Shedding light on the Sun-climate connection: evidence of a deeper Maunder Minimum from Bayesian modeling of satellite and proxy observations

Abstract

This dissertation investigates solar variability and its role in climate changes, addressing outstanding questions through four central research chapters. The first chapter re-examines global temperature responses to solar cycles, specifically whether disagreement exists between model predictions and observations of temperature response. The second chapter introduces a Bayesian hierarchical model for total solar irradiance (BTSI) to correct satellite biases and detect significant trends in total solar irradiance (TSI) over the past 40 years. The third chapter applies this model to the full satellite record, revealing a statistically significant decrease in TSI and implications for long-term TSI variability. The fourth chapter explores the drivers of TSI variations on long timescales, focusing on the roles of surface magnetic activity and the open magnetic field. We find that sunspots and facular indices alone are insufficient for capturing TSI variations, and propose a nonlinear empirical model incorporating open solar flux, sunspots, and facular indices. This model estimates greater solar variability than currently-used TSI reconstructions. Lastly, the dissertation assesses the ability of the BTSI TSI reconstruction to explain pre-industrial climate records with external radiative forcings, including solar forcing. Our results, which indicate substantial changes in solar variability since the Maunder Minimum, are consistent with studies from the early 2000s that observed agreement between radiative forcing and temperature changes over the past millennium when using solar forcing with large Maunder Minimum TSI variations. In conclusion, we highlight the need for further research to improve understanding of pre-industrial climate variability and thus better constrain climate sensitivity through paleoclimate observations.