Retrieving Chlorophyll-a Levels, Transparency and TSS Concentration from Multispectral Satellite Data by Using Artificial Neural Networks

AI and satellite data join forces to estimate water quality in Amazon reservoirs with high accuracy and minimal fieldwork.

Smart Water Monitoring in Remote Areas

How can we assess water quality in remote, vast, and sensitive areas without collecting physical samples every time? The answer comes from an intelligent system that combines satellite data and artificial neural networks to estimate key parameters such as chlorophyll-a, transparency, and total suspended solids (TSS) in water reservoirs.

Satellite Imaging Across the Hydrological Cycle

At the heart of the method lies the use of multispectral images from the Landsat 7 satellite, covering the full annual hydrological cycle (flood, emptying, dry, refill). The visible and near-infrared bands are calibrated and atmospherically corrected to accurately estimate surface water reflectance.

Training Neural Networks with Real and Remote Data

These images feed into an artificial neural network trained on more than 30 satellite images and validated with laboratory measurements collected between 2007 and 2014. The model, tailored for each of the seven sampling points in a large Amazonian hydroelectric reservoir, was tested using robust Leave-One-Out cross-validation techniques.

Accuracy Across Seasonal Variations

The results? Performance improves during dry periods when cloud cover is minimal and satellite imagery is clearer. However, even during other hydrological cycles, the average estimation error remains low, making the system reliable for environmental monitoring.

Scalable and Non-Invasive Environmental Surveillance

This approach offers a dual advantage: it provides accurate estimates without the need for constant on-site measurements and enables large-scale continuous monitoring of reservoirs that are critical for ecological balance and water resource management. A concrete example of how artificial intelligence and satellite observation can together transform the way we observe and protect our ecosystems.

Contacts

E-Mail:

info@gridplus.it
Phone:

+39 3387655663

Request information→

Useful links

Training

Authors

H. A. Nascimento Silva, G. Laneve, A. Rosato, M. Panella

February 19, 2018

A Review of the Enabling Methodologies for Knowledge Discovery from Smart Grids Data

A KDD-driven pipeline turns smart meter streams into multi-step load forecasts, benchmarking feature reduction and models.

Biomedical

Enhancing Autism Detection Through Gaze Analysis Using Eye Tracking Sensors and Data Attribution with Distillation in Deep Neural Networks

A deep learning model enhances early autism diagnosis by analyzing visual patterns with eye tracking.

Quantum computing

Quantum Generative Modeling via Straightforward State Preparation

A lightweight quantum generative model creates high-fidelity data samples with minimal parameters and efficient state preparation.

Quantum computing

Enhancing QAOA Ansatz via Multi-Parameterized Layer and Blockwise Optimization

A novel quantum-classical algorithm boosts QAOA performance with fewer layers, enabling real-world optimization on NISQ devices.

Renewable energy

A Deep Learning-based Approach for Battery Life Classification

A deep learning-based LSTM network accurately classifies battery health, optimizing energy storage and predictive maintenance.

Biomedical

An explainable fast deep neural network for emotion recognition

A fast, explainable deep neural network enhances emotion recognition by optimizing facial landmark analysis.

Renewable energy

Multi-label classification with imbalanced classes by fuzzy deep neural networks

A fuzzy deep neural network accurately classifies household appliances in real time using symbolic data and multi-label AI.

Quantum computing

Quantum enhanced knowledge distillation

Classical-to-quantum knowledge distillation boosts hybrid AI performance using efficient quantum circuits and reduced model sizes.

Quantum computing

A variational approach to quantum gated recurrent units

A faster and efficient Quantum Gated Recurrent Unit (QGRU) improves time series forecasting.

Aerospace

A Neural Network Symbolic Approach to Structural Health Monitoring in Aerospace Applications

A symbolic deep learning approach enhances structural health monitoring in aerospace achieving near-perfect damage classification.

Retrieving Chlorophyll-a Levels, Transparency and TSS Concentration from Multispectral Satellite Data by Using Artificial Neural Networks

Smart Water Monitoring in Remote Areas

Satellite Imaging Across the Hydrological Cycle

Training Neural Networks with Real and Remote Data

Accuracy Across Seasonal Variations

Scalable and Non-Invasive Environmental Surveillance

Contacts

Useful links

Authors

Recommended

Recommended

More articles to read

A Review of the Enabling Methodologies for Knowledge Discovery from Smart Grids Data

Enhancing Autism Detection Through Gaze Analysis Using Eye Tracking Sensors and Data Attribution with Distillation in Deep Neural Networks

Quantum Generative Modeling via Straightforward State Preparation

Enhancing QAOA Ansatz via Multi-Parameterized Layer and Blockwise Optimization

A Deep Learning-based Approach for Battery Life Classification

An explainable fast deep neural network for emotion recognition

Multi-label classification with imbalanced classes by fuzzy deep neural networks

Quantum enhanced knowledge distillation

A variational approach to quantum gated recurrent units

A Neural Network Symbolic Approach to Structural Health Monitoring in Aerospace Applications

Do you have a specific need?

Fill out the form and tell us about your project.
We'll propose the solution that best fits your context.

Do you have a specific need?

Fill out the form and tell us about your project. We'll propose the solution that best fits your context.

Retrieving Chlorophyll-a Levels, Transparency and TSS Concentration from Multispectral Satellite Data by Using Artificial Neural Networks

Smart Water Monitoring in Remote Areas

Satellite Imaging Across the Hydrological Cycle

Training Neural Networks with Real and Remote Data

Accuracy Across Seasonal Variations

Scalable and Non-Invasive Environmental Surveillance

Contacts

Useful links

Authors

Recommended

Recommended

More articles to read

A Review of the Enabling Methodologies for Knowledge Discovery from Smart Grids Data

Enhancing Autism Detection Through Gaze Analysis Using Eye Tracking Sensors and Data Attribution with Distillation in Deep Neural Networks

Quantum Generative Modeling via Straightforward State Preparation

Enhancing QAOA Ansatz via Multi-Parameterized Layer and Blockwise Optimization

A Deep Learning-based Approach for Battery Life Classification

An explainable fast deep neural network for emotion recognition

Multi-label classification with imbalanced classes by fuzzy deep neural networks

Quantum enhanced knowledge distillation

A variational approach to quantum gated recurrent units

A Neural Network Symbolic Approach to Structural Health Monitoring in Aerospace Applications

Do you have a specific need?

Fill out the form and tell us about your project.We'll propose the solution that best fits your context.

Do you have a specific need?

Fill out the form and tell us about your project. We'll propose the solution that best fits your context.

Fill out the form and tell us about your project.
We'll propose the solution that best fits your context.