Sensitivity of atmospheric rivers to aerosol treatment in regional climate simulations: insights from the AIRA identification algorithm

Abstract

Abstract. This study analyzed the sensitivity of atmospheric rivers (ARs) to aerosol treatment in regional climate simulations. Three experiments covering the Iberian Peninsula for the period from 1991 to 2010 were examined: (1) an experiment including prescribed aerosols (BASE); (2) an experiment including direct and semi-direct aerosol effects (ARI); and (3) an experiment including direct, semi-direct, and indirect aerosol effects (ARCI). A new regional-scale AR identification algorithm, AIRA, was developed and used to identify around 250 ARs in each experiment. The results showed that spring and autumn ARs were the most frequent, intense, and long-lasting and that ARs could explain up to 30 % of the total accumulated precipitation. The inclusion of aerosols was found to redistribute precipitation, with increases in the areas of AR occurrence. The analysis of common AR events showed that the differences between simulations were minimal in the most intense cases and that a negative correlation existed between mean direction and mean latitude differences. This implies that more zonal ARs in ARI or ARCI with respect to BASE could also be linked to northward deviations. The joint analysis and classification of dust and sea salt aerosol distributions allowed for the common events to be clustered into eight main aerosol configurations in ARI and ARCI. The sensitivity of ARs to different aerosol treatments was observed to be relevant, inducing spatial deviations and integrated water vapor transport (IVT) magnitude reinforcements/attenuations with respect to the BASE simulation depending on the aerosol configuration. Thus, the correct inclusion of aerosol effects is important for the simulation of AR behavior at both global and regional scales, which is essential for meteorological predictions and climate change projections.