Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Bacterial Phylum Cyanobacteria01:30

Bacterial Phylum Cyanobacteria

345
Cyanobacteria are a diverse group of oxygenic, phototrophic bacteria that played a pivotal role in converting Earth’s atmosphere from anoxic to oxygen-rich billions of years ago. They exhibit remarkable morphological diversity, ranging from unicellular forms to filamentous types, with cell sizes varying between 0.5 μm and 100 μm. Cyanobacteria are classified into five groups: Chroococcales (unicellular, dividing by binary fission), Pleurocapsales (unicellular, dividing by...
345

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Full-spectrum EEM end-member unmixing with statistical validation: A deep learning framework for organic pollution source apportionment in complex watersheds.

Journal of environmental management·2026
Same author

Multimodal artificial intelligence (AI) for estimating the sinking velocity of microplastic-microalgae aggregates using explainable AI and visualization.

Water research·2026
Same author

Thin-Film Eddy Current Sensors for Proximity Detection in Smart Capillary Pick-and-Place.

ACS sensors·2026
Same author

Adaptive reinforcement learning for energy-efficient high-recovery closed-circuit reverse osmosis.

Water research·2026
Same author

Advancing membrane fouling control via electrically driven microbubble generation: Phosphidation-enhanced electrocatalytic activity of Ni foam and numerical optimization of operational parameters.

Water research·2026
Same author

Graph neural network-based prediction and interpretation of Daphnia toxicity using distinct scale molecular representations.

Water research·2026

Related Experiment Video

Updated: Dec 10, 2025

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy
07:13

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy

Published on: February 25, 2021

4.2K

Using convolutional neural network for predicting cyanobacteria concentrations in river water.

JongCheol Pyo1, Lan Joo Park2, Yakov Pachepsky3

  • 1School of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology, 50 UNIST-gil, Eonyang-eup, Ulju-gun, Ulsan 689-798, Republic of Korea.

Water Research
|September 4, 2020
PubMed
Summary
This summary is machine-generated.

This study shows that convolutional neural networks (CNNs) can predict harmful cyanobacterial blooms using water quality data. CNNs offer a promising tool for forecasting algal blooms, with accuracy depending on data quality and observation density.

Keywords:
Convolutional neural networkEFDCMicrocystisPredictionSynthetic data

More Related Videos

Experimental Protocol for Detecting Cyanobacteria in Liquid and Solid Samples with an Antibody Microarray Chip
10:57

Experimental Protocol for Detecting Cyanobacteria in Liquid and Solid Samples with an Antibody Microarray Chip

Published on: February 7, 2017

9.4K
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
08:47

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

1.9K

Related Experiment Videos

Last Updated: Dec 10, 2025

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy
07:13

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy

Published on: February 25, 2021

4.2K
Experimental Protocol for Detecting Cyanobacteria in Liquid and Solid Samples with an Antibody Microarray Chip
10:57

Experimental Protocol for Detecting Cyanobacteria in Liquid and Solid Samples with an Antibody Microarray Chip

Published on: February 7, 2017

9.4K
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
08:47

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

1.9K

Area of Science:

  • Environmental Science
  • Water Quality Monitoring
  • Machine Learning Applications

Background:

  • Algal blooms, particularly harmful cyanobacterial blooms, pose significant threats to aquatic ecosystems and human health.
  • Predictive modeling is crucial for timely management and mitigation of algal bloom events.
  • Machine learning, specifically convolutional neural networks (CNNs), shows potential for advancing water quality prediction.

Purpose of the Study:

  • To investigate the capability of a CNN for predicting harmful cyanobacterial blooms (Microcystis) in a river section.
  • To evaluate the impact of forecast lead time, spatial observation density, and data noise on CNN prediction accuracy.
  • To assess the performance variations of the CNN model across the river reach.

Main Methods:

  • Generation of synthetic spatio-temporal water quality data using a 3D water quality model.
  • Application of a CNN model to predict cyanobacteria biomass from generated data.
  • Quantitative assessment of CNN performance using Nash-Sutcliffe Efficiency (NSE) under varying conditions (lead time, data density, noise).
  • Visualization of CNN results to understand spatial performance variations.

Main Results:

  • The CNN model achieved a high Nash-Sutcliffe Efficiency (NSE) of 0.87 for short-term nowcasting of Microcystis biomass.
  • Prediction accuracy decreased with increased forecast lead time, with NSE dropping to 0.58.
  • Increasing spatial observation density from 20% to 100% improved CNN prediction NSE from 0.70 to 0.84.
  • The model maintained acceptable accuracy (NSE = 0.76) even with 10% data noise.
  • Visualization revealed performance variations across the river reach.

Conclusions:

  • CNNs demonstrate a strong capability for predicting cyanobacterial blooms using high temporal frequency water quality data.
  • Model performance is sensitive to forecast lead time and spatial observation density, highlighting the importance of data acquisition strategies.
  • The CNN approach shows promise for operational monitoring and management of harmful algal blooms in riverine systems.