Investigating Nonlinear Quenching Effects on Polar Field Buildup in the Sun Using Physics-Informed Neural Networks

New Insights into Solar Magnetic Field Dynamics Using Advanced Neural Networks

Recent research has unveiled critical insights into the solar dynamo's mechanisms, focusing on the nonlinear feedback effects of tilt quenching (TQ) and latitude quenching (LQ) on polar field buildup. This study leverages Physics-Informed Neural Networks (PINN) to enhance the understanding of these processes, essential for predicting future solar cycles.

Methodology and Findings

The research involved systematically varying transport parameters to isolate the contributions of TQ and LQ to the buildup of the polar dipole. Key findings indicate that:

• TQ suppression intensifies with increasing diffusivity.
• LQ serves as the dominant factor in advection-dominated scenarios.
• The ratio of contributions from LQ to TQ ($ΔD_{\mathrm{LQ}}/ΔD_{\mathrm{TQ}}$) shows an inverse-square relationship with the dynamo effectivity range, enhancing previous empirical fits with improved precision.

Additionally, the study highlighted that a decay term is not necessary in the PINN setup, due to the training process's efficiency. The comparison between the traditional 1D SFT model and the PINN framework revealed that the latter achieved significantly lower error metrics and a more robust recovery of nonlinear trends.

Implications for Solar Cycle Prediction

These results underscore the complex interactions between LQ and TQ, explaining the observed alternations between weak and strong solar cycles. The findings position PINN as a promising tool in solar physics for predicting solar cycles.