Towards an algebraic method of solar cycle prediction II. Reducing the need for detailed input data with ARDoR

TL;DR

This paper introduces ARDoR, a new active region descriptor, to improve solar cycle prediction by focusing on a few large 'rogue' regions, reducing data requirements and enhancing prediction accuracy.

Contribution

It proposes ARDoR and demonstrates that considering only the top few active regions by ARDoR suffices for accurate solar dipole moment prediction, simplifying data needs.

Findings

Ranking active regions by ARDoR improves dipole moment prediction accuracy.

Considering only the top 5 regions reduces prediction errors significantly.

Stochastic effects are mainly driven by a few large active regions.

Abstract

An algebraic method for the reconstruction and potentially prediction of the solar dipole moment value at sunspot minimum (known to be a good predictor of the amplitude of the next solar cycle) was suggested in the first paper in this series. The method sums up the ultimate dipole moment contributions of individual active regions in a solar cycle: for this, detailed and reliable input data would in principle be needed for thousands of active regions in a solar cycle. To reduce the need for detailed input data, here we propose a new active region descriptor called ARDoR (Active Region Degree of Rogueness). In a detailed statistical analysis of a large number of activity cycles simulated with the 2$\times$2D dynamo model we demonstrate that ranking active regions by decreasing ARDoR, for a good reproduction of the solar dipole moment at the end of the cycle it is sufficient to consider the top $N$ regions on this list explicitly, where $N$ is a relatively low number, while for the other regions the ARDoR value may be set to zero. E.g., with $N=5$ the fraction of cycles where the dipole moment is reproduced with an error exceeding $\pm$30% is only 12%, significantly reduced with respect to the case $N=0$, i.e. ARDoR set to zero for all active regions, where this fraction is 26%. This indicates that stochastic effects on the intercycle variations of solar activity are dominated by the effect of a low number of large ``rogue'' active regions, rather than the combined effect of numerous small ARs. The method has a potential for future use in solar cycle prediction.