Artificial Intelligence Reconstructs Missing Sunspots and Forecasts a New Solar Minimum?

The retrodiction and prediction of solar activity are two closely-related problems in dynamo theory. We applied Machine Learning (ML) algorithms and analyses to the World Data Center’s newly constructed annual sunspot time series (1700-2019). This provides a unique model that gives insights into the various patterns of the Sun’s magnetic dynamo that drives solar activity maxima and minima. We found that the variability in the ∼ 11-year Sunspot Cycle is closely connected with 120-year oscillatory magnetic activity variations. We also identified a previously unknown/unreported feature in the sunspot record with a 5.5 year period. This 5.5-year pattern is also co-modulated by the 120-year oscillation, and also appears to influence the shape and energy/power content of individual 11-year cycles. Our ML algorithm was trained to recognize such underlying patterns. Our ML model provides a convincing hindcast of the full sunspot record from 1700-2019. It also suggests the possibility of missing sunspots during Sunspot Cycles -1, 0, and 1 (ca. 1730s-1760s). In addition, our ML model forecasts a new phase of extended solar minima, that started prior to Sunspot Cycle 24 (ca. 2008-2019), will persist until Sunspot Cycle 27 (2050 or so). Our ML model forecasts a peak annual sunspot number (SSN) of 95 with a probable range of 80 to 115 for Cycle 25 at around 2024 ± 1.