Traditionally, most studies of Arctic air pollution concentrated on the winter/spring period when pollution levels generally are highest, producing the so-called Arctic Haze. In summer, deposition is more efficient, transport is slower and, thus, pollution levels are lower – therefore, less attention was paid to this season. However, since more efficient deposition is an important reason why concentrations are lower in summer than in winter, this also means that the deposition load itself is relatively less reduced from winter to summer. For the radiative forcing of the Arctic climate, it is also important that in winter solar radiation is largely absent in the Arctic whereas in summer the sun is constantly above the horizon. Furthermore, highly reflective stratus cloud decks, above which absorption is enhanced, are frequent in summer.

BC is a particularly important agent for the radiative forcing of the Arctic. It absorbs radiation both in the atmosphere, as well as after deposition on snow/ice surfaces and might be of similar importance for the forcing of the Arctic climate as greenhouse gases (Hansen and Nazarenko, 2004). Stohl (2006) recently suggested that in summer BC from boreal forest fires, especially fires in Siberia, has a larger potential of being transported into the Arctic than BC from anthropogenic fossil fuel combustion. In a study of the boreal forest fires burning in Alaska and Canada in the summer of 2004, Stohl et al. (2006) indeed found that the concentrations of light aborbing aerosols and aerosol optical depths (AOD) at Barrow, Alert, Summit and Svalbard were strongly enhanced in fire plumes. At Barrow spectacularly high values were measured, higher than the largest values recorded during Arctic Haze conditions. At Summit, all AOD values over a 2-month period were above the normal background, due to continuous passages of forest fire plumes. It is also of interest that most of the Barrow data during fire episodes were removed from the official record by data screening algorithms actually designed for eliminating periods with local contamination. If that has occurred also during other years, the long-term Barrow data set may be significantly biased. 7

Here it is suggested to analyze aerosol light absorption, aerosol scattering, AOD, aerosol size distribution, levoglucosan (see WP 2) and carbon monoxide concentrations and other data from as many Arctic stations as possible and for several years. This will be done in a collaborative effort with the data producers, as in the Stohl et al. (2006) study and will include this time an additional analysis for selected years with the EMEP chemical transport model that can provide complementary information on the different aerosol components (see WP 7). Through a collaboration with Canadian colleagues (D. Lavoue, S. Sharma) we will get access to high-quality forest fire BC emission data with a resolution of better than 0.5º, hourly time resolution, and estimates of the fire energy and injection heights into the atmosphere, allowing FLEXPART forest fire plume transport simulations of unprecedented quality. Siberian forest fires emit even more BC than North American ones. Thus, it will also be important to investigate years with strong burning in Siberia (e.g., 2003). It is anticipated that this study may lead to a re-thinking of pollution and aerosol sources for the Arctic in summer. In another collaborative effort, carbon monoxide, aerosol light absorption and albedo data from Summit shall be analyzed in order to determine how strong an imprint the forest fire plumes left in the snow/ice. Summit is ideal for that purpose because forest fire plumes tend to be more pronounced at these high altitudes.

An interdisciplinary community, involving fire ecologists, meteorologists specialized in fire weather forecasts and satellite data experts, has come together in POLARCAT to specifically target forest fire plumes with research aircraft in summer 2008. This will allow, for the first time, to fully characterize forest fire plumes, including their microphysical properties and radiative effects, as they are transported into the Arctic, and to study the pyro-cumulonimbus phenomenon (Fromm and Servranckx, 2003). We will contribute to this community effort with analyses of the aircraft data.