On the importance of stratospheric ozone feedbacks for the simulation of solar effects on climate

An accurate assessment of the role of solar variability is a key step towards a proper quantification of natural and anthropogenic climate change. To this end, climate models and reconstructions spanning over the last century have been extensively used to quantify the solar contribution to climate variability. However, the role of natural variability in the detection of solar signals has often been overlooked, due to limited extension of observational records and model simulations. Moreover, owing to its large computational cost, the bulk of modeling studies to date have been performed without interactive stratospheric photochemistry: the impact of this simplification on the modeled climate system response to solar forcing remains largely unknown. Here we quantify this impact, by comparing the response of two model configurations, with and without interactive ozone chemistry. Using fully coupled 300-year long integrations, we first obtain robust surface temperature and precipitation responses to an idealized irradiance increase. Then, we show that the inclusion of interactive stratospheric chemistry significantly reduces the surface warming (by about one third) and the accompanying precipitation response. This behavior is linked to photochemically-induced stratospheric ozone changes, and their modulation of the surface solar radiation. Our results suggest that neglecting stratospheric photochemistry leads to a sizable overestimate of the surface response to changes in solar irradiance. Moreover, longer records than presently available are needed for a robust detection of solar signals in surface climate. This has implications for simulations of the climate in the Last Millennium, where models often use simplified treatments of stratospheric ozone that are inconsistent with the imposed solar forcing, and intrinsic (unforced) variability is not sufficiently accounted for.