Forecasting Continuum Intensity for Solar Active Region Emergence Prediction using Transformers

TL;DR

This paper introduces a Transformer-based model with an early detection architecture for predicting solar active region emergence, achieving earlier warnings and improved accuracy over traditional LSTM models, despite increased variance.

Contribution

The study develops a novel Transformer architecture with attention biases and a timing-aware loss for AR emergence prediction, outperforming LSTM baselines in early warning capability.

Findings

Transformer with early detection achieves 10.6% lower RMSE than LSTM.

Average warning time improved to 4.73 hours.

Model detects micro-changes, enabling earlier alerts despite higher variance.

Abstract

Early and accurate prediction of solar active region (AR) emergence is crucial for space weather forecasting. Building on established Long Short-Term Memory (LSTM) based approaches for forecasting the continuum intensity decrease associated with AR emergence, this work expands the modeling with new architectures and targets. We investigate a sliding-window Transformer architecture to forecast continuum intensity evolution up to 12 hours ahead using data from 46 ARs observed by SDO/HMI. We conduct a systematic ablation study to evaluate two key components: (1) the inclusion of a temporal 1D convolutional (Conv1D) front-end and (2) a novel `Early Detection' architecture featuring attention biases and a timing-aware loss function. Our best-performing model, combining the Early Detection architecture without the Conv1D layer, achieved a Root Mean Square Error (RMSE) of 0.1189 (representing a 10.6% improvement over the LSTM baseline) and an average advance warning time of 4.73 hours (timing difference of -4.73h), even under a stricter emergence criterion than previous studies. While the Transformer demonstrates superior aggregate timing and accuracy, we note that this high-sensitivity detection comes with increased variance compared to smoother baseline models. However, this volatility is a necessary trade-off for operational warning systems: the model's ability to detect micro-changes in precursor signals enables significantly earlier detection, outweighing the cost of increased noise. Our results demonstrate that Transformer architectures modified with early detection biases, when used without temporal smoothing layers, provide a high-sensitivity alternative for forecasting AR emergence that prioritizes advance warning over statistical smoothness.