Machine learning-based design and monitoring of algae blooms: Recent trends and future perspectives – A short review DOI

Abdul Gaffar Sheik,

Arvind Kumar,

Reeza Patnaik

и другие.

Critical Reviews in Environmental Science and Technology, Год журнала: 2023, Номер 54(7), С. 509 - 532

Опубликована: Сен. 7, 2023

AbstractMachine learning (ML) models are widely used methods for analyzing data from sensors and satellites to monitor climate change, predict natural disasters, protect wildlife. However, the application of these technologies monitoring managing algal blooms in freshwater environments is relatively new novel. The commonly (ABS) so far artificial neural networks (ANN), random forests (RF), support vector machine (SVM), data-driven modeling, long short-term memory (LSTM). In past, researchers have mostly worked on predicting effluent parameters, nutrients, microculture, area weather conditions, meteorological factors, ground waters, energy optimization, metallic substances using ML models. Most studies employed performance metrics like root mean squared error, peak signal, precision, determination coefficient as their primary model measures accuracy analysis, usage transfer, activation function. While there been some this topic, several research gaps still be addressed. most significant related limited different algae bloom scenarios, interpretability models, lack integration with existing systems. Keeping mind, review article has methodically arranged present an overview past studies, limitations, way forward toward prediction ABS, thus benefitting future area. This aims summarize that available, including benchmarking values.HighlightsReal-time dynamics essential mitigating blooms.Various complexities hinder applications current algorithms ABS.Activation transfer functions can selection ABS.Integrated drive feature engineering control ABS.Keywords: Activation-functionalgae bloomsmonitoringmachine learningperformance predictionHANDLING EDITORS: Hyunjung Kim Scott Bradford Disclosure statementNo potential conflict interest was reported by authors.

Язык: Английский

From Hydrometeorology to River Water Quality: Can a Deep Learning Model Predict Dissolved Oxygen at the Continental Scale? DOI
Wei Zhi, Dapeng Feng, Wen‐Ping Tsai

и другие.

Environmental Science & Technology, Год журнала: 2021, Номер 55(4), С. 2357 - 2368

Опубликована: Фев. 3, 2021

Dissolved oxygen (DO) reflects river metabolic pulses and is an essential water quality measure. Our capabilities of forecasting DO however remain elusive. Water data, specifically data here, often have large gaps sparse areal temporal coverage. Earth surface hydrometeorology on the other hand, become largely available. Here we ask: can a Long Short-Term Memory (LSTM) model learn about dynamics from intensive (daily) data? We used CAMELS-chem, new set with concentrations 236 minimally disturbed watersheds across U.S. The generally learns theory solubility captures its decreasing trend increasing temperature. It exhibits potential predicting in "chemically ungauged basins", defined as basins without any measurements broadly general. misses some peaks troughs when in-stream biogeochemical processes important. Surprisingly, does not perform better where more are Instead, it performs low variations streamflow DO, high runoff-ratio (>0.45), winter precipitation peaks. Results here suggest that collections at sparsely monitored areas to overcome issue scarcity, outstanding challenge community.

Язык: Английский

Процитировано

214
Recent advancement in water quality indicators for eutrophication in global freshwater lakes DOI Creative Commons
Keerthana Suresh, Ting Tang, Michelle T. H. van Vliet

и другие.

Environmental Research Letters, Год журнала: 2023, Номер 18(6), С. 063004 - 063004

Опубликована: Апрель 26, 2023

Abstract Eutrophication is a major global concern in lakes, caused by excessive nutrient loadings (nitrogen and phosphorus) from human activities likely exacerbated climate change. Present use of indicators to monitor assess lake eutrophication restricted water quality constituents (e.g. total phosphorus, nitrogen) does not necessarily represent environmental changes the anthropogenic influences within lake’s drainage basin. Nutrients interact multiple ways with climate, basin conditions socio-economic development, point-source, diffuse source pollutants), systems. It therefore essential account for complex feedback mechanisms non-linear interactions that exist between nutrients ecosystems assessments. However, lack set holistic understanding challenges such assessments, addition limited monitoring data available. In this review, we synthesize main freshwater basins only include but also sources, biogeochemical pathways responses emissions. We develop new causal network (i.e. links indicators) using DPSIR (drivers-pressure-state-impact-response) framework highlights interrelationships among provides perspective dynamics basins. further review 30 key drivers pressures seven cross-cutting themes: (i) hydro-climatology, (ii) socio-economy, (iii) land use, (iv) characteristics, (v) crop farming livestock, (vi) hydrology management, (vii) fishing aquaculture. This study indicates need more comprehensive systems, guide expansion networks, support integrated assessments manage eutrophication. Finally, proposed can be used managers decision-makers realistic targets sustainable management achieve clean all, line Sustainable Development Goal 6.

Язык: Английский

Процитировано

69
River water quality shaped by land–river connectivity in a changing climate DOI
Li Li, Julia L. A. Knapp, Anna Lintern

и другие.

Nature Climate Change, Год журнала: 2024, Номер 14(3), С. 225 - 237

Опубликована: Март 1, 2024

Язык: Английский

Процитировано

68
Harmful algal blooms in inland waters DOI
Lian Feng, Ying Wang, Xuejiao Hou

и другие.

Nature Reviews Earth & Environment, Год журнала: 2024, Номер 5(9), С. 631 - 644

Опубликована: Авг. 27, 2024

Язык: Английский

Процитировано

30
Global groundwater warming due to climate change DOI Creative Commons
Susanne A. Benz, Dylan J. Irvine, Gabriel C. Rau

и другие.

Nature Geoscience, Год журнала: 2024, Номер 17(6), С. 545 - 551

Опубликована: Июнь 1, 2024

Abstract Aquifers contain the largest store of unfrozen freshwater, making groundwater critical for life on Earth. Surprisingly little is known about how responds to surface warming across spatial and temporal scales. Focusing diffusive heat transport, we simulate current projected temperatures at global scale. We show that depth water table (excluding permafrost regions) conservatively warm average by 2.1 °C between 2000 2100 under a medium emissions pathway. However, regional shallow patterns vary substantially due variability in climate change depth. The lowest rates are mountain regions such as Andes or Rocky Mountains. illustrate increasing influences stream thermal regimes, groundwater-dependent ecosystems, aquatic biogeochemical processes, quality geothermal potential. Results indicate following pathway, 77 million 188 people live areas where exceeds highest threshold drinking set any country.

Язык: Английский

Процитировано

29
The Chesapeake Bay program modeling system: Overview and recommendations for future development DOI Creative Commons
Raleigh R. Hood, Gary W. Shenk, Rachel L. Dixon

и другие.

Ecological Modelling, Год журнала: 2021, Номер 456, С. 109635 - 109635

Опубликована: Июль 17, 2021

Язык: Английский

Процитировано

70
Climate Controls on River Chemistry DOI Creative Commons
Li Li, Bryn Stewart, Wei Zhi

и другие.

Earth s Future, Год журнала: 2022, Номер 10(6)

Опубликована: Май 16, 2022

Abstract How does climate control river chemistry? Existing literature has examined extensively the response of chemistry to short‐term weather conditions from event seasonal scales. Patterns and drivers long‐term, baseline have remained poorly understood. Here we compile analyze data 506 minimally impacted rivers (412,801 points) in contiguous United States (CAMELS‐Chem) identify patterns chemistry. Despite distinct sources diverse reaction characteristics, a universal pattern emerges for 16 major solutes at continental scale. Their long‐term mean concentrations ( C m ) decrease with discharge Q ), elevated arid climates lower humid climates, indicating overwhelming regulation by compared local Critical Zone characteristics such as lithology topography. To understand pattern, parsimonious watershed reactor model was solved bringing together hydrology (storage–discharge relationship) biogeochemical theories traditionally separate disciplines. The derivation steady state solutions lead power law form relationships. illuminates two competing processes that determine solute concentrations: production subsurface chemical weathering reactions, export (or removal) discharge, water flushing capacity dictated vegetation. In other words, watersheds function primarily reactors produce accumulate transporters climates. With space‐for‐time substitution, these results indicate places where dwindles warming climate, will elevate even without human perturbation, threatening quality aquatic ecosystems. Water deterioration therefore should be considered global calculation future risks.

Язык: Английский

Процитировано

57
Remote sensing for mapping algal blooms in freshwater lakes: a review DOI
Sílvia Beatriz Alves Rolim, Bijeesh Kozhikkodan Veettil,

Antônio Pedro Vieiro

и другие.

Environmental Science and Pollution Research, Год журнала: 2023, Номер 30(8), С. 19602 - 19616

Опубликована: Янв. 16, 2023

Язык: Английский

Процитировано

36
Can Artificial Intelligence Accelerate Fluid Mechanics Research? DOI Creative Commons
Dimitris Drikakis, Filippos Sofos

Fluids, Год журнала: 2023, Номер 8(7), С. 212 - 212

Опубликована: Июль 19, 2023

The significant growth of artificial intelligence (AI) methods in machine learning (ML) and deep (DL) has opened opportunities for fluid dynamics its applications science, engineering medicine. Developing AI encompass different challenges than with massive data, such as the Internet Things. For many scientific, biomedical problems, data are not massive, which poses limitations algorithmic challenges. This paper reviews ML DL research dynamics, presents discusses potential future directions.

Язык: Английский

Процитировано

32
Estimates of Lake Nitrogen, Phosphorus, and Chlorophyll‐a Concentrations to Characterize Harmful Algal Bloom Risk Across the United States DOI Creative Commons
Meredith M. Brehob, Michael J. Pennino, Amalia M. Handler

и другие.

Earth s Future, Год журнала: 2024, Номер 12(8)

Опубликована: Авг. 1, 2024

Abstract Excess nutrient pollution contributes to the formation of harmful algal blooms (HABs) that compromise fisheries and recreation can directly endanger human animal health via cyanotoxins. Efforts quantify occurrence, drivers, severity HABs across large areas is difficult due resource intensive nature field monitoring lake chlorophyll‐ a concentrations. To better characterize how nutrients interact with other environmental factors produce in freshwater systems, we used spatially explicit temporally matched climate, landscape, in‐lake characteristic, inventory data sets predict conterminous US (CONUS). Using nested modeling approach, three random forest (RF) models were trained explain spatiotemporal variation total nitrogen (TN), phosphorus (TP), concentrations EPA's National Lakes Assessment ( n = 2,062). Concentrations TN TP most important predictors and, variables, RF model accounted for 68% . We then these extrapolate predictions lakes without observations ∼112,000 CONUS. Risk high highest agriculturally dominated Midwest, but risk emerge hot spots country. These catchment lake‐specific results help managers identify potential may fuel blooms, prioritize at‐risk additional monitoring, optimize management protect end goals.

Язык: Английский

Процитировано

9