What question did this study set out to answer?

This research aims to improve forecasting accuracy of the Interplanetary Magnetic Field Bz.

February 19, 2026Solar Physics0 citations

Investigating a Multi-Modal Attention-Based Deep-Learning Framework for Long-Term IMF Bₙ Forecasting Based on CME Kinematics and Solar Magnetogram Data

TDTiar DaniUniversitas Gadjah Mada EWEdi WinarkoUniversitas Gadjah Mada LHLukman HeryawanUniversitas Gadjah Mada

Key Points

This research aims to improve forecasting accuracy of the Interplanetary Magnetic Field Bz.
Investigated a multi-modal deep learning framework for CME event-driven IMF Bz forecasting.
Developed the Bz 4SWx model using CME kinematic parameters and solar magnetogram data.
Compared performance against baseline models including a uni-modal MLP and CNN.
Bz 4SWx model achieved an MAE of 3.270 nT and RMSE of 4.124 nT.
The model demonstrated competitive timing precision compared to other multi-modal approaches.
Interpretability analysis confirmed magnetic signals as dominant predictive factors.

Abstract

Forecasting the Interplanetary Magnetic Field (IMF) \ (Bₙ\) component is a critical and persistent challenge in space weather, often termed the “ \ (Bₙ\) problem”. As Coronal Mass Ejections (CMEs) are the primary drivers of strong and sustained southward \ (Bₙ\) events, this study investigates a multi-modal deep-learning framework for CME event-driven IMF \ (Bₙ\) forecasts at 12-hour intervals up to a 96-hour lead time. We propose a novel attention-based architecture, \ (Bₙ\) 4SWx, to fuse CME kinematic parameters with their associated solar magnetograms. This model employs a dual-branch network enhanced by a Convolutional Block Attention Module (CBAM). To evaluate its effectiveness, we compared its performance against several baseline models, including a uni-modal MLP (numerics), a uni-modal CNN (images), and a naive concatenation-based fusion model. The \ (Bₙ\) 4SWx model achieved the best overall performance, yielding an MAE of 3. 270 nT, an RMSE of 4. 124 nT, and a bias of −2. 61 nT, with timing precision competitive with other multi-modal approaches. Interpretability analysis confirmed that while magnetograms provide the dominant predictive signal, CBAM was critical for dynamically focusing the model on relevant solar active regions. We conclude that attention mechanisms provide a powerful and interpretable framework for CME event-based IMF \ (Bₙ\) forecasting, representing a significant step toward resolving the persistent \ (Bₙ\) problem.

Ask AI

Helpful

Bookmark

View Full Paper