LAUSR

Abstract. Future sea ice retreat in the Arctic in summer and autumn is expected to affect both natural and anthropogenic aerosol emissions: sea ice acts as a barrier between the ocean and the atmosphere, and reducing it increases dimethyl sulfide and sea salt emissions. Additionally, a decrease in the area and thickness of sea ice could lead to enhanced Arctic ship traffic, for example due to shorter routes of cargo ships. Changes in the emissions of aerosol particles can then influence cloud properties, precipitation, surface albedo, and radiation. Next to changes in aerosol emissions, clouds will also be affected by increases in Arctic temperatures and humidities. In this study, we quantify how future aerosol radiative forcings and cloud radiative effects might change in the Arctic in late summer (July–August) and early autumn (September–October). Simulations were conducted for the years 2004 and 2050 with the global aerosol–climate model ECHAM6-HAM2. For 2050, simulations with and without additional ship emissions in the Arctic were carried out to quantify the impact of these emissions on the Arctic climate. In the future, sea salt as well as dimethyl sulfide emissions and burdens will increase in the Arctic. The increase in cloud condensation nuclei, which is due to changes in aerosol particles and meteorology, will enhance cloud droplet number concentrations over the Arctic Ocean ( +10  % in late summer and +29  % in early autumn; in-cloud values averaged between 75 and 90∘  N). Furthermore, both liquid and total water path will increase ( +10  % and +8  % in late summer; +34  % and +26  % in early autumn) since the specific humidity will be enhanced due to higher temperatures and the exposure of the ocean's surface. Changes in both aerosol radiative forcings and cloud radiative effects at the top of the atmosphere will not be dominated by the aerosol particles and clouds themselves but by the decrease in surface albedo (and by the increase in surface temperature for the longwave cloud radiative effect in early autumn). Mainly due to the reduction in sea ice, the aerosol radiative forcing will become less positive (decreasing from 0.53 to 0.36 W m −2 in late summer and from 0.15 to 0.11 W m −2 in early autumn). The decrease in sea ice is also mainly responsible for changes in the net cloud radiative effect, which will become more negative in late summer (changing from −36 to −46  W m −2 ). Therefore, the cooling component of both aerosols and clouds will gain importance in the future. We found that future Arctic ship emissions related to transport and oil and gas extraction ( Peters et al. ,  2011 ) will not have a large impact on clouds and radiation: changes in aerosols only become significant when we increase these ship emissions by a factor of 10. However, even with 10-fold ship emissions, the net aerosol radiative forcing shows no significant changes. Enhanced black carbon deposition on snow leads to a locally significant but very small increase in radiative forcing over the central Arctic Ocean in early autumn (no significant increase for average between 75 and 90∘  N). Furthermore, the 10-fold higher ship emissions increase the optical thickness and lifetime of clouds in late summer (net cloud radiative effect changing from −48 to −52  W m −2 ). These aerosol–cloud effects have a considerably larger influence on the radiative forcing than the direct effects of particles (both aerosol particles in the atmosphere and particles deposited on snow). In summary, future ship emissions of aerosols and their precursor gases might have a net cooling effect, which is small compared to other changes in future Arctic climate such as those caused by the decrease in surface albedo.