Abstract
We investigate the effects of solar forcing during summer on the North Atlantic climate in comprehensive simulations of the preindustrial last millennium. We use two Earth System Models forced only by variations in Total Solar Irradiance (TSI). Specifically, we examine how different statistical techniques commonly used in current literature, namely linear methods and composite techniques can condition our understanding of the effects of solar forcing on climate. We demonstrate that the results obtained are strongly shaped by internal model variability. Linear methods like regression and correlation are not suitable to separate solar impacts on summer climate from internal variability. Composite maps show a response of SSTs off the European coasts and atmospheric blocking-like pressure anomalies over the subpolar North Atlantic, with some model-dependent variations of its spatial patterns and extent. In the models analyzed, the relationship of TSI to the tropospheric and surface circulation is linked through a baroclinic response to diabatic heating at the ocean surface. A tendency toward blocking-like patterns over the middle and high latitudes might be subsequently created during summer and in high TSI periods.