Accuracy of data buoys for measurement of cyanobacteria, chlorophyll, and turbidity in a large lake (Lake Erie, North America): implications for estimation of cyanobacterial bloom parameters from water quality sonde measurements

A tale of two blooms: do ecological paradigms for algal bloom success and succession require revisiting?

Lake Erie algal bloom discussions have historically focused on cyanobacteria, with foundational "blooms like it hot" and "high nutrient" paradigms considered as primary drivers behind cyanobacterial bloom success. Yet, recent surveys have rediscovered winter-spring diatom blooms, introducing another key player in the Lake Erie eutrophication and algal bloom story which has been historically overlooked. These blooms (summer vs. winter) have been treated as solitary events separated by spatial and temporal gradients. However, new evidence suggests they may not be so isolated, linked in a manner that manifests as an algal bloom cycle. Equally notable are the emerging reports of cyanobacterial blooms in cold and/or oligotrophic freshwaters, which have been interpreted by some as shifts in classical bloom paradigms. These emerging bloom reports have led many to ask "what is a bloom?". Furthermore, questioning classic paradigms has caused others to wonder if we are overlooking additional factors which constrain bloom success. In light of emerging data and ideas, we revisited foundational concepts within the context of Lake Erie algal blooms and derived five key take-aways: 1) Additional bloom-formers (diatoms) need to be included in Lake Erie algal discussions, 2) The term "bloom" must be reinforced with a clear definition and quantitative metrics for each event, 3) Algal blooms should not be studied solitarily, 4) Shifts in physiochemical conditions serve as an alternative interpretation to potential shifts in ecological paradigms, 5) Additional factors which constrain bloom success and succession (i.e., pH and light) require consideration.

Diversity Patterns of Eukaryotic Phytoplankton in the Medog Section of the Yarlung Zangbo River

As one of the important biodiversity conservation areas in China, the ecosystem in the lower reaches of the Yarlung Zangbo River is fragile, and is particularly sensitive to global changes. To reveal the diversity pattern of phytoplankton, the metabarcode sequencing was employed in the Medog section of the lower reaches of the Yarlung Zangbo River during autumn 2019 in present study. The phytoplankton assemblies can be significantly divided into the main stem and the tributaries; there are significant differences in the phytoplankton biomass, alpha and beta diversity between the main stem and the tributaries. While both the main stem and the tributaries are affected by dispersal limitation, the phytoplankton assemblages in the entire lower reaches are primarily influenced by heterogeneous selection. Community dissimilarity and assembly process were significantly correlated with turbidity, electrical conductivity, and nitrogen nutrition. The tributaries were the main source of the increase in phytoplankton diversity in the lower reaches of the Yarlung Zangbo River. Such diversity pattern of phytoplankton in the lower reach may be caused by the special habitat in Medog, that is, the excessive flow velocity, and the significant spatial heterogeneity in physical and chemical factors between stem and tributaries. Based on the results and conclusions obtained in present study, continuous long-term monitoring is essential to assess and quantify the impact of global changes on phytoplankton.

Pulse Feature-Enhanced Classification of Microalgae and Cyanobacteria Using Polarized Light Scattering and Fluorescence Signals

Harmful algal blooms (HABs) pose a global threat to the biodiversity and stability of local aquatic ecosystems. Rapid and accurate classification of microalgae and cyanobacteria in water is increasingly desired for monitoring complex water environments. In this paper, we propose a pulse feature-enhanced classification (PFEC) method as a potential solution. Equipped with a rapid measurement prototype that simultaneously detects polarized light scattering and fluorescence signals of individual particles, PFEC allows for the extraction of 38 pulse features to improve the classification accuracy of microalgae, cyanobacteria, and other suspended particulate matter (SPM) to 89.03%. Compared with microscopic observation, PFEC reveals three phyla proportions in aquaculture samples with an average error of less than 14%. In this paper, PFEC is found to be more accurate than the pulse-average classification method, which is interpreted as pulse features carrying more detailed information about particles. The high consistency of the dominant and common species between PFEC and microscopy in all field samples also demonstrates the flexibility and robustness of the former. Moreover, the high Pearson correlation coefficient accounting for 0.958 between the cyanobacterial proportion obtained by PFEC and the cyanobacterial density given by microscopy implies that PFEC serves as a promising early warning tool for cyanobacterial blooms. The results of this work suggest that PFEC holds great potential for the rapid and accurate classification of microalgae and cyanobacteria in aquatic environment monitoring.

Environmental microcystin exposure triggers the poor prognosis of prostate cancer: Evidence from case-control, animal, and in vitro studies

Microcystin-leucine-arginine (MC-LR) is positively linked with multiple cancers in humans. However, the association between MC-LR and the risk and prognosis of prostate cancer has not been conducted in epidemiological studies. No reported studies have linked MC-LR exposure to the poor prognosis of prostate cancer by conducting experimental studies. The content of MC-LR was detected in most of the aquatic food in wet markets and supermarkets in Nanjing and posed a health risk for consumers. MC-LR levels in both prostate cancer tissues and serum were significantly higher than controls. The adjusted odds ratio (OR) for prostate cancer risk by serum MC-LR was 1.75 (95%CI: 1.21-2.52) in the whole subjects, and a positive correlation between MC-LR and advanced tumor stage was observed. Survival curve analysis indicated patients with higher MC-LR levels in tissues exhibited poorer overall survival. Human, animal, and cell studies confirmed that MC-LR exposure increases the expression of estrogen receptor-α (ERα) and promotes epithelial-mesenchymal transition (EMT) in prostate cancer. Moreover, MC-LR-induced decreased E-cadherin levels, increased vimentin levels, and increased migratory and invasive capacities of prostate cancer cells were markedly suppressed upon ERα knockdown. MC-LR-induced xenograft tumor growth and lung metastasis in BALB/c nude mice can be effectively alleviated with ERα knockdown. Our data demonstrated that MC-LR upregulated vimentin and downregulated E-cadherin through activating ERα, promoting migration and invasion of prostate cancer cells. Our findings highlight the role of MC-LR in prostate cancer, providing new perspectives to understand MC-LR-induced prostatic toxicity.

Impacts of invasive mussels on a large lake: Direct evidence from in situ control-volume experiments

Invasive dreissenid mussels have reengineered many freshwater ecosystems in North America and Europe. However, few studies have directly linked their filter feeding activity with ecological effects except in laboratory tests or small-scale field enclosures. We investigated in situ grazing on lake seston by dreissenid mussels (mainly quagga mussel Dreissena rostriformis bugensis) using a 'control volume' approach in the nearshore of eastern Lake Erie in 2016. Flow conditions were measured using an acoustic Doppler current profiler, surrounded by three vertical sampling stations that were arranged in a triangular configuration to collect time-integrated water samples from five different depths. Seston variables, including chlorophyll a, phaeopigment, particulate organic carbon and nitrogen, and particulate phosphorus, along with stoichiometric ratios and water flow over mussel colonies, were considered when estimating grazing rates. We observed suboptimal flow velocity for mussel grazing, i.e., 0.028 m s^-1 at 0.1 m above bottom (mab), and resuspension was deemed minimal. Water temperature (mean: 25.1 °C) and an unstratified water column were optimal for grazing. Concentration of seston was low (mean: 0.2 mg L^-1 particulate organic carbon) and decreased from surface to lakebed where noticeable depletion was observed. Grazing rates calculated at discrete depths varied substantially among trials, with maximum rates occurring at 0.25 or 0.5 mab. Positive grazing rates were restricted to 0.5 mab and below, defining an effective grazing zone (0.1-0.5 mab) in which the flow velocity, seston concentration, and water depth were consistently and positively correlated with grazing rates of different lake seston variables. Horizontal changes in stoichiometric ratios of seston were strongly associated with grazing rates, revealing higher uptake of particulate phosphorus than nitrogen and carbon. Our study supports the nearshore phosphorus shunt hypothesis, which posits that dreissenid mussels retain phosphorus on the lake bottom and contribute to a wide range of ecological effects on freshwater ecosystems.

Chlorophyll soft-sensor based on machine learning models for algal bloom predictions

Harmful algal blooms (HABs) are a growing concern to public health and aquatic ecosystems. Long-term water monitoring conducted by hand poses several limitations to the proper implementation of water safety plans. This work combines automatic high-frequency monitoring (AFHM) systems with machine learning (ML) techniques to build a data-driven chlorophyll-a (Chl-a) soft-sensor. Massive data for water temperature, pH, electrical conductivity (EC) and system battery were taken for three years at intervals of 15 min from two different areas of As Conchas freshwater reservoir (NW Spain). We designed a set of soft-sensors based on compact and energy efficient ML algorithms to infer Chl-a fluorescence by using low-cost input variables and to be deployed on buoys with limited battery and hardware resources. Input and output aggregations were applied in ML models to increase their inference performance. A component capable of triggering a 10 [Formula: see text]g/L Chl-a alert was also developed. The results showed that Chl-a soft-sensors could be a rapid and inexpensive tool to support manual sampling in water bodies at risk.

Development of a Risk Characterization Tool for Harmful Cyanobacteria Blooms on the Ohio River

A data-driven approach to characterizing the risk of cyanobacteria-based harmful algal blooms (cyanoHABs) was undertaken for the Ohio River. Twenty-five years of river discharge data were used to develop Bayesian regression models that are currently applicable to 20 sites spread-out along the entire 1579 km of the river’s length. Two site-level prediction models were developed based on the antecedent flow conditions of the two blooms that occurred on the river in 2015 and 2019: one predicts if the current year will have a bloom (the occurrence model), and another predicts bloom persistence (the persistence model). Predictors for both models were based on time-lagged average flow exceedances and a site’s characteristic residence time under low flow conditions. Model results are presented in terms of probabilities of occurrence or persistence with uncertainty. Although the occurrence of the 2019 bloom was well predicted with the modeling approach, the limited number of events constrained formal model validation. However, as a measure of performance, leave-one-out cross validation returned low misclassification rates, suggesting that future years with flow time series like the previous bloom years will be correctly predicted and characterized for persistence potential. The prediction probabilities are served in real time as a component of a risk characterization tool/web application. In addition to presenting the model’s results, the tool was designed with visualization options for studying water quality trends among eight river sites currently collecting data that could be associated with or indicative of bloom conditions. The tool is made accessible to river water quality professionals to support risk communication to stakeholders, as well as serving as a real-time water data monitoring utility.

Chlorophyll a estimation in lakes using multi-parameter sonde data

Algae blooms are of considerable concern in freshwater lakes and reservoirs worldwide. In-situ Chlorophyll a (Chl-a) fluorometers are widely used for rapid assessments of algae biomass. However, accurately converting Chl-a fluorescence to an equivalent concentration is challenging due to natural variations in the relationship as well as nonphotochemical quenching (NPQ) which occurs commonly in surface waters during daytime. This study is based on water quality data from a freshwater lake from October 2018 to December 2020. Initial analysis of sonde Chl-a fluorescence and laboratory extracted Chl-a concentrations shows that the two data sets exhibit a nonlinear relationship with positive correlation and significant errors. A bias correction method was next developed based on (1) concurrent sonde measurements of other water quality parameters (to account for nonlinearities) and (2) a bias correction approach for nonphotochemical quenching effects in surface waters. The new Chl-a model exhibits much improved accuracy, with a root mean square error (RMSE) less than 0.95 µg/L. The new method facilitates accurate Chl-a characterization in freshwater lakes and reservoirs based on readily obtainable in-situ fluorescence sonde measurements.

Retrieval of water quality parameters from hyperspectral images using a hybrid feedback deep factorization machine model

Environmental protection of water resources is of critical importance to daily life of human beings. In recent years, monitoring the variation of water quality using remote sensing techniques has become prevalent. Unmanned aerial vehicle (UAV) based remote sensing techniques have been applied to quantitative retrieval of concentrations of water quality parameters including phosphorus, nitrogen, chemical oxygen demand (COD), biochemical oxygen demand (BOD), and chlorophyll a (Chl-a), successfully and efficiently. In this study, a novel method with deep factorization machine, spatial distribution pattern analysis, and probabilistic analysis engaged, named hybrid feedback deep factorization machine (HF-DFM), has been developed to quantitatively estimate concentrations of water quality parameters based on hyperspectral reflectance data on large scale effectively. Our proposed method is a unified model for quantifying concentrations of water quality parameters with an end to end structure, which integrates UAV based optical remote sensing techniques and deep learning to estimate concentrations of water quality parameters. Furthermore, our proposed model was applied to real-time quantitative monitoring the variation of water quality of Mazhou River, Shenzhen, Guangdong, China. Finally, we evaluate the performance of proposed model on a real-world dataset in terms of root of mean squared error (RMSE), mean absolute percent error (MAPE), and coefficient of determination (R²). The experimental results show that our proposed model outperforms other state-of-the-art models with respect to RMSE, MAPE, and R², where resulting MAPEs for quantifying all water quality parameters range from 8.78% to 12.36%, and resulting R²s range from 0.81 to 0.93. It can serve as a useful tool for decision makers in effectively monitoring water quality of urban rivers.

The Lake Erie HABs Grab: A binational collaboration to characterize the western basin cyanobacterial harmful algal blooms at an unprecedented high-resolution spatial scale

Monitoring of cyanobacterial bloom biomass in large lakes at high resolution is made possible by remote sensing. However, monitoring cyanobacterial toxins is only feasible with grab samples, which, with only sporadic sampling, results in uncertainties in the spatial distribution of toxins. To address this issue, we conducted two intensive "HABs Grabs" of microcystin (MC)-producing Microcystis blooms in the western basin of Lake Erie. These were one-day sampling events during August of 2018 and 2019 in which 100 and 172 grab samples were collected, respectively, within a six-hour window covering up to 2,270 km² and analyzed using consistent methods to estimate the total mass of MC. The samples were analyzed for 57 parameters, including toxins, nutrients, chlorophyll, and genomics. There were an estimated 11,513 kg and 30,691 kg of MCs in the western basin during the 2018 and 2019 HABs Grabs, respectively. The bloom boundary poses substantial issues for spatial assessments because MC concentration varied by nearly two orders of magnitude over very short distances. The MC to chlorophyll ratio (MC:chl) varied by a factor up to 5.3 throughout the basin, which creates challenges for using MC:chl to predict MC concentrations. Many of the biomass metrics strongly correlated (r > 0.70) with each other except chlorophyll fluorescence and phycocyanin concentration. While MC and chlorophyll correlated well with total phosphorus and nitrogen concentrations, MC:chl correlated with dissolved inorganic nitrogen. More frequent MC data collection can overcome these issues, and models need to account for the MC:chl spatial heterogeneity when forecasting MCs.

Elevated pH Conditions Associated With Microcystis spp. Blooms Decrease Viability of the Cultured Diatom Fragilaria crotonensis and Natural Diatoms in Lake Erie

Cyanobacterial Harmful Algal Blooms (CyanoHABs) commonly increase water column pH to alkaline levels ≥9.2, and to as high as 11. This elevated pH has been suggested to confer a competitive advantage to cyanobacteria such as Microcystis aeruginosa. Yet, there is limited information regarding the restrictive effects bloom-induced pH levels may impose on this cyanobacterium’s competitors. Due to the pH-dependency of biosilicification processes, diatoms (which seasonally both precede and proceed Microcystis blooms in many fresh waters) may be unable to synthesize frustules at these pH levels. We assessed the effects of pH on the ecologically relevant diatom Fragilaria crotonensis in vitro, and on a Lake Erie diatom community in situ. In vitro assays revealed F. crotonensis monocultures exhibited lower growth rates and abundances when cultivated at a starting pH of 9.2 in comparison to pH 7.7. The suppressed growth trends in F. crotonensis were exacerbated when co-cultured with M. aeruginosa at pH conditions and cell densities that simulated a cyanobacteria bloom. Estimates demonstrated a significant decrease in silica (Si) deposition at alkaline pH in both in vitro F. crotonensis cultures and in situ Lake Erie diatom assemblages, after as little as 48 h of alkaline pH-exposure. These observations indicate elevated pH negatively affected growth rate and diatom silica deposition; in total providing a competitive disadvantage for diatoms. Our observations demonstrate pH likely plays a significant role in bloom succession, creating a potential to prolong summer Microcystis blooms and constrain diatom fall resurgence.

Sensitivity Analysis and Optimization of a Radiative Transfer Numerical Model for Turbid Lake Water

Remote sensing can detect and map algal blooms. The HydroLight (Sequoia Scientific Inc., Bellevue, Washington, DC, USA) model generates the reflectance profiles of various water bodies. However, the influence of model parameters has rarely been investigated for inland water. Moreover, the simulation time of the HydroLight model increases as the amount of input data increases, which limits the practicality of the HydroLight model. This study developed a graphical user interface (GUI) software for the sensitivity analysis of the HydroLight model through multiple executions. The GUI software stably performed parameter sensitivity analysis and substantially reduced the simulation time by up to 92%. The GUI software results for lake water show that the backscattering ratio was the most important parameter for estimating vertical reflectance profiles. Based on the sensitivity analysis results, parameter calibration of the HydroLight model was performed. The reflectance profiles obtained using the optimized parameters agreed with observed profiles, with R2 values of over 0.98. Thus, a strong relationship between the backscattering coefficient and the observed cyanobacteria genera cells was identified.

Predicting microcystin concentration action-level exceedances resulting from cyanobacterial blooms in selected lake sites in Ohio

Cyanobacterial harmful algal blooms and the toxins they produce are a global water-quality problem. Monitoring and prediction tools are needed to quickly predict cyanotoxin action-level exceedances in recreational and drinking waters used by the public. To address this need, data were collected at eight locations in Ohio, USA, to identify factors significantly related to observed concentrations of microcystins (a freshwater cyanotoxin) that could be used in two types of site-specific regression models. Real-time models include easily or continuously-measured factors that do not require that a sample be collected; comprehensive models use a combination of discrete sample-based measurements and real-time factors. The study sites included two recreational sites and six water treatment plant sites. Real-time models commonly included variables such as phycocyanin, pH, specific conductance, and streamflow or gage height. Many real-time factors were averages over time periods antecedent to the time the microcystin sample was collected, including water-quality data compiled from continuous monitors. Comprehensive models were useful at some sites with lagged variables for cyanobacterial toxin genes, dissolved nutrients, and (or) nitrogen to phosphorus ratios. Because models can be used for management decisions, important measures of model performance were sensitivity, specificity, and accuracy of estimates above or below the microcystin concentration threshold standard or action level. Sensitivity is how well the predictive tool correctly predicts exceedance of a threshold, an important measure for water-resource managers. Sensitivities > 90% at four Lake Erie water treatment plants indicated that models with continuous monitor data were especially promising. The planned next steps are to collect more data to build larger site-specific datasets and validate models before they can be used for management decisions.

Effectiveness of a fixed-depth sensor deployed from a buoy to estimate water-column cyanobacterial biomass depends on wind speed

Water quality sondes have the advantage of containing multiple sensors, extended deployment times, high temporal resolution, and telecommunication with stakeholder accessible data portals. However, sondes that are part of buoy deployments often suffer from typically being fixed at one depth. Because water treatment plants are interested in water quality at a depth of the water intake and other stakeholders (ex. boaters and swimmers) are interested in the surface, we examined whether a fixed depth of approximately 1 m could cause over- or under-estimation of cyanobacterial biomass. We sampled the vertical distribution of cyanobacteria adjacent to a water quality sonde buoy in the western basin of Lake Erie during the summers of 2015-2017. A comparison of buoy cyanobacteria RFU (Relative Fluorescence Unit) at 1 m to cyanobacteria chlorophyll a (chla) measured throughout the water column showed occurrences when the buoy both under and overestimated the cyanobacteria chla at specific depths. Largest differences between buoy measurements and at-depth grab samples occurred during low wind speeds (< 4.5 m/sec) because low winds allowed cyanobacteria to accumulate at the surface above the buoy's sonde. Higher wind speeds (> 4.5 m/sec) resulted in better agreement between the buoy and at-depth measurements. Averaging wind speeds 12 hr before sample collection decreased the difference between the buoy and at-depth samples for high wind speeds but not low speeds. We suggest that sondes should be placed at a depth of interest for the appropriate stakeholder group or deploy sondes with the ability to sample at various depths.

Monitoring of Cyanobacteria in Water Using Spectrophotometry and First Derivative of Absorbance

Management of cyanobacteria blooms and their negative impact on human and ecosystem health requires effective tools for monitoring their concentration in water bodies. This research investigated the potential of derivative spectrophotometry in detection and monitoring of cyanobacteria using toxigenic and non-toxigenic strains of Microcystis aeruginosa. Microcystis aeruginosa was quantified in deionized water and surface water using traditional spectrophotometry and the first derivative of absorbance. The first derivative of absorbance was effective in improving the signal of traditional spectrophotometry; however, it was not adequate in differentiating between signal and noise at low concentrations. Savitzky-Golay coefficients for first derivative were used to smooth the derivative spectra and improve the correlation between concentration and noise at low concentrations. Derivative spectrophotometry improved the detection limit as much as eight times in deionized water and as much as four times in surface water. The lowest detection limit measured in surface water with traditional spectrophotometry was 392,982 cells/mL, and the Savitzky-Golay first derivative of absorbance was 90,231 cells/mL. The method provided herein provides a promising tool in real-time monitoring of cyanobacteria concentrations and spectrophotometry offers the ability to measure water quality parameters together with cyanobacteria concentrations.