Green Bay: Spatial variation in water quality, and landscape correlations

Abstract

We conducted a high-resolution survey along the nearshore (369 km) in Green Bay using towed electronic instrumentation at approximately 15-m depth contour, with additional transects of the bay that were oriented cross-contour (49 km). Electronic sensor data provided an efficient characterization of a spatial pattern in water quality parameters. Nearshore water quality was correlated with adjacent landscape characterization. The regressions were able to explain over 80% of the alongshore variability for some parameters. The parameters with the strongest correlation were specific conductivity, beam attenuation, and chlorophyll. A clear feature of Green Bay is the loading introduced by the Fox River at the head of the bay. River loading sets up the conditions for a longitudinal gradient along the bay. Nutrient and chlorophyll gradients have persisted since first observed in monitoring surveys decades ago in spite of rapid flushing of the bay and efforts for remedial actions to restore areas of concern (AOCs). The water quality gradients were steepest in the 25-km closest to the mouth of the Fox River decreasing inversely with distance to where the bay opens to Lake Michigan. Summarized data from our 2010 tow and a concurrent National Coastal Condition Assessment (NCCA) survey compared to historical data (1971–1989) show a bay-wide rise in specific conductivity and chlorides, but only suggest highly variable total phosphorous and chlorophyll a in the inner bay. The tools employed (towed sensors, landuse characterization, and NCCA) can provide an efficient approach to a more regular and comprehensive bay-wide assessment.

Introduction

The large spatial scale of Green Bay (4212 km², Bertrand et al., 1976), basin hydrodynamic processes (Miller and Saylor, 1985, Miller and Saylor, 1993, Gottlieb et al., 1990), and multiple political jurisdictions have made it challenging in the past for agencies to make assessments and adequately address overall condition and spatial variability at the whole embayment scale. Yet monitoring at large spatial scales is required to sufficiently assess the entire bay, develop management plans, and evaluate responses to management plans. Spatially extensive studies have been used to address sediment quality, sinks, and loads across the entire bay with hundreds of collected samples (Manchester-Neesvig et al., 1996, Klump et al., 1997). Although similar extensive surveys for water quality have been conducted, little synthesized analysis has been published for the entire bay (Rockwell et al., 1980, and 1989–91 Green Bay Mass Balance project — GBMB). The waters of Green Bay are a shared resource for the states of Michigan and Wisconsin, and although state and local agencies conduct sampling at a number of local sampling sites (e.g., areas of concern AOCs) a coordinated routine monitoring of the whole of Green Bay is not presently being performed.

Sampling in Green Bay has been directed primarily towards the AOC areas in efforts to understand the extent of degradation, to direct remedial efforts, and to observe changes in conditions across time. The lower Green Bay and the lower Fox River for decades have been noted for eutrophication and pollution problems (Veith, 1972, Rousar and Beeton, 1973, Epstein et al., 1974, Sager and Wiersma, 1975, Bertrand et al., 1976). Similarly, the lower Menominee River, another major tributary to Green Bay, has a long history of pollution and contamination from the pulp and paper industry and chemical manufacturing facilities (Surber, 1953, Fitchko and Hutchinson, 1975, Marti and Armstrong, 1990). In contrast the upper Menominee River is in good ecological health (Riseng et al., 2010). The Fox and Menominee Rivers flow into and contribute to the condition of Green Bay and Lake Michigan (the 1st and 4th largest watershed tributaries to Lake Michigan). Remedial action plans (RAPs) within the AOCs have shown local progress (e.g., Uvaas and Baker, 2011). While these AOCs exist primarily as local problems on the scale of the Great Lakes, they also contribute to the condition to all of entire Green Bay. Contaminants from the two AOCs are transported throughout the bay by currents and circulation patterns (Gottlieb et al., 1990, Lathrop et al., 1990, Miller and Saylor, 1993, Martin et al., 1995). Additional nutrients and contaminants to the bay also are delivered by external loading from landscape activities in all the watersheds and at all spatial scales. The effect from these combined sources is not well known across the entire bay.

The objective of the US EPA — Midcontinent Ecology Division for the current study was to take advantage of co-occurring studies to survey Green Bay comprehensively. Our interest included the nearshore region which parallels other recent efforts to improve our understanding of this component of the Great Lakes ecosystem (Mackey and Goforth, 2005, Niemi et al., 2007, Kelly and Yurista, 2013). The interest includes efforts to assess embayments (small to large), and in this case a particularly large embayment that is physically distinct from the main lake. We were interested in applying new tools of in situ towed instrument arrays (Yurista and Kelly, 2009, Kelly and Yurista, 2013) to characterize a very large bay environment, including identifying large-scale patterns and improving the basic understanding of spatial variability in the Great Lakes nearshore and embayment regions. A related objective was to examine potential linkages of alongshore conditions with adjacent landscape variability in the bay, to parallel our recent efforts along most of the open shoreline of Lake Michigan (Yurista et al. 2015) as well as the other five Great Lakes (Kelly and Yurista, 2013).

The study was framed under a Coordinated Science and Monitoring Initiative for the Great Lakes (CSMI, Richardson et al., 2012) that is directed by both the US EPA Great Lakes National Program Office and Environment Canada. CSMI is a program to involve and coordinate academia, state, and federal agencies in conducting research to address critical and high priority science, research, and monitoring needs of the Lakewide Action and Management Plans for each Great Lake on a five-year rotating schedule. A second co-occurring survey effort was under the National Coastal Condition Assessment program (NCCA, EPA Office of Water) and is conducted every five years across the continental US. The NCCA formally included the Great Lakes for the first time in 2010. The NCCA survey complemented the CSMI year-of-Lake-Michigan studies.

The high resolution and traditional grab sample data collected through both survey efforts increased observational capacity with which to address some basic questions: 1) Can new survey approaches and new technology capture the character and variability of water quality conditions along the shoreline, 2) can we identify regional water quality patterns or spatial structure in the heterogeneity of a large embayment, 3) is there evidence alongshore conditions are correlated with adjacent landscape character?