Probabilistic detection of volcanic ash using a Bayesian approach.

Area of Science:

• Remote Sensing
• Geophysics
• Atmospheric Science

Background:

• Airborne volcanic ash poses significant risks to aviation, agriculture, and health.
• Monitoring ash clouds is crucial, especially using infrared satellite data, due to limitations of existing methods.
• Current techniques often lack certainty quantification and rely on subjective thresholds, leading to potential misidentification, e.g., with desert dust.

Purpose of the Study:

• To develop and present a novel probabilistic method for volcanic ash detection.
• To exploit contemporaneous atmospheric data for more accurate and physically based ash classification.
• To overcome limitations of existing binary classification methods and subjective thresholding.

Main Methods:

• Implementation of a Bayesian approach for ash detection.
• Calculation of probability density functions for ash observations.
• Demonstration using satellite data including land, sea, and desert dust scenarios.

Main Results:

• The proposed Bayesian method provides a probabilistic, physically based classification of volcanic ash.
• The technique effectively discriminates ash from other atmospheric phenomena like desert dust.
• The method offers a quantifiable measure of classification certainty for individual pixels.

Conclusions:

• The developed Bayesian scheme offers a robust and objective approach to volcanic ash detection.
• This remote sensing technique enhances the monitoring of volcanic ash hazards.
• The method's probabilistic output provides valuable insights into classification certainty.