Scrutinising the relationship between plage areas and sunspot areas and numbers

Context. Studies and reconstructions of past solar activity require data on all magnetic regions on the surface of the Sun (i.e. on dark sunspots as well as bright faculae-plage and network). Such data are also important for understanding the magnetic activity and variability of the Sun and Sun-like stars. The longest available direct faculae-plage datasets are white-light facular and Ca II K observations going back to 1874 and 1892, respectively. Prior to that time the only direct data available are for sunspots.
Aims: We reassess the relationship between plage areas and sunspot records (areas and numbers) since 1892, to allow reconstructions of facular-plage areas which can be employed for studies going further back in time to the period when solely sunspot observations are available.
Methods: We use the plage areas derived from 38 consistently processed Ca II K archives as well as the plage area composite based on these archives. The considered archives include both the well-known observatories (e.g. Coimbra, Kodaikanal, Meudon, Mt Wilson), and the less explored observatories (e.g. Kharkiv, Mees, and Upice). These data allow us to study the relationship between plage area and sunspot records (areas and number) over a period of 12 solar cycles and for different bandpasses.
Results: We find the relationship between plage and sunspot areas to be well represented by a power-law function; the relationship between the plage areas and the sunspot number is also best fit with a power-law function. We further find that the relationships depend on the bandwidth and the solar cycle strength. The reconstructions with a power-law relationship are in good agreement with the original plage area series, whereas employment of a cycle-strength-dependent relationship improves the reconstructions only marginally. We also estimate the error in the plage areas reconstructed from the sunspot areas or numbers. Performing the same analysis on other previously published plage area series, usually derived from a single archive with diverse processing techniques, returns different results when using different time series. This highlights the importance of applying a consistent processing to the various archives and demonstrates the uncertainties introduced by using previously published series for studies of past solar activity, including irradiance reconstructions.
Conclusions: Our results have implications for past solar activity and irradiance reconstructions and for stellar activity studies, which sometimes assume a linear dependence between plage and sunspot areas.