The mean solar butterfly diagram and poloidal field generation rate at the surface of the Sun

TL;DR

This study derives the mean solar butterfly diagram and surface poloidal field generation rate by averaging observational data across multiple cycles, providing insights compatible with mean-field dynamo models.

Contribution

It introduces a method to extract the expectation value of the butterfly diagram from observations, reducing noise and artifacts, and links it to the surface poloidal field generation rate.

Findings

Averaging removes realization noise and artifacts.

The flux emergence matches surface toroidal field and Joy's law.

Results support the Babcock-Leighton $ ext{α}$-effect in dynamo models.

Abstract

The difference between individual solar cycles in the magnetic butterfly diagram can mostly be ascribed to the stochasticity of the emergence process. We aim to obtain the expectation value of the butterfly diagram from observations of four cycles. This allows us to further determine the generation rate of the surface radial magnetic field. We use data from Wilcox Solar Observatory to generate time-latitude diagrams spanning cycles 21 to 24 of the surface radial and toroidal magnetic fields, symmetrize them across the equator and cycle-average them. From the mean butterfly diagram and surface toroidal field we then infer the mean poloidal field generation rate at the surface of the Sun. The averaging procedure removes realization noise from individual cycles. The amount of emerging flux required to account for the evolution of the surface radial field is found to match that provided by the observed surface toroidal field and Joy's law. Cycle-averaging butterfly diagrams removes realization noise and artefacts due to imperfect scale separation, and corresponds to an ensemble average that can be interpreted in the mean-field framework. The result can then be directly compared to $\alpha\Omega$-type dynamo models. The Babcock-Leighton $\alpha$-effect is consistent with observations, a result that can be appreciated only if the observational data is averaged in some way.