Reconstructing solar magnetic fields from historical observations

IV. Testing the reconstruction method

¹ ReSoLVE Centre of Excellence, Space Climate Research Unit, University of Oulu, PO Box 3000, 90014 Oulu, Finland
e-mail: iiro.virtanen@oulu.fi
² National Solar Observatory, Boulder, CO 80303, USA
³ Pulkovo Astronomical Observatory, Russian Academy of Sciences, Pulkovskoye Shosse 65, Saint Petersburg 196140, Russian Federation
⁴ Department of Mathematical Sciences, Durham University, Durham DH1 3LE, UK

Received: 3 April 2019
Accepted: 28 May 2019

Abstract

Aims. The evolution of the photospheric magnetic field has only been regularly observed since the 1970s. The absence of earlier observations severely limits our ability to understand the long-term evolution of solar magnetic fields, especially the polar fields that are important drivers of space weather. Here, we test the possibility to reconstruct the large-scale solar magnetic fields from Ca II K line observations and sunspot magnetic field observations, and to create synoptic maps of the photospheric magnetic field for times before modern-time magnetographic observations.

Methods. We reconstructed active regions from Ca II K line synoptic maps and assigned them magnetic polarities using sunspot magnetic field observations. We used the reconstructed active regions as input in a surface flux transport simulation to produce synoptic maps of the photospheric magnetic field. We compared the simulated field with the observed field in 1975−1985 in order to test and validate our method.

Results. The reconstruction very accurately reproduces the long-term evolution of the large-scale field, including the poleward flux surges and the strength of polar fields. The reconstruction has slightly less emerging flux because a few weak active regions are missing, but it includes the large active regions that are the most important for the large-scale evolution of the field. Although our reconstruction method is very robust, individual reconstructed active regions may be slightly inaccurate in terms of area, total flux, or polarity, which leads to some uncertainty in the simulation. However, due to the randomness of these inaccuracies and the lack of long-term memory in the simulation, these problems do not significantly affect the long-term evolution of the large-scale field.