A New Tool for CME Arrival Time Prediction Using Machine Learning Algorithms: CAT-PUMA

Abstract

Coronal Mass Ejections (CMEs) are arguably the most violent eruptions in the Solar System. CMEs can cause severe disturbances in the interplanetary space and even affect human activities in many respects, causing damages to infrastructure and losses of revenue. Fast and accurate prediction of CME arrival time is then vital to minimize the disruption CMEs may cause when interacting with geospace. In this paper, we propose a new approach for partial-/full-halo CME Arrival Time Prediction Using Machine learning Algorithms (CAT-PUMA). Via detailed analysis of the CME features and solar wind parameters, we build a prediction engine taking advantage of 182 previously observed geo-effective partial-/full-halo CMEs and using algorithms of the Support Vector Machine (SVM). We demonstrate that CAT-PUMA is accurate and fast. In particular, predictions after applying CAT-PUMA to a test set, that is unknown to the engine, show a mean absolute prediction error $\sim$5.9 hours of the CME arrival time, with 54% of the predictions having absolute errors less than 5.9 hours. Comparison with other models reveals that CAT-PUMA has a more accurate prediction for 77% of the events investigated; and can be carried out very fast, i.e. within minutes after providing the necessary input parameters of a CME. A practical guide containing the CAT-PUMA engine and the source code of two examples are available in the Appendix, allowing the community to perform their own applications for prediction using CAT-PUMA.

arXiv comment13 pages, 7 figures, 2 tables; accepted by the Astrophysical Journal

References in corpus (25)

• A Model for Solar Coronal Mass Ejections
• Solar Flare Prediction Using SDO/HMI Vector Magnetic Field Data with a Machine-Learning Algorithm
• Observation of An Evolving Magnetic Flux Rope Prior To and During A Solar Eruption
• Automated LASCO CME catalog for solar cycle 23: are CMEs scale invariant?
• The Interaction of Successive Coronal Mass Ejections: A Review
• On a Coronal Blowout Jet: The First Observation of a Simultaneously Produced Bubble-like CME and a Jet-like CME in a Solar Event
• Solar Flare Prediction Model with Three Machine-Learning Algorithms Using Ultraviolet Brightening and Vector Magnetogram
• Connecting speeds, directions and arrival times of 22 coronal mass ejections from the Sun to 1 AU
• History and Development of Coronal Mass Ejections as a Key Player in Solar Terrestrial Relationship
• Sigmoid-to-Flux-Rope Transition Leading to A Loop-Like Coronal Mass Ejection
• First SDO AIA Observations of a Global Coronal EUV "Wave": Multiple Components and "Ripples"
• Quantitative Analysis of CME Deflections in the Corona
• Predicting Coronal Mass Ejections Using Machine Learning Methods
• Three-dimensional evolution of flux rope CMEs and its relation to the local orientation of the heliospheric current sheet
• Global Trends of CME Deflections Based on CME and Solar Parameters
• Satistical Study of the Interplanetary Coronal Mass Ejections from 1996 to 2014
• Super-elastic Collision of Large-scale Magnetized Plasmoids in The Heliosphere
• A solar type II radio burst from CME-coronal ray interaction: simultaneous radio and EUV imaging
• Can solar wind viscous drag account for CME deceleration?
• Full halo coronal mass ejections: Do we need to correct the projection effect in terms of velocity?
• Could the collision of CMEs in the heliosphere be super-elastic? --- Validation through three-dimensional simulations
• A Solar Coronal Jet Event Triggers A Coronal Mass Ejection
• Source Region of the Decameter - Hectometric Type II Radio Burst: Shock - Streamer Interaction Region
• On Understanding the Nature of Collision of Coronal Mass Ejections Observed by \textit{STEREO}
• Solar jet-coronal hole collision and a related coronal mass ejection

Cited by in corpus (11)

• The Challenge of Machine Learning in Space Weather Nowcasting and Forecasting
• Surveying the reach and maturity of machine learning and artificial intelligence in astronomy
• Fast Solar Image Classification Using Deep Learning and its Importance for Automation in Solar Physics
• 2D solar wind speeds from 6 to 26 solar radii in solar cycle 24 by using Fourier filtering
• Modeling Interplanetary Expansion and Deformation of CMEs with ANTEATR-PARADE I: Relative Contribution of Different Forces
• Propagating Conditions and the Time of ICMEs Arrival: A Comparison of the Effective Acceleration Model with ENLIL and DBEM Models
• Timing of the solar wind propagation delay between L1 and Earth based on machine learning
• CAMEL. II. A 3D Coronal Mass Ejection Catalog Based on Coronal Mass Ejection Automatic Detection with Deep Learning
• Observation and modeling of complex transient structure in heliosphere followed by geomagnetic storm on May 10-11, 2024
• Automatic detection of large-scale flux ropes and their geoeffectiveness with a machine learning approach
• A catalogue of observed geo-effective CME/ICME characteristics