Great Lakes coastal fish habitat classification and assessment

Basin-scale assessment of fish habitat in Great Lakes coastal ecosystems would increase our ability to prioritize fish habitat management and restoration actions. As a first step in this direction, we identified key habitat factors associated with highest probability of occurrence for several societally and ecologically important coastal fish species as well as community metrics, using data from the Great Lakes Aquatic Habitat Framework (GLAHF), Great Lakes Environmental Indicators (GLEI) and Coastal Wetland Monitoring Program (CWMP). Secondly, we assessed whether species-specific habitat was threatened by watershed-level anthropogenic stressors. In the southern Great Lakes, key habitat factors for determining presence/absence of several species of coastal fish were chlorophyll concentrations, turbidity, and wave height, whereas in the northern ecoprovince temperature was the major habitat driver for most of the species modeled. Habitat factors best explaining fish richness and diversity were bottom slope and chlorophyll a. These models could likely be further improved with addition of high-resolution submerged macrophyte complexity data which are currently unavailable at the basin-wide scale. Proportion of invasive species was correlated primarily with increasing maximum observed inorganic turbidity and chlorophyll a. We also demonstrate that preferred habitat for several coastal species and high-diversity areas overlap with areas of high watershed stress. Great Lakes coastal wetland fish are a large contributor to ecosystem services as well as commercial and recreational fishery harvest, and scalable basin-wide habitat models developed in this study may be useful for informing management actions targeting specific species or overall coastal fish biodiversity.     

An algorithm to retrieve chlorophyll,dissolved organic carbon,and suspended minerals from Great Lakes satellite data

An algorithm that utilizes individual lake hydro-optical (HO) models has been developed for the Great Lakes that uses SeaWiFS, MODIS, or MERIS satellite data to estimate concentrations of chlorophyll, dissolved organic carbon, and suspended minerals. The Color Producing Agent Algorithm (CPA-A) uses a specific HO model for each lake. The HO models provide absorption functions for the Color Producing Agents (CPAs) (chlorophyll (chl), colored dissolved organic matter (as dissolved organic carbon, doc), and suspended minerals (sm)) as well as backscatter for the chlorophyll, and suspended mineral parameters. These models were generated using simultaneous optical data collected with in situ measurements of CPAs collected during research cruises in the Great Lakes using regression analysis as well as using specific absorption and backscatter coefficients at specific chl, doc, and sm concentrations. A single average HO model for the Great Lakes was found to generate insufficiently accurate concentrations for Lakes Michigan, Erie, Superior and Huron. These new individual lake retrievals were evaluated with respect to EPA in situ field observations, as well as compared to the widely used OC3 MODIS retrieval. The new algorithm retrievals provided slightly more accurate chl values for Lakes Michigan, Superior, Huron, and Ontario than those obtained using the OC3 approach as well as providing additional concentration information on doc and sm. The CPA-A chl retrieval for Lake Erie is quite robust, producing reliable chl values in the reported EPA concentration ranges. Atmospheric correction approaches were also evaluated in this study.     

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 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.     

A new look at lake-effect snowfall trends in the Laurentian Great Lakes using a temporally homogeneous data set

Snowfall data are subject to quality issues that affect their usefulness for detection of climate trends. A new analysis of lake-effect snowfall trends utilizes a restricted set of stations identified as suitable for trends analysis based on a careful quality assessment of long-term observation stations in the lake-effect snowbelts of the Laurentian Great Lakes. An upward trend in snowfall was found in two (Superior and Michigan) of the four snowbelt areas. The trends for Lakes Erie and Ontario depended on the period of analysis. Although these results are qualitatively similar to outcomes of other recent studies, the magnitude of the upward trend is about half as large as trends in previous findings. The upward trend in snowfall was accompanied by an upward trend in liquid water equivalent for Superior and Michigan, while no trend was observed for Erie and Ontario. Air temperature has also trended upward for Superior and Michigan, suggesting that warmer surface waters and less ice cover are contributing to the upward snowfall trends by enhancing lake heat and moisture fluxes during cold air outbreaks. However, a more comprehensive study is needed to definitely determine cause and effect. Overall, this study finds that trends in lake-effect snowfall are not as large as was believed based on prior research.     

Hierarchical multi-scale classification of nearshore aquatic habitats of the Great Lakes: Western Lake Erie

Classification is a valuable conservation tool for examining natural resource status and problems and is being developed for coastal aquatic habitats. We present an objective, multi-scale hydrospatial framework for nearshore areas of the Great Lakes. The hydrospatial framework consists of spatial units at eight hierarchical scales from the North American Continent to the individual 270-m spatial cell. Characterization of spatial units based on fish abundance and diversity provides a fish-guided classification of aquatic areas at each spatial scale and demonstrates how classifications may be generated from that framework. Those classification units then provide information about habitat, as well as biotic conditions, which can be compared, contrasted, and hierarchically related spatially. Examples within several representative coastal or open water zones of the Western Lake Erie pilot area highlight potential application of this classification system to management problems. This classification system can assist natural resource managers with planning and establishing priorities for aquatic habitat protection, developing rehabilitation strategies, or identifying special management actions.     

Widespread prevalence of hypoxia and the classification of hypoxic conditions in the Laurentian Great Lakes

Aquatic hypoxia within the Laurentian Great Lakes has contributed to various adverse ecological consequences and stimulated research interest in recent decades. An analysis of published peer-reviewed journal articles from 2000 to 2020 demonstrates an increasing trend of studies related to hypoxia in the Laurentian Great Lakes. However, the majority of these studies (78%) focus on Lake Erie and in particular the well-documented hypolimnetic hypoxic conditions that develop in the central basin of Lake Erie. This hypoxic zone is relatively large (up to 1.5 million ha), has substantial ecological effects, and motivates monitoring programs and water quality improvement initiatives. Nonetheless, the hypoxic zone in the central basin of Lake Erie is only one of over twenty documented hypoxic zones in the Laurentian Great Lakes. Moreover, hypoxic conditions in the Great Lakes are quite diverse. Here, we define and characterize a four-fold classification of Great Lakes hypoxic conditions: 1) hypolimnetic hypoxia, 2) over-winter hypoxia, 3) diel hypoxia, and 4) episodic hypoxia. We suggest that Great Lakes research and monitoring programs should seek to more broadly document hypoxic conditions and develop models to predict the temporal and spatial occurrence of hypoxia. Such efforts are particularly timely as future climatic conditions contributing to warmer temperatures, longer and more intense stratified periods, increased spring nutrient loading and more variable allocthonous inputs are expected to exacerbate three of the four hypoxic conditions described for the Great Lakes (hypolimnetic, diel, and episodic hypoxia).     

Satellite SAR Remote Sensing of Great Lakes Ice Cover,Part 2. Ice Classification and Mapping

During the 1997 winter season, shipborne polarimetric backscatter measurements of Great Lakes (freshwater) ice types using the Jet Propulsion Laboratory C-band scatterometer, together with surface-based ice physical characterization measurements and environmental parameters, were acquired concurrently with Earth Resource Satellite 2 (ERS-2) and RADARSAT Synthetic Aperture Radar (SAR) data. This polarimetric data set, composed of over 20 variations of different ice types measured at incident angles from 0° to 60° for all polarizations, was processed to radar cross-section to establish a library of signatures (look-up table) for different ice types. The library is used in the computer classification of calibrated satellite SAR data. Computer analysis of ERS-2 and RADARSAT ScanSAR images of Great Lakes ice cover using a supervised classification technique indicates that different ice types in the ice cover can be identified and mapped, and that wind speed and direction can have an influence on the classification of water as ice based on single frequency, single polarization data. Once satellite SAR data are classified into ice types, the ice map provides important and necessary input for environmental protection and management, ice control and ice breaking operations, and ice forecasting and modeling efforts.     

Identifying and classifying Great Lakes ship transits using a state machine approach

Invasive, aquatic organisms have entered the North America Great Lakes from ships' ballast water, often originating from Europe. Current approaches for preventing the introduction of such organisms in ballast water include ballast water treatment (BWT) or ballast water exchange (BWE). This paper examines BWE, which is conducted in (1) waters >200 nautical miles (nm) from shore, or (2) waters >50?nm from shore and >200?m deep. We used historical records of ships transiting from Europe to the Great Lakes during one year (2014) to determine the duration (in days) that ships were within waters that met the criteria for BWE. Ship paths were classified based upon transitions between location-assigned “states” (e.g., from European waters across the North Atlantic Ocean to North America), and from these state transitions, four types of routes were identified. On average, ships sailing these routes had between 3.5 and 4.7 d to perform BWE in areas >200?nm from shore and 4.7 to 6.2 d when >50?nm from shore and >200-m deep water. Conducting BWE in daylight hours, if deemed necessary for safety, shortened the time window for BWE, especially in winter months, by approximately 50–70%. The state “machine” approach could, in the future, be used to identify ships from specific regions (e.g., ports within waterways at high risk of harboring potentially invasive species). Reshaping the definition of regional boundaries based upon time-of-year, water temperature, or other variables would further refine the ability to identify high-risk transits.     

Mapping invasive Phragmites australis in the coastal Great Lakes with ALOS PALSAR satellite imagery for decision support

The invasive variety of Phragmites australis (common reed) forms dense stands that can cause negative impacts on coastal Great Lakes wetlands including habitat degradation and reduced biological diversity. Early treatment is key to controlling Phragmites, therefore a map of the current distribution is needed. ALOS PALSAR imagery was used to produce the first basin-wide distribution map showing the extent of large, dense invasive Phragmites-dominated habitats in wetlands and other coastal ecosystems along the U.S. shore of the Great Lakes. PALSAR is a satellite imaging radar sensor that is sensitive to differences in plant biomass and inundation patterns, allowing for the detection and delineation of these tall (up to 5 m), high density, high biomass invasive Phragmites stands. Classification was based on multi-season ALOS PALSAR L-band (23 cm wavelength) HH and HV polarization data. Seasonal (spring, summer, and fall) datasets were used to improve discrimination of Phragmites by taking advantage of phenological changes in vegetation and inundation patterns over the seasons. Extensive field collections of training and randomly selected validation data were conducted in 2010–2011 to aid in mapping and for accuracy assessments. Overall basin-wide map accuracy was 87%, with 86% producer's accuracy and 43% user's accuracy for invasive Phragmites. The invasive Phragmites maps are being used to identify major environmental drivers of this invader's distribution, to assess areas vulnerable to new invasion, and to provide information to regional stakeholders through a decision support tool.     

Risk-based classification and interactive map of watersheds contributing anthropogenic stress to Laurentian Great Lakes coastal ecosystems

We describe development anthropogenic stress indices for coastal margins of the Laurentian Great Lakes basin. Indices were derived based on the response of species assemblages to watershed-scale stress from agriculture and urbanization. Metrics were calculated for five groups of wetland biota: diatoms, wetland vegetation, aquatic invertebrates, fishes, and birds. Previously published community change points of these assemblages were used to classify each watershed as ‘least-disturbed’, ‘at-risk’, or ‘degraded’ based on community response to these stressors. The end products of this work are an on-line map utility and downloadable data that characterize the degree of agricultural land use and development in all watersheds of the US and Canadian Great Lakes basin. Discrepancies between the observed biological condition and putative anthropogenic stress can be used to determine if a site is more degraded than predicted based on watershed characteristics, or if remediation efforts are having beneficial impacts on site condition. This study provides a landscape-scale evaluation of wetland condition that is a critical first step for multi-scale assessments to help prioritize conservation or restoration efforts.     

Sensitivity of fish density estimates to standard analytical procedures applied to Great Lakes hydroacoustic data

Standardized methods of data collection and analysis ensure quality and facilitate comparisons among systems. We evaluated the importance of three recommendations from the Standard Operating Procedure for hydroacoustics in the Laurentian Great Lakes (GLSOP) on density estimates of target species: noise subtraction; setting volume backscattering strength (S_v) thresholds from user-defined minimum target strength (TS) of interest (TS-based S_v threshold); and calculations of an index for multiple targets (N_v index) to identify and remove biased TS values. Eliminating noise had the predictable effect of decreasing density estimates in most lakes. Using the TS-based S_v threshold decreased fish densities in the middle and lower layers in the deepest lakes with abundant invertebrates (e.g., Mysis diluviana). Correcting for biased in situ TS increased measured density up to 86% in the shallower lakes, which had the highest fish densities. The current recommendations by the GLSOP significantly influence acoustic density estimates, but the degree of importance is lake dependent. Applying GLSOP recommendations, whether in the Laurentian Great Lakes or elsewhere, will improve our ability to compare results among lakes. We recommend further development of standards, including minimum TS and analytical cell size, for reducing the effect of biased in situ TS on density estimates.     

Satellite SAR Remote Sensing of Great Lakes Ice Cover,Part 1. Ice Backscatter Signatures at C Band

For remote sensing of Great Lakes ice cover, a field experiment campaign was conducted in the 1997 winter season across the Straits of Mackinac and Lake Superior. The campaign was coordinated in two expeditions on two different United States Coast Guard icebreaker vessels, the Biscayne Bay in February and the Mackinaw in March. Aboard these icebreakers, the Jet Propulsion Laboratory C-band polarimetric scatterometer was used to measure backscatter signatures of various ice types and open water at incidence angles from 0° to 60°. The radar measurements include incidence angles and polarizations of spaceborne Synthetic Aperture Radars (SAR) on ERS, RADARSAT, and Envisat satellites. The radar data together with in situ measurements form a signature library that can be used to interpret SAR data for ice classification and mapping. Results are presented for backscatter signatures of Great Lakes ice types from thin lake ice to thick brash ice with different snow-cover and surface conditions. The signature library indicates that several ice types can be identified with multi-polarization SAR data; however, single-polarization data can result in misclassification of ice and open water at different ranges of incidence angle and wind conditions. For incidence angles larger than 30°, thick brash ice, the most difficult for icebreaking operations and the most hazardous for ship navigation, can be uniquely identified by co-polarized backscatter for all wind conditions below the gale force.     

Mapping and monitoring the extent of submerged aquatic vegetation in the Laurentian Great Lakes with multi-scale satellite remote sensing

A satellite-based algorithm intended to map submerged aquatic vegetation (SAV), which was mostly the nuisance algae Cladophora, for the Laurentian Great Lakes has been developed and successfully demonstrated in test areas in Lakes Michigan and Ontario. The new Submerged Aquatic Vegetation Mapping Algorithm (SAVMA) first utilizes deep water (opaque) radiance values to correct shallow water values (transparent) so that depth invariant reflectance values for all three visible Landsat bands of the lake bottom can be classified. Combinations of two bands are then used to generate a depth invariant bottom type index. The algorithm then maps the lake bottom into three types: sand, dense SAV, less dense SAV by thresh-holding the depth invariant reflectance values. The SAVMA also generates a biomass estimate by assigning an average dry weight obtained by field sampling to both the dense and less dense SAV areas identified by the algorithm.The algorithm performance was successfully evaluated on Lake Michigan at the Sleeping Bear Dunes National Lakeshore (SBDNL) using Cladophora locations provided by diver survey as well as from an independent National Park Service monitoring. The SAVMA correctly mapped Cladophora to an approximate accuracy of 85%, where the misclassification was a result of mixed pixels due to the resolution of the Landsat data and imprecise atmospheric corrections. The algorithm was also successfully evaluated in Lake Ontario near Pickering, ON. The SAVMA was then used to generate both short and long term time-series analyses of Cladophora extent at SBDNL.     

Evaluation of regional climate simulations over the Great Lakes region driven by three global data sets

The performance of regional climate simulations is evaluated for the Great Lakes region. Three 10-year (1990–1999) current-climate simulations are performed using the MM5 regional climate model (RCM) with 36-km horizontal resolution. The simulations employed identical configuration and physical parameterizations, but different lateral boundary conditions and sea-surface temperatures derived from the NCEP Global Reanalysis and output from the CCSM3 and GISS general circulation models (GCMs). The simulation results are compared to the North American Regional Reanalysis (NARR). The three RCM simulations appeared to be more accurate in winter and least accurate in summer, and more accurate aloft than near the surface. The reanalysis-constrained simulation adequately captured the spatial distribution and seasonal cycle of the observed surface-air temperature and precipitation, but it produced consistently across all seasons a cold bias that is generally larger over the lakes than over land and a wet bias due to an overestimation of non-convective precipitation. The simulated seasonal cycle of moisture–flux convergence over the region was in very good agreement with NARR. The two GCM-driven runs adequately simulated the spatial and seasonal variation of temperature, but overestimated cold-season precipitation and underestimated summer precipitation, reversing the observed annual precipitation cycle. The GISS-driven run failed to simulate the prevailing low-level flow and moisture convergence patterns. All three RCM simulations successfully captured the impact of the Great Lakes on the region's climate, especially on winter precipitation, a significant improvement over coarse-resolution GCM simulations over the region.     

Deep Lake Explorer: A web application for crowdsourcing the classification of benthic underwater video from the Laurentian Great Lakes

Underwater video is increasingly used to study aspects of the Great Lakes benthos including the abundance of round goby and dreissenid mussels. The introduction of these species has resulted in major ecological shifts in the Great Lakes, but the abundance and impacts of these species have heretofore been underassessed due to limitations of monitoring methods. Underwater video (UVID) can “sample” hard bottom sites where grab samplers cannot. Efficient use of UVID data requires affordable and accurate classification and analysis tools. Deep Lake Explorer (DLE) is a web application developed to support crowdsourced classification of UVID collected in the Great Lakes. Volunteers (i.e., the crowd) used DLE to classify 199 videos collected in the Niagara River, Lake Huron, and Lake Ontario for the presence of round gobies, dreissenid mussels, or aquatic vegetation, and for dominant substrate type. We compared DLE classification results to expert classification of the same videos to evaluate accuracy. DLE had the lowest agreement with expert classification for hard substrate (77%), and highest agreement for vegetation presence (90%), with intermediate agreement for round goby and mussel presence (89% and 79%, respectively). Video quality in the application, video processing, abundance of species of interest, volunteer experience, and task complexity may have affected accuracy. We provide recommendations for future crowdsourcing projects like DLE, which can increase timeliness and decrease costs for classification but may come with tradeoffs in accuracy and completeness.     

Stimulating a Great Lakes coastal wetland seed bank using portable cofferdams: implications for habitat rehabilitation

Coastal wetland seed banks exposed by low lake levels or through management actions fuel the reestablishment of emergent plant assemblages (i.e., wetland habitat) critical to Great Lakes aquatic biota. This project explored the effectiveness of using portable, water-filled cofferdams as a management tool to promote the natural growth of emergent vegetation from the seed bank in a Lake Erie coastal wetland. A series of dams stretching approximately 450 m was installed temporarily to isolate hydrologically a 10-ha corner of the Crane Creek wetland complex from Lake Erie. The test area was dewatered in 2004 to mimic a low-water year, and vegetation sampling characterized the wetland seed bank response at low, middle, and high elevations in areas open to and protected from bird and mammal herbivory. The nearly two-month drawdown stimulated a rapid seed-bank-driven response by 45 plant taxa. Herbivory had little effect on plant species richness, regardless of the location along an elevation gradient. Inundation contributed to the replacement of immature emergent plant species with submersed aquatic species after the dams failed and were removed prematurely. This study revealed a number of important issues that must be considered for effective long-term implementation of portable cofferdam technology to stimulate wetland seed banks, including duration of dewatering, product size, source of clean water, replacement of damaged dams, and regular maintenance. This technology is a potentially important tool in the arsenal used by resource managers seeking to rehabilitate the functions and values of Great Lakes coastal wetland habitats.     

Ecological data as a resource for invasive species management in U.S. Great Lakes coastal wetlands

The coordinated use of ecological data is critical to the proper management of invasive species in the coastal wetlands of the Laurentian Great Lakes. Researchers and government programs have been increasingly calling for the use of data in management activities to increase the likelihood of success and add transparency in decision making. Web-enabled databases have the potential to provide managers working in Great Lakes coastal wetlands with relevant data to support management decisions. To assess the potential value of these databases to managers in Laurentian Great Lakes states, we surveyed wetland managers to determine their current data usage as well as their future data interests and catalogued the online databases currently available. Surveys were disseminated via email to managers in 56 different organizations overseeing invasive species management efforts in Great Lakes coastal wetlands; 46 responses were included in this analysis. Of the survey respondents, all reported using raw biotic data for decision making, (i.e. presence of target species) but many indicated that they would prefer to incorporate a greater variety of data, as well as more complex information. Our survey found that managers used web-enabled databases, but most databases that we catalogued only provided presence data for wetland biota. We concluded that databases can provide the types of data sought by invasive species managers but have unmet potential to be integrated into responsive management processes.