Variability of the Sun's luminosity places constraints on the thermal equilibrium of the convection zone
Luminosity, which is the total amount of radiant energy emitted by an object, is one of the most critical quantities in astrophysics for characterizing stars. Equally important is the temporal evolution of a star's luminosity because of its intimate connection with the stellar energy budget, large-scale convective motion, and heat storage in the stellar interior. Here, we model the solar luminosity by extending a semi-empirical total solar irradiance (TSI) model that uses solar-surface magnetism to reconstruct solar irradiance over the entire 4π solid angle around the Sun. This model was constrained by comparing its output to the irradiance in the Earth's direction with the measured TSI. Comparing the solar luminosity to the TSI on timescales from days to for cycles 23 and 24, we find poor agreement on short timescales (< solar rotation). On longer timescales, however, we find good agreement between the luminosity model and the TSI, which suggests that the extrapolation of luminosities to multi-cycle timescales based on TSI reconstructions may be possible. We show that the solar luminosity is not constant but varies in phase with the solar cycle. This variation has an amplitude of 0.14% from minimum to maximum for solar cycle 23. Considering the energetics in the solar convection zone, it is therefore obvious that a steady-state input from the radiative zone at the solar minimum level would lead to a gradual reduction in the energy content in the convection zone over multi-century timescales. We show that the luminosity at the base of the convection zone should be approximately 0.032% higher than that at the solar surface during solar minimum to maintain net energy equilibrium through the solar cycle. These different energy-input scenarios place constraints on the long-term evolution of the total solar irradiance and its impact on the solar forcing of climate variability.