Nonlinear mean-field dynamo and prediction of solar activity

TL;DR

This paper develops a nonlinear mean-field dynamo model incorporating magnetic helicity conservation and sunspot formation mechanisms, combined with neural networks, to predict solar activity with good accuracy on a monthly scale.

Contribution

It introduces a novel prediction method combining a nonlinear dynamo model with neural networks, accounting for magnetic helicity and sunspot formation.

Findings

Predicted solar activity aligns well with observed Wolf numbers.

The model effectively captures the nonlinear dynamics of the solar magnetic field.

Monthly predictions show good agreement with actual solar activity data.

Abstract

We apply a nonlinear mean-field dynamo model which includes a budget equation for the dynamics of Wolf numbers to predict solar activity. This dynamo model takes into account the algebraic and dynamic nonlinearities of the alpha effect, where the equation for the dynamic nonlinearity is derived from the conservation law for the magnetic helicity. The budget equation for the evolution of the Wolf number is based on a formation mechanism of sunspots related to the negative effective magnetic pressure instability. This instability redistributes the magnetic flux produced by the mean-field dynamo. To predict solar activity on the time scale of one month we use a method based on a combination of the numerical solution of the nonlinear mean-field dynamo equations and the artificial neural network. A comparison of the results of the prediction of the solar activity with the observed Wolf numbers demonstrates a good agreement between the forecast and observations.