4.7 Article

Prediction of phytoplankton biomass and identification of key influencing factors using interpretable machine learning models

相关参考文献

注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article Energy & Fuels

Predictive control for the operation of cascade pumping stations in water supply canal systems considering energy consumption and costs

Lingzhong Kong et al.

Summary: This study presents a closed-loop predictive control method for the operation of pumping stations in water supply canal systems, considering operation safety and energy consumption. The method uses a linear prediction model derived from the hydrodynamic model, data assimilation, and an optimization control model to obtain the optimal pump control strategy. The results showed that the proposed method could reduce the system's operating energy and cost while ensuring the safe operation of the system.

APPLIED ENERGY (2023)

Article Environmental Sciences

Assessment of the spatiotemporal water quality variations in the Middle Route of China's South-to-North Water Diversion Project by multivariate analysis

Yi Xu et al.

Summary: In this study, the water quality of China's South-to-North Water Diversion Project (MRP) was compared between different stages and spatial locations. The results showed significant improvement in water quality during the current stage. Principal component analysis and hierarchical cluster analysis identified key factors influencing water quality variations. The findings of this study provide useful information for local administrations to control pollution and protect the water quality of the MRP.

ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH (2023)

Article Biodiversity Conservation

Risk assessment of sectional water quality based on deterioration rate of water quality indicators: A case study of the main canal of the Middle Route of South-to-North Water Diversion Project

Ming Tang et al.

Summary: This study establishes probability density models for eight water quality indicators in the Middle Route of South-to-North Water Diversion Project of China (MRSNWDPC) based on real-time monitoring data. By defining water quality risks and calculating risk levels, it can predict water quality risks in the next month. The study finds that water quality deterioration exists in the MRSNWDPC, with higher risks observed for pH, dissolved oxygen, and BOD5. The risk assessment method proposed in this study can serve as an early warning system for water quality management in the MRSNWDPC.

ECOLOGICAL INDICATORS (2022)

Article Environmental Sciences

Spatial and temporal characteristics, influencing factors and prediction models of water quality and algae in early stage of Middle Route of South-North Water Diversion Project

Zheng Jing et al.

Summary: The study analyzes the change trends and distribution characteristics of main water ecological environment elements in the main canal of the South-to-North Water Diversion Project Middle Route at the early stage. The main factors influencing algae are investigated by principal component analysis, and the relationship between environmental factors, hydrodynamic, water quality, and algae in the coupled canal-sluice system is determined. The results may provide technical support for water quality forecasting and algae risk warning in the project.

ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH (2022)

Article Environmental Sciences

Towards better process management in wastewater treatment plants: Process analytics based on SHAP values for tree-based machine learning methods

Dong Wang et al.

Summary: Understanding pollutant removal mechanisms in WWTPs is crucial for controlling effluent quality. An upgraded ML framework with tree-based models and advanced SHAP interpretation system was used to study cause-and-effect relationships. Results showed XGBoost models to be optimal for TSSe and PO4(e) labels, providing insights for model comparison and interpretation in WWTP applications.

JOURNAL OF ENVIRONMENTAL MANAGEMENT (2022)

Article Engineering, Civil

Prediction of estuarine water quality using interpretable machine learning approach

Shuo Wang et al.

Summary: Estuaries are major sources of pollution in coastal areas, and estuarine water quality prediction models can provide early warnings to prevent disasters. This study developed machine learning models to predict NH4+-N in the Xiaoqing River estuary, China, finding a strong nonlinear relationship between estuarine NH4+-N and NH4+-N of the upper reaches. The explanations from the model highlighted the critical impact of two monitoring stations in the upper reaches on estuarine water quality.

JOURNAL OF HYDROLOGY (2022)

Article Engineering, Environmental

A hybrid framework for forecasting monthly reservoir inflow based on machine learning techniques with dynamic climate forecasts, satellite-based data, and climate phenomenon information

Di Tian et al.

Summary: This study developed a hybrid framework for reservoir inflow forecast, which integrated new datasets using machine learning models. The performance was evaluated in two headwater reservoirs, showing the best models and input variables for accurate inflow forecasts. The approach has potential for improving reservoir inflow forecasts globally.

STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT (2022)

Article Environmental Sciences

Predictive modeling of groundwater nitrate pollution and evaluating its main impact factors using random forest

Song He et al.

Summary: This study successfully predicted groundwater nitrate concentrations in the Yinchuan Region using a random forest model, identifying factors such as distance to the Yellow River, meteorological elements, water level elevation, urban land, and arable land as key influencers. Urban land was found to have a greater impact on groundwater nitrate concentrations compared to arable land due to intense urban land expansion from 2000 to 2015.

CHEMOSPHERE (2022)

Article Environmental Sciences

Fine-scale modeling of the urban heat island: A comparison of multiple linear regression and random forest approaches

Gabriel Yoshikazu Oukawa et al.

Summary: Characterizing the spatiotemporal variability of the Urban Heat Island (UHI) and its drivers is crucial for creating healthier cities and enhancing urban resilience to climate change. This study developed regression and random forest models to analyze and predict the UHI intensity using air temperature as the response variable. Results showed that anticyclonic circulations favored the largest UHI, while cyclonic circulations dampened its development. The random forest models outperformed the regression models in capturing and mapping the fine-scale spatiotemporal variability of air temperature.

SCIENCE OF THE TOTAL ENVIRONMENT (2022)

Article Construction & Building Technology

Prediction of rapid chloride penetration resistance of metakaolin based high strength concrete using light GBM and XGBoost models by incorporating SHAP analysis

Anas Abdulalim Alabdullah et al.

Summary: This study investigates the non-linear capabilities of two machine learning prediction models, Light GBM and XGBoost, for predicting RCPT values. The study found that LightGBM surpasses XGBoost in prediction accuracy and that the W/B ratio and MK replacement are of significant importance in resisting chloride penetration.

CONSTRUCTION AND BUILDING MATERIALS (2022)

Article Environmental Sciences

Decoding river pollution trends and their landscape determinants in an ecologically fragile karst basin using a machine learning model

Guoyu Xu et al.

Summary: This study developed a machine learning method and applied it to predict water quality in a karst watershed, revealing potential factors influencing water quality impairments. The research findings are important for decision-making in water quality restoration projects and bridging the science:policy gap.

ENVIRONMENTAL RESEARCH (2022)

Article Environmental Sciences

Interpretation of ensemble learning to predict water quality using explainable artificial intelligence

Jungsu Park et al.

Summary: This study developed an XGBoost ensemble machine learning model to predict chlorophyll-a (Chl-a) concentration and explored the effect of input variable selection on model performance. Using explainable artificial intelligence (XAI) algorithms, the study provided interpretable analyses of model predictions. The results showed that selecting input variables based on the Shapley value (SHAP) algorithm improved model stability and helped reduce the cost of water quality analysis.

SCIENCE OF THE TOTAL ENVIRONMENT (2022)

Article Engineering, Environmental

A comprehensive method for improvement of water quality index (WQI) models for coastal water quality assessment

Md Galal Uddin et al.

Summary: This study presents an improved water quality index (WQI) model for assessing coastal water quality in Cork Harbour, Ireland. The model uses machine learning and sub-indexing methods to provide a more objective and data-driven evaluation of water quality, while reducing issues of ambiguity and interference.

WATER RESEARCH (2022)

Article Water Resources

Analysis of runoff generation driving factors based on hydrological model and interpretable machine learning method

Shuo Wang et al.

Summary: The study focused on identifying the driving factors of temporal and spatial variation in runoff in Xiaoqing River Basin, Shandong Province, China. The traditional machine learning model lacks transparency and interpretability, which hinders the widespread application of machine learning in hydrology. An interpretable machine learning method using extreme gradient boosting (XGBoost) and Shapely additive explanations (SHAP) was applied to analyze the impact of driving factors on runoff generation. The study found that climatic features and land use combination (or slope) have significant effects on runoff generation, with rainfall at high elevations (or slope) having a stronger impact than that at low elevations. The study provides new insights and a method for factor analysis of runoff.

JOURNAL OF HYDROLOGY-REGIONAL STUDIES (2022)

Article Construction & Building Technology

Predicting mechanical properties of carbon nanotube-reinforced cementitious nanocomposites using interpretable ensemble learning models

Hossein Adel et al.

Summary: Machine learning techniques were utilized to predict compressive and flexural strength of cementitious nanocomposites with carbon nanotubes. Among the ensemble learning-based algorithms and standalone models tested, XGBoost was found to provide more reliable and accurate results, effectively mapping the relationships between input variables. Additionally, SHAP analysis was performed to explain the contribution of each input variable to the output of XGBoost.

CONSTRUCTION AND BUILDING MATERIALS (2022)

Article Environmental Sciences

A framework for exploring environmental risk of the longest inter-basin water diversion project under the influence of multiple factors: A case study in China

Chi Zhang et al.

Summary: This article introduces a multivariate risk assessment method based on Vine Copula theory and Monte Carlo simulation to evaluate the risk of algal proliferation in the Middle Route of the South-to-North Water Diversion Project of China. The results demonstrate that water temperature can serve as an appropriate early-warning indicator of algal proliferation, and corresponding early-warning intervals and control measures are proposed.

JOURNAL OF ENVIRONMENTAL MANAGEMENT (2022)

Article Environmental Sciences

Spatiotemporal dynamics and anthropologically dominated drivers of chlorophyll-a, TN and TP concentrations in the Pearl River Estuary based on retrieval algorithm and random forest regression

Xingyu Yuan et al.

Summary: In this study, a retrieval algorithm was developed to predict water quality parameters in the Pearl River Estuary using in situ measurements and remote sensing images. Turbidity and pH were identified as important indicators for explaining water quality variation, and agricultural pollution was the main pollution source. Light limitation caused by high turbidity resulted in decreased algae productivity.

ENVIRONMENTAL RESEARCH (2022)

Article Biodiversity Conservation

Comparison of random forest and multiple linear regression models for estimation of soil extracellular enzyme activities in agricultural reclaimed coastal saline land

Xuefeng Xie et al.

Summary: This study used models to estimate soil amylase and urease activities, and found that soil water content and total nitrogen are the main factors influencing their activities. The random forest model outperformed the multiple linear regressions model in estimating soil enzyme activities.

ECOLOGICAL INDICATORS (2021)

Article Biodiversity Conservation

Visual quality evaluation model of an urban river landscape based on random forest

Xin Li et al.

Summary: This paper established a quantitative landscape index system using deep learning and analyzed the visual quality of an urban river on-water landscape. A prediction model was developed to analyze the impact of factors such as urban construction level, destructive index, hard revetment visibility, and green visibility index on visual quality. The green visibility index was positively correlated, while other factors were negatively correlated with visual quality.

ECOLOGICAL INDICATORS (2021)

Article Biodiversity Conservation

Combined Generalized Additive model and Random Forest to evaluate the influence of environmental factors on phytoplankton biomass in a large eutrophic lake

Jinpeng Zhang et al.

Summary: This study investigated the drivers and response relationships of phytoplankton biomass variations in Lake Okeechobee using nonlinear methods. Results showed spatial heterogeneity in phytoplankton biomass distribution and a three-peak-two-valley variation trend over time. The importance of inorganic nutrients and light availability factors as predictors of phytoplankton biomass was highlighted at both lake-wide and local scales.

ECOLOGICAL INDICATORS (2021)

Article Ecology

Interpretable machine learning to forecast hypoxia in a lagoon

Dimitris Politikos et al.

Summary: Dissolved oxygen is a key indicator in aquatic ecosystems, and low levels may lead to high risk for aquatic organisms. The study found that the Extremely Randomized Trees (EXT) algorithm slightly outperformed the others in forecasting hypoxia, with pH, water temperature, chlorophyll, and salinity being the top explanatory variables.

ECOLOGICAL INFORMATICS (2021)

Article Environmental Sciences

Spatial heterogeneity modeling of water quality based on random forest regression and model interpretation

Feier Wang et al.

Summary: The study used random forest regression (RFR) models and SHAP interpretation to successfully establish a water quality distribution model for the Taihu Lake basin. Results showed better water quality in urban areas compared to rural areas, leading to an unexpected positive relationship between water quality and population density due to differences in sewage collection and treatment.

ENVIRONMENTAL RESEARCH (2021)

Article Environmental Sciences

Application of Machine Learning for eutrophication analysis and algal bloom prediction in an urban river: A 10-year study of the Han River, South Korea

Quang Viet Ly et al.

Summary: This study used machine learning algorithms to predict algal blooms in the Han River and successfully identified the importance of water quality parameters in algal bloom prediction, with the adaptive neuro-fuzzy inference system showing the best performance.

SCIENCE OF THE TOTAL ENVIRONMENT (2021)

Article Engineering, Civil

Ecological scheduling of the middle route of south-to-north water diversion project based on a reinforcement learning model

Jie Zhu et al.

Summary: By building an ecological scheduling model, this study successfully addressed the issue of abnormal algae proliferation in the Middle Route of the South-to-North Water Diversion project, ensuring comprehensive protection of water quality and supply safety in the canal.

JOURNAL OF HYDROLOGY (2021)

Article Environmental Sciences

Temporal and spatial variation in water quality in the Three Gorges Reservoir from 1998 to 2018

Rong Xiang et al.

Summary: The overall water quality of the Three Gorges Reservoir was good, but showed a slight decline before full impoundment. The low water level period had the worst water quality, and the midstream area exhibited the poorest water quality. Total nitrogen and total phosphorus were identified as key pollution indices.

SCIENCE OF THE TOTAL ENVIRONMENT (2021)

Article Engineering, Environmental

Water quality variation in tributaries of the Three Gorges Reservoir from 2000 to 2015

Rong Xiang et al.

Summary: The study investigated the spatiotemporal variations in water quality of the Three Gorges Reservoir tributaries, revealing that the water quality was mainly affected by the mainstream backwater movement. Overall, the water quality of tributaries was good and showed an improving trend, with upstream tributaries having better water quality compared to downstream areas and estuaries.

WATER RESEARCH (2021)

Article Computer Science, Interdisciplinary Applications

A novel multi-source data fusion method based on Bayesian inference for accurate estimation of chlorophyll-a concentration over eutrophic lakes

Cheng Chen et al.

Summary: A novel multi-source data fusion method based on Bayesian inference was proposed for accurate chlorophyll-a concentration estimation in Lake Taihu. The multiplicative error model was found to be more suitable as compared to the additive model. The Bayesian inference method outperformed other data fusion algorithms, showing the largest correlation coefficients and smallest root mean square error, and successfully captured the spatial distribution of chlorophyll-a concentrations in Lake Taihu.

ENVIRONMENTAL MODELLING & SOFTWARE (2021)

Article Environmental Sciences

Analysis of spatio-temporal variation in phytoplankton and its relationship with water quality parameters in the South-to-North Water Diversion Project of China

Xizhi Nong et al.

Summary: The Middle Route of the South-to-North Water Diversion Project of China (MRSNWDPC) is the longest trans-basin water diversion project in the world, and its water quality security and algae control strategy are crucial for project management.

ENVIRONMENTAL MONITORING AND ASSESSMENT (2021)

Article Water Resources

Hydrological probabilistic forecasting based on deep learning and Bayesian optimization algorithm

Haijun Bai et al.

Summary: A hybrid model combining XGB and GPR with BOA optimization is proposed for runoff probabilistic forecasting, achieving high accuracy and reliability on runoff prediction problems by integrating these two methods and optimizing hyper-parameters using BOA.

HYDROLOGY RESEARCH (2021)

Article Green & Sustainable Science & Technology

Efficiency analysis of metacoupling of water transfer based on the parallel data envelopment analysis model: A case of the South-North Water Transfer Project-Middle Route in China

Qingmu Su et al.

Summary: The deepening of globalization and rapid socio-economic development have increased the demand for water resources, leading to water shortages. Cross-regional water transfer is an effective measure to address this issue. The impact of metacoupling in water transfer projects needs further exploration. A model and evaluation framework have been developed to calculate overall efficiency and optimize it by adjusting water resource proportions.

JOURNAL OF CLEANER PRODUCTION (2021)

Article Geosciences, Multidisciplinary

Machine-learning methods for stream water temperature prediction

Moritz Feigl et al.

Summary: This study evaluates the performance of six different machine-learning models for daily water temperature prediction in 10 Austrian catchments, with FNNs and XGBoost performing best in majority of the catchments. The importance of hyperparameter optimization is highlighted, showing varied performance for different models.

HYDROLOGY AND EARTH SYSTEM SCIENCES (2021)

Article Environmental Sciences

Application of feature selection and regression models for chlorophyll-a prediction in a shallow lake

Xue Li et al.

ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH (2018)

Article Ecology

Modeling lake trophic state: a random forest approach

Jeffrey W. Hollister et al.

ECOSPHERE (2016)

Article Biology

The dynamics of temperature and light on the growth of phytoplankton

Ming Chen et al.

JOURNAL OF THEORETICAL BIOLOGY (2015)

Article Chemistry, Physical

Prediction of Crude Oil Asphaltene Precipitation Using Support Vector Regression

Seyyed Reza Na'imi et al.

JOURNAL OF DISPERSION SCIENCE AND TECHNOLOGY (2014)

Article Computer Science, Interdisciplinary Applications

Monthly river flow forecasting using artificial neural network and support vector regression models coupled with wavelet transform

Aman Mohammad Kalteh

COMPUTERS & GEOSCIENCES (2013)

Article Environmental Sciences

Effect of flow velocity on phytoplankton biomass and composition in a freshwater lake

Li Feipeng et al.

SCIENCE OF THE TOTAL ENVIRONMENT (2013)

Article Agricultural Engineering

Role and functions of beneficial microorganisms in sustainable aquaculture

Qunlan Zhou et al.

BIORESOURCE TECHNOLOGY (2009)

Article Computer Science, Artificial Intelligence

Random forests

L Breiman

MACHINE LEARNING (2001)