Comment on wcd-2023-9

Abstract

<strong class="journal-contentHeaderColor">Abstract.</strong> Studies concerning solar-terrestrial connections over the last decades claim to have found evidence that the quasi-decadal solar cycle can have an influence on the dynamics in the middle atmosphere in the Northern Hemisphere during winter season. It has been argued that feedbacks between the intensity of the UV part of the solar spectrum and low latitude stratospheric ozone may produce anomalies in meridional temperature gradients which have the potential to alter the zonal mean flow in mid to high latitudes. Interactions between the zonal wind and planetary waves can lead to a downward propagation of the anomalies, produced in the middle atmosphere, down to the troposphere. More recently it has been proposed that the projection of possible decadal surface solar signals on the North Atlantic Oscillation might lead to a synchronization of the latter via the 11-year solar cycle. Furthermore, it has been claimed that a realistic representation of the solar cycle in climate models may lead to a significant increase of the decadal prediction skill. These conclusions have been debated controversial since then and a confirmation from other modelling groups is missing. In this paper we aim for an unbiased evaluation of possible solar imprints from the middle atmosphere to the surface and with that from head to toe. Thus, we analyze model output from historical ensemble simulations conducted with the state-of-the-art Earth system model MPI-ESM-HR. The target of these simulations was to isolate the most crucial model physics to foster basic research on decadal climate prediction and to develop an operational ensemble decadal prediction system within the MiKlip framework. Based on correlations and multiple linear regression analysis we show that the MPI-ESM-HR simulates a realistic, statistically significant and robust shortwave heating rate and temperature response at the tropical stratopause, which is known from existing studies. However, the dynamical response to this initial radiative signal in the NH during the boreal winter season is rather weak. In this context we find a slight strengthening of the polar vortex in midwinter season during solar maximum conditions in the ensemble mean, which basically agrees with the so-called &ldquo;top-down&rdquo; mechanism. The individual ensemble members, however, show a large spread in the dynamical response with opposite sign in response to the solar cycle. We also analyze the possible surface responses to the 11-year solar cycle and review the proposed synchronization between the solar forcing and the North Atlantic Oscillation. We find that the westerly wind anomalies in the lower troposphere as well as the anomalies in the mean sea level pressure are most likely independent from the seasonal march in the middle atmosphere since they mimic positive and negative phases of the Arctic- and North Atlantic Oscillation rather sporadically than in a systematic way. Finally, by applying lead/lag correlations, our results indicate that the proposed synchronization between the solar cycle and the decadal component of the North Atlantic Oscillation might rather be interpreted as a statistical artefact than a plausible physical connection between the solar forcing and quasi-decadal variations in the troposphere.