Space-time chlorophyll-a retrieval in optically complex waters that accounts for remote sensing and modeling uncertainties and improves remote estimation accuracy

Employing hybrid deep learning for near-real-time forecasts of sensor-based algal parameters in a Microcystis bloom-dominated lake

Harmful cyanobacterial blooms (CyanoHABs) are increasingly impacting the ecosystem of lakes, reservoirs and estuaries globally. The integration of real-time monitoring and deep learning technology has opened up new horizons for early warnings of CyanoHABs. However, unlike traditional methods such as pigment quantification or microscopy counting, the high-frequency data from in-situ fluorometric sensors display unpredictable fluctuations and variability, posing a challenge for predictive models to discern underlying trends within the time-series sequence. This study introduces a hybrid framework for near-real-time CyanoHABs predictions in a cyanobacterium Microcystis-dominated lake - Lake Dianchi, China. The proposed model was validated using hourly Chlorophyll-a (Chl a) concentrations and algal cell densities. Our results demonstrate that applying decomposition-based singular spectrum analysis (SSA) significantly enhances the prediction accuracy of subsequent CyanoHABs models, particularly in the case of temporal convolutional network (TCN). Comparative experiments revealed that the SSA-TCN model outperforms other SSA-based deep learning models for predicting Chl a (R2 = 0.45-0.93, RMSE = 2.29-5.89 μg/L) and algal cell density (R2 = 0.63-0.89, RMSE = 9489.39-16,015.37 cells/mL) at one to four steps ahead predictions. The forecast of bloom intensities achieved a remarkable accuracy of 98.56 % and an average precision rate of 94.04 % ± 0.05 %. In addition, scenarios involving various input combinations of environmental factors demonstrated that water temperature emerged as the most effective driver for CyanoHABs predictions, with a mean RMSE of 2.94 ± 0.12 μg/L, MAE of 1.55 ± 0.09 μg/L, and R2 of 0.83 ± 0.01. Overall, the newly developed approach underscores the potential of a well-designed hybrid deep-learning framework for accurately predicting sensor-based algal parameters. It offers novel perspectives for managing CyanoHABs through online monitoring and artificial intelligence in aquatic ecosystems.

Prediction of chlorophyll a and risk assessment of water blooms in Poyang Lake based on a machine learning method

Four different methods were used to identify the important factors influencing chlorophyll-a (Chl-a) content: correlation analysis (CC-NMI), principal component analysis (PCA), decision tree (DT), and random forest recursive feature elimination (RF-RFE). Considering the relationship between Chl-a and its active and passive factors, we established machine learning combination models based on multiple linear regression (MLR), multi-layer perceptron (MLP), and support vector regression (SVR) to predict Chl-a content for Poyang Lake, China. Then, the predictive effects of different combination models were compared and evaluated from multiple perspectives. Considering the actual needs for eutrophication prevention and control, the concept of risk probability was then introduced to assess the risk degree of risk associated with water blooms in Poyang Lake. The results indicated that the mean R2 for the Chl-a predictions using the MLR, MLP, and SVR models was 0.21, 0.61, and 0.75, respectively. Consequently, the SVR model demonstrated higher precision and more accurate predictions. Compared to other methods, integrating the SVR model with the RF-RFE method significantly improved the prediction accuracy, with the R2 increasing to 0.94. For Poyang Lake, 8.8% of random samples indicated a low risk level with a water bloom probability of 21.1%-36.5%; one sample indicated a medium risk level with a risk probability of 45.5%. The research results offer valuable insights for predicting eutrophication and conducting risk assessments for Poyang Lake. They also provide reliable scientific support for making decisions about eutrophication in lakes and reservoirs. Therefore, the results hold significant theoretical importance, practical value, and potential for widespread application.

Knowledge-guided mixture density network for chlorophyll-a retrieval and associated pixel-by-pixel uncertainty assessment in optically variable inland waters

Machine learning has been increasingly used to retrieve chlorophyll-a (Chl-a) in optically variable waters. However, without the guidance of physical principles or expert knowledge, machine learning may produce biased mapping relationships, or waste considerable time searching for physically infeasible hyperparameter domains. In addition, most Chl-a retrieval models cannot evaluate retrieval uncertainty when ground observations are not available, and the retrieval uncertainty is crucial for understanding the model limitations and evaluating the reliability of retrieval results. In this study, we developed a novel knowledge-guided mixture density network to retrieve Chl-a in optically variable inland waters based on Sentinel-3 Ocean and Land Color Instrument (OLCI) imagery. The proposed method embedded prior knowledge derived from spectral shape classification into the mixture density network. Compared to another deterministic model, the knowledge-guided mixture density network outputted the conditional distribution of Chl-a given an input spectrum, enabling us to estimate the optimal retrieval and the associated uncertainty. The proposed method showed favorable correspondence with the field Chl-a, with root mean square error (RMSE) of 6.56 μg/L, and mean absolute percentage error (MAPE) of 43.64 %. Calibrated against Sentinel-3 OLCI spectrum, the proposed method also performed well when applied to field spectrum (RMSE = 4.58 μg/L, MAPE = 72.70 %), suggesting its effectiveness and good generalization. The proposed method provided the standard deviation of each estimated Chl-a, which enabled us to inspect the reliability of the estimated results and understand the model limitations. Overall, the proposed method improved the Chl-a retrieval in terms of model accuracy and uncertainty evaluation, providing a more comprehensive Chl-a observation of inland waters.

A new framework for water quality forecasting coupling causal inference, time-frequency analysis and uncertainty quantification

Accurate forecasting of water quality variables in river systems is crucial for relevant administrators to identify potential water quality degradation issues and take countermeasures promptly. However, pure data-driven forecasting models are often insufficient to deal with the highly varying periodicity of water quality in today's more complex environment. This study presents a new holistic framework for time-series forecasting of water quality parameters by combining advanced deep learning algorithms (i.e., Long Short-Term Memory (LSTM) and Informer) with causal inference, time-frequency analysis, and uncertainty quantification. The framework was demonstrated for total nitrogen (TN) forecasting in the largest artificial lakes in Asia (i.e., the Danjiangkou Reservoir, China) with six-year monitoring data from January 2017 to June 2022. The results showed that the pre-processing techniques based on causal inference and wavelet decomposition can significantly improve the performance of deep learning algorithms. Compared to the individual LSTM and Informer models, wavelet-coupled approaches diminished well the apparent forecasting errors of TN concentrations, with 24.39%, 32.68%, and 41.26% reduction at most in the average, standard deviation, and maximum values of the errors, respectively. In addition, a post-processing algorithm based on the Copula function and Bayesian theory was designed to quantify the uncertainty of predictions. With the help of this algorithm, each deterministic prediction of our model can correspond to a range of possible outputs. The 95% forecast confidence interval covered almost all the observations, which proves a measure of the reliability and robustness of the predictions. This study provides rich scientific references for applying advanced data-driven methods in time-series forecasting tasks and a practical methodological framework for water resources management and similar projects.

Monitor water quality through retrieving water quality parameters from hyperspectral images using graph convolution network with superposition of multi-point effect: A case study in Maozhou River

Quantitative prediction by unmanned aerial vehicle (UAV) remote sensing on water quality parameters (WQPs) including phosphorus, nitrogen, chemical oxygen demand (COD), biochemical oxygen demand (BOD), and chlorophyll a (Chl-a), total suspended solids (TSS), and turbidity provides a flexible and effective approach to monitor the variation in water quality. In this study, a deep learning-based method integrating graph convolution network (GCN), gravity model variant, and dual feedback machine involving parametric probability analysis and spatial distribution pattern analysis, named Graph Convolution Network with Superposition of Multi-point Effect (SMPE-GCN) has been developed to calculate concentrations of WQPs through UAV hyperspectral reflectance data on large scale efficiently. With an end-to-end structure, our proposed method has been applied to assisting environmental protection department to trace potential pollution sources in real time. The proposed method is trained on a real-world dataset and its effectiveness is validated on an equal amount of testing dataset with respect to three evaluation metrics including root of mean squared error (RMSE), mean absolute percent error (MAPE), and coefficient of determination (R²). The experimental results demonstrate that our proposed model achieves better performance in comparison with state-of-the-art baseline models in terms of RMSE, MAPE, and R². The proposed method is applicable for quantifying seven various WQPs and has achieved good performance for each WQP. The resulting MAPE ranges from 7.16% to 10.96% and R² ranges from 0.80 to 0.94 for all WQPs. This approach brings a novel and systematic insight into real-time quantitative water quality monitoring of urban rivers, and provides a unified framework for in-situ data acquisition, feature engineering, data conversion, and data modeling for further research. It provides fundamental support to assist environmental managers to efficiently monitor water quality of urban rivers.

Sample based water quality monitoring of coastal seas: How significant is the information loss in patchy time series compared to continuous ones?

The high temporal and spatial variability of tidal dominated coastal areas poses a challenge for characterising water quality. Water quality monitoring relies often on information collected by water sampling from a vessel or by satellites, and covers limited time periods and therefore limited tidal and meteorological conditions. To assess the loss of information from discrete sampling, continuous time series of one year (suspended particulate matter (SPM) concentration, SPM flux and Chlorophyll a (Chl) concentration) were used. Eight different schemes of sampling into these time series were applied that are typical for many monitoring programs. They differ in the time between sampling events (synodic or half-synodic) and the duration of the sampling (tidal cycle, half a tidal cycle, one or more samples). The information loss was quantified by applying a bootstrap method to calculate the mean and standard deviation over the considered period. These were then compared with the true mean calculated from the continuous series. The probability to match the true mean within a certain margin depends on the sampling period and the season, but it is always low, especially if the allowed uncertainty is stringent (e.g., ±2.5 % about the true mean). For the SPM concentration this probability is lower than 10 % and for Chl concentration lower than 20 %. Similarly, conclusions arise for the detection of trends in a 20 year time series of SPM concentration with an artificial yearly increase of 0.5 %. None of the sampling schemes was able to assess statistical significant interannual trends with probabilities above 60 %. Further, the significant trends overestimated the increase by a factor 2 to 8. Here, present modus operandi is thus inadequate for basic trend detection, but may be acceptable for the more marine, lower turbid areas where higher probabilities were obtained in this study.

V, Vijayan NT, Subramanian M, Gauns MU. Applications of Machine Learning in Chemical and Biological Oceanography

Machine learning (ML) refers to computer algorithms that predict a meaningful output or categorize complex systems based on a large amount of data. ML is applied in various areas including natural science, engineering, space exploration, and even gaming development. This review focuses on the use of machine learning in the field of chemical and biological oceanography. In the prediction of global fixed nitrogen levels, partial carbon dioxide pressure, and other chemical properties, the application of ML is a promising tool. Machine learning is also utilized in the field of biological oceanography to detect planktonic forms from various images (i.e., microscopy, FlowCAM, and video recorders), spectrometers, and other signal processing techniques. Moreover, ML successfully classified the mammals using their acoustics, detecting endangered mammalian and fish species in a specific environment. Most importantly, using environmental data, the ML proved to be an effective method for predicting hypoxic conditions and harmful algal bloom events, an essential measurement in terms of environmental monitoring. Furthermore, machine learning was used to construct a number of databases for various species that will be useful to other researchers, and the creation of new algorithms will help the marine research community better comprehend the chemistry and biology of the ocean.

Algal community structure prediction by machine learning

The algal community structure is vital for aquatic management. However, the complicated environmental and biological processes make modeling challenging. To cope with this difficulty, we investigated using random forests (RF) to predict phytoplankton community shifting based on multi-source environmental factors (including physicochemical, hydrological, and meteorological variables). The RF models robustly predicted the algal communities composed by 13 major classes (Bray-Curtis dissimilarity = 9.2 ± 7.0%, validation NRMSE mostly <10%), with accurate simulations to the total biomass (validation R² > 0.74) in Norway's largest lake, Lake Mjosa. The importance analysis showed that the hydro-meteorological variables (Standardized MSE and Node Purity mostly >0.5) were the most influential factors in regulating the phytoplankton. Furthermore, an in-depth ecological interpretation uncovered the interactive stress-response effect on the algal community learned by the RF models. The interpretation results disclosed that the environmental drivers (i.e., temperature, lake inflow, and nutrients) can jointly pose strong influence on the algal community shifts. This study highlighted the power of machine learning in predicting complex algal community structures and provided insights into the model interpretability.

Monitoring trophic status using in situ data and Sentinel-2 MSI algorithm: lesson from Lake Malombe, Malawi

With excessive nutrient enrichment exacerbated by anthropogenic drivers, many standing water bodies are changing from oligotrophic to mesotrophic, eutrophic, and finally hypertrophic-negatively affecting ecosystem functioning, biodiversity, and human populations. Efforts have been devoted to developing novel algorithms for estimating chlorophyll-a (chl-a), cyno-blooms, and floating vegetation. However, to this date, little research has focused on freshwater lakes in the data-scarce Sub-Saharan African countries such as Malawi. We, therefore, estimated the trophic status of Lake Malombe in Malawi-a lake likely to be affected by eutrophication and algal bloom-emerging threats to freshwater ecosystem functioning globally-especially with the onset of climatic and anthropogenic drivers. We integrated in situ data with high-resolution Sentinel-2 Multispectral Imagery Analysis (MSI). We independently assessed the remote sensing technique using in situ data and tested the model at multiple stages. The scatter plot showed that most points were in the 95% confidence interval. The validation results between the measured in situ chl-a concentrations and the Sentinel-2 MSI-based chl-a retrieval had a root mean square error (RMSE) of 2.88 µg/L. The chl-a concentrations retrieved from MSI images were consistent with in situ data, indicating that the normalized difference chlorophyll index (NDCI) algorithm estimated chl-a concentrations in Lake Malombe with acceptable accuracy. Dissolved oxygen (DO), sulfate (SO₄^2-), nitrite [Formula: see text], soluble reactive phosphorous [Formula: see text]), total dissolved solids (TDS), and chl-a, except for temperatures from the hot-dry-season, cold-dry-windy-season, and rainy-season, were significantly different (P < 0.05). The Sentinel-2 MSI imagery analysis also depicted similar results, with high chl-a concentration reported in March (rainy season) and October (hot-dry season) and the lowest from May to August (cold-dry-windy season). On the contrary, the ANOVA results for water quality parameters from all five points had P > 0.05. The correlation matrix showed coefficients of (0.798 < r < 0.930, n = 30, P < 0.005), suggesting that Lake Malombe is homogenous. Our results demonstrate that integrating remote sensing based on MSI imagery and in situ data to estimate chl-a can provide an effective tool for monitoring eutrophication in small, medium, and large standing waterbodies-crucial information required to respond to global ecological and climatic dynamics.

Application and recent progress of inland water monitoring using remote sensing techniques

Hyperspectral remote sensing, which retrieves the water quality parameters by direct high-resolution analysis of the electromagnetic spectrum reflected from the water surface, has been widely applied for inland water quality detection. Such a new approach provides an opportunity to generate real-time data from water with the noncontact method, largely improving working efficiency. By summarizing the development and current applications of hyperspectral remote sensing, we compare the relative merits of varying remote sensing platforms, popular inversion models, and the application of hyperspectral monitoring of chlorophyll-a (Chl-a), transparency, total suspended solids (TSS), colored dissolved organic matter (CDOM), phycocyanin (PC), total phosphorus (TP), and total nitrogen (TN) water quality parameters. Most studies have focused on spaceborne remote sensing, which is usually used to monitor large waterbodies for Chl-a and other water quality parameters with optical properties; semiempirical, bio-optical, and semianalytical models are frequently used. With the rapid development of aerospace technology and near-surface remote sensing, the spectral resolution of remote sensing imaging technology has been dramatically improved and has begun to be applied to small waterbodies. In the future, the multiplatform linkage monitoring approach may become a new research direction. Advanced computer technology has also enabled machine learning models to be applied to water quality parameter inversion, and machine learning models have higher robustness than the three commonly used models mentioned above. Although nitrogen and phosphorus, with nonoptical properties, have also received attention and research from some scholars in recent years, the uncertainty of their mechanisms makes it necessary to maintain a cautious attitude when treating such research.

Algal bloom forecasting with time-frequency analysis: A hybrid deep learning approach

The rapid emergence of deep learning long-short-term-memory (LSTM) technique presents a promising solution to algal bloom forecasting. However, the discontinuous and non-stationary processes within algal dynamics still largely limit the functions of LSTMs. To overcome this challenge, an advanced time-frequency wavelet analysis (WA) technique was introduced to enhance the prediction accuracy of LSTMs. Herein, the novel hybrid approach (named WLSTM) successfully decreased the algal forecasting inaccuracy of classic LSTMs by 41% ± 8% in Lake Mendota (Wisconsin, USA), with powerful one-step-ahead predictions at hourly, daily, and monthly time resolutions (R² = 0.976, 0.878, and 0.814, respectively). In addition, the WLSTM outperformed the other two widely used algal forecasting approaches - deep neural network (DNN), and autoregressive-integrated-moving-average (ARIMA) model, represented by average 72% and 85% decrease in root-mean-square-error, respectively. Furthermore, the WLSTM was implemented in an experimentally fertilized lake (Lake Tuesday, Michigan) for a multi-step forecasting examination. It satisfactorily forecasted the algal fluctuations involving substantial peak and extreme values (average R² > 0.900) and presented accurate judgment outcomes to their bloom levels with high accuracy > 95% on average. This work highlighted the utility of deep learning approaches in effective early-warning for algal blooms, and demonstrated an important direction for improving the adaptability of conventional deep learning approaches to the aquatic problems.

Optimal Band Selection for Airborne Hyperspectral Imagery to Retrieve a Wide Range of Cyanobacterial Pigment Concentration Using a Data-Driven Approach

Understanding the concentration and distribution of cyanobacteria blooms is an important aspect of managing water quality problems and protecting aquatic ecosystems. Airborne hyperspectral imagery (HSI)—which has high temporal, spatial, and spectral resolutions—is widely used to remotely sense cyanobacteria bloom, and it provides the distribution of the bloom over a wide area. In this study, we determined the input spectral bands that were relevant in effectively estimating the main two pigments (PC, Phycocyanin; Chl-a, Chlorophyll-a) of cyanobacteria by applying data-driven algorithms to HSI and then evaluating the change in the spatio-temporal distribution of cyanobacteria. The input variables for the algorithms consisted of reflectance band ratios associated with the optical properties of PC and Chl-a, which were calculated by the selected hyperspectral bands using a feature selection method. The selected input variable was composed of six reflectance bands (465.7–589.6, 603.6–631.8, 641.2–655.35, 664.8–679.0, 698.0–712.3, and 731.4–784.1 nm). The artificial neural network showed the best results for the estimation of the two pigments with average coefficients of determination 0.80 and 0.74. This study proposes relevant input spectral information and an algorithm that can effectively detect the occurrence of cyanobacteria in the weir pool along the Geum river, South Korea. The algorithm is expected to help establish a preemptive response to the formation of cyanobacterial blooms, and to contribute to the preparation of suitable water quality management plans for freshwater environments.

Spatiotemporal BME characterization and mapping of sea surface chlorophyll in Chesapeake Bay (USA) using auxiliary sea surface temperature data

Improving the spatiotemporal coverage of remote sensing (RS) products, such as sea surface chlorophyll concentration (SSCC), can offer a better understanding of the spatiotemporal SSCC distribution for ocean management purposes. In the first part of this work, 834 in-situ SSCC measurements of the SeaBASS-NASA (National Aeronautics and Space Administration) during 2002-2016 served as the empirical dataset. A moving window with ±3 days and ±0.5° centered at each of the in-situ SSCC measurements established a search neighborhood for Moderate Resolution Imaging Spectroradiometer Level 2 (MODIS L2) SSCC and MODIS L2 sea surface temperature (SST) data, and the matched SSCC and SST data were used for building a linear SSCC-SST relationship. The unmatched SST was introduced to the linear model for generating soft SSCC data with uniform distributions. The inherent spatiotemporal dependency of the SSCC distribution was then represented by the Bayesian maximum entropy (BME) method, which incorporated the soft SSCC data as auxiliary variable for SSCC estimation and mapping purposes. The results showed that a 75.3% accuracy improvement of remote SSCC retrieval in terms of R² can be achieved by BME-based method compared to the original MODIS L2 product. Subsequently, the BME-based method was applied to obtain daily SSCC dataset in Chesapeake Bay (USA) during the period 2010-2019. It was found that the SSCC distribution exhibited a decreasing spatial trend from the upper bay to the outer bay, whereas decreasing and increasing temporal trends were detected during the periods 2011-2014 and 2016-2019, respectively. The generalized Cauchy process was used to quantitatively describe the autocorrelation SSCC function in the Chesapeake Bay. The results showed that the outer bay exhibited the strongest long-range dependence among the four sub-regions, whereas the middle bay exhibited the weakest long-range dependence. Finally, one-point and two-point stochastic site indicators (SSIs) were employed to explore the spatiotemporal SSCC characteristics in Chesapeake Bay. The one-point SSI results showed that nearly 100% of the upper, middle and the lower bay areas experienced a high SSCC level (>5 mg/m³) during the entire study period. The area with SSCC >5 mg/m³ in the outer bay increased a lot during the winter season, but the area with SSCC >10 or 20 mg/m³ decreased significantly in the upper, middle and lower bay. Simultaneously, the SSCC dispersion in these areas was rather small during the winter season. On the other hand, the two-point SSI results showed that although the SSCC levels differ among the four sub-regions, but the SSCC connectivity structures between pairs of points also displayed some similarities in terms of their spatiotemporal dependency. In conclusion, the proposed BME-based method was shown to be a promising remote SSCC mapping technique that exhibited a powerful ability to improve both accuracy and coverage of RS products. The SSIs can be also used to explore the spatiotemporal characteristics of a variety of natural attributes in waters.

Machine learning in natural and engineered water systems

Water resources of desired quality and quantity are the foundation for human survival and sustainable development. To better protect the water environment and conserve water resources, efficient water management, purification, and transportation are of critical importance. In recent years, machine learning (ML) has exhibited its practicability, reliability, and high efficiency in numerous applications; furthermore, it has solved conventional and emerging problems in both natural and engineered water systems. For example, ML can predict various water quality indicators in situ and real-time by considering the complex interactions among water-related variables. ML approaches can also solve emerging pollution problems with proven rules or universal mechanisms summarized from the related research. Moreover, by applying image recognition technology to analyze the relationships between image information and physicochemical properties of the research object, ML can effectively identify and characterize specific contaminants. In view of the bright prospects of ML, this review comprehensively summarizes the development of ML applications in natural and engineered water systems. First, the concept and modeling steps of ML are briefly introduced, including data preparation, algorithm selection and model evaluation. In addition, comprehensive applications of ML in recent studies, including predicting water quality, mapping groundwater contaminants, classifying water resources, tracing contaminant sources, and evaluating pollutant toxicity in natural water systems, as well as modeling treatment techniques, assisting characterization analysis, purifying and distributing drinking water, and collecting and treating sewage water in engineered water systems, are summarized. Finally, the advantages and disadvantages of commonly used algorithms are analyzed according to their structures and mechanisms, and recommendations on the selection of ML algorithms for different studies, as well as prospects on the application and development of ML in water science are proposed. This review provides references for solving a wider range of water-related problems and brings further insights into the intelligent development of water science.

Retrieval of Water Quality from UAV-Borne Hyperspectral Imagery: A Comparative Study of Machine Learning Algorithms

The rapidly increasing world population and human activities accelerate the crisis of the limited freshwater resources. Water quality must be monitored for the sustainability of freshwater resources. Unmanned aerial vehicle (UAV)-borne hyperspectral data can capture fine features of water bodies, which have been widely used for monitoring water quality. In this study, nine machine learning algorithms are systematically evaluated for the inversion of water quality parameters including chlorophyll-a (Chl-a) and suspended solids (SS) with UAV-borne hyperspectral data. In comparing the experimental results of the machine learning model on the water quality parameters, we can observe that the prediction performance of the Catboost regression (CBR) model is the best. However, the prediction performances of the Multi-layer Perceptron regression (MLPR) and Elastic net (EN) models are very unsatisfactory, indicating that the MLPR and EN models are not suitable for the inversion of water quality parameters. In addition, the water quality distribution map is generated, which can be used to identify polluted areas of water bodies.

Inversion of Chlorophyll-a Concentration in Donghu Lake Based on Machine Learning Algorithm

Machine learning algorithm, as an important method for numerical modeling, has been widely used for chlorophyll-a concentration inversion modeling. In this work, a variety of models were built by applying five kinds of datasets and adopting back propagation neural network (BPNN), extreme learning machine (ELM), support vector machine (SVM). The results revealed that modeling with multi-factor datasets has the possibility to improve the accuracy of inversion model, and seven band combinations are better than seven single bands when modeling, Besides, SVM is more suitable than BPNN and ELM for chlorophyll-a concentration inversion modeling of Donghu Lake. The SVM model based on seven three-band combination dataset (SVM3) is the best inversion one among all multi-factor models that the mean relative error (MRE), mean absolute error (MAE), root mean square error (RMSE) of the SVM model based on single-factor dataset (SF-SVM) are 30.82%, 9.44 μg/L and 12.66 μg/L, respectively. SF-SVM performs best in single-factor models, MRE, MAE, RMSE of SF-SVM are 28.63%, 13.69 μg/L and 16.49 μg/L, respectively. In addition, the simulation effect of SVM3 is better than that of SF-SVM. On the whole, an effective model for retrieving chlorophyll-a concentration has been built based on machine learning algorithm, and our work provides a reliable basis and promotion for exploring accurate and applicable chlorophyll-a inversion model.

Inversion and distribution of total suspended matter in water based on remote sensing images-A case study on Yuqiao Reservoir, China

In this paper, Yuqiao Reservoir is taken as the research object. The total suspended matter (TSM) produced by the economic development in the upper reaches of the reservoir and its surrounding areas has brought great ecological harm to the safe operation of the reservoir. Satellite remote sensing technology provides a good way to obtain the temporal and spatial variation of TSM in the study area. Two field surveys were carried out in the Yuqiao Reservoir, a total of 44 sampling points collected in the two tests. The spectral data and concentration of TSM were obtained. We developed and validated a robust empirical model to estimate the concentration of TSM in the water of the Yuqiao Reservoir for the first time. The TSM distribution map of the Yuqiao Reservoir in 2013-2018 is retrieved based on Landsat 8 OLI images. This paper analyzes the spatial distribution characteristics of TSM concentration in the Yuqiao Reservoir for several years, as well as the interannual, seasonal, and monthly variation laws and development trends. The results show that the spatial distribution of TSM in Yuqiao Reservoir shows a decreasing trend from the periphery to the center; the interannual changes are mainly as follows: The annual change trend of TSM in Yuqiao Reservoir is not obvious; the seasonal changes are significant: the highest in summer (higher than 40 mg/L), the second in autumn, and the lowest in spring and winter (lower than 15 mg/L); and the monthly changes show regular fluctuations: In a year cycle, the concentration of TSM generally shows an inverted V-shaped trend; that is, TSM increases gradually from January to August and decreases gradually from August to December. The research results of this paper can be applied to other similar types of land water bodies, which will promote the wide application of Landsat 8 OLI images in the monitoring of TSM in lakes, rivers, and reservoirs in different regions across China, and provide data support for the scientific management of the safe operation of research areas. PRACTITIONER POINTS: The monitoring model of TSM in Yuqiao Reservoir was built for the first time. Temporal and spatial analysis of TSM concentration in Yuqiao Reservoir for the first time. The concentration of TSM is in Yuqiao Reservoir greatly affected by wind speed and precipitation.