Search Results (22)

Environmental monitoring systems require robust uncertainty quantification for effective decision-making in complex ecological processes. Harmful algal blooms represent a critical challenge where prediction uncertainty directly impacts resource allocation and response timing, yet current remote sensing-based prediction systems provide only deterministic classifications without confidence measures. This gap between algorithmic predictions and actionable risk assessment limits operational utility for stakeholders managing water quality under varying risk tolerances. This study developed a transferable probabilistic forecasting framework integrating Sentinel-2 multispectral imagery with quantile regression and ensemble machine learning to generate continuous confidence indicators for cyanobacteria density prediction, demonstrated through its application to Lake Okeechobee, Florida. The methodology combines spectral indices extracted from Sentinel-2 data with XGBoost for quantile regression at 0.05, 0.50, and 0.95 probability levels, and LightGBM for multi-horizon temporal forecasting. Sentinel-2’s 13 spectral bands spanning visible to shortwave infrared wavelengths, combined with its 5-day revisit frequency provide a spectrally rich and temporally dense input space that is well-suited to gradient boosting methods such as XGBoost, which can exploit complex nonlinear interactions among spectral features to distinguish cyanobacterial signatures from background water constituents. LightGBM achieved mean absolute percentage errors of 2.9% for 10-day forecasts and 5.7% for 20-day forecasts, outperforming conventional regression models. The framework generates 90% prediction intervals that enable reliable risk classifications for operational bloom management. This approach bridges the gap between satellite-based algal bloom detection and actionable decision-making by quantifying predictive uncertainty, representing a shift from binary classifications to probability-based environmental monitoring systems that accommodate varying stakeholder risk tolerances in water quality management applications. Full article

Inland water pollution poses significant risks to aquatic environments, affecting ecological and human health while increasing drinking-water treatment costs. Continuous monitoring using reliable techniques such as satellite remote sensing is essential. This study investigates the spatial and temporal water quality dynamics of Lake Okeechobee by integrating satellite imagery with in situ water samples. The objectives are (1) to analyze water quality trends and (2) to develop linear regression models for predicting Chlorophyll-a (Chla), turbidity, and Trophic State Index (TSI) for eutrophication monitoring. Landsat 8 and 9 imagery, coupled with water quality data from monitoring stations in the South Florida Water Management District’s DBHydro database, provided measurements of Chla, total nitrogen (TN), total phosphorus (TP), and turbidity. Statistical analyses determined optimal regression models with adjusted R² values of 0.69 and 0.93 for individual band Chla and turbidity models and 0.65, 0.82, and 0.66 for spectral band ratio models of Chla, turbidity, and TSI, respectively. Northwestern and southwestern peripheries of Lake Okeechobee exhibited Chla concentrations exceeding the 20 µg/L threshold, while turbidity peaked near the lake’s center. From 2013 to 2019, TN increased significantly, followed by a slight decline from 2019 to 2023, whereas TP exhibited no clear trend. The TSI ranged from mesotrophic to hypereutrophic states. This study demonstrates satellite remote sensing as an efficient tool for monitoring water quality changes, guiding restoration measures, and identifying highly contaminated zones through threshold-based segmentation of satellite-derived water quality maps, rather than assessing the lake as a uniform whole. Full article

In southwest Florida, the Caloosahatchee River flows from Lake Okeechobee into a biologically productive tidal estuarine system. A combination of excess water during the wet season, insufficient fresh water in the dry season, and poor quality of the river water are damaging the estuarine ecosystem. To better control the quality and quantity of the water entering the estuary, reservoirs are being constructed to store excess, poor quality water during the wet season and return it to the river for discharge into the estuary at an appropriate time. This stored water is enriched in nutrients and organic carbon. Because of the subtropical nature of the climate in southwest Florida and potential increases in temperature in the future, the return flow of water from the reservoirs must be treated before it can be returned to the river. Hence, an experimental water treatment system was developed and operated to compare biological treatment processes consisting of solely wetland plants versus adding some engineered processes, including slow sand filtration and a combination of slow sand filtration and ultraviolet (UV) treatment. These three treatment trains were operated and monitored through a seasonal cycle in 2021–2022. All three treatment methods significantly reduced the concentrations of nutrients and total organic carbon. While the enhanced engineered wetlands’ treatment trains did slightly outperform the wetland train, a comparison of the three process trains showed no statistically significant difference. It was concluded that upscaling of the slow sand filtration and UV process could improve the treatment efficiency, but this change would have to be evaluated within a framework of long-term economic benefits. It was also concluded that the Caloosahatchee River water quality is quite enriched in nutrients so that reservoir storage would increase the organic carbon concentrations, making it imperative that it be treated before being returned to the river. It was also discovered that the green alga Cladophora sp. grew rapidly in the biological treatment tubs and will present a significant challenge for the treatment of the reservoir discharge water using the currently proposed alum treatment. Full article

Cyanobacterial harmful algal blooms (cyanoHABs) occur in fresh water globally. These can degrade water quality and produce toxins, resulting in ecological and economic damages. Thus, short-term management methods (i.e., algaecides) are necessary to rapidly mitigate the negative impacts of cyanoHABs. In this study, we assess the efficacy of a hydrogen peroxide-based algaecide (PAK^® 27) on a Microcystis dominated bloom which occurred within the Pahokee Marina on Lake Okeechobee, Florida, USA. We observed a significant reduction in chlorophyll a (96.81%), phycocyanin (93.17%), and Microcystis cell counts (99.92%), and a substantial reduction in microcystins (86.7%) 48 h after treatment (HAT). Additionally, there was a significant shift in bacterial community structure 48 HAT, which coincided with an increase in the relative abundance of photosynthetic protists. These results indicate that hydrogen peroxide-based algaecides are an effective treatment method for cyanoHAB control and highlight their effects on non-target microorganisms (i.e., bacteria and protists). Full article

A very detailed water budget analysis was conducted on Lake Trafford in South Florida. The inflow was dominated by surface water influx via five canals (61%), with groundwater influx constituting 12% and direct rainfall constituting 27%. Lake discharge was dominated by sheet flow (69%) and evapotranspiration (30.5%), with groundwater recharge of the hydraulically connected unconfined aquifer accounting for only 0.5%. The removal of 30 M tons (4.4 × 10⁶ m³) of organic sediment impacted the groundwater influx, causing enhanced groundwater flow into the deeper parts of the lake and mixed flow along the banks, creating a rather unusual pattern. The large number of groundwater seepage meters used during this investigation led to a very reliable set of measurements with occasional failure of only a few meters. A distinctive relationship was found between the wet-season lake stage, heavy rainfall events, and pulses of exiting sheet flow from the lake. Estimation of the evapotranspiration loss using data collected from a weather station on the lake allowed the use of three different models, which, when averaged, produced results comparable to Lake Okeechobee (South Florida). A limitation of this investigation was the inability to directly measure sheet-flow discharges, which had to be estimated as a residual within the calculated water budget. Full article

 Despite years of efforts to improve water quality, harmful algal blooms remain a chronic phenomenon, with devastating environmental, economic, and social impacts in many regions worldwide. In this study, we assessed the complexity of nutrient pollution attributed to harmful algal blooms in South Florida (USA) by analyzing 20 years of flow and nutrient data within two headwater basins in the Lake Okeechobee (LO) watershed. The study used an established advanced regression method, the Weighted Regression on Time, Discharge, and Season (WRTDS) method, as an analysis framework to examine the impact of nutrient management practices on water quality trends. The WRTDS method produced total phosphorus (TP) and total nitrogen (TN) concentration and flux trends, which were then compared with existing and historic nutrient management records within the basin. Results from this study highlight divergences in progress to improve water quality. Nutrient management practices only had a weak impact on TP and TN flux trends in one of the two basins, where TP flux decreased 2% per year, and TN flux decreased 0.1% per year. TP and TN flux increased in the second basin. Variances of improvement between the two basins are likely attributable to differences in contemporary point source loading and legacy nutrient pools from non-point source inputs 20 years or more before the analysis period. The long-lasting impacts of legacy nutrients also emphasize a need for investments in technologies and practices that can withdraw nutrients from enriched soil and water.  Full article

Satellite-based monitoring of cyanobacterial harmful algal blooms (CyanoHABs) heavily utilizes historical Envisat-MERIS and current Sentinel-OLCI observations due to the availability of the 620 nm and 709 nm bands. The permanent loss of communication with Envisat in April 2012 created an observational gap from 2012 until the operationalization of OLCI in 2016. Although MODIS-Terra has been used to bridge the gap from 2012 to 2015, differences in band architecture and the absence of the 709 nm band have complicated generating a consistent and continuous CyanoHAB monitoring product. Moreover, several Terra bands often saturate during extreme high-concentration CyanoHAB events. This study trained a fully connected deep network (CyanNet) to model MERIS-Cyanobacteria Index (CI)—a key satellite algorithm for detecting and quantifying cyanobacteria. The network was trained with Rayleigh-corrected surface reflectance at 12 Terra bands from 2002–2008, 2010–2012, and 2017–2021 and validated with data from 2009 and 2016 in Lake Okeechobee. Model performance was satisfactory, with a ~17% median difference in Lake Okeechobee annual bloom magnitude. The median difference was ~36% with 10-day Chlorophyll-a time series data, with differences often due to variations in data availability, clouds or glint. Without further regional training, the same network performed well in Lake Apopka, Lake George, and western Lake Erie. Validation success, especially in Lake Erie, shows the generalizability of CyanNet and transferability to other geographic regions. Full article

Water is an essential resource for the survival of living beings. Remote-sensing data provides the best possible way to detect water bodies and monitor change over time. With a surplus amount of remote sensing data, machine learning approaches have become an effective and efficient way to detect and monitor surface water bodies. This research focused on utilizing remote sensing and machine learning approaches to monitor changes in the surface water of Lake Okeechobee, Florida, USA. This investigation used two sources of remotely sensed data, Landsat 7, and Landsat 8, for 2002 and 2022, respectively. Two machine learning algorithms, support vector machine (SVM) and random forest (RF), were adopted, considering their power and robustness, among other factors, for supervised classification. Both algorithms provided an accuracy of over 92% and a kappa statistic exceeding 0.8. Further, we used image differencing techniques to track changes across two decades. The SVM suggested an increase of 85 km², and RF indicated an expansion of 52 km² in the surface water area. This study explicitly demonstrates how dynamic natural resources are, especially water sources. Thus, it can provide a foundation for research that further explores environmental assessments and sustainable water resource planning in Lake Okeechobee. Full article

Harmful cyanobacterial blooms of the toxin-producing Microcystis have become a growing problem for Southwest Florida freshwater bodies. Recently, a 2016 bloom in Lake Okeechobee and a 2018 bloom in the Caloosahatchee River both led to the declaration of a state of emergency for the state of Florida. Fast-acting suppression methods are needed to protect residents and wildlife. Hydrogen peroxide and L-lysine have shown promising results in selectively inhibiting the growth of Microcystis aeruginosa and are more ecologically friendly due to fast degradation in water or the biological enhancement of nontarget organisms, respectively. We further explored the use of hydrogen peroxide, L-lysine, and combined treatments of both chemicals, which have never been tested before, for the rapid suppression of Microcystis. We assessed the susceptibility of seven M. aeruginosa strains and six other phytoplankton (Cyanobium spp., Synechococcus sp., Dolichospermum planctonica, Mychonastes homosphaera, and Chromochloris zofingiensis) commonly found in Florida, and revealed that susceptibility was diverse. All three treatments were effective at inhibiting the growth of M. aeruginosa, mixed treatments (16.7 mg/L hydrogen peroxide: 8 mg/L L-lysine) were most effective with a median growth inhibition ratio of 94.2% on the last day of the experiment, while hydrogen peroxide (16.7 mg/L) (83.8%) and L-lysine (8 mg/L) (78.5%) were less so. We found axenic M. aeruginosa to be significantly more sensitive to hydrogen peroxide when compared with nonaxenic strains (p < 0.01, n = 18). L-lysine was found to be significantly more toxic to M. aeruginosa than other examined cyanobacteria and chlorophyte strains at the end of the experiment (p < 0.001, n = 33), demonstrating its specificity to this cyanobacterium, while hydrogen peroxide and mixed treatments had varying effects on the other tested phytoplankton. Full article

A 72 h small-scale trial was conducted in enclosed mesocosms in the Lake Okeechobee waterway to evaluate the effectiveness of a USEPA-registered peroxide-based algaecide (formulated as sodium carbonate peroxyhydrate) for controlling a natural cyanobacteria population. Mesocosms were initially subjected to either no algaecide or the maximum label rate of 10 mg H₂O₂·L⁻¹. A subset of mesocosms were then subjected to a sequential application of 5 mg H₂O₂·L⁻¹ at 48 h after initial treatment. Following application, peroxide concentrations rapidly decreased and were undetectable by 48 h. At 24 h after treatment, significant decreases in all biomass indicators were observed (compared to untreated mesocosms), including extracted chlorophyll a, microscopic counts (total phytoplankton and total cyanobacteria), and cyanobacteria-specific 16S rRNA gene copies by over 71%. Although peroxide treatment reduced cyanobacteria biomass, there was no change in overall community structure and the remaining population was still dominated by cyanobacteria (>90%). After 48 h exposure, some biomass recovered in single application mesocosms resulting in only a 32–45% reduction in biomass. Repeated peroxide dosing resulted in the greatest efficacy, which had a sustained (60–91%) decrease in all biomass indicators for the entire study. While a single application of the peroxide was effective in the first 24 h, a sequential treatment is likely necessary to sustain efficacy when using this approach to manage cyanobacteria in the field. Results of this study support that this peroxide-based algaecide is a strong candidate to continue with scalable field trials to assess its potential future utility for operational management programs in the Lake Okeechobee waterway. Full article

Soil phosphorus (P) built up due to past management practices, legacy P, in the Lake Okeechobee Watershed (LOW) in south-central Florida, U.S.A., is often discussed as the root cause of lake eutrophication. Improvement of the lake’s water quality requires the identification of critical P sources and quantifying their contributions. We performed a global sensitivity analysis of the Watershed Assessment Model (WAM), a common evaluation tool in LOW environmental planning, using the Morris method. A pre-calibrated WAM setup (Baseline) of the LOW sub-watershed, Taylor Creek Nubbin Slough (TCNS), was used as a test case. Eight scenarios were formulated to estimate the contributions of various P sources. The Morris analysis indicated that total phosphorus (TP) loads were highly sensitive to legacy P in improved pastures, the major land use covering 46.2% of TCNS. The scenario modeling revealed that legacy P, inorganic fertilizers, and other sources contribute 63%, 10%, and 32%, respectively, to the Baseline TP load of 111.3 metric tons/y to the lake. Improved pastures, dairies, citrus, and field crops are the top TP load contributors. Our results have important implications for water quality improvement plans in the LOW and highlighted the need for accurate spatial mapping of legacy P and incorporation of such information in modeling efforts for watersheds demonstrating legacy P problems. Full article

Anthropogenic developments in coastal watersheds cause significant ecological changes to estuaries. Since estuaries respond to inputs on relatively long time scales, robust analyses of long-term data should be employed to account for seasonality, internal cycling, and climatological cycles. This study characterizes the water quality of a highly managed coastal basin, the St. Lucie Estuary Basin, FL, USA, from 1999 to 2019 to detect spatiotemporal differences in the estuary’s water quality and its tributaries. The estuary is artificially connected to Lake Okeechobee, so it receives fresh water from an external basin. Monthly water samples collected from November 1999 to October 2019 were assessed using principal component analysis, correlation analysis, and the Seasonal Kendall trend test. Nitrogen, phosphorus, color, total suspended solids, and turbidity concentrations varied seasonally and spatially. Inflows from Lake Okeechobee were characterized by high turbidity, while higher phosphorus concentrations characterized inflows from tributaries within the basin. Differences among tributaries within the basin may be attributed to flow regimes (e.g., significant releases vs. steady flow) and land use (e.g., pasture vs. row crops). Decreasing trends for orthophosphate, total phosphorus, and color and increasing trends for dissolved oxygen were found over the long term. Decreases in nutrient concentrations over time could be due to local mitigation efforts. Understanding the differences in water quality between the tributaries of the St. Lucie Estuary is essential for the overall water quality management of the estuary. Full article

Operational coarse-resolution satellite thermal sensors designed for global oceans are often insufficient for evaluating surface temperature of small water bodies. Here, the quality of the thermal data, collected by the ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS), over several South Florida estuaries, Chesapeake Bay, and Lake Okeechobee is evaluated using both in situ and Moderate Resolution Imaging Spectroradiometer (MODIS) Sea Surface Temperature (SST) data. Overall, for SST between ~6 and ~32 °C, ECOSTRESS LST (Land Surface Temperature, used as a surrogate for SST in this study) appears to be slightly underestimated, with the underestimation being more severe at night (−1.13 °C) than during the day (−0.64 °C), in spring and summer (−1.25 ± 1.39 °C) than in autumn and winter (−0.57 ± 0.98 °C), and after May 2019 when two of the five bands failed. The root-mean-square uncertainties of ECOSTRESS SST are generally within 1–2 °C. Spatial analysis further suggests that ECOSTRESS SST covers waters closer to shore and reveals more spatial features than MODIS, with comparable image noise. From these observations, after proper georeferencing and empirical correction of the negative bias, ECOSTRESS SST may be used to evaluate the thermal environments of small water bodies, thus filling gaps in the coarse-resolution satellite data. Full article

The state of water quality of lakes is highly related to watershed processes which will be responsible for the delivery of sediment, nutrients, and other pollutants to receiving water bodies. The spatiotemporal variability of water quality parameters along with the seasonal changes were studied for Lake Okeechobee, South Florida. The dynamics of selected four water quality parameters: total phosphate (TP), total Kjeldahl nitrogen (TKN), total suspended solid (TSS), and chlorophyll-a (chl-a) were analyzed using data from satellites and water quality monitoring stations. Statistical approaches were used to establish correlation between reflectance and observed water quality records. Landsat Thematic Mapper (TM) data (2000 and 2007) and Landsat Operational Land Imager (OLI) in 2015 in dry and wet seasons were used in the analysis of water quality variability in Lake Okeechobee. Water quality parameters were collected from twenty-six (26) monitoring stations for model development and validation. In the regression model developed, individual bands, band ratios and various combination of bands were used to establish correlation, and hence generate the models. A stepwise multiple linear regression (MLR) approach was employed and the results showed that for the dry season, higher coefficient of determination (R²) were found (R² = 0.84 for chl-a and R² = 0.67 for TSS) between observed water quality data and the reflectance data from the remotely-sensed data. For the wet season, the R² values were moderate (R² = 0.48 for chl-a and R² = 0.60 for TSS). It was also found that strong correlation was found for TP and TKN with chl-a, TSS, and selected band ratios. Total phosphate and TKN were estimated using best-fit multiple linear regression models as a function of reflectance data from Landsat TM and OLI, and ground data. This analysis showed a high coefficient of determination in dry season (R² = 0.92 for TP and R² = 0.94 for TKN) and in wet season (R² = 0.89 for TP and R² = 0.93 for TKN). Based on the findings, the Multiple linear regression (MLR) model can be a useful tool for monitoring large lakes like Lake Okeechobee and also predict the spatiotemporal variability of both optically active (Chl-a and TSS) and inactive water (nutrients) quality parameters. Full article

Lake Okeechobee is one of the largest freshwater lakes in the United States. As a eutrophic lake, it has frequent algal blooms composed predominantly of the cyanobacterium genus Microcystis. Many of the algal blooms are associated with the resuspension of a thixotropic benthic mud containing legacy nutrients. Since Lake Okeechobee has an area of 1732 km² (40–50 km radius) and a mean depth of only 2.7 m, there is sufficient fetch and shallow water depth to allow frequent wind, wave, and current generated events, which cause sediment resuspension. Three types of mud exist in the lake including an immobile dark-colored, consolidated mud, a brownish-colored mud, which is poorly consolidated and mobile, and a dark-colored thixotropic, highly mobile mud that is a mixture of organic matter and clay-sized minerals. Altogether, these muds contain an estimated 4.6 × 10⁶ kg of total phosphorus and commensurate high amounts of labile nitrogen. The thixotropic mud covers most of the lakebed and contains the suitable nutrient ratios to trigger algal blooms. A bioassay analysis of the thixotropic mud compared to the consolidated mud showed that it produced up to 50% more nutrient mass compared to the consolidated mud. The thixotropic mud does not consolidate, thus remains mobile. The mobility is maintained by the dynamics of the algal blooms and bacterial decay of extracellular secretions (transparent exopolymer particles) that bind sediment, transfer it to the bottom, and undergo bacterial digestion causing gas emissions, thus maintaining the organic/sediment matrix in suspension. Despite major efforts to control external nutrient loading into the lake, the high frequency of algal blooms will continue until the muds bearing legacy nutrients are removed from the lake. Full article