Characterization of Volcanic Cloud Components Using Machine Learning Techniques and SEVIRI Infrared Images

Area of Science:

• Geosciences
• Atmospheric Science
• Remote Sensing

Background:

• Volcanic eruptions release particles and gases, forming hazardous volcanic clouds.
• Volcanic clouds pose risks to human health and air travel.
• Satellite infrared sensors are effective for monitoring volcanic clouds.

Purpose of the Study:

• To develop a machine learning approach for detecting and classifying volcanic clouds using satellite infrared images.
• To automate the monitoring and mapping of volcanic ash clouds.

Main Methods:

• Implemented a supervised support vector machine (SVM) algorithm.
• Utilized thermal infrared (TIR) bands from the geostationary MSG-SEVIRI satellite.
• Applied the algorithm to data from Mt. Etna explosive events (2020-2022).

Main Results:

• Achieved an accuracy of 0.86 in detecting and mapping volcanic ash clouds.
• Demonstrated the efficiency of the ML approach using TIR bands.
• Enabled automatic detection, tracking, and mapping in near real-time.

Conclusions:

• The developed ML approach is highly efficient for volcanic cloud monitoring.
• This method provides accurate and automated detection of volcanic ash.
• The approach supports near real-time hazard assessment and air traffic safety.