at a site on Lake Huron and then for a data set developed in 2006 for a site on Lake Michigan. Model performance under the strikingly different forcing conditions (depth, light, phosphorus levels) characteristic of these two sites affirms the widespread applicability of the tool. The GLCM is then extended to examine the impacts of ecosystem perturbation (dreissenid colonization) on Cladophora growth and to future approaches to monitoring and management.     

Distribution and abundance of nesting common and Caspian terns on the North American Great Lakes, 1976 to 1999

Canadian and US federal wildlife agencies completed three surveys (1976–1980, 1989–1991, and 1997–2000) to census colonial waterbirds breeding on the Great Lakes. We here summarize and comment on nest numbers and colony site distribution of common terns (Sterna hirundo) and Caspian terns (Hydroprogne caspia). Common terns are in serious trouble on the Great Lakes. Numbers declined with substantial losses in nests (− 19.1%) and colony sites (− 23.2%) between the first and third censuses. An increase in numbers at US sites (+ 26.6%) did not compensate for losses (− 33.1%) at Canadian sites. Caspian terns increased in nest numbers (+ 65.9%) and colony sites (+ 50.0%) over the same period. The increase at US sites (136.5%) was greater than at Canadian sites (11.5%). Most (70.7%, n = 186) common tern sites had nests during only one census; 17 sites (6.5%) had nests during all censuses. In contrast, 9 of 33 (27.2%) Caspian tern sites had nests during all censuses and contained a majority of nests (50–82%) in each census. Pairs of both species nested on natural substrates across the Great Lakes. Common terns nested mostly on artificial (human-constructed) substrates on the lower Great Lakes. We identify site characteristics that may have contributed to long-term (three census) occupancy by common terns (small size, artificial substrates, absence of ring-billed gulls) and Caspian terns (natural substrates on large, remote islands). We suggest an urgent need for protection and conservation of common tern colonies and identify specific priority sites for implementation of management protocols.     

Metazoan Parasites of Introduced Round and Tubenose Gobies in the Great Lakes: Support for the “Enemy Release Hypothesis”

Recent invasion theory has hypothesized that newly established exotic species may initially be free of their native parasites, augmenting their population success. Others have hypothesized that invaders may introduce exotic parasites to native species and/or may become hosts to native parasites in their new habitats. Our study analyzed the parasites of two exotic Eurasian gobies that were detected in the Great Lakes in 1990: the round goby Apollonia melanostoma and the tubenose goby Proterorhinus semilunaris. We compared our results from the central region of their introduced ranges in Lakes Huron, St. Clair, and Erie with other studies in the Great Lakes over the past decade, as well as Eurasian native and nonindigenous habitats. Results showed that goby-specific metazoan parasites were absent in the Great Lakes, and all but one species were represented only as larvae, suggesting that adult parasites presently are poorly-adapted to the new gobies as hosts. Seven parasitic species are known to infest the tubenose goby in the Great Lakes, including our new finding of the acanthocephalan Southwellina hispida, and all are rare. We provide the first findings of four parasite species in the round goby and clarified two others, totaling 22 in the Great Lakes—with most being rare. In contrast, 72 round goby parasites occur in the Black Sea region. Trematodes are the most common parasitic group of the round goby in the Great Lakes, as in their native Black Sea range and Baltic Sea introduction. Holarctic trematode Diplostomum spathaceum larvae, which are one of two widely distributed species shared with Eurasia, were found in round goby eyes from all Great Lakes localities except Lake Huron proper. Our study and others reveal no overall increases in parasitism of the invasive gobies over the past decade after their establishment in the Great Lakes. In conclusion, the parasite “load” on the invasive gobies appears relatively low in comparison with their native habitats, lending support to the “enemy release hypothesis.”     

Distribution of nuisance Cladophora in the lower Great Lakes: Patterns with land use, near shore water quality and dreissenid abundance

Selected shorelines and offshore shoals in Lakes Erie, Huron and Ontario were surveyed with a high frequency hydroacoustic system to investigate current spatial patterns of nuisance benthic filamentous algal (e.g., Cladophora) cover and stand height. Cladophora reached nuisance levels at all sites in Lakes Erie and Ontario, but not in Lake Huron or Georgian Bay. Despite clear gradients in coastal land cover, near shore water quality gradients were generally weak, and for Lakes Erie and Ontario, measures of near shore water quality were similar to that at offshore shoals. Hierarchical partitioning analysis suggested that while dreissenid mussel abundance appeared to be important in determining the magnitude of Cladophora standing crop, the joint contribution of catchment land cover, near shore water quality (nutrient levels and suspended matter) and dreissenid mussel abundance explained nearly 95% of the total variance in nuisance Cladophora standing crop observed in this study. Although the results from this study are necessarily correlative in nature and definition of causal relationships is not possible, these results provide corroborating evidence from sites across a gradient within and across the lower Great Lakes that is consistent with the operation of the near shore shunt model.     

Great Lakes Cladophora in the 21st century: same algae—different ecosystem

Nuisance growth of the attached, green alga Cladophora was considered to have been abated by phosphorus management programs mandated under the Great Lakes Water Quality Agreement. The apparent resurgence of nuisance growth in Lakes Erie, Michigan and Ontario has been linked conceptually to ecosystem alterations engineered by invasive dreissenid mussels (Dreissena polymorpha and Dreissenabugensis). Here, we apply contemporary modeling tools and historical water quality data sets in quantifying the impact of long-term changes in phosphorus loading and dreissenid-mediated changes in water clarity on the distribution and production of Cladophora. It is concluded that reductions in phosphorus loading in the pre-dreissenid period achieved the desired effect, as model simulations were consistent with the biomass declines reported from the early 1970s to the early 1980s. These declines were, however, largely offset by dreissenid-driven changes in water clarity that extended the depth of colonization by Cladophora, increasing total production. We were not able to isolate and quantify the significance of dreissenid mediation of phosphorus cycling using the historical database. Phosphorus management remains the appropriate mechanism for reducing nuisance levels of Cladophora growth. The development of action plans will require an improved understanding of nearshore phosphorus dynamics such as might be obtained through regular monitoring of soluble reactive phosphorus levels, internal phosphorus content and Cladophora biomass in impacted nearshore regions of the Great Lakes.     

Bio-optical properties and primary production of Lake Michigan: Insights from 13-years of SeaWiFS imagery

Thirteen years of SeaWiFS data (1998–2010) from the early spring isothermal period (March–April) were used to determine trends of water attenuation coefficient (K_dPAR), chlorophyll a (Chl a), Photosynthetic Available Radiation (PAR), and modeled primary production in southern Lake Michigan. Surface PAR values remained unchanged between 1998 and 2010, but there was an 18–22% drop in K_dPAR during the March/April isothermal period as water clarity increased. This transparency increase was accompanied by a 41–53% decline in Chl a concentration (μg · L^− 1) and a 42–46% decline in modeled primary production (Great Lakes Primary Production Model). These changes were most pronounced in 2001–2003 which coincided with the period of initial colonization of the quagga mussels. Statistically significant spatial differences were noted in Chl a (μg · L^− 1) concentrations between mid-depth (z = 30–90 m deep), and offshore (z > 90 m deep) waters. Chl a concentrations in the mid-depth region (30–90 m) decreased at a higher rate compared to offshore waters (> 90 m) likely as a result of filtration activities of quagga mussel.     

Is reduced benthic flux related to the Diporeia decline? Analysis of spring blooms and whiting events in Lake Ontario

Benthic monitoring by USGS off the southern shore of Lake Ontario from October 1993 to October 1995 provides a detailed view of the early stages of the decline of the native amphipod Diporeia. A loss of the 1994 and 1995 year classes of Diporeia preceded the disappearance of the native amphipod at sites near Oswego and Rochester at depths from 55 to 130 m. In succeeding years, Diporeia populations continued to decline in Lake Ontario and were nearly extirpated by 2008. Explanations for Diporeia's decline in the Great Lakes include several hypotheses often linked to the introduction and expansion of exotic zebra and quagga mussels (Dreissena sp.). We compare the timeline of the Diporeia decline in Lake Ontario with trends in two sources of organic matter to the sediments — spring diatom blooms and late summer whiting events. The 1994–95 decline of Diporeia coincided with localized dreissenid effects on phytoplankton in the nearshore and a year (April 1994 to May 1995) of decreased flux of organic carbon recorded by sediment traps moored offshore of Oswego. Later declines of profundal (> 90 m) Diporeia populations in 2003 were poorly associated with trends in spring algal blooms and late summer whiting events.     

Investigation of viruses in Diporeia spp. from the Laurentian Great Lakes and Owasco Lake as potential stressors of declining populations

The benthic amphipod Diporeia is an ecologically and biogeochemically important constituent of deep freshwater lakes in North America. The proliferation of dreissenid mussels in the mid-1990s coincided with a sharp decrease in Diporeia populations in several Laurentian Great Lakes; however the ultimate cause and mechanisms of their decline are still unknown. Here we examined the composition of DNA viruses associated with Diporeia collected from populations of Lake Michigan that had declined and stable populations in Lake Superior and Owasco Lake (Finger Lake in central New York State). Viral metagenomic libraries from Owasco Lake and Lake Superior were comprised primarily of bacteriophages, which may infect bacteria within the amphipod microbiome. In contrast, the metagenomic library from Lake Michigan contained well-represented ssDNA circular viral genomes. The prevalence and viral load of one putative Type V ssDNA circular virus (LM29173) that recruited almost 30% of total viral sequence reads in the Lake Michigan library was analyzed by quantitative PCR. The prevalence of LM29173 was over two orders of magnitude greater in Lake Michigan compared to the other two lakes. Although further research is necessary to establish the pathology and epidemiological extent of viral-Diporeia interactions, our data suggest that viruses may be numerically significant constituents of the Diporeia microbiome, and if pathogenic some of these viruses may be a stressor of Great Lakes Diporeia populations. Our data further indicate that special attention should be given to the circovirus that was prevalent in the declining Michigan population but uncommon in the other two lakes.     

Shifting status and distribution of range margin chorus frog (Pseudacris) populations in eastern Great Lakes watersheds

Contiguous with their range across major rivers into Canada, two different species of chorus frogs are now thought to inhabit the Great Lakes watersheds of New York. Pseudacris triseriata is found along the western Lake Ontario and Lake Erie plains while P. maculata (tentatively a new frog species in NY) inhabits the lowlands of eastern Lake Ontario and the St. Lawrence River. Both species are on their extreme range margins in NY. In 2010 we detected distributional declines of both putative chorus frog species based on a broad survey following standardized occupancy detection protocols. Causes are unclear but could relate to reforestation and urbanization of formerly more extensive agricultural lands, climate change, pathogenic fungal outbreaks and/or the contaminant effects of intensive agriculture. On the other hand, the prior overestimation of ranges because of misidentification may have inflated earlier perceived distributions (positive survey bias) because false positives are problematical for this cryptic frog. At broad geographical scales, chorus frog (meta)populations are highly dynamic and are likely shifting their ranges in response to rapidly changing overall environmental conditions in the northeastern U.S. and Canada.     

Quantification of cormorant litter and nutrient deposition to Great Lakes island ecosystems

Ornithogenic nutrients derived from waterbirds such as the double-crested cormorant (Phalacrocorax auritus, Lesson) have been linked to habitat change within nesting colonies. For the islands of Lake Erie, where increasing cormorant populations and subsequent habitat change have spurred management activity, estimates of the quantity and chemical characteristics of avian-derived contributions are lacking. To evaluate the quantity and chemical characteristics of ornithogenic litterfall beneath a double-crested cormorant colony on a western Lake Erie island we investigated the mass of material and nutrient composition (PO₄^3 −, NO₃⁻ and NH₄⁺) reaching the forest floor under three nest densities (Low: 1–96 nests ha^− 1; Medium: 97–255 nests ha^− 1 and High: > 255 nests ha^− 1). As expected, litterfall (total mass) input differed among nest densities with the most substantial input (225.05 g/m² week^− 1) measured under High nest density conditions. Nutrient concentrations also showed increases with nest density to a point, where mean PO₄^3 − and NH₄⁺ concentrations showed no differences between Medium and High nest density sites. As well, NO₃⁻ concentrations were highest under Medium density, with no differences in this nutrient observed between Low and High density. Collectively, litterfall nutrient composition was similar to those linked to habitat changes in other waterbird colonies. Similarities in the concentrations of several nutrients between Medium and High nest density categories suggest that management actions aimed at reducing allochthonous nutrient contributions should try to sustain nest density at or below 96 nests ha^− 1.