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

You might also read

Related Articles

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

Sort by
Same author

MicroRAG: Development of a Novel Artificial Intelligence Retrieval-Augmented Generation Model for Microsurgery Clinical Decision Support.

Microsurgery·2025
See all related articles

Related Experiment Video

Updated: Jan 13, 2026

Reefshape: A System for the Efficient Collection and Automated Processing of Time-Series Underwater Photogrammetry Data for Benthic Habitat Monitoring
13:35

Reefshape: A System for the Efficient Collection and Automated Processing of Time-Series Underwater Photogrammetry Data for Benthic Habitat Monitoring

Published on: June 13, 2025

1.3K

Spatio-Temporal Machine Learning for Marine Pollution Prediction: A Multi-Modal Approach for Hotspot Detection and

Sarthak Pattnaik1, Eugene Pinsky1

  • 1Department of Computer Science, Metropolitan College, Boston University, Boston, MA 02215, USA.

Toxics
|October 28, 2025
PubMed
Summary

Marine pollution in Pacific Islands is predictable. Machine learning models accurately forecast pollution types and hotspots, aiding conservation efforts during peak seasons, particularly in June.

Keywords:
LSTM networksPacific Oceandeep learningenvironmental managementhotspot detectionmarine pollution predictionpollution classificationseasonal patternsspatio-temporal analysis

More Related Videos

Author Spotlight: Unveiling Plankton Response to Climate Change Through Time-Series Data and Artistic Expression
08:15

Author Spotlight: Unveiling Plankton Response to Climate Change Through Time-Series Data and Artistic Expression

Published on: July 28, 2023

1.8K
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

4.4K

Related Experiment Videos

Last Updated: Jan 13, 2026

Reefshape: A System for the Efficient Collection and Automated Processing of Time-Series Underwater Photogrammetry Data for Benthic Habitat Monitoring
13:35

Reefshape: A System for the Efficient Collection and Automated Processing of Time-Series Underwater Photogrammetry Data for Benthic Habitat Monitoring

Published on: June 13, 2025

1.3K
Author Spotlight: Unveiling Plankton Response to Climate Change Through Time-Series Data and Artistic Expression
08:15

Author Spotlight: Unveiling Plankton Response to Climate Change Through Time-Series Data and Artistic Expression

Published on: July 28, 2023

1.8K
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

4.4K

Area of Science:

  • Marine Biology
  • Environmental Science
  • Data Science

Background:

  • Marine pollution incidents severely impact Pacific Island ecosystems and communities.
  • Effective environmental management requires advanced predictive capabilities for these incidents.

Purpose of the Study:

  • To develop predictive models for marine pollution type classification, hotspot identification, and seasonal pattern forecasting in Pacific Island nations.
  • To establish a comprehensive baseline for marine pollution patterns in the region.

Main Methods:

  • Analysis of 8133 marine pollution incidents from 2001-2014 across 25 Pacific Island nations.
  • Application of machine learning for pollution type classification and pattern analysis.
  • Temporal and geographic analysis to identify pollution hotspots and seasonal dependencies.

Main Results:

  • Papua New Guinea identified as the dominant pollution hotspot (51.9% of incidents), with plastic waste dumping being the primary type (78.8%).
  • Machine learning achieved 99.1% accuracy in predicting pollution types, with material composition, season, and location as key predictors.
  • Peak pollution activity observed in June (755 average incidents), coinciding with critical fish breeding and vulnerable marine ecological periods.

Conclusions:

  • Marine pollution in Pacific Island nations exhibits predictable patterns based on type, location, and season.
  • Machine learning provides a validated approach for proactive marine pollution monitoring and targeted conservation.
  • The findings offer scalable solutions for protecting ocean ecosystems and informing policy for the region.