Factors influencing the phosphorus distribution near the mouth of the Grand River, Ontario, Lake Erie

Journal of Great Lakes Research 42 (2016) 549–564

Contents lists available at ScienceDirect

Journal of Great Lakes Research journal homepage: www.elsevier.com/locate/jglr

Factors influencing the phosphorus distribution near the mouth of the Grand River, Ontario, Lake Erie Krista M. Chomicki a,⁎, E. Todd Howell b, Emma Defield a, Amanda Dumas a, William D. Taylor a a b

Department of Biology, University of Waterloo, 200 University Ave W, Waterloo, ON, N2L 3G3, Canada Environmental Monitoring and Reporting Branch, Ontario Ministry of the Environment and Climate Change, 125 Resources Rd., Toronto, ON, M9P 3V6, Canada

a r t i c l e

i n f o

Article history: Received 10 September 2015 Accepted 15 March 2016 Communicated by Craig Stow Keywords: Nutrient dynamics Nearshore Phosphorus Speciation Dreissenids Cladophora

a b s t r a c t Phosphorus distribution in the nearshore of Lake Erie near the mouth of the Grand River, Ontario, reflects the extent of the mixing area between the river and the lake, with elevated concentrations observed directly within the river plume decreasing as the plume is mixed with the nearshore waters. Easterly alongshore currents were dominant within the area and affected the spatial distribution of phosphorus (P). Suspended solids concentration declined by an order of magnitude between the river and lake, and particulate P (PP) transitioned from being largely organic phosphorus and non-apatite inorganic phosphorus (NAIP) to predominantly NAIP only. Dominant processes transitioned from PP transport in suspension or resuspension in the river below Dunnville Dam to consumption and sedimentation in the lower reaches of the river and the nearshore. Higher dreissenid mussel density and mussel phosphorus content were at times associated with the mixing area of the Grand River, suggesting that river P influences the local ecology (e.g., Cladophora and mussel growth). Mussels and Cladophora in the study area are estimated to contain 8.6 and up to 4.9 tons of phosphorus in the standing biomass, respectively, which can be supplied by the Grand River in approximately 16–25 days. © 2016 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved.

Introduction Despite the phosphorus control programs initiated in the 1970s that led to reductions in total phosphorus (TP) loadings and water column concentrations in Lake Erie (Lesht and Rockwell, 1985, 1987; Rosa, 1987; Rockwell et al., 1989; Dolan and Chapra, 2012), nuisance benthic algae are once again an issue in nearshore regions of eastern Lake Erie (e.g., Higgins et al., 2005; Depew et al., 2011; EC and USEPA, 2014). Since 1996, Lake Erie has experienced an increase in seasonal average phytoplankton biomass throughout the lake (Conroy et al., 2005) with cyanobacterial blooms in the western basin and the nearshore of the central and eastern basins (Budd et al., 2001; Vanderploeg et al., 2001; Vincent et al., 2004; Conroy and Culver, 2005). However, despite increases in biomass and blooms in the western basin, eastern basin nearshore regions have been experiencing a decrease in chlorophyll a and primary production in comparison to the offshore (Depew et al., 2006). Offshore waters of the eastern basin of Lake Erie are oligotrophic to oligo-mesotrophic with low TP (e.g b13 μg/L: Makarewicz and Bertram, 1991; North et al., 2012; Dove and Chapra, 2015) and moderate to high water clarity (e.g., average summer Secchi depth of 5–11 m since 2000; Dove and Chapra, 2015). Phosphorus (P) is the limiting ⁎ Corresponding author at: Toronto and Region Conservation Authority, 5 Shoreham Drive, Toronto, ON M3N 1S4, Canada. Tel.: +1 416 661 6600x5857, +1 416 235 6567. E-mail addresses: [email protected], [email protected] (K.M. Chomicki).

nutrient for phytoplankton and Cladophora, the dominant nuisance benthic alga, even though TP can be elevated in the nearshore (e.g., Schwab et al., 2009). Phosphorus dynamics within these nearshore regions can be affected by a number of mechanisms including tributary discharge (Makarewicz et al., 2012), shoreline erosion, sediment resuspension (Mayer and Manning, 1989), and biological activity (Hecky et al., 2004). However, it is difficult to apportion the contributions of these mechanisms to the observed nutrient gradients in the coastal regions of Lake Erie. Tributaries can strongly affect water quality in the nearshore (Baker, 1985; Chen and Driscoll, 2009). Nutrient dynamics and mixing zones are shaped by temperature gradients between the tributary plume and the lake (e.g., Murthy et al., 1986) and features of alongshore circulation (e.g., Rao and Schwab, 2007; Howell et al., 2012; Howell et al., 2014). Modeling studies on river plumes entering the Great Lakes indicate that they can periodically extend far into the offshore (e.g., Ji et al., 2002; Chen et al., 2004), with seasonal changes in river flow influencing the nearshore. Environmental conditions in eastern Lake Erie adjacent to the mouth of the Grand River are highly dynamic as a result of the external loading, physical forcing, and internal cycling (transport, uptake and release of phosphorus within the nearshore) that shape nearshore water quality patterns. Spatial gradients in nutrient chemistry and biological activity are observed when river water enters the lake, as nutrients are diluted and assimilated. The size and orientation of the mixing area between

http://dx.doi.org/10.1016/j.jglr.2016.03.014 0380-1330/© 2016 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved.

550

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

the river and the lake are determined by the physical mixing of the two water masses and lake circulation. He et al. (2006) highlight the importance of wind-driven coastal currents to the movement of the Grand River (Ontario) plume in the eastern basin and determined that the frequent current reversals limit the extent that the plume travels. As the physical shoreline laterally constrains water movement and diverts current flow parallel to the shore (Rao and Schwab, 2007), nutrient-rich areas capable of sustaining high productivity are created. Conductivity in the nearshore can be used to map the extent of horizontal mixing and dispersion of the plume (Rao and Schwab, 2007) that is likely the cause of the elevated TP concentrations reported by Nicholls et al. (2001) and the broad scale productivity gradient observed near the river mouth by Nicholls et al. (1983). The TP load entering Lake Erie is comprised of dissolved and particulate phosphorus. It is generally accepted that dissolved phosphorus (DP) is more bioavailable than particulate phosphorus (PP) and can be accessed by biota. Particulate phosphorus, however, dominates the TP load. Baker et al. (2014) determined up to 30% of the PP entering Lake Erie from Ohio rivers was bioavailable; it is also conceivable that some of the recalcitrant portion of PP can become available as a nutrient source after being assimilated and excreted by mussels (Ozersky et al., 2009) or other particle feeders. The bioavailability of PP is likely important to understanding the nutrient supply to the nearshore. In eastern Lake Erie, shoreline materials are easily eroded clay and silt and contribute turbidity and TP to the nearshore. Suspended solids (SS) in surface waters can contain more bioavailable P than bottom waters (e.g., Mayer and Manning, 1989). Particle size, phosphorus speciation, and particle geochemistry affect the availability of particulate P and rate of soluble reactive phosphorus (SRP) release from tributary particulate matter (DePinto et al., 1981). However, PP also enters the water column by sediment resuspension, especially during late fall and winter storms. Studies in Lake Erie (Williams et al., 1980) have shown that algal uptake of P from lake sediments is positively related to the amount non-apatite inorganic phosphorus. Therefore, particulate phosphorus speciation is likely important to the understanding of nutrient supply to the biota in nearshore regions. Dreissenid mussels act as both sources and sinks of phosphorus. Conroy et al. (2005) suggest that mussels increase phosphorus nutrient fluxes and might facilitate phytoplankton growth in western Lake Erie, while Zhang et al. (2011) conclude that these nutrient contributions by mussel excretion are concentrated in the bottom waters. Ozersky et al. (2009) found that the SRP excreted from mussels was similar to the amount required by Cladophora communities. Dreissenid biomass can retain appreciable amounts of phosphorus over the colonized lakebed (Pennuto et al., 2012, 2014). Bedrock heavily colonized by mussels has been observed at depths of less than 20 m in the vicinity of the Grand River confluence with Lake Erie, east and west of the river mouth. The proximity of the lakebed colonized by dreissenid mussels with gradients of water quality resulting from the river mixing with the lake suggests that interactions between the mixing plume and the biological cover of lakebed may influence P speciation and fate. The Grand River is the largest tributary from Canada flowing to Lake Erie and is a major nutrient input to the eastern basin. Despite its potential importance, the influence of the Grand River on nutrient distributions in the adjacent nearshore of Lake Erie is not well understood. While the effects of winter and early spring plumes on primary production (e.g., EEGLE: Episodic Events—Great Lakes Experiment; Green and Eadie, 2004) and the importance of late winter–spring plumes and storms on nearshore–offshore transport (Rao et al., 2002) have been examined in the Great Lakes, summer pulses have not been a focus due to their smaller contribution to load. However, in the western and central basins Michalak et al. (2013) found that extreme precipitation and runoff events extending until June can impact nearshore nutrient regimes; events such as these could provide nutrients for nuisance algae. One might expect that inputs during the growing season might be disproportionately important to the growth of nuisance algae. The

influence of the river on the adjacent shores of the lake, and the spatial and temporal scales over which the Grand River discharge affects environmental and ecological conditions in the nearshore environment, remain unclear. Recently, the Annex 4 Objectives and Targets Task Team (US-EPA and Environment Canada, 2015) concluded that while reductions in P loads are expected to result in reductions of Cladophora, they were unable to make specific recommendations without additional research. This paper compiles data from a number of programs to examine the distribution of SRP, dissolved P and particulate P at the mouth of the Grand River and the adjacent waters of Lake Erie to better define the Grand River’s zone of influence. We use the relationship between TP and suspended sediments, and data on particulate P speciation, currents, and particle-size distribution, to make inferences about the fate and impact of riverine P. We also examine the distribution of dreissenids and Cladophora in the nearshore and discuss the possible role of dreissenid mussels in modifying the effect of riverine nutrients on nuisance algae and in the fate of the loaded phosphorus. Methods Site description The Grand River, located in Southern Ontario, Canada, drains a watershed of nearly 7000 km2. While land use in the basin is primarily agricultural (approximately 70%: Depew et al., 2011), the watershed includes several cities and towns including Kitchener, Waterloo, Cambridge, and Guelph that also affect river quality. They are among the fastest-growing urban areas in Ontario (Jyrkama and Sykes, 2007) and collectively are home to nearly one million residents. Increasing urbanization and nutrient loading have the potential to threaten the water quality of the river (e.g., Winter and Duthie, 2000), the water supply to communities along the Grand, and the nearshore of Lake Erie. Water quality in the lower river is affected by land use both locally and upstream, and influenced by the local geomorphology (e.g., low gradient and clay-rich soils). The turbid waters of the lower river are eutrophic, being high in TP, nitrates, suspended solids and chlorophyll a (e.g., TP up to ~ 160 μg P/L: MacDougall and Ryan, 2012, and median concentrations of 256 μg P/L reported in Venkiteswaran et al., 2014). Other indicators of degraded water quality include high levels of chloride (MacDougall and Ryan, 2012) and low night-time dissolved oxygen (Rosamond et al., 2011; Venkiteswaran et al., 2014). The river has a high discharge compared to all other tributaries of the eastern basin (average of 60 m3/s versus 14 m3/s). Because nutrient concentration gradients in the mixing area of the Grand River and Lake Erie are dependent on the water quality in the river, the fate of nutrients in the Grand River plume is an important aspect of the nutrient regime of the eastern basin. Of particular importance is the large amount of phosphorus the Grand contributes, a significant portion of which comes from sewage treatment plant effluent and agricultural runoff. Annual TP loads from the Grand River to Lake Erie have not been published recently; however, estimates from 1994 (320 metric T/yr; reported in Schwab et al., 2009) and used in the TP loading analysis by Dolan and McGunagle (2005) and Dolan and Chapra (2012) as well as subsequent estimates by Depew et al. (2011; 200–220 T/yr) indicate that the Grand River accounts for a large proportion of the load entering the eastern basin. Water quality Multiple years of water quality data were available for this analysis. In 2001, ship surveys were conducted in the lower Grand River (below Dunnville Dam) and in the adjacent nearshore environment in Lake Erie on successive days (Fig. 1). Ship surveys were conducted 5 times per year throughout the ice-free season between April and November 2001. Surveys followed a pre-defined survey track at a speed of b10 km/h and were limited by the water depth needed by the survey vessels (N 2.5 m water depth). Geo-referenced measurements were

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

551

Fig. 1. Map of the mouth of the Grand River study area, identifying 2010 phosphorus partitioning sample locations (GR u/s, GR mouth, LE east, LE west, LE o/s) and chemistry sampling locations, 2010 benthic survey transects (T1 to T4), 2001 example of the survey track followed (grey lines), 2001 water quality sampling locations (green dots), and the separation of the nearshore into zones for comparative purposes.

made every 5 to 10 m by sensors deployed directly in the lake at 1.5 m depth attached to a frame mounted on a pole. Sensors logged measurements of conductivity, temperature, beam transmission (660 nm), and chlorophyll a. Temperature and conductivity were measured with an Ocean Sensors 3000 CT. Chlorophyll a fluorescence was determined with a Chelsea Aquatracka II probe and beam transmission with a Chelsea transmissometer (0.25 m path). Suspended solids were estimated from the beam attenuation coefficients by linear regression against lab SS values for each survey. Discrete water samples were simultaneously collected for lab-based chemical analysis (Figs. 1 and 2). Water was drawn using a peristaltic pump from an intake located adjacent to the sensors. Samples were transported to the laboratory on ice or dry ice until processing. Whole water samples were analyzed for silicate, SRP, suspended solids, turbidity, chloride, conductivity, TP, and nitrate + nitrite. Nitrate + nitrite and SRP samples were settled prior to chemical analysis. Samples were used to calibrate continuous data where applicable. To convert field measurements of chlorophyll a fluorescence to extracted chlorophyll, aliquots of up to 1 L were filtered through 1.2 μm nylon filters immediately upon collection. The filters were stored on dry ice until delivered to the lab for analysis. Similarly, for each day of survey, field-measured conductivity was corrected using linear regression between paired field and laboratory samples. All laboratory-based water analyses were conducted at the Ontario Ministry of Environment’s Laboratory Services Branch using standard OME methods for surface waters (e.g., Chow et al., 2010). Water quality results were summarized in zones (1–7) operationally defined roughly based on bathymetry. In 2010, five sampling locations were selected for collection of PP samples during spring, summer, and fall (i.e., May, July, October, November; Fig. 1). Surface (1.5 m) and bottom waters (approximately 1 m above the lakebed) were collected and filtered onto pre-ashed and pre-weighed GF/C filters. The filters were stored on ice then frozen until analysis. Discrete water samples were collected at the same time for water quality analyses using methods described above. At the same locations, a LISST-100X submersible particle size analyzer recorded

profiles of in situ particle concentrations at 32 specific log-spaced angle ranges using a red 670 nm diode laser with a silicon detector. Filters with particulate material were analyzed for PP fractions using a method similar to Bostan et al. (2000). For Total PP, filters were ashed at 500 °C, extracted in 1 M HCl, and analyzed for PO4 using the ascorbic acid method (Murphy and Riley, 1962). For non-apatite inorganic P (NAIP), filters were extracted in 1 M NaOH, neutralized, and analyzed for PO4. The residue was extracted in 1 M HCl and the PO4 present was assumed to be apatite-P (AP). Other filters were extracted first with 1 M HCl, and the PO4 in that extract was used as an estimate of inorganic P (IP). The residue was digested with persulfate in a boiling water bath, and the resulting PO4 liberated was used to estimate organic P (OP). In October and November, samples were collected in triplicate, allowing all of the fractions to be estimated directly. In May and July, only single filters were collected, and the filter was split into two, and total PP was not estimated directly. Instead, OP and IP fractions were summed to estimate total PP. Dreissenid mussel density and Cladophora biomass Benthic surveys were completed in July 2010 along four transects adjacent to the mouth the Grand River (Fig. 1). Four sample sites were selected along each transect at depths of 3, 6, 10, and 18 m. Mussels were collected within three 0.15 m2 quadrats at each depth along each transect where divers removed mussels by hand or scraped the surface of the quadrat while applying an air-lift to collect the samples into mesh bags. Samples were frozen on dry ice in the field and then freeze-dried for analysis. Dreissenid mussels were counted, weighed, and measured to estimate population densities, size distribution, and biomass. In the case of shells cracked or broken during processing, the morphometry of the shell was taken into account to estimate the length of the whole shell. Approximately 30 mussel samples were dissected for soft tissue P and shell P analysis, attempting to maintain the size distribution of the entire sample. After dissection, shell and tissue samples were baked at 450 °C for 4 h. Samples were analyzed for TP as above

552

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

Fig. 2. Nearshore surface conductivity, suspended solids, and temperature spatial maps at the mouth of the Grand River in 2001. The numeric values represent SRP, TP, and nitrate + nitrite concentrations at discrete locations in columns from left to right. The panels a) to d) horizontally illustrate results for surveys completed in April, May, July, and August.

after persulfate digestion. Cladophora samples were collected at the same sites as the mussels to determine Cladophora biomass. Filaments were removed from the quadrats, frozen in the field with dry ice, and then freeze-dried for determination of dry weight. Deployed sensors for physical measurements Bottom-moored Acoustic Doppler Current Profilers (ADCP–RDI Workhorse 600 kHz) were deployed within 2 km of the shoreline east of the Grand River mouth in 2004 and 2010 and also west of the river mouth in 2010 (Fig. 3). Additional ADCPs were also placed approximately 5 km offshore of the Grand River mouth in 2010 and 2004. Lake currents were recorded at 30-min intervals from 1 m bins between April and October or November. Currents were resolved into alongshore and cross-shore vectors using a shoreline azimuth angle of 84° with the exception of the far west ADCP which was rotated to align with 102°. The raw data were averaged over 12-h intervals for analysis.

Temperature sensors (Onset Stowaway Tidbits) were placed on the river bed at 3 locations between Dunnville Dam and the mouth of the Grand River. Additional sensors were placed through the water column at selected sites in the adjacent nearshore of Lake Erie extending approximately 2.5 km southeast and southwest of the river mouth from April to November 2001. Antifouling sensors for conductivity (ALEC Electronics CompactCLW) were deployed in 2010 at sites LE west and LE east at depths of 3.9 and 3.7 m, respectively (Fig. 1), from May to November. Data analysis Surface maps of water quality were produced by Kriging UTM georeferenced field sensor data coordinates (i.e., interpolating the data between survey track measurements) using the software package Environmental Visualization System Pro v 9.42 (EVS). The data were not detrended and are only meant to describe the spatial distribution of the field data. The Kriged data were displayed in Arcmap 9.1.3

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

553

Fig. 3. Near surface water current velocities at ADCP deployment sites in 2004 (squares) and 2010 (circles). The current data are vector averages over 12-h intervals with alongshore current velocities in blue and cross-shore velocities in red. Data are for a 2-m depth interval beginning approximately 2 m from the lake surface.

software using 1:10,000 Ontario base maps (Ontario Ministry of Natural Resources). Analyses of covariance (ANCOVAs) were completed on monthly TP and SS data for each of the study locations using Systat version 10 to complete a residual analysis and provide insight to the dominant processes in the region. To identify natural groupings of particle size fractions measured by the LISST-100X submersible particle size analyzer, principal components analysis (PCA) was completed on the particle size data using princomp in the vegan 2.0-0 package of R (version 2.13.1). The data were not standardized. Selected axes were rotated using varimax to produce new variables that summarized the size distribution. Redundancy analysis (RDA) was completed using the biplot add-in to Excel developed by Eric Smith at Virginia Tech to identify which particulate phosphorus fractions were related to particle size. Data were centered, standardized, and plotted as a symmetric biplot. Tributary discharge Grand River flows were attained from the Grand River Information Network (maps.grandriver.ca). Watershed areas were estimated using the Ontario Flow Assessment Tools III, created by the Ministry of Natural Resources and Forestry. Flow estimates in 2001 and 2010 at York were prorated to the mouth of the Grand River where it enters Lake Erie.

Results Mixing areas in the nearshore Conductivity, a tracer of river influence, decreased from the lower Grand River along the shoreline of the lake and into the offshore (Fig. 2). It was highly correlated with chloride within each sampling event (R2 N0.99). Average chloride concentrations varied by up to a factor of 5 between the lower Grand River and the nearshore (Tables 1 and 2). Mixing areas, identified by intermediate conductivity, are evident at the mouth of the River in Splatt’s Bay (zone 1 in Fig. 1) and generally followed the shoreline extending approximately 1 km from the shoreline (zones 2 and 5 in Figs. 1 and 2). In 2010, conductivity sites LE west and LE east (west and east of the Grand River mouth) ranged from 278 to 396 μS/cm and from 280 to 366 μS/cm (Fig. 4), respectively. West of the Grand River mouth (LE west), conductivity was similar to the offshore in summer and fall but was elevated throughout the spring. In contrast, conductivity was elevated in comparison to the outer portion of the nearshore throughout the spring to fall period east of the river mouth (LE east). Average daily discharge of the Grand River ranged between 14 m3/s and 782 m3/s between April and November 2001 and between 23 m3/s and 299 m3/s between April and November 2010 (Figs. 4 and 5). Compared to 2001, there were more high discharge events above

554

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

Table 1 Mean concentration and standard deviation of water quality parameters in surface samples among the different zones in the nearshore in 2001. The discrete samples from the 5 surveys completed between April and November 2001 are averaged. SRP is soluble reactive phosphorus. General locations of the samples: Zones 2–4 are west of the river mouth, and Zones 5–7 are east of the mouth (see Fig. 1), where E = East, W = West, R = River, and SB = Splatt’s Bay. Zone General location

4

3

2

River

1

5

6

Water quality parameter

W

W

W

R

SB

E

E

E

16.1 ± 0.8 0.6 ± 0.2 272 ± 3 1.5 ± 0.9 185 ± 28 0.7 ± 0.3 7.3 ± 2.6 0.1 ± 0.0 15.1 ± 1.4

16.4 ± 1.1 1.0 ± 0.5 274 ± 6 2.1 ± 0.8 187 ± 44 1.2 ± 1.9 9.7 ± 3.5 0.1 ± 0.1 11.8 ± 1.1

20.6 ± 5.7 2.6 ± 2.2 314 ± 56 4.1 ± 3.0 299 ± 137 1.0 ± 0.6 10.4 ± 3.3 0.1 ± 0.0 5.1 ± 1.8

80.1 ± 24.6 32.1 ± 7.4 800 ± 85 34.9 ± 41.3 2446 ± 1874 14.8 ± 9.7 98.0 ± 4.0 1.1 ± 1.1 5.4 ± 0.2

29.6 ± 18.6 7.9 ± 9.7 389 ± 149 10.2 ± 16.6 627 ± 745 3.8 ± 4.1 29.8 ± 24.9 0.3 ± 0.3 8.5 ± 9.6

22.6 ± 6.2 2.3 ± 1.6 329 ± 52 3.4 ± 6.0 447 ± 400 4.1 ± 4.1 14.8 ± 7.6 0.2 ± 0.2 4.5 ± 1.1

17.3 ± 2.1 2.0 ± 2.3 293 ± 46 1.6 ± 1.2 245 ± 151 3.0 ± 3.8 11.7 ± 6.6 0.2 ± 0.2 11.9 ± 2.3

17.0 ± 1.1 0.9 ± 0.5 280 ± 11 1.8 ± 1.0 245 ± 138 2.4 ± 3.4 14.0 ± 11.1 0.1 ± 0.2 17.5 ± 2.8

Chloride (mg/L) Suspended solids (mg/L) Conductivity (μS/cm) Chlorophyll a (μg/L) Nitrate + nitrite (μg N/L) SRP (μg P/L) Total phosphorus (μg P/L) Silicate (mg/L) Depth (m)

Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD

100 m3/s in the summer of 2010. Despite this, the average discharge between April and November was similar between the two years (2001, 53 m3/s; 2010, 51 m3/s). Lake current data for the surface layer indicate that alongshore transport dominated in both 2004 and 2010, with flow to the east prevailing (Fig. 3). Flow reversals were frequent; however, the westward flows (negative in Fig. 3) were weak in comparison with those to the east. The cross-shore flows were mostly weak compared with alongshore flows, with exceptions. Quiescent conditions were observed at a shallow water deployment protected by a reef west of the river (Fig. 3). Strong offshore flows were detected at a site east of the river mouth near the shoreline, likely stemming from the curvature of the adjacent shoreline. The alongshore velocities were greater at the shoreline sites than at sites located more than 5 km from the shoreline in both 2004 and 2010. Considering all deployment sites and the two years of data, maximum 12-h flows in the eastward and westward directions were 30 cm/s and 39 cm/s, respectively. Flows greater than 20 cm/s were observed only in the late summer and fall. The difference in temperature between the Grand River and the nearshore of Lake Erie can be used to understand the interaction between the two water masses. Near the Dunnville Dam, where the river is little influenced by the lake, water temperature was elevated in comparison to the lake from April until mid-September, after which the lake temperature exceeded the temperatures observed in the river (Fig. 5a). The difference in surface temperature between the river and nearshore east and west of the river mouth varies in a similar pattern; however, the temperatures to the east were slightly warmer until mid-August after which temperatures are similar (Fig. 5b). The cooler temperatures in the more shoreward parts of the nearshore compared with the river again indicate a positive buoyancy of the river discharge and a warming effect on the nearshore until late summer (Fig. 5d). After October, temperatures at all lake sites are similar. The contrasting mixing regime between early and late seasons in the nearshore is evident from the temperature differences between the surface, 10 and 18 m depths at the offshore station. From May to midSeptember, strong but intermittent stratification is evident contrasting with mostly well-mixed conditions after mid-September. During spring and early summer, river discharge is more likely to disperse over the broader nearshore in a surface mixed layer in contrast to the fall when dilution with the full water column or a sinking plume is likely. Nearshore nutrient distributions Nutrient concentrations in the Grand River are higher and more variable than in the adjacent nearshore (Tables 1 and 2). In 2001, nutrient concentrations in the nearshore mixing areas located less than 1 km from the shoreline (zones 1, 2, and 5) are generally greater than more lakeward regions of the nearshore (e.g., zones 4 and 7). Given the strong

7

relationship between conductivity and river discharge we can infer that maps of conductivity (Figs. 2a–d and 4) also depict the spatial– temporal variation in nutrients and other water quality features with elevated TP and nitrate + nitrite often corresponding to higher levels of conductivity. Conductivity, chlorophyll a, TP, SS, and nitrate + nitrite frequently declined away from the shoreline (Fig. 2, Tables 1 and 2). A repeated pattern was a sharp drop in concentration away from the immediate shoreline with little or subtle decline beyond 1 to 3 km offshore. Mean nutrient concentrations at the mouth of the river decrease by at least 50% from the mean river concentrations. Soluble reactive phosphorus concentrations were greatest near the shore, with higher mean concentrations observed east of the Grand River mouth in comparison to the west. Concentrations tend to be greatest in Splatt’s Bay and along the shoreline. Total phosphorus concentrations were 7 to 24 times higher at the mouth of the Grand River than in the nearshore. The annual mean ± SE) river mouth TP ranged between 98 ± 4 μg P/L in 2001 and 66 ± 16 μg P/L in 2010. However, elevated TP was often observed within the immediate mixing area between the river confluence with the nearshore (30 ± 25 μg P/L in 2001; Table 1, Fig. 2). An exception was the high TP in November throughout the nearshore, reaching 40 μg P/L approximately 5 km from the shoreline; the cooler temperature near the shoreline in comparison to 5 km from shore suggests these high concentrations may have been due to an upwelling event in part and/or wind-driven resuspension (turbidity 2 FTU; Secchi 3 m) and shoreline erosion (the survey was curtailed due to rough lake conditions). The highest TP generally corresponded with high conductivity, as expected. Concentrations of TP in the nearshore decrease moving through zones 2 to 4 and 5 to 7. TP values in the east were higher than the west (Fig. 2). We used conductivity quartiles to summarize the effect of river influence (high conductivity quartile) and low land influence (low conductivity quartile). Overall, mean TP values were greatest in the high conductivity quartiles and decrease from spring to fall. In the mid- and low-conductivity quartiles, TP (and SRP) values increase in late summer and fall (Table 3). Silicate was highest in the river, decreased moving from zones 5 to 7 east of the river mouth, and fairly uniform in the nearshore west of the Grand River. Concentrations of silicate in the lake were generally an order of magnitude lower than in the river (i.e., ~0.1 mg/L vs ~1 mg/L), except in Splatt’s Bay at the mouth of the Grand River. Similarly, the highest levels of particulate material observed in the lower Grand River (32 ± 7 mg/L) were from 4 to 54 times higher than samples from the nearshore (Table 1). Variability among areas and dates was low in the nearshore, zones 2 through 7, in comparison to the river and its confluence with Lake Erie at Splatt’s Bay (zone 1). Elevated concentrations of TP and nitrate were also observed in the 2010 discrete samples collected in the river and by the shoreline west of

E

18.2 ± 1.0 1.2 ± 0.5 287 ± 9 1.2 ± 0.5 156 ± 71 2.9 ± 1.2 5.0 ± 1.8 0.1 ± 0.1 20.8 ± 0.4 48.5 ± 16.3 20.3 ± 2.2 1.6 ± 0.4 305 ± 13 4.5 ± 6.9 300 ± 101 3.4 ± 1.7 5.5 ± 2.4 0.2 ± 0.1 8.9 ± 0.4 55.0b ± 13.6 27.0 3.7 391 1.2 1410 8.7 14.0 0.5 8.3 67.0 51.5 ± 26.2 28.1 ± 17.6 633 ± 245 16.9 ± 7.5 3163 ± 2025 5.2 ± 1.7 61.7 ± 35.9 0.8 ± 0.6 5.4 ± 0.4 20.0 ± 7.6

LE o/s

60.3 ± 33.4 38.3 ± 3.6 687 ± 127 36.3 ± 34.9 2945 ± 474 8.4 ± 4.7 78.5 ± 14.8 1.7 ± 0.7 4.3 ± 1.6 20.9 ± 5.7 17.8 ± 0.4 1.5 ± 0.5 282 ± 3 2.0 ± 0.5 136 ± 50 2.8 ± 1.1 4.5 ± 1.3 0.1 ± 0.1 20.8 ± 0.4 51.3 ± 9.8

Fig. 4. Conductivity measured at sites LE west and LE east in 2010 with x-axis labels at the first of the month. Sensors were deployed at 3.9 m and 3.7 m depth with site depths of 7.2 and 8.0 m, at LE west and LE east, respectively (see Fig. 1 for locations). Data are for daily average conductivity. The dotted line gives nearshore background conductivity as inferred from average surface conductivity (282 μS/cm) at LE offshore. Average daily flow from the Grand River is illustrated by dashed line in the lower panel.

the river mouth (Table 2) and support the water quality patterns observed in 2001. Mean concentrations of nitrate + nitrite and TP were lower in 2010 than 2001 when comparing similar site locations (e.g., annual mean (± SD) river mouth TP: 2001 = 98 ± 4 μg P/L; 2010 = 66 ± 16 μg P/L). In 2010, mean nutrient concentrations were mostly similar in the surface and bottom samples with the exception of TP and SRP (Table 2). Concentrations of TP in water samples collected 1 m above the lakebed within the mixing area were more variable than surface samples although the means were similar. Similarly, comparisons between SRP in the surface and bottom waters were variable, with greater SRP observed in the bottom waters at GR upstream (below Dunnville Dam) and at LE east (within the mixing area between the lake and river mouth). The river was well mixed below the dam during sampling events and became stratified by the river mouth in July and October. In the lake, stratification was apparent from May to October with the exception of west of the river mouth (Fig. 5).

28.4 3.1 410 2.0 1800 6.1 23.0 0.6 8.3 71.3 62.7 ± 18.9 33.6 ± 7.4 719 ± 104 25.9 ± 27.6 3550 ± 1058 6.4 ± 4.5 76.0 ± 20.4 1.2 ± 0.9 3.2 ± 0.2 23.3 ± 7.3

Covariation of TP with conductivity and SS

b

a

Only 2 sampling dates. Only 3 sampling dates.

Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Chloride (mg/L) Suspended solids (mg/L) Conductivity (μS/cm) Chlorophyll a (μg/L) Nitrate + nitrite (μg N/L) SRP (μg P/L) Total phosphorus (μg P/L) Silicate (mg/L) Depth (m) DP/TP (%)

62.6 ± 21.0 25.0 ± 5.6 718 ± 118 24.1 ± 34.6 3605 ± 1059 8.6 ± 6.1 65.8 ± 15.7 1.1 ± 0.7 5.3 ± 0.3 26.1 ± 4.6

18.0 ± 1.0 1.5 ± 0.6 283 ± 10 1.4 ± 0.3 159 ± 108 2.1 ± 1.2 5.3 ± 1.5 0.1 ± 0.1 8.4 ± 0.4 51.4 ± 3.6

19.4 ± 1.8 8.2 ± 13.8 300 ± 20 1.6 ± 0.6 311 ± 237 2.8 ± 0.9 5.3 ± 1.3 0.1 ± 0.1 8.9 ± 0.4 61.7 ± 11.4

R E

GR mouth GR u/s LE o/s LE east

E W W R

LE west (J-N) LE west (M) GR mouth

R

GR u/s

Water quality parameter

General location

555

18.0 ± 0.8 2.0 ± 0.6 285 ± 7 1.4 ± 0.2 151 ± 74 2.2 ± 1.1 4.7 ± 1.2 0.1 ± 0.1 8.4 ± 0.3 45.3 ± 9.5

E

LE east LE west (J-N)

W W

LE west (M)

R

Zone (Bottom waters)

b a

Zone (Surface waters)

Table 2 Mean concentration and standard deviation of water quality parameters in surface and bottom samples (~1 m above lakebed) among the different discrete sampling locations in the nearshore in 2010. The discrete samples from 4 sampling events between May and November 2010 are averaged unless otherwise noted. LE west 2010 data is partitioned into May and July–November. SRP is soluble reactive phosphorus. General locations of the samples: LE west is west of the river mouth, and LE east is east of the mouth (see Fig. 1), where E = East, W = West, R = River, u/s = upstream, and o/s = offshore.

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

TP covaried strongly with conductivity in each of the monthly surveys conducted both in the lower river alone and the river mouth and the nearshore in 2001 (R2 range = 0.822 to 0.981, n = 11–24 for lake surveys and n = 14–16 for river surveys). The slopes of the relationships were significantly different among surveys as determined by ANCOVA analysis (p values: river = 0.013, lake = 0.018). The pattern in the residuals in TP from conductivity suggests greater variability around the prediction of the river data compared to the river mouth and lake survey data with a general decrease from the shore to the offshore. We attempted to use the TP–conductivity relationships to diagnose the processes affecting the fate of P in the nearshore. We expected simple mixing between lake and river water to produce a linear relationship between TP and conductivity (dilution: zero residuals), while sedimentation and consumption should produce faster TP loss (negative residuals) and resuspension of particulate material should augment TP (positive residuals).

556

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

Fig. 5. Water temperatures in the lower Grand River and adjacent nearshore of Lake Erie in 2001. (a) Surface temperature in the river approximately 5 km upstream of the river mouth and 6.6 km offshore of the river mouth as red and blue lines, respectively. The black hatched line illustrates the 2001 average daily flow from the Grand River. (b) Difference between river temperature and lake surface temperature (positive indicates the lake is warmer) at sensors 2.3 km southeast and 2.6 km southwest of the river mouth, red and blue lines, respectively. (c) Temperatures at the surface and 10 and 18 m depths at sensors 6.6 km offshore of the river mouth. (d) Difference in surface temperature between sensor 6.6 km offshore of the river mouth and at sensors 2.3 km southeast and 2.6 km southwest of the river mouth.

At the river sites, the high variability in TP residuals (Fig. 6) obscured interpretations. Because their conductivity was high, this residual variation is not likely due to mixing with lake water, but rather due to the coarse nature of the particulate P at these sites (causing high variation among samples) or local sources and processes (causing intense spatial heterogeneity). Nonetheless, on most occasions the residual plots were “U-shaped” suggesting that TP declined more rapidly than would be expected by dilution by lake influx near the river mouth and implicating sedimentation and/or consumption. Samples with low or intermediate conductivity had high TP values on some dates, for example August and November, suggesting re-suspension of particulates rather than a direct river source. Relationships between TP and suspended solids were generally strong with 90% of the Pearson correlations greater than 0.75 among surveys. Linear regressions between TP and suspended solids among surveys were stronger for the lake and river mouth than the river alone surveys (lake R2 = 0.794 to 0.993; river R2 = 0.224 to

0.869), and they were significantly different from each other (ANCOVA p ≤ 0.011; Table 4). These relationships were significantly different between months for both the lake and river samples (lake p b 0.001, river p = 0.044 with the removal of two statistical outliers). Dissolved phosphorus and P content of the particulate matter can be roughly inferred using the TP–SS regressions created from the river and lake surveys in 2001. Because there is no PP when suspended solids are zero, the intercept serves as an estimate of the dissolved fraction within TP. The intercepts of these TP–SS plots illustrate that the lake has low DP (b10 μg P/L, with the exception of November), while the river has a higher DP (N30 μg P/L; Table 4). In 2010, the percent of measured DP in TP in surface waters was lower in the river (average b 26% of TP), in comparison to the lake (average DP N51% of TP; Table 2). Bottom waters in the lake were also higher than bottom waters in the river (lake: DPN 49% of TP; river: DP b21% of TP). Phosphorus content in the suspended solids was inferred from the slope of TP–SS regressions; the P content of particulates from the lake

Table 3 Mean TP and SRP concentrations within high, mid, and low conductivity quartiles, N75%, between 25% and 75%, and b25%, respectively. High conductivity quartile Survey

Mean TP (μg P/L)

Mean SRP (μg P/L)

April 2001 May 2001 July 2001 August 2001 November 2001

50.3 52.4 37.6 30.0 32.0

5.5 3.6 1.2 2.3 9.5

Conductivity range (μS/cm) N332 N374 N310 N376 N329

Mid conductivity quartile

Low conductivity quartile

Mean TP (μg P/L)

Mean SRP (μg P/L)

Conductivity range (μS/cm)

Mean TP (μg P/L)

Mean SRP (μg P/L)

12.8 8.5 8.3 16.4 29.0

1.9 1.2 0.5 4.1 7.7

269 - 332 273 - 374 279 - 310 274 - 376 273 - 329

8.7 9.0 6.7 13.7 24.0

0.7 1.0 0.5 3.4 8.2

Conductivity range (μS/cm) b 269 b 273 b 279 b 274 b 273

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

557

Fig. 6. TP residuals separated by zone for (a) April, (b) May, (c) July, (d) August, and (e) November. Two outliers have been removed. River sample is taken on day of lake survey, and river survey generally occurred one day prior to lake survey. Zone locations are displayed in Fig. 1. River surveys are illustrated as U, M, L (Upper, mid, and lower reaches below Dunnville Dam), and Lk (lake).

(N2.1 μg P/mg, or N 0.21%) was higher than that of particulates in the river (b 2.0 μg P/mg, or b0.20%).

Particulate phosphorus fractions Suspended solids sampled in 2010 had higher particulate OP and NAIP in the river than in the lake with the exception of November (Fig. 7). In the river samples (GR upstream and GR mouth) the proportion of the PP as OP tended to be higher than NAIP, which contrasts with the lake

samples (LE east, LE west and LE offshore) where there was typically a higher fraction of NAIP. Samples from the bottom waters were mostly similar to the surface waters except east of the river mouth in November (LE east) and towards the offshore in May (LE offshore). Apatite was the smallest PP fraction and was mostly restricted to lake samples. River concentrations of NAIP, OP, and AP varied seasonally, decreasing in the fall when tributary discharge was low with little variability (Fig. 5); however, lake concentrations were similar from May to November with a few exceptions (Fig. 7). Seasonally, although concentrations differed, the

558

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

Table 4 Dissolved Phosphorus (DP) and sedimentary phosphorus content (i.e., the percentage of suspended solids that is phosphorus) characteristics in the lower Grand River and the adjacent nearshore of Lake Erie based on TP–SS empirical relationships from 2001 survey dates for each river and lake survey. The DP intercept columns show the intercept of TP (y-axis) regressed against SS (x-axis), where TP (μg/L) = P content (μg/mg) × SS (mg/L) + DP (μg/L). When SS = 0 mg/L, the phosphorus present must be dissolved. Month April ⁎ May ⁎ July ⁎ August ⁎ November ⁎

River DP (μg P/L)⁎⁎ (intercept) 43.3 29.9 63.3 65.8 53.6

Lake DP (μg P/L)⁎⁎ (intercept)

River P content (μg P/mg SS) (slope)

Lake P content (μg P/mg SS) (slope)

River P content (%)

Lake P content (%)

River R2

Lake R2

1.4 2.0 1.1 1.2 0.9

2.1 3.0 3.3 2.9 2.3

0.14 0.20 0.11 0.12 0.09

0.21 0.30 0.33 0.29 0.23

0.869 0.826 0.559 0.701 0.224

0.989 0.993 0.975 0.881 0.794

8.0 6.0 5.2 5.3 21.2

⁎ Lake TP–SS slopes are significantly different each month, and river TP–SS slopes are significantly different between months at p b 0.05. ⁎⁎ Lake and river TP–SS relationships are significantly different at p b 0.05 on each survey date.

ratio of NAIP to OP was consistent within the river and at the mouth (e.g., July: ~0.7; October: ~ 1.2; November: ~0.5), while the lake ratio was more variable. Covariation of particle size with phosphorus fractions Particle-size fractions were assessed to explore their ability to transport bioavailable PP. The 32 particle size fractions from 1.44 to 231 μm were aggregated into natural groupings using principal component analysis of logged concentrations. We found that 90% of the variation is explained by the first two components (axis 1: 62.5%; axis 2: 27.9%) once one outlier sample (identified by Systat) was removed. A varimax rotation of the loadings of the first two components revealed that the particle data had the following natural groupings: (a) b 3.76 μm (clay), (b) 4.43– 7.24 μm (silt), (c) 8.54–23 μm (silt), (d) 27.1–52.4 μm (silt), and (e) 61.7–231 μm (very fine to fine sand). Redundancy analysis between the particle size groups and the particulate phosphorus partitioning data (Fig. 8) suggests that NAIP is most strongly related to the 4.43–7.24 μm silt fraction (linear regression R2 = 0.677), and OP is related to the 8.54–23 μm silt fractions (linear regression R2 = 0.770). However, AP does not strongly relate to any of the particle size fractions (Fig. 8; all linear regressions with particle size fractions R2 b 0.081). Dreissenid mussel and Cladophora distribution East of the Grand River mouth, average mussel densities generally decreased eastward from the river mouth with the exception of the sites at 6 m depth (Table 5). Slightly lower densities were observed west of the river mouth at Transect 1 in comparison to Transect 2 directly east of the river mouth; average mussel densities ranged from 296 to 2669 individuals/m2 at Transect 1 compared with 0 to 3380 individuals/m2 east of the river mouth. The average density across all depths in Transect 2 (2141 individuals/m2) was greater than the average density from Transects 1, 3, and 4. A general trend of increasing mean shell length with depth occurred both east and west of the river mouth in Transects 1 to 4 with the exception of the shallower depths in Transects 1 and 3. Overall, average shell lengths were greatest in Transect 2 (mean of Transect 2: 14.7 mm; maximum: 25.4 mm) when all of the depths were considered collectively. Mean shell-free dry weight (SFDW) increased from the shoreline to the offshore with one exception in Transect 2 where the mean SFDW at the shallowest sample depth exceeded that at 6 m. Maximum and mean concentrations were greatest in Transect 2 at 8.5 and 3.5 g/m2, respectively. Mean soft tissue phosphorus concentrations of Dreissena were greater than mean shell phosphorus concentrations (Table 5). Average shell phosphorus ranged between 20.7 μg P/g DW and 82.6 μg P/g DW and generally increased with depth, with the exception of Transect 3. Similar to density, length, and weight, maximum and mean shell P concentrations were greatest in Transect 2 at 100.8 and 59.5 μg P/g DW, respectively, when data were compiled for the entire transect. Average tissue phosphorus ranged between 10.7 mg P/g AFDW and 23.5 mg P/g

AFDW. Although no clear relationships were apparent among depths, the average concentration for Transect 2 (when considered as a whole) was greater than those for the other transects. Similarly, tissue from Transect 2 overall contained higher average P concentrations than the other transects. Cladophora biomass was greatest at 3 m depth and decreased with depth (maximum values observed at 3 m vs 6 m sites; 104 g DW/m2 to 7.6 g DW/m2; Table 5). The highest biomass was observed at 3 m depth at Transect 3; elevated biomass exceeding 50 g DW/m2 was also observed at Transects 1 and 4. No biomass was observed at 10 m or deeper with the exception of Transect 1. Discussion Grand River mixing zone and nutrient patterns in the nearshore The variability in nearshore water quality adjacent to the Grand River mouth was strongly shaped by the discharge of nutrient-rich and turbid water from the river. The highest concentrations of TP, SRP, and nitrate + nitrite in the vicinity of the Grand River mouth were within the river plume, which is affected by the wind-driven coastal current (He et al., 2006). Eddies (Rao and Schwab, 2007) and frequent reversals of the current limit the alongshore extent of the mixing area creating a coastal band of nutrient-rich water distinguishable by its high conductivity (He et al., 2006). The dominant direction of the nearshore surface currents is eastward and consequently higher nutrient concentrations were often detected east of the river mouth. The water quality of the lower Grand River was influenced by exchange with the lake as seiches and/or wind-driven surface currents moving lake water into the lower river and river water over colder lake water. Chloride concentration, and water quality in general, was more variable near the river mouth compared to the more upstream reaches. Evidence of lake intrusion into the river includes concurrent drops in nearshore and river temperature (e.g., late May; Fig. 5). Within the river, lake effects have been noted to 8 km upstream at the Dunnville Dam, including changes in river level, water chemistry, and temperature (GRCA, 2013; MacDougall and Ryan, 2012). These authors note that storm surges resuspend sediments in the lower Grand River creating elevated total suspended solids and sediment bound TP. Temperature gradients between the river discharge and the nearshore strongly influence features of the mixing zone. During spring and early summer, the typically warmer and therefore buoyant river discharge caused vertical heterogeneity the mixing zone. As well, warm river discharge may have a lesser tendency to mix with offshore water; for example, this was evident in the shallow nearshore area east of the river during the warm spring of 2001. Warm river water has been observed regularly at a Ministry of the Environment and Climate Change (MOECC) monitoring station approximately 2.2 km from the river mouth during April and May sampling events (Howell, unpublished data). Greater alongshore spreading of a relatively warm and coherent plume in spring and early summer could increase its impact on nearshore biota at this time.

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

559

Nitrate concentrations were elevated in the river, but were lower than those noted by Rott et al. (1994) with the exception of the spring 2010 samples. TP and SRP at the mouth of the Grand River fell within the range found in past studies of the Grand River (Rott et al., 1994). In 2001, TP, nitrates, and suspended solids near the mouth of the river (Table 1) were lower than observed at the Provincial Water Quality Monitoring Network monitoring station at Dunnville Dam (PWQMN station ID 16018403502, Ontario Ministry of the Environment). In the mixing area of the plume and east of the river, SRP concentrations (Tables 1 and 2) were elevated and at levels that could support nuisance algal growth (e.g., Higgins et al., 2006). However, west of the river mouth, SRP was often below the detection limit (0.5 μg P/L). In 2001, the SRP at sites furthest from the shoreline and east of the river mouth were greater than annual basin-wide spring means reported by Barbiero et al. (2006) of approximately 2–3 μg P/L and Makarewicz et al. (2000) of 0.5–4. 6 μg P/L. Nearshore TP concentrations outside of the plume were also similar to the spring open-lake concentrations collected during Environment Canada’s Great Lakes Surveillance Program (Environment Canada, unpublished; Dove and Chapra, 2015) with average TP at stations 4 to 5 km offshore just below the meso-oligotrophic lake objective of 10 μg P/L of Annex 3 of the previous Great Lakes Water Quality Agreement. Away from the immediate river mixing zone nearshore TP was generally not much higher than the offshore. The similarity to basin-wide offshore concentrations is not surprising as the Grand River is the only major source to the eastern basin (Fraser, 1987) and historically contributed an estimated 25% of the tributary loadings to the eastern basin (Nicholls et al., 1983). Although high loads are observed in the spring, and to some extent in the fall (e.g., Schwab et al., 2009), significant pulses associated with storm related discharge events have been observed in other parts of Lake Erie in late spring to early summer (e.g., Michalak et al., 2013). Inter-annual TP variation is expected and likely reflects differences in external loadings to the basin (e.g., Eastern basin loading between 2001 and 2008 ranges between 768 and 1126.5 MTA; Dolan and Chapra, 2012). TP has decreased significantly in the eastern basin since the 1970s, although the trends are less pronounced than in the central and western basins (North et al., 2012). The spring data from 2001 are similar to the pre- and post-dreissenid invasion spring TP in the eastern basin reported by North et al. (2012) and lower than the average historical eastern basin concentrations reported by Chapra and Sonzogni (1979) and El-Shaarawi (1987). Declines in eastern basin summer TP since dreissenid invasion have also been reported (Makarewicz et al., 2000). The nearshore segment east of the Grand River mouth is more affected by river discharge. Conductivity mapping in 2002 indicates that direct effects of the mixing zone could be detected at least 8 km east from the mouth (Howell and Hobson, 2002). Current measurements in 2004 and 2010 identify a more frequent eastward orientation of the mixing area and also reveal a prevalent shore-parallel circulation in the surface layer. Seasonal temperature patterns in 2001 demonstrate that water discharged from the river will be as warm or warmer than the adjacent nearshore, with the exception of the fall, and likely to be well represented by surface measurements of currents and chemistry. Seasonal conductivity at a sensor east of the river in 2010 was elevated over ambient lake levels throughout its 6-month deployment demonstrating the enriching effect of the river. The lower frequency of westward alongshore flow events do not preclude significant influences of the river on the nutrient regime to the west of the river mouth, as was evident in the May 2001 and again in the 2010 chemistry data. Nearshore variability there is

Fig. 7. Concentrations of particulate phosphorus fractions in (a) May, (b) July, (c) October, and (d) November with site locations displayed on Fig. 1. NAIP is non-apatite inorganic phosphorus, AP is apatite-bound phosphorus, and OP is organic phosphorus, GR u/s = Grand River upstream, GR mouth = Grand River mouth, LE west = Lake Erie west of the mouth, LE east = Lake Erie east of the mouth, LE o/s = Lake Erie offshore, T = surface samples, B = bottom samples.

560

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

Fig. 8. Redundancy analysis between particulate phosphorus fractions and particle density in the Grand River and the nearshore of Lake Erie. Samples are labeled according to month of collection (May (M), July (J), October (O), November (N)), site (1 = west of mouth, 2 = east of mouth, 3 = offshore, 4 = mouth of river, 5 = upstream), and water column position (top, T, or bottom, B).

high, dominated by more lake-like conditions much of the time with occasional enrichment from the river. The enrichment is less extensive moving away from the river to the west than it is to the east. Table 5 Mean dreissenid valve length, density, shell free dry mass, and phosphorus concentrations in the tissues and shells at different depths along four transects in the nearshore region of the Grand River, Lake Erie, 2010. ns = no sample. Bottom rows are Cladophora biomass as dry weight.

Mean shell length (mm)

Mean density (individuals/m2)

Mean shell free dry mass (g/m2)

Soft tissue [P] (mg P/g AFDW)

Shell [P] (μg P/g DW)

Cladophora biomass (g DW/m2)

Depth

1

3m 6m 10 m 18 m 3m 6m 10 m 18 m 3m 6m 10 m 18 m 3m 6m 10 m 18 m 3m 6m 10 m 18 m 3m 6m 10 m 18 m

12.2 9.4 14.4 21.2 327 942 2669 296 ⁎

⁎ Sample condition insufficient for P analysis.

7.5 12.2 21.3 ⁎ 10.7 23.5 13.2 ⁎ 20.7 57.1 41.7 79.1 49.5 32.4 0

Transect number 2 3 9.8 10.6 13.9 24.5 3300 1167 3380 716 14.2 10.5 33 35.2 16.9 23.2 21.6 22.6 39 40.5 82.6 75.8 6.9 0 0 0

13.9 13.1 19.2 ns 1016 2080 1747 0 15.2 15.2 40.7 0 20.5 15.8 20.6 ns 50 42.7 74.9 ns 104.4 13.2 0 ns

4 12.9 ns 19.7 ns 1040 0 882 0 8.2 0 21.6 0 20 ns 15.9 ns 33.5 ns 57.4 ns 61 7.6 0 ns

Fate of riverine P The analysis of TP residuals from the linear regressions predicting TP from conductivity in the lower Grand River and the nearshore suggests that processes in the lower river transition from formation, resuspension (positive residuals), and dilution (near-zero residuals) of PP to consumption and settling (negative residuals) closer to the mouth and in the nearshore. Resuspension (as indicated by positive residuals) may also affect water column PP levels in the nearshore at times. Kuntz (2008) observed that dissolved nutrients in the river are transformed into phytoplankton in the reservoir above Dunnville Dam, and the river below Dunnville Dam has also been found to be productive (e.g., DeBruyn et al., 2004). It is likely that as water flows below Dunnville Dam dissolved nutrients will continue to be incorporated into biomass and particulate. The 2010 particulate P data indicate that there is proportionately more OP in the river particulate phosphorus than in the lake, suggesting high productivity or wash-in of organic material to the river. Consumption of PP in the lower river may account for the lower phosphorus per unit mass of SS observed in the river (0.1%) in comparison to the lake (0.3%; Table 4). The higher proportion of negative residuals in TP from conductivity in the lower river (Fig. 6: 65% negative, all surveys; N80% negative excluding summer surveys) suggests, in general, a net loss of TP occurs in the lowest reaches of the Grand River, particularly in the spring and fall. Loss of particulate P could be a result of sedimentation of particles as water velocity and turbulence decrease, or biological uptake of particles by benthos. Mayer and Manning (1989) suggest a settling of coarser inorganic particles as river flow declines nears the mouth. Because conductivity was not substantially different among the river locations, yet it was different between river and lake, dilution with lake water is not likely the cause of the pattern of TP residuals in the lower Grand River. Greater chlorophyll a, bacterial, and viral concentrations have been

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

observed within the lower Grand River in comparison to the mouth of the river (DeBruyn et al., 2004); despite that, P is apparently being lost. The nearshore samples by the Grand River mouth suggest that the phosphorus being resuspended in the nearshore is predominantly NAIP. The sharp drop in the proportion of OP from the river to the lake across all three nearshore sites suggests that either OP is lost rapidly upon the river water entering the lake, or there is a NAIP reservoir in the nearshore that undergoes resuspension and settling. Simple dilution of river particulates by lake particulates would be expected to a yield a gradient of decreasing OP proportions away from the river. Increases in OP may be observed with large scale decay or benthic microalgal resuspension; however, it does not appear that an event such as this has been captured. Redundancy analysis suggests the OP fraction is associated with a coarse silt fraction (8.5 to 23 μm), suggesting these particles and much of the OP settles out rapidly upon reaching the lake. The smaller silt particles (4.4 to 7.2 μm) containing AP appear to stay in suspension longer. The RDA also suggests that river particulates in summer are high in OP, while in spring and fall they have more NAIP. AP appears to be associated with the finest fraction and river samples on the biplot; however, the relationship between AP and the finest fraction is poor (linear regression R2 = 0.01). Low NAIP concentrations are observed east of the Grand River mouth in comparison to the sampling location furthest from the shore, which may be related to dreissenid and Cladophora distributions in the nearshore. DePinto et al. (1981) estimated for the Maumee River plume that it would take approximately 10 days for half of the particle volume (median particle diameter of 3 μm) to settle to the bottom at a mean depth of 7.3 m. Therefore, it is highly probable that suspended sediments would be transported a considerable distance before settling, allowing for appreciable release of bioavailable phosphorus in the water column. One might expect that PP associated with suspended solids 1 m above the lakebed would be different from near-surface PP, assuming that PP undergoes processing and diagenesis on and at the sediment surface; however, this does not appear to be the case at the nearshore sites. This could reflect the balance between consumption by algae and resuspension of fecal matter from dreissenids; alternatively, weak water column stratification in the nearshore and periodic water column mixing could erode water column patterns in particulate processing as particulates settle. The MOECC periodically measures SS and TP one meter above the lakebed at a reference station 2.2 km away from the river mouth. The data show a relationship between SS and TP (TP (μg P/L) = 4.8 SS (mg/L)–1.3, R2 = 0.699, n = 13 from 1998 to 2010, one statistical outlier removed; unpublished data). Using this relationship, we can estimate the maximum amount of P that could potentially be deposited to the sediments from the water column assuming (1) P can be predicted from the suspended sediments, (2) the water quality 1 m above the lakebed is representative of what can sediment, and (3) there is little exchange between the dissolved and particulate phases. Using this reasoning, 3.5 mg P/g of particulate is available for sedimentation 1 m above the lake bed and our PP samples from the base of water column suggest this PP could be bioavailable. This is close to the estimated P content of 2.1 to 3.3 μg P/mg SS for lake stations based on similar TP to SS regressions in Table 4. Baker et al. (2014) determined that 26–30% of the PP in Ohio tributaries entering Lake Erie was bioavailable. DePinto et al. (1981) suggested that riverine SS will settle or be transported to depth prior to the release of its bioavailable phosphorus. However, they completed their study prior to invasion of dreissenid mussels to the nearshore environment. Since that time, dreissenid mussels have re-engineered the littoral areas with hard substrate (Hecky et al., 2004). Dreissenids intercept particulate material that was transported to the offshore and transform some of it into dissolved P accessible by benthic algae such as Cladophora (e.g., Ozersky et al., 2009; Zhang et al., 2011).

561

There is a transition zone at the confluence of the river mouth and the nearshore of Lake Erie where TP deposition, consumption, and resuspension processes are occurring concurrently resulting in net loss or gain of TP from the water column that is spatially and temporally variable. Although TP is elevated in comparison to the remainder of the study area, on most occasions (65% of all occasions, or N 80% excluding summer), our analysis of residuals of TP from conductivity suggest that phosphorus is settling or being consumed. Substrate maps (St. Jacques and Rukavina, 1973) indicate an area of fine sediment (sandy mud and silty sand) to the east of the river mouth that corresponds with the mixing area. Our PP analysis suggests that phosphorus associated with fine sediments settling in the mixing area of the plume is mostly OP and NAIP. This may comprise an enriched food supply for benthic organisms. Phosphorus-containing organic particles captured by filter feeding dreissenids may be mineralized by the mussels or deposited as feces and pseudofeces that may be mineralized by deposit feeders or microbes, liberating soluble phosphorus available to benthic algae such as Cladophora. Higher biomass of benthic invertebrates has been observed at the mouth of the Grand River even before the arrival or dreissenids (Barton, 1983). Outside of the plume’s more frequent eastward path, lower nutrient concentrations prevail but are supplemented by periodic reversals of alongshore currents and possibly resuspension events re-introducing particulate P into the moving water column. Ongoing erosion at shallow depths and strong resuspension events create occasions when TP concentrations are higher than expected given the conductivity. While Fig. 6 does not capture many of these events, episodes of high TP and low conductivity are evident. For example, in August and November, elevated TP at the river mouth and in Zone 1 (Splatt’s Bay) occur in low conductivity conditions. The presence of AP in particulate material from the lake during the October and November surveys could suggest an increased contribution of resuspended material at that time. Surficial sediments within Lake Erie are predominantly silt and clay (Thomas et al., 1976), and major storms are the predominant driver of their resuspension and transport (Lick et al., 1994). During August from 1998 to 2010, approximately every 3 years, the MOECC collected composite samples of surficial sediment (0–3 cm) at their reference station measuring nutrient and metal concentrations. Phosphorus concentrations ranged from 0.72 to 1.1 mg P/g (MOECC, unpublished data). This is only slightly lower than the inferred P of particulate material in the water samples collected in the Grand River in 2001 ranging from 0.9 to 2.0 and averaging 1.3 mg P/g (Table 4). The concentration of TP in sediments offshore of the mouth of the Grand is similar to historical TP concentrations in surficial sediments between Port Dover and Port Colborne observed by Williams et al. (1976; range: 0.8–1.26 mg P/g). The inferred P content of water column particulate material, which may include resuspended surficial sediments, in the lake available for deposition via settling (Table 4) is approximately 3 times the P content of surficial sediment alone. This suggests that the release of DP from the particulates does occur, or that phosphorus bound to suspended solids is intercepted prior to reaching the lakebed. With high concentrations of NAIP and fine particles at the base of the water column, it is probable that there is sedimentation and resuspension of bioavailable phosphorus within this nepheloid layer. It is important to assess the bioavailability of phosphorus settling to the lakebed to understand its potential as a nutrient source to the dreissenid mussel and Cladophora communities in the nearshore. Different forms of phosphorus may be related to different size fractions of sediment and these will settle at different rates. As noted, OP will settle out of the water column first and NAIP will be transported further. The association of NAIP and OP with fine-grained particles in Lake Erie was also observed by Williams et al. (1976). Decreasing grain size with increasing phosphorus content was also observed by Viner (1982) and Mudroch and Duncan (1986). Similarly, Stone and English (1993) found higher NAIP and OP concentrations in small particles in

562

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

two Lake Erie tributaries draining into the western and central basins, while in contrast AP concentrations decreased with decreasing grain size. They also found that the potential release of bioavailable phosphorus was most significant in the b8 μm size fraction of tributary sediments. Williams et al. (1980) cultured the green alga Scenedesmus quadricauda using suspended solids collected during a high runoff period from the Grand River and also from a number of other runoff and erosional bluff locations in Lakes Ontario and Erie. They found increased algal growth in samples from both Lakes Ontario and Erie with additions of NAIP and OP, while decreases occurred with increased AP. Similarly, the biological availability of NAIP is considered to be high in Ohio tributaries entering the southern shores of Lake Erie, while the availability of apatite is low (Logan et al., 1979). If we assume that the NAIP and OP speciation has not changed since Williams et al. (1980), bioavailable phosphorus from the Grand River is likely distributed along the shoreline in the form of fine particulates and could be a potential nutrient source for dreissenids and Cladophora. Implications of the Grand River discharge for dreissenid mussels and Cladophora Phosphorus sequestration and release by the lakebed can affect, and be affected by, a variety of benthic organisms. Dreissenid mussels are benthic filter feeders that excel at capturing particulates near the bottom of the water column and repackaging and transforming them (e.g., Holland et al., 1995; Klerks et al., 1996; Zhang et al., 2011). It is conceivable that the loading of particulate material and nutrients along with variability in water quality over the nearshore adjacent to the Grand River mouth affects the distribution and productivity of both mussels and Cladophora. The mixing and dilution of the Grand River discharge creates gradients in water column particulate material and phytoplankton (as inferred from chlorophyll a) and presumably the food supply for the filter-feeding dreissenids. Assuming the supply of particulates in the discharge augments the food supply for a food-limited population, particularly along the area most frequently affected by the mixing plume, it might be expected that the pattern in lake mixing may mirror biomass and phosphorus content of dreissenid mussels on the lakebed. Exposure to physical disturbance from wave action may also influence densities and/or biomass. While the pattern is inconclusive, mussel densities appeared to be elevated along the more frequent eastward path of the Grand River plume closer to the shoreline. The phosphorus content of mussel tissue and shells were variable, however, consistent with prior analyses of western basin Lake Erie shell and soft tissue by Arnott and Vanni (1996). Maximum values of tissue and shell P were higher on the two eastward transects nearest the river mouth. Dreissena densities and biomass in 2010 were lower than the average eastern basin densities observed by Patterson et al. (2005; density = 9481 mussels/m2). There are a number of factors that may contribute to the apparent difference. Patterson et al. (2005) used 17 sites for the entire eastern basin in comparison to the 16 sites we sampled from the Grand River region. Our quadrats are larger (0.15 m2) than the 0.0214 m2 airlifts on rocky substrata ≤ 10 m deep and the 0.0225 m2 Ekman grabs at 20 m they employed in 2002. Their sites were all located between Turkey and Peacock points, west of the Grand River. Since the Grand River is the largest input in the eastern basin, it is possible that the Patterson et al. (2005) sites are not representative of the Grand River region. Densities by the Grand River mouth likely reflect the riverine inputs and the physical dynamics of the area. High turbidity is periodically observed by the mouth of the Grand, which may not be conducive to mussel growth. This may be aggravated by siltation of the hard substrate. Based on diver observations, there is abundant soft sediment in the area. It is also possible that densities in 2010 have declined since the 2002 surveys reported in Patterson et al. (2005); these authors noted a 53% decline between 1998 and 2002.

Sedimentation of particulate material originating in the Grand River discharge is also a likely mechanism for the flux of phosphorus to the dreissenid-colonized lakebed. More suspended solids originating from the Grand River settle east and west of the river mouth than in deeper waters further offshore. The sedimenting PP is predominately OP, which can be expected to contribute to benthic production. Cladophora is prevalent on the nearshore lakebed of eastern Lake Erie, including the nearshore adjacent to the Grand River mouth. The growth of Cladophora is limited by phosphorus (Higgins et al., 2008), suggesting that the augmentation of phosphorus levels in the nearshore adjacent to the Grand River mouth might enhance the growth of Cladophora in a pattern consistent with the river mixing tendencies. The biomass of Cladophora over the study area was high (reaching in excess of 50 g/m2 dry weight) in three of the four transects. There was no clear pattern in abundance among transects except that the east transect nearest the river mouth had little Cladophora. The shallow depth distribution of biomass (little past 6 m except at Transect 1) is consistent with the high light requirements of Cladophora for optimal growth (e.g., Higgins et al., 2008). Higgins et al. (2005) found Cladophora biomass of 180 g/m2 at depths of 0–2 m at Splatt’s Bay in 1995. The absence of Cladophora east of Splatt’s Bay at Transect 2, an area frequently affected by the turbid plume of the river, is possibly a result of light limitation. Areas of mussel colonization are likely sources for DP and sinks for PP. Filtration by mussels can result in decreases in TP, chlorophyll, and SS in the nearshore adjacent to mussel beds (Fahnenstiel et al., 1995; Nicholls et al., 1999, 2001). Turner (2010) noted that the flux of SRP from mussels was variable but increased when the mussels were located on sediments vs higher within the water column. Similarly, Martin (2010) measured high SRP adjacent to mussel beds while Ozersky et al. (2009) noted that, in Lake Ontario, SRP excreted by dreissenid mussels could support observed Cladophora growth. Armenio et al. (2016) found that Cladophora grown in close proximity to Dreissena mussels have elevated growth rates and assimilate more P than Cladophora grown without Dreissena. Mussels may also affect nearshore environments by increasing the hard surface area that Cladophora can colonize, improving water clarity, and also increasing the bioavailability of phosphorus. Cladophora would therefore benefit from both the SRP released by mussels and the substrate afforded by the mussels themselves (Higgins et al., 2008). Despite the awareness that mussels are potentially impacting the phosphorus distribution in the nearshore, particularly near the lakebed, the degree that they alter water column SRP concentrations near the Grand River is not known. The density of mussels and their phosphorus content can be estimated for the study area using the distribution of hard substrate, the bathymetry of the region, and mussel P content in 4 depth zones (0 to 5 m, 5–10 m, 10–15 m, and 15–20 m). We assume that mussels are only growing on the hard substrata (approximately 45 km2 of the total study area in Fig. 1) and that the densities and P content observed within the quadrats were representative of the zones. The average mussel density across the hard substrata of the Grand River nearshore was calculated as approximately 1705 individuals/m2 with 66% of the individuals within the 5 to 10 m bathymetry contours. Over the hard substrata of the study area, approximately 8.6 metric tons of P would be contained with the mussels, mostly within 10 to 15 m. Using the loadings of Depew et al. (2011), we estimate that the Grand River can supply 8.6 metric tons of P in 14–16 days. Using data on the whole eastern basin from Patterson et al. (2005), and the assumptions that the tissue:shell mass and the average shell and tissue P from the current study are representative, we estimate that the Grand River would take 77–85 days to supply the estimated 46.8 metric tons of contained in dreissenids. Using the same suitable substrata area for Cladophora biomass as for mussels, and using an internal Cladophora P content ranging between 0.052% and 0.23% DM (e.g., Higgins et al., 2005ab, Ontario Ministry of the Environment, 1998), approximately 1.1 to 4.9 metric tons of

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

phosphorus could be contained within the Cladophora which corresponds to 2–9 days of loading from the Grand River (based on average daily loading). Put another way, the standing stock of mussels and Cladophora is a small fraction (~4 to 7%) of the estimated river load entering Lake Erie (~200–220 T/yr; Depew et al., 2011), but the phosphorus contained within them can be supplied by the Grand River in 16–25 days; this leaves the balance of riverine P available, in addition to the re-introduction of large amounts of phosphorus to the water column through die-off and sloughing events, for distribution and ecosystem sustenance beyond the nearshore region of the Grand River. The spatial and temporal heterogeneity near the mouth of the Grand River and the adjacent shoreline presents challenges to efforts to predict the biological outcomes of nutrient loading to the nearshore. A recent indication of this is the absence to date of phosphorus loading targets to abate the overabundance of Cladophora in the eastern basin as part of the Lake Erie Annex 4 phosphorus objectives (US-EPA and Environment Canada, 2015). Difficulty in quantitatively relating the shoreline loading of bioavailable phosphorus to Cladophora growth remains. Our work does not provide an answer to dealing with nearshore nutrient variability but clearly points to the necessity of using fine-scale and depth variable modeling approaches for quantifying physical conditions, water quality, and biological outcomes in an integrative manner that can inform nutrient management. Conclusion Spatial patterns of nutrients in the nearshore of Lake Erie near the Grand River are controlled by the plume of the river and by the physical dynamics of the nearshore. Nutrient and suspended solids concentrations were elevated in the river compared to the lake and declined with distance away from the river mouth and river mixing area. The mixing area with elevated nutrients is usually east of the Grand River mouth due to the prevalence of eastward alongshore currents in the region predominantly within 2–3 km of the shoreline. Although elevated phosphorus concentrations are observed close to the shoreline, mean background TP levels are generally below the meso-oligotrophic lake objective of 10 μg/L of Annex 3 of the previous Great Lakes Water Quality Act. Zones of sedimentation and resuspension were identified, with dominant processes transitioning from P production or resuspension in the upper reaches of the river to consumption and sedimentation near the river mouth and in the nearshore. Particulate phosphorus in the lower river was mostly OP and NAIP; however, in the lake, the particulates were predominantly NAIP. NAIP was associated with finer particles, with possibly lower sedimentation velocities. The high variability in particle size distribution and its relationship to eutrophication potential suggest that future studies should not only consider total particulate phosphorus loading but also consider particle size, bioavailability, and transport potential. The nutrient patterns observed in the nearshore likely influence the local ecology, with higher mussel and Cladophora densities and higher mussel phosphorus content within the Grand River-nearshore mixing zone. Acknowledgments We would like to thank past and present field crews of the Great Lakes Unit at the Environmental Monitoring and Reporting Branch. We are also grateful to Ralph Smith, who allowed us to work in his lab, Amy Hennesey who provided valuable methodology and spectrophotometry assistance, and Mohamed Mohamed for helpful comments of previous drafts. We thank Bruce Grey and Great Lakes Field group at Environment Canada for collection of the dreissenid and Cladophora samples. Helpful feedback provided by anonymous reviewers greatly improved this manuscript. Funding was provided by the Canada Ontario Agreement and a NSERC Canada Discovery Grant.

563

References Armenio, P.M., Mayer, C.M., Heckathorn, S.A., Bridgeman, T.B., Panek, S.E., 2016. Resource contributions from dreissenid mussels to the benthic algae Lyngbya wollei (Cyanobacteria) and Cladophora glomerata (Chlorophyta). Hydrobiologia 763, 35–51. Arnott, D.L., Vanni, M.J., 1996. Nitrogen and phosphorus recycling by the zebra mussel (Dreissena polymorpha) in the western basin of Lake Erie. Can. J. Fish. Aquat. Sci. 53, 646–659. Baker, D.B., 1985. Regional water quality impacts of intensive row-crop agriculture: a Lake Erie basin case study. J. Soil Water Conserv. 40, 125–132. Baker, D.B., Confesor, R., Ewing, D.E., Johnson, L.T., Kramer, J.W., Merryfield, B.J., 2014. Phosphorus loading to Lake Erie from Maumee, Sandusky and Cuyahoga rivers. The importance of bioavailability 40, pp. 502–517. Barbiero, R.P., Rockwell, D.C., Warren, G.J., Tuchman, M.L., 2006. Changes in spring phytoplankton communities and nutrient dynamics in the eastern basin of Lake Erie since the invasion of Dreissena spp. Can. J. Fish. Aquat. Sci. 63, 1549–1563. Barton, D.R., 1983. A Survey of Benthic Macroinvertebrates Near the Mouth of the Grand River, Lake Erie, 1981. Ontario Ministry of the Environment. Queen’s Printer for Ontario. Bostan, V., Dominik, J., Bostina, M., Pardos, M., 2000. Forms of particulate phosphorus in suspension and in bottom sediment in the Danube Delta. Lakes Reserv. Res. Manag. 5, 105–110. http://dx.doi.org/10.1046/j.1440-1770.2000.00104.x. Budd, J.W., Beeton, A.M., Stumpf, R.P., Culver, D.A., Kerfoot, W.C., 2001. Satellite observations of Microcyctis blooms in western Lake Erie. Verh. Int. Ver. Theor. Angew. Limnol. 27, 3787–3793. Chapra, S.C., Sonzogni, W.C., 1979. Great Lakes total phosphorus budget for the mid 1970s. J. WPCF 51 (10), 2524–2533. Chen, C., Wang, L., Ji, R., Budd, J.W., Schwab, D.J., Beletsky, D., Fahnenstiel, G.L., Vanderploeg, H., Eadie, B.J., Cotner, J.B., 2004. Impacts of suspended sediment on the ecosystem in Lake Michigan: a comparison between the 1998 and 1999 plume event. J. Geophys. Res. 109, C10S05 (18 pp.). Chen, X., Driscoll, C.T., 2009. Watershed land use controls on chemical inputs to Lake Ontario embayments. J. Environ. Qual. 38, 2084–2095. Chow, J., Abbey, A.I., Khan, Z., Dermicheva, S., Jennings, W., Wilson, P., 2010. 2009 performance report: general chemistry and microbiology analyses section. Ontario Ministry of the Environment Report. Queen’s Printer for Ontario, Ontario, Canada. Conroy, J.D., Culver, D.A., 2005. Do dreissenids affect Lake Erie ecosystem stability processes? Am. Midl. Nat. 153, 20–32. Conroy, J.D., Edwards, W.J., Pontius, R.A., Kane, D.D., Zhang, H., Shea, J.F., Richey, J.N., Culver, D.A., 2005. Soluble nitrogen and phosphorus excretion of exotic freshwater mussels (Dreissena spp.): potential impacts for nutrient re-mineralization in western Lake Erie. Freshw. Biol. 50, 1146–1162. DeBruyn, J.M., Leigh-Bell, J.A., McKay, R.M.L., Bourbonniere, R.A., Wilhelm, S.W., 2004. Microbial distributions and the impact of phosphorus on bacterial activity in Lake Erie. J. Great Lakes Res. 30 (1), 166–183. Depew, D.C., Guildford, S.J., Smith, R.E.H., 2006. Nearshore–offshore comparison of chlorophyll a and phytoplankton production in the dreissenid colonized eastern basin of Lake Erie. Can. J. Fish. Aquat. Sci. 63, 1115–1129. Depew, D.C., Houben, A.J., Guildford, S.J., Hecky, R.E., 2011. Distribution of nuisance Cladophora in the lower Great Lakes: patterns with land use, near shore water quality and dreissenid abundance. J. Great Lakes Res. 37, 656–671. DePinto, J.V., Young, T.C., Martin, S.C., 1981. Algal-available phosphorus in suspended sediments from lower Great Lakes tributaries. J. Great Lakes Res. 7 (3), 311–325. Dolan, D.M., Chapra, S.C., 2012. Great Lakes total phosphorus revisited: 1. Loading analysis and update (1994–2008). J. Great Lakes Res. 38 (4), 730–740. Dolan, D.M., McGunagle, K.P., 2005. Lake Erie total phosphorus loading analysis and update: 1996–2002. J. Great Lakes Res. 31 (Sup.1), 11–22. Dove, A., Chapra, S.C., 2015. Long-term trends of nutrients and trophic response variables for the Great Lakes. Limnol. Oceanogr. 60, 696–721. http://dx.doi.org/10.1002/lno. 10055. El-Shaarawi, A.H., 1987. Water quality changes in Lake Erie, 1968–1980. J. Great Lakes Res. 13 (4), 674–683. Environment Canada and the U.S. Environmental Protection Agency, 2014g. State of the Great Lakes 2011. Cat No. En161-3/1-2011E-PDF. EPA 950-R-13-002. Available at http://binational.net. Fahnenstiel, G.L., Bridgeman, T.B., Lang, G.A., McCormick, M.J., Nalepa, R.F., 1995. Phytoplankton productivity in Saginaw Bay, Lake Huron: effects of zebra mussel (Dreissena polymorpha) colonization. J. Great Lakes Res. 21, 465–475. Fraser, A.S., 1987. Tributary and point source total phosphorus loading to Lake Erie. J. Great Lakes Res. 13, 659–666. GRCA, 2013. A Framework for Identifying Indicators of Water Resource Conditions— Support of Ecological Health by Water Resources in the Grand River–Lake Erie Interface. Retrieved from http://www.grandriver.ca/waterplan/WaterResource IndicatorFramework_V2.pdf. Green, S.A., Eadie, B.J., 2004. Introduction to special section: transport and transformation of biogeochemically important materials in coastal waters. J. Geophys. Res. 109, C10S01. http://dx.doi.org/10.1029/2004JC002697. He, C., Rao, Y.R., Skafel, M.G., Howell, T., 2006. Numerical modelling of the Grand River plume in Lake Erie during unstratified period. Water Qual. Res. J. Can. 41 (1), 16–23. Hecky, R.E., Smith, R.E.H., Barton, D.R., Guildford, S.J., Taylor, W.D., Charlton, M.N., Howell, T., 2004. The nearshore phosphorus shunt: a consequence of ecosystem engineering by dreissenids in the Laurentian Great Lakes. Can. J. Fish. Aquat. Sci. 61, 1285–1293. Higgins, S.N., Hecky, R.E., Guildford, S.J., 2006. Environmental controls of Cladophora growth dynamics in eastern Lake Erie: application of the Cladophora growth model (CGM). J. Great Lakes Res. 32, 629–644.

564

K.M. Chomicki et al. / Journal of Great Lakes Research 42 (2016) 549–564

Higgins, S.N., Howell, E.T., Hecky, R.E., Guildford, S.J., Smith, R.E., 2005. The wall of green: the status of Cladophora glomerata on the northern shores of Lake Erie’s eastern basin, 1995–2002. J. Great Lakes Res. 31, 547–563. Higgins, S.N., Malkin, S.Y., Howell, E.T., Guildford, S.J., Campbell, L., Hiriart-Baer, V., Hecky, R.E., 2008. An ecological review of Cladophora glomerata (Chlorophyta) in the Laurentian Great Lakes. J. Phycol. 44, 839–854. Holland, R.E., Johengen, T.H., Beeton, A.M., 1995. Trends in nutrient concentrations in Hatchery Bay, western Lake Erie, before and after Dreissena polymorpha. Can. J. Fish. Aquat. Sci. 52, 1202–1209. Howell, E.T., Barton, D.R., Fistsch, C.-L., Kaltenecker, G., 2014. Fine-scale nutrient enrichment and water quality on the rural shores of Southeast Lake Huron. J. Great Lakes Res. 40 (1), 126–140. Howell, E.T., Chomicki, K.M., Kaltenecker, G., 2012. Tributary discharge, lake circulation and lake biology as drivers of water quality in the Canadian Nearshore of Lake Ontario. J. Great Lakes Res. 38, 47–61. Howell, T., Hobson, G., 2002. Technical Memorandum: Water Quality On the Lake Erie Shores of Wainfleet Township in 2002. Water Monitoring Section Environmental Monitoring and Reporting Branch Ontario Ministry of the Environment. Ji, R., Chen, C., Budd, J.W., Schwab, D.J., Beletsky, D., Fahnenstiel, G.L., Johengen, T.H., Vanderploeg, H.A., Eadie, B.J., Cotner, J.B., Gardner, W., Bundy, M.H., 2002. Influences of suspended sediments on the ecosystem in Lake Michigan: a 3-D coupled biophysical modeling experiment. Ecol. Model. 152, 169–190. Jyrkama, M.I., Sykes, J.F., 2007. The impact of climate change on spatially varying groundwater recharge in the grand river watershed (Ontario). J. Hydrol. 338, 237–250. Klerks, P.L., Fraleigh, P.C., Lawnizak, J.E., 1996. Effects of zebra mussels (Dreissena polymorpha) on seston levels and sediment deposition in western Lake Erie. Can. J. Fish. Aquat. Sci. 53, 2284–2291. Kuntz, T., 2008. System and Plankton Metabolism in the Lower Grand River, Ontario M.Sc. thesis University of Waterloo, Waterloo, ON. Lesht, B.M., Rockwell, D.C., 1985. The state of the middle Great Lakes: results of the 1983 water quality survey of Lakes Erie, Huron and Michigan. Report No. ANL/ER-85-2. Environmental Research Division, Argonne National Laboratory, Argonne, IL. Lesht, B.M., Rockwell, D.C., 1987. The state of the middle Great Lakes: results of the 1984 water quality survey of Lakes Erie, Huron and Michigan. Report No. ANL/ER-87-1. Environmental Research Division, Argonne National Laboratory, Argonne, IL. Lick, W., Lick, J., Ziegler, C.K., 1994. The resuspension and transport of fine-grained sediments in Lake Erie. J. Great Lakes Res. 20 (4), 599–612. Logan, T.J., Oloya, T.O., Yaksich, S.M., 1979. Phosphate characteristics and bioavailability of suspended sediments from streams draining into Lake Erie. J. Great Lakes Res. 5 (2), 112–123. MacDougall, T.M., Ryan, P.A., 2012. An Assessment of Aquatic Habitat in the Southern Grand River, Ontario: Water Quality, Lower Trophic Levels, and Fish Communities. Lake Erie Management Unit, Provincial Services Division, Fish and Wildlife Branch, Ontario Ministry of Natural Resources, Port Dover, Ontario (141 pp. + appendices). Makarewicz, J.C., Bertram, P., 1991. Evidence for the restoration of the Lake Erie ecosystem. Bioscience 41, 216–223. Makarewicz, J.C., Bertram, P., Lewis, T.W., 2000. Chemistry of offshore surface waters of Lake Erie: pre- and post-Dreissena introduction (1983–1993). J. Great Lakes Res. 26 (1), 82–93. Makarewicz, J.C., Booty, W.G., Bowen, G.S., 2012. Tributary phosphorus loading to Lake Ontario. J. Great Lakes Res. 38 (Sup. 4), 14–20. Martin, G., 2010. Nutrient Sources for Excessive Growth of Benthic Algae in Lake Ontario as Inferred by the Distribution of SRP M.Sc. thesis University of Waterloo, Waterloo, ON. Mayer, R., Manning, P.G., 1989. Variability of phosphorus forms in suspended solids at the Lake Erie–Grand River confluence. J. Great Lakes Res. 15 (4), 687–699. Michalak, A.M., Anderson, E.J., Beletsky, D., Boland, S., Bosch, N.S., Bridgeman, T.B., Chaffin, J.D., Cho, K., Confesor, R., Daloglu, I., DePinto, J.V., Evans, M., Fahnenstiel, G.L., He, L., Ho, J.C., Jenkins, L., Johengen, T.H., Kuo, K.C., LaPorte, E., Liu, X., McWilliams, M.R., Moore, M.R., Posselt, D.J., Richards, R.P., Scavia, D., Steiner, A.L., Verhamme, E., Wright, D.M., Zagorski, M.A., 2013. Record-setting algal bloom in Lake Erie caused by agricultural and meteorological trends consistent with expected future conditions. PNAS 110 (16), 6448–6452. Mudroch, A., Duncan, G., 1986. Distribution of metals in different size fractions of sediment from the Niagara River. J. Great Lakes Res. 11, 193–200. Murphy, J., Riley, J.P., 1962. A modified single solution method for the determination of phosphate in natural waters. Anal. Chim. Acta 27, 31–36. Murthy, C.R., Simons, T.J., Lam, D.C.L., 1986. Dynamic and transport modeling of the Niagara River plume in Lake Ontario. Rapp. P.V. Reun. Cons. Int. Explor. Mer. 186, 150–164. Nicholls, K.H., Hopkins, G.J., Standke, S.J., 1999. Reduced chlorophyll to phosphorus ratios in nearshore Great Lakes waters coincide with the establishment of dreissenid mussels. Can. J. Fish. Aquat. Sci. 56, 153–161. Nicholls, K.H., Hopkins, G.J., Standke, S.J., Nakamoto, L., 2001. Trends in total phosphorus in Canadian nearshore waters of the Laurentian Great Lakes: 1976–1999. J. Great Lakes Res. 27, 402–422.

Nicholls, K.H., Taylor, R., Hamdy, Y., 1983. The influence of the Grand River on phytoplankton near the northeastern shore of Lake Erie during 1979. Arch. Hydrobiol. 98, 146–172. North, R.L., Smith, R.E.H., Hecky, R.E., Depew, D.C., Leon, L.F., Charlton, M.N., Guildford, S.J., 2012. Distribution of seston and nutrient concentrations in the eastern basin of Lake Erie pre- and post-dreissenid mussel invasion. J. Great Lakes Res. 38, 463–476. Ontario Ministry of the Environment, 1998. Occurrence of the Alga Cladophora Along the North Shore of Eastern Lake Erie in 1995. Queen’s Printer for Ontario. Ozersky, T., Malkin, S.Y., Barton, D.R., Hecky, R.E., 2009. Dreissenid phosphorus excretion can sustain C. glomerata growth along a portion of Lake Ontario shoreline. J. Great Lakes Res. 35 (3), 321–328. http://dx.doi.org/10.1016/j.jglr2009.05.001. Patterson, M.W.R., Ciborowski, J.J.H., Barton, D.R., 2005. The distribution and abundance of Dreissena species (Dreissenidae) in Lake Erie, 2002. J. Great Lakes Res. 31 (Suppl.2), 223–237. Pennuto, C.M., Burlakova, L.E., Karatayev, A.Y., Kramer, J., Fischer, A., Mayer, C., 2014. Spatiotemporal characteristics of nitrogen and phosphorus in the benthos of nearshore Lake Erie. J. Great Lakes Res. 40 (3), 541–549. Pennuto, C.M., Howell, E.T., Lewis, T.K., Makarewicz, J.C., 2012. Dreissena population status in nearshore Lake Ontario. J. Great Lakes Res. 38 (Sup. 4), 161–170. Rao, Y.R., Schwab, D.J., 2007. Transport and mixing between the coastal and offshore waters in the Great Lakes: a review. J. Great Lakes Res. 33, 202–218. Rao, Y.R., Murthy, C.R., McCormick, M.J., Miller, G.S., Saylor, J.H., 2002. Observations of circulation and coastal exchange characteristics in southern Lake Michigan during 2000 winter season. Geophys. Res. Lett. 29 (0). http://dx.doi.org/10.1029/2002GL014895. Rockwell, D.C., Salisbury, D.K., Lesht, B.M., 1989. Water quality in the middle Great Lakes: results of the 1985 USEPA survey of Lakes Erie, Huron and Michigan. U.S.EPA Report: EPA-905/6/89-001. Rosa, F., 1987. Lake Erie central basin total phosphorus trend analysis from 1968 to 1982. J. Great Lakes Res. 13 (4), 667–673. Rosamond, M.S., Thuss, S.J., Schiff, S.L., Elgood, R.J., 2011. Coupled cycles of dissolved oxygen and nitrous oxide in rivers along a trophic gradient in Southern Ontario, Canada. J. Environ. Qual. 40, 256–270. Rott, E., Duthie, H.C., Pipp, E., 1994. Monitoring organic pollution and eutrophication in the Grand River, Ontario, by means of diatoms. Can. J. Fish. Aquat. Sci. 55 (6), 1443–1453. Schwab, D.J., Beletsky, D., DePinto, J., Dolan, D.M., 2009. A hydrodynamic approach to modeling phosphorus distribution in Lake Erie. J. Great Lakes Res. 35, 50–60. St. Jacques, D.A., Rukavina, N.A., 1973. Lake Erie nearshore sediments—Mohawk point to Port Burwell, Ontario. Proc. 16th Conf. Great Lakes Res.International Association of Great Lakes Research, pp. 454–467 Stone, M., English, M.C., 1993. Geochemical composition, phosphorus speciation and mass transport of fine-grained sediment in two Lake Erie tributaries. Hydrobiologia 253, 17–29. Thomas, R.L., Jaquet, J.-M., Kemp, A.L.W., Lewis, C.F.M., 1976. Surficial sediments of Lake Erie. J. Fish. Res. Board Can. 33, 385–403. Turner, C.B., 2010. Influence of zebra (Dreissena polymorpha) and quagga (Dreissena rostriformis) mussel invasions on benthic nutrient and oxygen dynamics. Can. J. Fish. Aquat. Sci. 67, 1899–1908. US-EPA, Environment Canada, 2015. Recommended phosphorus loading targets for Lake Erie. Annex 4 Objectives and Target Development Task Team Final Report to the Nutrients Annex Subcommittee (Retrieved from https://www.epa.gov/sites/production/ files/2015-06/documents/report-recommended-phosphorus-loading-targets-lakeerie-201505.pdf). Vanderploeg, H.A., Liebig, J.R., Carmichael, W.W., Agy, M.A., Johengen, T.H., Fahnenstiel, G.L., Nalepa, T.F., 2001. Zebra mussel (Dreissena polymorpha) selective filtration promoted toxic Microcystis blooms in Saginaw Bay (Lake Huron) and Lake Erie. Can. J. Fish. Aquat. Sci. 58, 1208–1221. Venkiteswaran, J.J., Rosamond, M.S., Schiff, S.L., 2014. Nonlinear response of riverine N2O fluxes to oxygen and temperature. Environ. Sci. Technol. 48 (3), 1566–1573. Vincent, R.K., Qin, X., McKay, R.M.L., Miner, J., Czajkowski, K., Savino, J., Bridgeman, T., 2004. Phycocyanin detection from LANDSAT TM data for mapping cyanobacterial blooms in Lake Erie. Remote Sens. Environ. 89, 381–392. Viner, A.B., 1982. A qualitative assessment of the nutrient phosphate transported by particulates in a tropical river. Rev. Hydrobiol. Trop. 15, 3–8. Williams, J.D.H., Jaquet, J.-M., Thomas, R.L., 1976. Forms of phosphorus in the surficial sediments of Lake Erie. J. Fish. Res. Board Can. 33, 413–429. Williams, J.D.H., Shear, H., Thomas, R.L., 1980. Availability to Scenedesmus quadricauda of different forms of phosphorus and sedimentary materials from the Great Lakes. Limnol. Oceanogr. 25, 1–11. Winter, J.G., Duthie, H.C., 2000. Stream epilithic, epiphytic diatoms: habitat fidelity and use in biomonitoring. Aquat. Ecol. 34, 345–353. Zhang, H., Culver, D.A., Boegman, L., 2011. Dreissenids in Lake Erie: an algal filter or a fertilizer? Aquat. Invasions 6 (2), 175–194.