Sign in Subscribe
Machine learning

Machine Learning Model Accurately Detects Eutrophication in Reservoirs Using Remote Sensing Data

Strategic Insights

— 1 min read

Research conducted by the University of Oviedo has demonstrated the efficacy of a machine learning model in detecting eutrophication in reservoirs using remote sensing data. The study, published in the IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, utilized satellite imagery to monitor 20 reservoirs from 2017 to 2024, resulting in a dataset of 496 images paired with corresponding in situ measurements.

The machine learning model, integrated with deep learning techniques, achieved balanced accuracy scores above 0.89 and accuracy levels above 0.92 across the entire reservoir dataset. The study concluded that remote sensing systems can effectively detect variations in water quality, characterize the trophic status of reservoirs, and monitor changes in their environmental conditions over time.

Key Takeaways:

  • The machine learning model utilized remote sensing data from satellite imagery to monitor 20 reservoirs from 2017 to 2024.
  • The dataset consisted of 496 images paired with corresponding in situ measurements, from which 198 multispectral indices were computed for analysis.
  • The model achieved balanced accuracy scores above 0.89 and accuracy levels above 0.92 across the entire reservoir dataset.
  • The study concluded that remote sensing systems can effectively detect variations in water quality and characterize the trophic status of reservoirs.
  • The model's strong ability to distinguish between eutrophic and noneutrophic conditions was demonstrated.
  • The study highlighted the potential of remote sensing systems in monitoring changes in environmental conditions over time.

Statistics:

  • 20 reservoirs were monitored from 2017 to 2024.
  • 496 images were paired with corresponding in situ measurements.
  • 198 multispectral indices were computed for analysis.
  • The model achieved balanced accuracy scores above 0.89.
  • Accuracy levels above 0.92 were achieved across the entire reservoir dataset.

Sources:

  • NewsRx. New Machine Learning Study Findings Have Been Reported by Researchers at University of Oviedo (Remote Sensing and Machine Learning for Eutrophication Detection: Assessing the Trophic State in Reservoirs Using Multispectral Indices and Deep ...). Life Science Weekly. August 5, 2025; p 2363.
  • IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. Remote Sensing and Machine Learning for Eutrophication Detection: Assessing the Trophic State in Reservoirs Using Multispectral Indices and Deep Learning. 2025, 18():16206-16226.
  • IEEE (publisher).

Read more