Simulating and Predicting Solar Cycles Using a Flux‐Transport Dynamo

Abstract

We construct a predictive tool based on a Babcock-Leighton-type flux-transport dynamo model of a solar cycle, run the model by updating the surface magnetic source using old cycles’ data since cycle 12, and show that the model can correctly simulate the relative peaks of cycles 16-23. The simulations use the first four cycles to load the meridional circulation conveyor belt to create the Sun’s memory about its past magnetic fields. Extending the simulation into the future, we predict that cycle 24 will be 30%-50% stronger than the current cycle 23. These simulations and predictions are robust for a wide range of convection zone magnetic diffusivity values between 3*10^10 and 2*10^11 cm2 s-1. Our model predictions are the same for three different treatments of the unknown surface magnetic source for the cycles to be predicted, namely (1) assuming some cyclic pattern, (2) incorporating ‘‘zero’’ surface source, or (3) including a surface source derived from the self-excited version of the dynamo model. Technique 3, for treating the surface source for cycles to be predicted, also shows significant skill in predicting two cycles ahead. Analyzing the evolution of magnetic field patterns over a full magnetic cycle, we show that the key to success of our prediction model lies in the formation of a ‘‘seed’’ for producing cycle n from the combination of latitudinal fields at high latitudes from three past cycles, n - 1, n - 2, and n - 3, instead of the previous cycle’s polar fields. These re- sults have many implications for both solar and stellar dynamo modeling.