In the early 2000s, harmful algal blooms and hypoxia returned to Lake Erie following a period of ecosystem recovery from the late 1980s through the 1990s. This corresponds to a drastic increase in dissolved reactive phosphorus loads and flow-weighted mean concentrations in the major tributaries to the Western Lake Erie Basin. However, there is substantial variability in suspended sediment and nutrient yields across Lake Erie tributaries. It is well known that agricultural and urban land uses lead to increased watershed sediment and nutrient yields, especially in the absence of proper management practices. yet attributes such as watershed soil types and slope can also affect yields. We examined the influence of watershed land use, hydrology, soil type, and slope on annual nutrient and sediment yields from tributaries to Lake Erie monitored as part of the Heidelberg Tributary Loading Program. A minimum of one sample and, during storm runoff, up to three samples a day are analyzed for all major nutrients and suspended sediments. The 5-year average annual yields across these watersheds exhibited distinct geographic patterns from west to east, with high suspended sediment but lower nutrient yields from the eastern most tributaries (Cuyahoga, Old Woman Creek, Huron) and the lowest sediment along with lower nutrient yields from the western most tributaries (Raisin, Tiffin, Lost). The Maumee, Portage, Sandusky, along with select subwatersheds tended to have intermediate sediment yields with high nutrient yields. Altogether, the % cultivated crops and poorly drained soil in the watershed increased nutrient yields whereas a higher % slope and lower % pasture increased sediment yields. These patterns highlighted unique regional differences that can help guide management decisions for these watersheds that ultimately would improve the health of Lake Erie.

A mass balance model is presented that links the total phosphorus concentration in lakes to the water residence time, R_w (lake volume divided by the annual water inflow) and the total phosphorus residence time, R_p (average standing stock of lake total phosphorous divided by the external annual total phosphorus input). Following a change in the external load, the lake total phosphorus concentration asymptotically approaches a new value that depends on the R_p:R_w ratio, with the rate of approach controlled by R_p. We applied this approach to a recent reanalysis of the water and total phosphorus budgets of the Lake Erie system of the Laurentian Great Lakes for the 2003-2016 period. We generated load–response relationships and response matrices that relate the steady state total phosphorus concentrations to external total phosphorus loads, for the whole Lake Erie system and for the individual basins (Lake St. Clair, western basin, central basin, eastern basin) and connecting channels (St. Clair River, Detroit River). These relationships and matrices provide a simple but robust framework to gauge the potential response of total phosphorus concentrations to total phosphorus load reductions, such as the 40% reduction proposed for Lake Erie under the Canada-United States Great Lakes Water Quality Agreement. The mass balance analysis further highlights the importance of inter-basin total phosphorus transfers. For example, around 70% of the total phosphorus concentration in the eastern basin is contributed by inflow from the central basin. Consequently, total phosphorus load abatements in watersheds upstream of the eastern basin, rather than in the direct watershed itself, will have the greatest impact on the eastern basin's concentration. Overall, our results illustrate how simple mass balance calculations can provide useful guidance to efforts to manage phosphorus enrichment of lakes.

The east basin of Lake Erie continues to suffer from blooms of filamentous green algae (primarily Cladophora). Potential management action through reduced phosphorus loadings have and continue to focus on the dissolved inorganic phosphorus pool but other potentially bioavailable phosphorus pools are not always considered. In this study, we describe the dissolved organic phosphorus pool in eastern Lake Erie, with an extensive sampling of four transects along the north shore of the eastern basin from May to September 2019. The dissolved organic phosphorus pool was characterized and quantified using sequential enzymatic hydrolysis to provide information on the enzymatically hydrolysable phosphorus fraction and component monoester P, diester P and a phytase hydrolysable component. These dissolved organic phosphorus fractions were compared to the soluble reactive phosphorus and total dissolved phosphorus pools. Results from this study revealed that a significant fraction (up to 63%) of the dissolved organic phosphorus pool is potentially bioavailable via enzymatic hydrolysis. Vertical differences in soluble reactive phosphorus, total dissolved phosphorus, and phosphate diester pools were also observed when comparing water column with near lake-bed samples, suggesting that the new benthic ecosystem since dreissenid colonization may be a greater source of dissolved inorganic phosphorus to the east basin than prior to colonization.

This study reports directly measured nitrification rates in the water column of western Lake Erie, which is affected by annual cyanobacterial harmful algal blooms, and across all three Lake Erie basins. Over three field seasons, ¹⁵NH₄ ⁺ stable isotope tracers were employed to quantify nitrification rates, and relative abundances of ammonia-oxidizing bacteria and ammonia-oxidizing archaea were determined via qPCR. Nitrification rates ranged from undetectable to 1,270 nmol L^-1 d^-1 and were generally greatest in the western basin near the Maumee River mouth (a major nutrient source). Nitrification rates were highest in early summer, and often lowest during peak cyanobacterial harmful algal blooms months (August and September), before increasing again in October. In the western basin, nitrification was negatively correlated with cyanobacterial biomass. There were no consistent differences in nitrification rates between the three Lake Erie basins. Over the three years in western Lake Erie, ammonia-oxidizing bacteria and ammonia-oxidizing archaea were often present in high and similar abundances, but overall, ammonia-oxidizing bacteria exceeded ammonia-oxidizing archaea, particularly in 2017. No relationships were observed between nitrification rates and ammonia-oxidizing bacteria and ammonia-oxidizing archaea abundances. Thus, despite abundant ammonia-oxidizer DNA, lower nitrification rates during cyanobacterial harmful algal blooms suggest that nitrifiers were poor competitors for regenerated and available NH₄ ⁺ during these blooms, as also observed in similar systems. Low nitrification rates during cyanobacterial harmful algal blooms could limit system nitrogen removal via denitrification, a natural pathway for its removal and a valuable ecosystem service. Lower denitrification rates allow more bioavailable nitrogen to remain in the system and support biomass and microcystin production; therefore, these results help explain how non-nitrogen-fixing cyanobacterial harmful algal blooms persist, despite low bioavailable nitrogen concentrations during these blooms, and support management efforts to reduce external nitrogen loading to eutrophic systems.

Over the past two decades, western Lake Erie has experienced recurring summer cyanobacterial blooms that pose severe negative impacts on human, animal, and ecological health. Previous research has identified a strong correlation between annual cyanobacterial bloom intensity and preceding spring (March-July) phosphorus loading from the Maumee river, the largest tributary to western Lake Erie, which is used to predict upcoming summer bloom severity. Maumee river spring phosphorus load, however, does not explain all the variation of bloom severity between years. Considering additional environmental parameters may help to better capture the physical and biogeochemical processes that regulate bloom severity, eventually leading to improved cyanobacterial forecasts which serve as an early warning for Lake Erie stakeholders. We aggregated various environmental parameters that may influence western Lake Erie cyanobacterial blooms to examine these factors as potential predictors for annual bloom severity. These included nitrogen and phosphorus loading from the Maumee river, freshwater discharge from the primary rivers and tributaries (Detroit, Huron, Raisin, Maumee, and Portage rivers), seasonal lake surface water temperature (mean winter, spring, and summer temperature), and Lake Erie winter ice extent and duration from 2002-2022. Empirical model results show that spring phosphorus loading, as total bioavailable phosphorus, from the Maumee river remains the dominant environmental factor controlling cyanobacterial blooms. However, additional environmental factors, such as Maumee river winter phosphorus loads and Lake Erie winter ice extent and timing, are likely important in modulating bloom severity, particularly in years with moderate phosphorus loads. Finally, we suggest incorporating mechanistic or rule-based models, in addition to empirical models, to better understand and predict annual cyanobacterial bloom severity. The updated models not only improve seasonal forecast accuracy which provides advanced warning of bloom severity to Lake Erie stakeholders, but also helps identify which factors we can better manage to reduce the frequency of severe blooms.

Numerical models are commonly used tools to simulate hydrodynamics and water quality of lakes. Model dimensionality (0D, 1D, 2D, or 3D) requires different simplification levels of physical-biogeochemical processes, computational power and calibration strategies and metrics against observations. To investigate these modelling considerations, the 1D (vertical) Aquatic Ecosystem Dynamics – General Lake Model and the 3D Aquatic Ecosystem Model were applied to western Lake Erie in 2008 and 2011-14. The performance of the models was evaluated by comparing the simulations against observations of water temperature, total phosphorus, orthophosphate, nitrate, total chlorophyll-a and cyanobacteria at three stations located along a transect from the Maumee River mouth to mid-basin, as well as to the basin-averaged cyanobacteria index. The 3D model showed better skill in qualitatively reproducing seasonal and spatial variations of nutrients and phytoplankton and had lower average root-mean-square error, especially through the algal plume near the Maumee River mouth. however, the horizontally averaged 1D model performed better in qualitatively capturing the cyanobacteria bloom years, as this model was extensively calibrated to basin-average values. We conclude that models should be selected and calibrated to provide the required decision support information, rather than the highest resolution or lowest error metrics at discrete sites.

To examine the potential impact of invasive dreissenid mussels on in situ populations of phytoplankton and nutrients in western Lake Erie, we combined mussel population estimates from a 2018 survey, results from mussel excretion, grazing, and in situ growth experiments, along with nutrient measurements on collected lake water. We calculated the proportion of the water column filtered per day, based on both clearance rates from grazing experiments and mussel biomass. In most cases the water column was filtered less than once per day. Based on mussel densities from nearby survey sites, we found that mussels could be expected to clear less than 5% of phytoplankton from the water column each day. We combined measurements of nitrogen and phosphorus excretion by mussels with survey densities and found that concentrations of nitrogen and phosphorus from excretion were much less than the ambient inorganic nitrogen and phosphorus measured throughout the season. Despite the modest potential impact that we measured, spatial variability in mussel density and temporal variability in nutrients and seston suggest that more substantial impact likely occurs in some conditions. Lastly, we used a mass balance approach to compare flows of nitrogen and phosphorus attributable to mussel assimilation, growth, and excretion. The proportion of assimilated nitrogen (0.01-0.21) and phosphorus (0.007-0.08) due to growth changed markedly throughout the season, but the excretion rate sometimes exceeded the apparent assimilation rate. These differences in growth:assimilation suggest changes in food quantity or quality, fluctuations in growth rates over time, or other physiological effects can lead to short-term imbalance in nutrient cycling by mussels, which could lead to locally important impacts on phytoplankton and algal blooms. Moreover, this work underscores the importance of mapping mussel densities at fine spatial scales and across interannual variation.

Herein we provide experimental evidence for the effects invasive mussels (Dreissena) grazing can exert on a natural assemblage of plankton that included both hetero- and photo-trophic components in western Lake Erie. Five mussel feeding experiments were performed seasonally in 2018 (May, June, July August, October). Pre- and post-grazing samples were collected from seven 20-L mesocosms (3 control, 4 mussel), and analyzed using microscope cell counts. Results from our experiments showed that Dreissena were active grazers of plankton on all dates, with significant correspondence between mussel clearance rates measured using microscopy versus size-specific chlorophyll and FluoroProbe based estimates (Spearman rank correlation, r=0.45, and r=0.48, respectively, p<0.05). Clearance rates (ml mg^-1 h^-1) were variable among taxonomic groups and seemed to track the abundance of ambient plankton assemblage (range 1.70 to 25.00, mean ± SE 11.70 ± 1.33). Dreissena grazed consistently on nano-sized hetero- and phototrophic plankton that constituted key trophic linkages in the Lake Erie foodweb. The most actively grazed plankton were phototrophic cryptophytes (Rhodomonas minuta, Cryptomonas spp.), centric diatoms (Cyclotella sp. and Discotella spp.), and non-pigmented chrysomonads (Chromulina sp., Ochromonas sp.). Conversely, clearance rates were low for cyanobacteria (e.g. Microcystis), dinoflagellates (Gymnodinium varians), and some colonial chlorophytes (Desmodesmus, Pediastrum); these plankton groups occurred during specific temporal windows (one-way ANOVA, p<0.05). Our results indicate the potential mussel grazing can suppress typical, non-harmful plankton species in the nanoplankton size range, thereby favoring the occurrence of less-edible, larger cyanobacteria and chlorophyte species in western Lake Erie.

We measured Quagga Mussel grazing-induced changes in seston concentration in 20-L laboratory mesocosms containing lake water, during May-October, as part of a study to investigate the present role of mussel feeding and nutrient excretion in affecting phytoplankton composition. A variety of measuring methodologies including size-fractionated chlorophyll (< 2, 2-20, and >20 µm), phytoplankton groups by FluoroProbe, cyanobacterial genera by 16S rRNA gene sequencing, and particulate C, N, and P gave us different insights into seasonal phytoplankton dynamics and mussel selective feeding and assimilation. High clearance rates and/or high assimilation rates were seen during late May-early July across all size categories, with high rates seen particularly in cryptophytes and diatoms as revealed by FluoroProbe. Starting in late July cyanophytes dominated the phytoplankton and feeding rate on them was generally zero or low. Sequencing analyses suggested that there was a diverse cyanobacterial community present, with Microcystis, contrary to expectation, dominating only in late June. Cyanobium was the dominant genera at most times during summer-fall, and as expected, Microcystis was rejected by mussels relative to other cyanobacteria. Planktothrix, dominant in spring, was readily ingested as well as Anabaena in late June. We show that a combination of methods is helpful to make progress in understanding plankton succession and grazing. These methods were important adjuncts to microscopic analysis (Carrick et al, this issue). Our results support the hypothesis that dreissenid mussels, when abundant, can affect seasonal succession of phytoplankton shifting composition to cyanobacteria and even changes within the cyanobacterial community; however, impacts are likely modest now due low mussel biomass (Carter et al., this issue). Assimilation of C, N, and P was generally high, which is important for mussel population maintenance.

Invasive Quagga Mussels (Dreissena rostriformis bugensis) and Zebra Mussels (D. polymorpha) have been present in the Great Lakes for almost four decades and have caused substantial economic and environmental impacts. Factors that influence dreissenid mussel growth and population dynamics warrant more study, particularly for Quagga Mussels. We conducted a five-month field experiment in western Lake Erie to measure multiple growth metrics for Quagga Mussels at two sites with different conditions using mussels contained in cages on instrumented moorings. We also quantified dreissenid mussels that colonized into the cages and surveyed dreissenid mussels in the sediments at each site. By multiple measures of growth except for one, growth rates were similar between the two sites despite different chlorophyll a and turbidity levels as well as notable differences in the density and size distribution of mussels found in the sediments at these sites. The growth rates were approximately 0.03 mm d^-1 for 12-mm Quagga Mussels and these rates declined with increasing initial shell length. Specific growth rate did not differ between shell and tissue measures or between sites, but both shell and tissue specific growth rates were much higher for smaller mussels. Site WE2, which is closer to Maumee River, had very few dreissenid mussels present in the sediments surrounding the mooring despite having a potential for growth comparable to that at WE4. The high level of colonization on the mooring at WE2 indicated that conditions at the sediment surface inhibit mussel settlement; adjacent benthic surveys indicated that very few mussels survive beyond the juvenile stage. Ultimately, the results from this study provide useful mussel growth parameters and indicate substrate or other limitations for dreissenid mussel populations in western Lake Erie.