Measurements of the chemical composition of the air and precipitation have been made at Svalbard for a long time. Sulphate concentrations in the air have been measured since 1977, first at Ny Ålesund and since 1989 at Mount Zeppelin. At Barrow, an equally long record exists for light scattering. As summarized in the White Paper, trends for most available species (e.g., light scattering and absorption at Barrow, sulphur at Zeppelin) at both stations are decreasing but a possible trend reversal has occurred since the end of the 1990s. Of interest are also light hydrocarbons, which can be used as markers for the different source categories contributing to pollution, and ozone, for which relatively long records exist from Svalbard.

We suggest a complete re-analysis of the data from Svalbard with regard to pollution sources. For this, FLEXPART will be run in backward mode every 3 hours and 20 days backward in time over the entire period of more than 30 years with available data, using ECMWF input data (re-analyses and operational analyses for the last few years). The results of these calculations will be made available via websites (results for the years 1999-2005 are already available). Using Bayesian inversion, the source regions of sulphur and BC in particular will be determined for different periods of the record to analyze how the source regions have changed over the past few decades. It is expected that large changes in pollution source regions will be found, for instance a strong decrease of sulphur pollution transported from western Europe in the 1980s and an abrupt drop of pollution from eastern European countries and the former U.S.S.R around the year 1989. Yet, to date these changes have not been documented fully on a source region basis (but see Sharma et al., 2004), partly because trajectory analyses are not accurate for the long transport distances involved, and partly because variability in air transport masks effects of emission changes. Also of interest will be how the source regions of sulphur, BC, and other species (e.g., carbon monoxide as a tracer) have changed more recently (since about 2000 and into the IPY period), as the economies in eastern European countries are now growing. Southeast Asia is now the dominant source of BC worldwide. Koch and Hansen (2005), in a model study, recently suggested that it is also the largest source of BC for the Arctic. This was questioned by Stohl (2006) in another model study. The light absorption measurements from Zeppelin, in combination with the backward calculations and results from a 3-D aerosol transport model and a global climate model (see WP 7), will hopefully allow a clarification.

Levoglucosan is the most recognized marker for tracing emissions of particulate matter from biomass burning. It is emitted in high concentrations, not present in the vapor phase, associated with fine aerosols exclusively, and is not selectively removed from the atmosphere. These qualities enable detection of biomass burning plumes far from their sources. For POLARCAT, daily levoglucosan measurements for about 14 months will be made to study the influence of North American and Siberian wildfires in summer (see WP 4), as well as of domestic wood burning in winter, on aerosol concentrations at Svalbard. Aerosols will be collected at Zeppelin using a high-volume sampler and levoglucosan analysis will be performed using HPLC/HRMS (Dye et al., 2005; Yttri et 6 al., 2005). Collaboration with Johan Ström, Stockholm University (BC and aerosol size distributions) will allow constructing detailed aerosol and ozone budgets for Zeppelin.