Saved in:
Bibliographic Details
Main Authors: Chatterjee, Snehamoy, Waite, Greg, Paheding, Sidike, Bowman, Luke
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.21803
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908288618070016
author Chatterjee, Snehamoy
Waite, Greg
Paheding, Sidike
Bowman, Luke
author_facet Chatterjee, Snehamoy
Waite, Greg
Paheding, Sidike
Bowman, Luke
contents Forecasting volcanic activity is critical for hazard assessment and risk mitigation. Volcanic Radiative Power (VPR), derived from thermal remote sensing data, serves as an essential indicator of volcanic activity. In this study, we employ Bayesian Regularized Neural Networks (BRNN) to predict future VPR values based on historical data from Fuego Volcano, comparing its performance against Scaled Conjugate Gradient (SCG) and Levenberg-Marquardt (LM) models. The results indicate that BRNN outperforms SCG and LM, achieving the lowest mean squared error (1.77E+16) and the highest R-squared value (0.50), demonstrating its superior ability to capture VPR variability while minimizing overfitting. Despite these promising results, challenges remain in improving the model's predictive accuracy. Future research should focus on integrating additional geophysical parameters, such as seismic and gas emission data, to enhance forecasting precision. The findings highlight the potential of machine learning models, particularly BRNN, in advancing volcanic activity forecasting, contributing to more effective early warning systems for volcanic hazards.
format Preprint
id arxiv_https___arxiv_org_abs_2503_21803
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Forecasting Volcanic Radiative Power (VPR) at Fuego Volcano Using Bayesian Regularized Neural Network
Chatterjee, Snehamoy
Waite, Greg
Paheding, Sidike
Bowman, Luke
Machine Learning
Artificial Intelligence
Signal Processing
Atmospheric and Oceanic Physics
Forecasting volcanic activity is critical for hazard assessment and risk mitigation. Volcanic Radiative Power (VPR), derived from thermal remote sensing data, serves as an essential indicator of volcanic activity. In this study, we employ Bayesian Regularized Neural Networks (BRNN) to predict future VPR values based on historical data from Fuego Volcano, comparing its performance against Scaled Conjugate Gradient (SCG) and Levenberg-Marquardt (LM) models. The results indicate that BRNN outperforms SCG and LM, achieving the lowest mean squared error (1.77E+16) and the highest R-squared value (0.50), demonstrating its superior ability to capture VPR variability while minimizing overfitting. Despite these promising results, challenges remain in improving the model's predictive accuracy. Future research should focus on integrating additional geophysical parameters, such as seismic and gas emission data, to enhance forecasting precision. The findings highlight the potential of machine learning models, particularly BRNN, in advancing volcanic activity forecasting, contributing to more effective early warning systems for volcanic hazards.
title Forecasting Volcanic Radiative Power (VPR) at Fuego Volcano Using Bayesian Regularized Neural Network
topic Machine Learning
Artificial Intelligence
Signal Processing
Atmospheric and Oceanic Physics
url https://arxiv.org/abs/2503.21803