Managing the Impact of Multiple Stressors in Saginaw Bay

Multi-stressors logo


In 2007,NOAA's Center for Sponsored Coastal Ocean Research (CSCOR) housed under NOAA's National Centers for Coastal Oceans Science (NCCOS) awarded a grant to NOAA's Great Lakes Environmental Research Laboratory (GLERL) and an multidisciplinary team of research partners to conduct the project, Managing the Impact of Multiple Stressors in Saginaw Bay.

The NCCOS-funded project has supported 5 years of research (2008-2012) focused on the effect of multiple stressors on Saginaw Bay, a large embayment on the southwest side of Lake Huron (Figure 1). In the conduct of this research, GLERL scientists collaborated with partners from the University of Michigan's Cooperative Institute for Limnology and Ecosystems Research (CILER), Michigan State University, Michigan's Department of Natural Resources (DNR) and Department of Environmental Quality (DEQ), consulting firms and other agencies and academic institutions. This collaborative research project was designed to investigate the interactions among land, water, and biota, including invasive species, and how the Saginaw Bay ecosystem is changing in response to these interactions.

An important component of this research was application of the knowledge gained on how human-induced stressors (like nutrient loading, erosion and sedimentation, lake level declines, and invasive species) affect the Saginaw Bay ecosystem for the purpose of developing and implementing effective management solutions. The results are relevant for decision-making on management issues related to the water quality, invasive species, and the fisheries of Saginaw Bay.

The culmination of our 5-year research project is presented in the June 2014 special issue of the Journal of Great Lakes Research (JGLR): The Continuing Effects of Multiple Stressors in Saginaw Bay (Volume 40, Supplement 1, Pages 1-204). This JGLR issue provides an updated picture on the state of the Saginaw Bay at a time when the ratification of the 2012 updated Great Lakes Water Quality Agreement (GLWQA) provides a new catalyst for action, and continued funding under the Great Lakes Restoration Initiative offers hope that actions pursuant to the new Agreement can be accomplished.

Important Conclusions:


In 1972 the United States and Canada signed the GLWQA to protect the aquatic environment common to the two counties, known geologically as the Laurentian Great Lakes. This agreement expresses the intent of each country to restore and enhance the quality of a water resource that represents 20% of the world's fresh water. The GLWQA was revised in 1978 and again in 2012. One of the major thrusts of the agreement has been to solve the problem of eutrophication (high level of productivity of plant life in aquatic ecosystems) and its consequences. The GLWQA set forth an overall policy of reducing phosphorus loading to help mitigate the eutrophication related problems. The process of eutrophication involves increases in the population of certain algae, stimulated by the presence of excess phosphorus in a lake, leading to the depletion of oxygen in the water and the consequential deterioration of the lake.

The multiple stressor research study used the Adaptive Integrated Framework (AIF) methodology for managing impacts of multiple stressors in coastal ecosystems. In renegotiating the 2012 GLWQA, the target phosphorus load reevaluation was called for to provide information needed to guide resource managers charged with implementing the Agreement.


Figure 1: Satellite image of Saginaw Bay

Sidebar A: What do we mean by multiple stressors?

Harmful conditions usually associated with human activities, such as excess nutrients, erosion and sedimentation, abnormal temperature changes, invasive species, toxic contaminants, and overfishing. The combined effect of these stressors impairs ecological health and causes the loss of many valued ecosystem features and services.

Like many coastal areas around the world, Saginaw Bay is negatively impacted by changing conditions associated with human activity both on the land and water. Negative "stressors" on the bay include excess nutrients, erosion and sedimentation, variable lake levels, invasive species, toxic contaminants, overfishing, among others. The combined effect of these stressors has impaired the bay's water quality and ecological health, resulting in the loss of many valued ecosystem features and services including: The water quality and ecological impairments caused by multiple stressors pose complex, sometimes unpredictable challenges for ecosystem management and stewardship. Our multiple stressors research project on Saginaw Bay was designed to help us better understand and predict the ecosystem dynamics-critical in developing science-based management strategies to overcome these challenges.

Problems Caused by Multiple Stressors in Saginaw Bay

Figure 2: Saginaw Bay and the bay's overall location in the Great Lakes.
Credit: NOAA

Saginaw Bay is a shallow, productive (elevated plant growth), and highly stressed bay that extends from the western shore of Lake Huron (Figure 2). More than half of the land use in the area surrounding the bay is agricultural, and the entire Saginaw Bay watershed comprises approximately 15% of the total land area in Michigan.


The effect of multiple stressors in Saginaw Bay has increased over the years as human land use has intensified. The run-off from the land surrounding Saginaw Bay transports nutrients (e.g. phosphorus and nitrogen contained in fertilizer and other sources), sediments, and toxic contaminants (e.g. PCBs) into Saginaw Bay. Among the most serious problems caused by pollutant run-off from the land includes excessive algal growth, degradation of fish and wildlife populations, and human health risks. While designated as an Area of Concern (AOC) by the International Joint Commission (IJC), Saginaw Bay remains a freshwater resource for about 500,000 people, providing beneficial uses such as drinking water, recreation such as swimming, fishing, and boating, and aesthetics, among other benefits.


A priority concern in the Saginaw Bay watershed is the run-off of phosphorus-applied to farmland in fertilizers to maximize crop growth-draining from the land into the bay. As phosphorus stimulates plant growth for crops, this nutrient also stimulates algal growth in the bay. While some phosphorus is needed in the water for the growth of aquatic plants, too much of this nutrient can stimulate the growth of nuisance and toxic algae, namely cyanobacteria. Historically, in response to this water quality problem plaguing Saginaw Bay and other areas of the Great Lakes, the GLWQA established an aggressive phosphorus reduction program in 1978. The GLWQA provides guidance on strategies to reduce the amount of phosphorus used on farmland and the discharge of phosphorus from municipalities and industrial sources. Under the 1978 GLWQA, phosphorus levels in detergents were also reduced.

As a result of less phosphorus draining into the bay, the frequency and occurrence of harmful algal blooms (caused by cyanobacteria) appeared to decrease in the late 1980s and early 1990s. This improvement, however, was short-lived, and within a decade's time, large blooms of cyanobacteria were found, including toxic colonial cyanobacteria, and benthic filamentous algae, such as Cladophora. In addition, human health risks emerged with the production of toxins by some strains of the cyanobacteria, such as Microcystis, that can affect the liver and skin.

Lake Level Declines

Another stressor contributing to water quality problems in Saginaw Bay has been the decline of water levels in Lakes Michigan and Huron that began in the late 1990s. Evidence suggests that a primary driver for this decline is directly related to increased evaporation in the Great Lakes watershed, a possible consequence of global climate change. In the summer, on the western shore of the inner bay, algae washes up on the expanding beach area and then decomposes. The accumulation of decomposing algae, referred to as "muck," can cause taste and order problems and degrade aesthetic qualities of the bay as well as the economic value of shoreline property.

Sidebar B: Benthic macroinvertebrates are organisms characterized by the absence of a backbone, and visible to the eye without the aid of a microscope. The term "benthic" means "bottom-living", organisms that live on, under, and around rocks and sediment on the bottoms of lakes, rivers, and streams.

Figure 3: Muck along the shoreline in Saginaw Bay near Bay City.
Credit: NOAA (Vijay Kannappan)

Invasive Species

The establishment of non-native species, such as invasive mussels, round goby, and the spiny waterflea, is another major source of stress in Saginaw Bay impacting the food web structure of the ecosystem. Research indicates that invasive quagga and zebra mussels (dreissenid mussels) have fundamentally changed the ecosystem by filtering large quantities of phytoplankton (microscopic green aquatic plants) from the water. As a result, native populations of zooplankton (microscopic aquatic animals) and benthic macroinvertebrate communities (Sidebar B) that would normally feed on phytoplankton have declined, thus decreasing the amount of food available for many native fish species. For instance, the growth of juvenile yellow perch is limited, which impacts the recreational fishery.

The filtration of phytoplankton by dreissenid mussels from the water column has increased water clarity in Saginaw Bay. While it may sound like good news that the water is clearer, the bad news is that, because of the clearer water, more sunlight reaches the bottom of the shallow bay, stimulating more algal growth. As a result, algae accumulate in the shallow areas along the shoreline and then decay, contributing to the problem of muck accumulation (Figure 3).

Persistence of Water Quality Problems in Saginaw Bay

To mitigate the problems associated with high phosphorus levels, the 1978 amendment to the GLWQA established a 440 metric tons/year Total Phosphorus (TP) target load in the Great Lakes region. Following the 1978 GLWQA, the reduced levels of phosphorus, assessed along most of the Great Lakes shorelines, were thought to have significantly reduced beach fouling (deposition of algae on the beaches) in the 1980s-1990s. However, this was not the case in the Saginaw Bay area. Park ranger logs from the Bay City Recreation Area (BCRA), located northwest of the mouth of the Saginaw River (Figure 2), reveal that muck continued to be a regular problem throughout the 1980s and early 90s. Water quality problems that were observed in the bay during this time frame included algal blooms as well as drinking water taste and odor problems associated with muck.

Sidebar C: Top Research Objectives
  • Water Quality
    • Predict and manage muck deposition on beaches
    • Predict and manage E. coli/pathogens outbreaks
    • Determine what type of management efforts or policy changes would be effective in reducing the impacts of contaminants in Saginaw Bay (i.e. dredging of hot spots)
    • Manage sediment loading
    • Manage and understand the impacts of agriculture in Saginaw Bay (i.e. nutrient loads, sedimentation, E. coli)
  • Fisheries
    • Management of dominant percid (walleye and yellow perch) populations and their associated fisheries
    • Manage for a diverse ecosystem, fish community, and fishery
    • Restoration of native species and control of non-natives
    • Manage for long-term sustainability of fisheries
Recognizing these persistent problems in Saginaw Bay and other areas of the Great Lakes, the GLWQA was updated in 2012 to mandate TP load reduction targets for watersheds that have a "significant localized impact" on nearshore waters of the Great Lakes. The results from our research have provided important data, information and sampling experience that can be used to evaluate progress in reaching the phosphorus loading target established under the 1978 GLWQA and the updated 2012 Agreement.

Research Summary: Managing the Impact of Multiple Stressors in Saginaw Bay

The following research summary provides an overview of NOAA GLERL's five-year project, Managing the Impact of Multiple Stressors in Saginaw Bay. Detailed descriptions of the approaches, methods, results, and conclusions of this research are available in papers published in the peer-reviewed literature and also in the June 2014 special issue of the JGLR, The Continuing Effects of Multiple Stressors in Saginaw Bay (Volume 40, Supplement 1, Pages 1-204).

Research Goals and Objectives

Project Timeline: 2008-2012

Project Design

GLERL collaborated with the University of Michigan Cooperative Institute for Limnology and Ecosystems Research (CILER), Michigan State University, Michigan DNR and DEQ, and other project partners to determine the combined effect of the stressors at work in Saginaw Bay. Data were collected to examine how these stressors interact to impact the Saginaw Bay ecosystem. An important aspect of this research was determining the quantitative relationship between stressors and their cumulative impact on Saginaw Bay, as estimated through the development of models. For example, researchers calculated the mathematical relationship between the measured amounts of phosphorus loading into the bay and changes in the level of the algal growth over time. This model can be used to evaluate the effect of different management practices related to phosphorus loading.

Figure 4: "Adaptive Integrative Framework" (AIF): A continuous feedback loop interconnecting management actions and scientific understanding of observed changes in the ecosystem

The AIF approach (Figure 4) was useful in studying the response of the Saginaw Bay ecosystem to multiple stressors, providing overall guidance to build and apply modeling to better understand and forecast the cumulative effect of climate change, land use, and invasive species on the ecosystem. Specifically, project researchers and partners examined the cumulative effect of multiple stressors on important management goals and objectives established under the GLWQA as related to water quality, human health, and fish production. They also used the AIF to integrate and organize complex data and information to advance our understanding of current ecosystem changes and forecast future conditions. This information is considered valuable in the process of making sound management decisions. The feedback loop integrated in the AIF serves to evaluate the effectiveness of management practices and identify the adjustments needed to make improvements.

Another important element was engagement with community stakeholders. Throughout the life of this project, research was conducted in coordination with local organizations under the guidance of representatives from Michigan DNR and DEQ. The interactive environment established served in transferring scientifically-based information from researchers to resource managers and policy makers. In addition, interactions with the local stakeholder groups helped in sharing research results with the public and establishing a better understanding of the goals and challenges involved in effectively studying, understanding, and managing Saginaw Bay.

Project Team

Project Findings

In addition to enhancing our understanding of the interactions in Saginaw Bay and the surrounding watershed, findings from our research will provide guidance in the decision making process to improve the bay's water quality and ecosystem health. Monitoring focused specifically on the following issues:

1. Total Phosphorus Loading

The assessment of total phosphorus (TP) loading into Saginaw Bay was an important component of the multiple stressor project, given the long history of eutrophication problems, particularly in the inner bay (Figure 5). These problems have included excessive algal production — including cyanobacterial (blue green algae) blooms — drinking water taste and odor issues, and decaying organic matter deposits along the beaches. Phosphorus loading has been identified as the major stressor causing excessive algal production, leading to eutrophication (Sidebar D).

Sidebar D: Process of Eutrophication Total Phosphorus (TP) is an essential nutrient for plants and animals, but in high, or eutrophic, levels, it can negatively impact the water quality. Eutrophication can include algal blooms, accelerated plant growth, and low dissolved oxygen from the decomposition of additional vegetation. An acceptable range for TP is 10 μg/L to 40 μg/L. Probable contributors to TP eutrophication include run-off from croplands and fertilized lawns, wastewater treatment plants, run-off from animal manure storage areas, disturbed land areas, drained wetlands, and decomposition of organic matter.
Credit: U.S. Environmental Protection Agency Factsheet: Total Phosphorus

Figure 5: Map of Saginaw Bay showing the location of sampling sites for TP and chlorophyll a. Also shown is the separation of the bay into an inner and outer region (denoted by a dashed black line). This separation is due to physical features that make the two sections distinct. The outer bay (mean depth of ~14 meters) is cold and clear with water quality characteristics similar to the main body of Lake Huron. The inner bay (mean depth of ~ 5 meters) is warm, eutrophic, and turbid (cloudiness of water sample) with thermal stratification of the water.

Our research on phosphorus loading into Saginaw Bay for the multiple stressor project was focused on evaluating the current status of the bay's water quality, particularly in the context of the objectives established under the 1978 GLWQA to address eutrophication problems in the Great Lakes:

Annual TP Loading into Saginaw Bay

The calculations of phosphorus loading (mass per unit time) were based on both stream flow (volume/time) and phosphorus concentration (mass/volume) measurements in the tributaries entering Saginaw Bay. As the largest tributary entering Saginaw Bay, the Saginaw River had the best data records, contributing an estimated 70-80% of the TP load into the Bay. In monitoring for TP, there was limited availability of both flow and TP concentration measurements in the tributaries entering Saginaw Bay. As a result of this sparse data record, annual TP load estimates were considered highly uncertain. To estimate annual TP loads over time (1968-2010) and quantify the level of uncertainty of the loading calculations, a Bayesian hierarchical model (Figure 6) was developed for the Saginaw River.

Overall, results indicated that the estimated annual Saginaw River TP load (updated through 2011) generally exceeded the GLWQA target of 440 metric tons (shown by the horizontal orange line in Figure 6). Compliance with the GLWQA fell short, even when TP loading data was based solely on the Saginaw River (one of several tributaries flowing into Saginaw Bay). While results did indicate decreased TP loading for a limited period following the 1978 GLWQA, these declines ceased by the early 1990s, showing considerable yearly variability observed since then.

Also shown in Figure 6 are the preliminary predictions made for TP data (red squares) for 2010 and 2011, using only Saginaw River flow data before TP concentration measurements were available. When TP concentration data did become available, the estimates for both years (2010 and 2011) were updated, incorporating the actual TP measurements. The results indicated that the Bayesian hierarchical model provided preliminary predictions on TP concentration (using Saginaw River flow data) that fell within uncertainty limits.

TP declines (Figure 6) observed early in the late 1970s and early 1980s have been largely attributed to decreased TP concentrations in the tributaries when regulatory controls of point sources (e.g. waste water treatment plants and industrial discharges) became effective. The yearly variability seen from approximately the mid-1980s has been attributed largely to differences in the annual river flow. For instance, the lowest annual TP loads occurred in the late 1990s to early 2000s, a period when the river was experiencing slightly lower flows.

Figure 6: Estimates of the Saginaw River Annual TP Loads Over Time (1968 to 2010) - The blue dots in the graph above represent the mean of the Bayesian predictive distribution, bars represent +/- 2 standard deviations (variation from the mean). Red squares and lines indicate predicted estimates for 2010 and 2011using Saginaw River flow data without TP concentration data. The horizontal orange line denotes 440 metric tons/year target phosphorus load established under the 1978 amendment of the GLWQA.

Results indicated that following the TP abatement program under the 1978 amendment of the GLWQA, the high levels of TP loading seen in the late 1960s had decreased by the early 1980s. By the early 1990s, however, the declines in TP ceased, with considerable yearly variability since that timeframe.

It is important to recognize that in conducting this research, TP monitoring was limited to 10-12 samples per year. To help resource managers determine with more certainty how much TP is actually loading into Saginaw Bay, it would be necessary to increase the frequency of monitoring. As an outcome from this research, it has been recommended that resource managers determine the level of monitoring needed to reach the certainty levels that would improve decision making for the management of TP loading into the Saginaw Bay watershed.

Another important finding from the multiple stressor research, based on modeling calculations, was that point sources of TP contributed a relatively low proportion of the total annual phosphorus load to the bay. Most of the TP was found to come from non-point sources, however, the distribution of these sources within the watershed was not well-characterized (Sidebar E).

Total Phosphorus and Chlorophyll a Concentrations:

Figure 7: The data plots compare total phosphorus (TP) concentrations (upper) with chlorophyll a concentrations (lower) versus time. These plots indicate that both TP and chlorophyll a declined in the late 1970s to the early 1980s approximately in concurrence with decreasing TP loads. Since the mid 1980s, concentrations of both have been fairly stable.

To evaluate the effect of TP loading on water quality in Saginaw Bay, researchers compared concentrations of TP and chlorophyll a from 1968-2010 (Figure 7). Chlorophyll a is a photosynthetic pigment used as an index to estimate phytoplankton biomass and related levels of algal productivity (rate of plant growth). A direct relationship exists between the concentration of TP and phytoplankton production, implying that increased phosphorus loading causes higher levels of phytoplankton production and chlorophyll a. This direct relationship also implies that phosphorus is a limiting factor to phytoplankton production; that is to say, decreased TP loading results in reduced algal growth.

As shown in Figure 7, monitoring results for the inner bay (see Saginaw Bay map - Figure 5) indicated that both TP and chlorophyll a concentrations declined in the late 1970s to the early 1980s, approximately in concurrence with estimated TP loads reductions. While the concentrations of TP and chlorophyll a have been relatively stable since approximately mid-1980s, their concentrations have not met the GLWQA objectives (denoted by the orange horizontal line in Figure 7) of 15 micrograms/liter of phosphorus and 3.6 micrograms/liter of chlorophyll a.

Additional results shown in Figures 8 and 9 indicate that both TP and chlorophyll a concentrations were higher in the river plume samples than the non-plume samples from the inner bay. It was also found that outer bay concentrations of TP and chlorophyll a are considerably lower than inner-bay concentrations. Data collected from 2008 to 2010 in the inner bay indicate that the GLWQA target of 15 microgram/L TP was exceeded 47% within the river plume and 32% in the non-plume samples, while no outer bay samples exceeded the limit. These results demonstrated the primary role of the Saginaw River in contributing to increased levels of TP loading and chlorophyll a concentrations.

Figure 8: Box and whisker plot sample distribution of TP in various time periods and locations in Saginaw Bay. Boxes indicate middle 50% of observations, dots indicate sample averages, horizontal lines indicate sample medians, whiskers indicate extreme values, box width is proportional to sample size. Sample size (N) and sample concentration averages (TP microgram/liter) are shown above the plot. Horizontal dashed line indicates the 15 micrograms/liter reference concentration based on the GLWQA objective for TP

Figure 9: Box and whisker plot sample distribution of chlorophyll a in various time periods and locations in Saginaw Bay. Boxes indicate middle 50% of observations, dots indicate sample averages, horizontal lines indicate sample medians, whiskers indicate extreme values, box width is proportional to sample size. Sample size (N) and sample concentration averages (chlorophyll a microgram/liter) are shown above the plot. Horizontal dashed line indicates the 3.6 micrograms/liter reference concentration based on the GLWQA objective for chlorophyll a.

Sidebar E: Point and Non-Point Source Pollution

A point source of pollution is a single, identifiable source of pollution such as a pipe or a drain. Industrial wastes are commonly discharged to rivers in this way.

Non-point sources of pollution are often termed 'diffuse' pollution and refer to those inputs and impacts from a wide area and are not easily attributed to a single source. They are often associated with particular land uses. In urban areas, for example, stormwater run-off is often contaminated with car oil, dust, and animal feces. In rural areas, where farming occurs, non-point sources of pollution include pesticides, fertilizers, animal manure, and soil that drains into streams in rainfall run-off.

U.S. Environmental Protection Agency Fact Sheet: What Is Nonpoint Source Pollution?

Sidebar F: Chlorophyll a is a photosynthetic pigment in green plants and is used as an index to estimate phytoplankton biomass and related levels of algal productivity (level of plant growth).

Sidebar G: River Plume

When a stream empties into a lake, a distinct color difference can be observed between the river and lake water. The area that appears to be an extension of "river-colored" water into a lake is called a "plume." The plume appears more brown in color than the lake because the river carries more suspended particles such as silt and clay.

Above is a 2013 aerial view of Saginaw Bay with visible river plumes flowing into Saginaw Bay.

Credit: Partnership for the Saginaw Bay Watershed
Overall, results from our multiple stressors research have indicated that a well-defined, direct relationship exists between chlorophyll a and TP concentrations, implying that phosphorus continues to be a limiting factor for primary production in Saginaw Bay. It follows that while increased loading of phosphorus will result in increased levels of phytoplankton growth, a reduction in phosphorus input from the watershed will reduce phytoplankton growth. Our researchers have recommended that future work in determining the mathematical relationship between phosphorus loading, chlorophyll a concentrations, and phytoplankton growth may provide valuable guidance in developing enforceable, numerical phosphorus loading objectives called for in the 2012 amendment of the GLWQA.

An unexpected finding from our research was the relationship found between dreissenid mussel invasion and the concentrations of TP and chlorophyll a in Saginaw Bay. The zebra mussel (Dreissenid polymorpha)-first discovered in Saginaw Bay in 1991-reached its peak population in the early 2000s, which then declined as the quagga mussels (Dreissena rostriformis bugensis) moved in (Figure 21).

In Saginaw Bay, dreissenid invasions have generally been associated with declines in TP and chlorophyll a concentrations, particularly in those areas not affected by the Saginaw River plume. These declines have been attributed to the mussels' filtration of phytoplankton (containing phosphorus and chlorophyll) from the water column. The nearshore shunt hypothesis suggests that as dreissenids consume phytoplankton in the nearshore, the phosphorus becomes "trapped" within the tissue of the mussels, thus reducing levels of phosphorus available for export to the offshore zone. This shunt process provides an explanation for the low levels of phosphorus found in the outer bay of Saginaw Bay.

2. Saginaw Bay Muck: What We've Learned and What's Unknown

In Saginaw Bay, beach fouling material, referred to as muck, is composed of different types of decomposing algae, macrophytes and unidentifiable organic matter (Figure 3). The reduction in phosphorus levels, following the 1978 GLWQA, significantly diminished the level of beach fouling caused by the accumulation of muck in many areas of Great Lakes shoreline, as observed during the 1980s-1990s. However, in Saginaw Bay, this was not the case. The log observations maintained by a park ranger in the BCRA( located northwest of the mouth of Saginaw River) (Figure 10), revealed that "thick, stinky muck" was a regular problem in Saginaw Bay throughout this interim period. The ranger also mentioned wind as a factor causing muck accumulation on the beach.

Overall, the descriptions in the BCRA log suggest that the phosphorus reductions attained in Saginaw Bay (following the target load reductions under the 1978 GLWQA) did not result in a decrease in beach muck comparable to observations of other Great Lakes beaches during the same timeframe.

Muck Study: To determine the dynamics of beach fouling from muck in Saginaw Bay, researchers conducted surveys of beach muck and benthic filamentous algae in Saginaw Bay based on the following objectives:

Specifically, we examined how light and nutrient conditions influenced benthic algal growth to help inform management decisions in addressing problems caused by muck accumulation on the beaches of Saginaw Bay.

Beach Survey: Each summer from 2009 to 2012, an observational study of 7-10 sites was conducted along the southwestern Saginaw Bay shoreline, between the Saginaw River and Linwood, Mich. (Figures 10 and 13). At these sites, researchers monitored the volume, frequency, and composition of detrital wash-up of muck (Figures 11 and 12).

Figures 11 and 12: Surveying muck on the beach of Saginaw Bay.
Credit: Scott Peacor, Michigan State University

Figure 10 (top): Muck survey sampling sites along the southwest shoreline of Saginaw Bay, as denoted.

Figure 13 (bottom): Locations identified in Saginaw Bay where frequent, severe shoreline fouling events occurred. Also depicted is the region of benthic algal growth identified by bay-wide observations.
Credit: NOAA.

Our results suggest that muck composition differs across the summer season and between years. We observed the presence of Cladophora, other types of filamentous green algae (e.g. Mougeotia, Oedogonium, Spirogyra, Zygnema, Chara), as well as macrophytes and wood chips in varying proportions over time. We also observed a difference in the state of decay of the beach muck, affecting our ability to identify the composition of the muck over the study period. Along areas of severe deposition, primarily near the BCRA, fresh muck deposition with identifiable composition occurred early in the season and only occasionally thereafter. More often, we found muck deposition that contained material so badly decomposed that the composition could not be determined. However, as the survey moved northward along the shoreline, the detritus was less decomposed and more consistently identifiable.

Sidebar H: Comparison of the quagga (left) and zebra mussel (right). The quagga mussel has been able to displace the zebra mussel primarily due to its ability to colonize soft substrate, such as sandy lake bottoms. Zebra mussel colonization is confined to hard surfaces, such as rocky bottoms and concrete structures.

Sidebar I: Macrophytes:

Macrophytes are aquatic plants that grow in or near the water. They provide cover for fish and habitat for aquatic invertebrates, produce oxygen and act as food for some fish and wildlife. However, excessively high nutrient levels may create an overabundance of macrophytes, causing the accumulation of muck and other eutrophication problems.

Sidebar J: Examples of two species of algae, Spirogyra and Cladophora found in Saginaw Bay.
Credit: Scott Peacor, Michigan State University

Sidebar K: The benthic community includes organisms living in the benthic zone, the ecological region at the lowest level of a body of water, including the sediment surface and some sub-surface layers. Benthos, members of the benthic community, generally live in close relationship with the substrate bottom.

FIB are bacteria such as E. coli and Enterococci, which live in the gut of warm blooded animals and are introduced into the environment through fecal matter. While most FIB are harmless to humans, the presence of FIB indicates that pathogens also found in fecal matter and harmful to humans may be present.

Figure 14: How old is muck? Results from the beach surveys indicated that muck originates from algae produced in the same year.
Credit: Scott Peacor, Michigan State University

Our researchers found that muck originates from algae produced in the same year. It was also observed that Spirogyra composition shifted to Cladophora as the season progressed (Sidebar J). In characterizing beach fouling muck, we found that the material was dominated by different sources at different times, reflecting the diverse benthic community (Sidebar K) of the bay.

While we discovered extensive amounts of Cladophora in the bay that sometimes constituted a significant proportion of the muck, the observed muck deposits on the beach often contained other types of filamentous algae as well as large amounts of decaying macrophytes. The evidence for varied composition of muck in Saginaw Bay suggested a more complex process than in most other Great Lakes areas where the source of muck was found to be primarily composed of Cladophora.

Saginaw Bay Survey: In 2009-2010, our researchers conducted a survey to determine the location, composition, and factors limiting the growth of the benthic filamentous algal community of Saginaw Bay. The study area was an oblong region, approximately 20 km long and 3.5 km wide in the southwest region of the inner bay, extending 5 km west of the Saginaw River to the northern reach of Pinconning, Mich., ranging from 2.0 to 5.0 meters in depth (Figure 13). This region was chosen for both accessibility and optimal growth of Cladophora and other benthic algae.

In July of 2009, sloughed Cladophora was observed in some areas causing 100% coverage of the benthos (bottom dwelling organisms) (Figure 15). During this same time period, our beach observations indicated that muck deposition events included substantial amounts of degraded Cladophora which was not found elsewhere in the bay. These observations suggest that the decaying Cladophora on the beach originated from muck precursors from the sampling area in the southwest region of the bay (Figure 15 and 13). Additionally, circulation modeling indicated that currents moving over this area of heavy algal growth converge north of the Saginaw River, where muck deposition is most severe (Figure 13).

Figures 15: Underwater photos of benthic green filamentous algal growth at a sampling site near Pinconning, Mich.: 4.0 m depth on 6/14/2010.
Credit: Scott Peacor, Michigan State University

An important focus of the 2009-2010 survey in Saginaw Bay was the analysis of benthic filamentous algae collected from the southwest quadrant of the bay to determine if light and phosphorus were growth limiting factors. Benthic light conditions and tissue phosphorus measurements were made on algal samples from a variety of light and phosphorus conditions. Measurements-compared to published light and phosphorus levels documented to cause stress-indicated deficiencies in both light and phosphorus levels. These results suggest that the benthic algae in the southwest region of the bay were both light and phosphorus stressed, limiting the growth of the algal muck precursors.

Another finding of particular interest generated from our study was the increased light penetration resulting from the dreissenid mussel invasion. The mussels' filtration of algae from the water column increased the clarity of water, allowing for light to reach the benthos at greater depths. Results indicated that the depth of potential Cladophora habitat increased approximately 1.75 meters due to the dreissenid invasion.

The management implications from the results of this study are mixed with both good and bad news. Since phosphorus limitation was shown to be common among benthic filamentous algae, primarily Cladophora, there is potential for these muck sources to be controlled with further reductions of phosphorus from nonpoint sources. Results, however, also indicated that there were numerous constituents in muck, including macrophytes. The presence of macrophytes could complicate efforts to control muck fouling of beaches, since this aquatic weed obtains its source of phosphorus from in-lake sediments rather than more controllable nonpoint sources.

Fecal Indicator Bacteria: Muck not only causes problems to the public as a nuisance weed, but the microorganisms our researchers found harbored in muck could pose human health risks. In conjunction with the beach survey discussed above, muck samples were analyzed for fecal indicator bacteria (FIB). The presence of FIB indicated the possible presence of pathogens that are also found in fecal matter and can be harmful to humans. Results indicated that the wet muck collected near the water's edge tested high for FIB levels. In sampling dry muck collected further up on the beach and in beach sand, the levels of FIB were lower than in the wet muck. We also observed that muck samples experimentally exposed to sunlight for several days showed declining FIB levels over time (Figure 16). These results suggest that wet muck harbors FIB and may be a bacteria source to nearby areas.

Figure 16: Effect of 96 hours of sunlight exposure on the persistence (survival) of several Fecal Indicator Organisms (FIBs; E. coli, Enterococci, and C. perfringens) within freshly deposited wet muck collected from Tobico Beach, at the BCRA. In this experiment, a mat of muck was moved to an area of direct sunlight and away from possible contamination from wave action. The muck was spread to a thickness of 15 cm, and 5 g was collected daily from the start of the experiment August 27-31, 2010 and analyzed for FIB.
Credit: Scott Peacor, Michigan State University

3. Invasive Species and Foodweb Dynamics

Since the 1970s, Saginaw Bay has experienced the introduction and spread of numerous non-native aquatic invasive species, including spiny waterflea (Bythotrephes), white perch (Morone americana), round goby (Neogobius melanostomus), and zebra/quagga mussels (Dreissena polymorpha and D. bugensis), among others. Moreover, the alewife (Alosa pseudoharengus), an invasive fish species that has been a dominant component of the ecosystem since the 1950s, crashed in 2003 and has been essentially absent from the system ever since.

Figure 17: Great Lakes Aquatic Invasive Species.
Credit: NOAA

Results from the multiple stressors project indicate that the invasion of non-native species coupled with other negative stressors (e.g., chemical contamination, over-fishing, nutrient and sediment loading) have taken a toll on Saginaw Bay's ecosystem. Key findings from our research demonstrate the following changes in the foodweb.

Zooplankton Community (add photo panel: calanoid copepods, daphnia, Bosminidae, Leptodora, Bythotrephes)

Our researchers observed a significant shift in the zooplankton community from the 1990s to 2009-2010. Among these changes is a shift in the dominant species in the community and seasonal differences in overall abundance and size distribution. For example, Calanoid copepods increased in proportional abundance as did some of the large-bodied taxa, such as Daphnia galeata. Relatively small Bosminidae were the most abundant taxa during both time periods, but their proportional abundance was greater during the 1990s than 2009-2013.

Figure 18: Monthly trends in zooplankton abundance and composition between the two time periods of 1991-1996 and 2009-2010.
Credit: Pothoven, S.A., T.O. Hook, T.F. Nalepa, M.V. Thomas, and J. Dyble. Aquatic Ecology.

Overall, the Saginaw Bay zooplankton community appears to have shifted from species that are tolerant of eutrophic conditions (increased nutrient levels cause increased productivity of the system, such as phytoplankton growth) to those species more inclined to inhabit oligotrophic conditions (lower nutrient levels decrease productivity of the system and water clarity increases). Both reduced nutrient loading, and dreissenid filtering of the water may have contributed to this "oligotrophication" of the zooplankton assemblage (e.g., increased abundance of calanoid copepods).

Sidebar M - Zooplankton (click image to enlarge): Zooplankton are small aquatic organisms with animal-like traits that are suspended in the water and either drift with the currents or swim weakly. The name plankton comes from the Greek word 'planktos' which means 'wanderer' or 'drifter'. Zooplankton are a vital component of freshwater food webs. The smallest zooplankton are eaten by the larger zooplankton which, in turn, are eaten by small fish, aquatic insects and so on. Herbivorous zooplankton graze on phytoplankton or algae, and help maintain the natural balance of algae.

To view photos of Great Lakes zooplankton, refer to: Great Lakes Water Life Photo Gallery: Crustacean Zooplankton of the Great Lakes

Figure 19: Percent composition of zooplankton community between two time periods in Saginaw Bay indicating a shift to more oligotrophic zooplankton assemblage.
Key: CYC: cyclopoid copepods, CAL: calanoid copepods, BO: Bosminidae, DA: Daphnia, CLAD: other cladocerans.
Credit: Aquatic Ecology: Pothoven, S.A., T.O. Hook, T.F. Nalepa, M.V. Thomas, and J. Dyble

In addition, Saginaw Bay's herbaceous zooplankton community appears to have shifted in terms of their seasonal abundance. During the 1990s, zooplankton abundance was greatest in late spring and early summer (June), whereas during 2009-2010, zooplankton abundance was greatest during the fall (September-October). This shift may have important implications for survival and subsequent recruitment of young fish, such as yellow perch, which rely on zooplankton as prey during their early life stages in late spring and early summer.

Predatory zooplankton patterns also have shifted in Saginaw Bay from the 1990s to 2009-2010. Two taxa of predatory zooplankton (Leptodora and Bythotrephes) were present in the ecosystem during both of these time periods. While the densities of the invasive Bythotrephes increased significantly between the two time periods, the densities of Leptodora decreased. While Leptodora peaked in abundance during June in both 1991-1996 and 2009-2010, a subsequent peak in late summer was only evident in the 1990s. The high density of Bythotrephes likely contributes to increased predation on herbaceous zooplankton species.

Figure 20: Increase in Bythotrephes density from the time period of 1991-1996 to 2009-10
Credit: Aquatic Ecology: Pothoven, S.A., T.O. Hook, T.F. Nalepa, M.V. Thomas, and J. Dyble

Benthic Invertebrates

Researchers collected benthic invertebrates (bottom dwelling aquatic animals without a backbone) on a monthly basis from several locations in Saginaw Bay. They found greater spatial than temporal (e.g., seasonal) variation in benthic invertebrate assemblages. The dominant invertebrates included the quagga mussels and chironomids. For the most part, Saginaw Bay continues to be dominated by benthic invertebrate species tolerant of eutrophic and poor water conditions. However, some sensitive species, including the burrowing mayfly Hexagenia, were documented. These results indicate that the benthic community may be marginally responding to decreased nutrient loading and improved water quality.

Another significant change our researchers observed in Saginaw Bay between the early 1990s and first decade of the 21st century, is a significant decline in the annual dreissenid mussel density. In 1992, the mussel population in Saginaw Bay peaked at 31,334 mussels/meters2. By 2009, their densities fell precipitously to 421 mussels/meters2 (Figure 21). A potential factor contributing to this change is predation by the invasive fish round goby. However, preliminary bioenergetics estimates of goby consumption suggest that such predatory control is unlikely. Another observed change to the invasive mussel population over this time span is the shift of species composition from zebra to quagga mussels.

Figure 21: Annual Dreissenid Mussel Density

Fish Community: 1970-2011

The Saginaw Bay fish community changed dramatically from 1970 to 2011, as documented in the annual fall trawling survey conducted by the Michigan DNR during this period. By analyzing data from this long-term survey, we documented changes in the relative abundances of many fish species in Saginaw Bay. From 1970 to 2011, the frequency of occurrence of many fish species highly tolerant of eutrophic conditions declined, while occurrence of species moderately tolerant or sensitive to eutrophic conditions increased (Figure 22). These shifts in the Saginaw fish assemblage have occurred coincidentally with declines in nutrient loading and overall productivity in Saginaw Bay. In addition, the observed shifts in assemblage composition (e.g., reductions in eutrophic tolerant species) are indicative that the fish community has responded to improved water and habitat quality related to diminished eutrophic conditions.

Figure 22: Abundance of fish species in response to a range of eutrophic conditions.
Credit: Hook, T.O.

In addition to analyzing community changes, we specifically focused our analyses of data from the DNR's long-term program to help us better understand the dynamics influencing early-life performance of yellow perch and walleye in Saginaw Bay. (Figure 23)

Figure 23: Michigan DNRE Fall Trawl Survey.
Credit: Hook, T.O.

Walleye (larvae photo on right; Credit - NOAA): Past research has demonstrated that before the collapse of alewives, this invasive species likely limited walleye recruitment in Saginaw Bay. Alewives, as aggressive planktivores, not only compete with larval walleye and perch (also planktivores) for food, but may also prey on these larvae. Thus, when the alewife population collapsed, it was found that the early life survival rates of walleye (as well as perch) increased dramatically. However, it is unclear what currently limits walleye recruitment success in Saginaw Bay.

Based on historical analyses of walleye, relative abundances and mean length of age-0 walleye were strongly, positively related to the relative abundances of age-1 and age-2 walleye in subsequent years. These findings suggest that factors influencing growth and survival of walleye before fall of their first year of life are strong determinants of year-class strength (both in terms of abundance and mean size).

Perch: In the analysis of historical data (1970-2008) on the population dynamics of perch, researchers found that yellow perch year-class strength (both in terms of abundance and mean size) is set by the fall of age-1 yellow perch. While a positive relationship exists between the relative abundance and mean length of age-0 yellow perch and that of age-1 yellow perch the next year, such a positive relationship was not found between age-0 yellow perch and age-2 yellow perch. However, relative abundance and mean length of age-1 yellow perch was consistently, strongly related to abundance and mean length of age-2 yellow perch; suggesting that yellow perch recruitment is set by fall of age-1 (i.e., second year of life). A subsequent analysis to identify the factors influencing yellow perch recruitment from 1970 to 2011 was inconclusive. That is, none of the environmental factors considered were strong predictors of yellow perch recruitment success.

Field Surveys on Mechanisms of Fish Recruitment and Trophic Interactions: To determine dynamics in Saginaw Bay's fish community for the multiple stressors project, our researchers conducted surveys from April 2009 through June 2011. The surveys were designed to track two annual cohorts (a group of individual fishes born in the same spawning season) in 2009 and 2010 from just after hatching through survival to following spring. Specifically, we were tracking young cohorts of yellow perch and walleye and documenting their feeding, growth, and survival. Given their economic value, walleye and perch represent the most important fisheries in Saginaw Bay. Additionally, due to their potential ecological importance, we also studied round goby, trout-perch, and lake whitefish.

Walleye: Field survey results for walleye indicated that they grew rapidly from larval to juvenile stages. As the larval walleye transitioned into juveniles, the composition of their diet shifted from herbaceous (plant-eating) zooplankton to larger invertebrates (predatory zooplankton, Bythotrephes and Leptodora, and benthic invertebrates, Chironomidae). Then, as piscivores, walleye consume a diet high in fish (rainbow smelt, shiners, and round goby). We found that the relatively recent invaders Bythotrephes and round goby were important diet components for young walleye. Another significant finding was that yellow perch constituted a relatively unimportant diet component for age-0 walleye. As adults, however, walleye fed almost exclusively on fish, including yellow perch along with round goby. The importance of round goby is noteworthy, since this invasive fish has become a key forage species. It is thought that the round goby is supporting the walleye population in a manner similar to the alewife before their crash (Figures 24 and 25).

Figure 24: The diets of age-0 walleye (2009-2010).
Credit: Hook, T.O.

Figure 25: The diets of age-1 walleye (2009-2010).
Credit: Hook, T.O.

Yellow Perch: Field survey results (2009-2011) for yellow perch show that both the larval and age-0 consumed a diverse diet. Larval perch ate small-bodied zooplankton and later stage age-0 perch expanded their diet to include larger bodied zooplankton and benthic invertebrates, with the most common zooplankton prey to be Daphnia (Figure 26).

Figure 26: Age-0 Yellow Perch Diet Biomass.
Credit: Hook, T.O.

A significant trend observed during this field study was that the well-fed larval and age-0 perch maintained a fairly rapid growth during the late spring and early summer. This growth rate, however, slowed down in the late summer and fall. The trend was found for both growth in biomass and growth in total body energy. Evidence also suggested that young yellow perch lost a great deal of energy over winter, which could contribute to size-dependent mortality (e.g., more vulnerable to walleye predators). It was found that the mean lengths of young yellow perch found in the stomachs of piscivorous walleye were consistently lower than the young yellow perch in the environment (collected in trawling nets), suggesting that the smallest individuals were most like to be consumed by predators. Our researchers concluded that the slowed growth of the yellow perch could contribute to poor survival.

Other Fish Species: The round goby has proven its reputation as an aggressive invasive fish by proliferating throughout Saginaw Bay following its introduction discovered in late 1992. In becoming an integral member of the fish community, the round goby has provided an important diet item for yellow perch and walleye. Our researchers also observed that the round goby preferentially consumed chironomid (non-biting midges) larvae in the bay. Since the walleye and yellow perch also feed on chironomid larvae, the goby has the ability to compete with these native fish species for food. Given the aggressive behavior of round goby, this invasive fish could threaten the availability of food for native fisheries in Saginaw Bay, such as young walleye, yellow perch, lake whitefish (pictured to the right; Credit - NOAA), trout-perch, among others.

We also documented rapid growth for young lake whitefish in Saginaw Bay. Compared to open Lake Huron, Saginaw Bay contains a relatively high abundance of preferred prey for whitefish (such as zooplankton and benthic invertebrates), which may contribute to high growth potential for young whitefish. Simultaneously, no evidence was found of other fish preying upon young whitefish. The management implications of high growth and low predation risk are positive, since whitefish support important commercial fisheries in Saginaw Bay as well as the main basin of Lake Huron. Strong whitefish recruitment could potentially support an important nursery for the rest of Lake Huron.

Governmental and Societal Relevance of Research

An important outcome of our multiple stressor project, Managing the Impact of Multiple Stressors in Saginaw Bay, is the development of a blueprint to improve decision-making on the management of coastal ecosystems, such as Saginaw Bay, based on scientific research and observations characterizing the bay's ecosystem. The Adaptive Integrated Framework (AIF) (Figure 4) provides for the exchange of information between agency managers, modelers, scientists, and interested stakeholders to improve overall management and optimize management strategies, model development, and program monitoring.

In the words of Craig Stow, guest editor of the JGLR special issue The Continuing Effects of Multiple Stressors in Saginaw Bay (Volume 40, Supplement 1, Pages 1-204) "adaptive management is an active process, advancing monitoring, learning, and refining management actions based on improved understanding of system behavior and detecting changes that occur." An adaptive approach to management uses an iterative approach with continuous feedback between management action and scientific understanding of observed changes, and thereby improving management decisions.

The success of this approach depends upon strong communication between resource managers and researchers such that 1) managers identify, prioritize, and convey their information needs to researchers, and 2) researchers incorporate those needs into innovative research approaches. The relationship between managers and researchers should serve as a mutually beneficial process, facilitating assessment of specific environmental problems/stressors and long-term management of ecosystems.

Another key to adaptive management, illustrated by our multiple stressors research in Saginaw Bay, is long-term monitoring. Monitoring on a continuous, regular and long term basis is critical in documenting historic trends and identifying potential mechanisms driving these trends. The results from long term monitoring programs play an important role in determining the role that stressors play in disrupting historic trends and how management may (or may not) mitigate the impact of stressors.

Our research conducted in Saginaw Bay has provided a valuable opportunity to develop, evaluate, and operationalize the Adaptive Integrated Framework (AIF). Throughout the multiple stressors project, the stage has been set for scientists and resource managers to work together to: 1) assess the status of key ecosystem properties and characteristics, 2) make and test quantitative predictions of how ecosystems will respond to specific stressors based on modeling, and 3) develop management strategies based on such predictions as well as the results. While it is generally accepted that uncertainty exists throughout this process, progress has been made under the AIF to better define where and why uncertainty exists and develop a process to quantify and then to decrease this level of uncertainty. Use of the AIF approach will advance more effective management of the land, water, and living resources of Saginaw Bay.

The impact of phosphorus loading on Saginaw Bay's water quality is a good example of how project outcomes support governmental policy priorities, such as those under the GLWQA. As noted in the research results section of this summary, the target of 440 metric tons/year of phosphorus load - originally developed under the 1978 GLWQA - has almost never been met. This explains, at least in part, why levels of total phosphorus, chlorophyll a, and water clarity were regularly exceeded in the sections of Saginaw Bay that were monitored. Model simulations based on monitoring results indicate that reduction in phosphorus levels is a critical factor to reduce algal production and the impacts of eutrophication. Determining the target load reductions of phosphorus will require further monitoring coupled with predictive modeling based on an adaptive management approach.