Background. The Arctic troposphere is a unique environment within the earth’s atmospheric system. Its uniqueness stems from generally cold temperatures, a prolonged period of darkness followed by a period of continuous light, underlying snow and ice, and a low tropopause above. While there are virtually no anthropogenic pollution sources within the Arctic itself, it is impacted by emissions from many of the world's largest industrial regions (e.g., Rahn et al., 1977; Barrie, 1986) . Initial research into Arctic Haze found that emissions from northern Eurasia were most significant, though recent evidence suggests that contributions from rapidly developing economies in eastern Asia are growing in importance (Koch and Hansen, 2005). In summer, the arctic free troposphere frequently receives “pollution” from boreal forest fires (e.g., Mauzerall et al., 1996; Dibb et al., 1996) . The composition of the arctic troposphere is further influenced by snow to air exchange of key trace chemicals followed by homogeneous and heterogeneous reactions (e.g., Honrath et al., 1999; Dibb and Arsenault, 2002; Dibb et al., 2002) .

Atmospheric chemistry research in the Arctic has tended to come in waves targeting largely separate questions. The discovery of Arctic Haze in the late 1950s led to the development of a network of surface observatories. Between 1983 and 1991 four AGASP airborne campaigns complemented the surface network by documenting the vertical and horizontal distribution and composition of haze in the lower troposphere over the western sector of the Arctic in late winter and spring. Sampling at Alert and Barrow in support of haze investigations discovered severe ozone depletion events (ODEs) in the boundary layer over the Arctic Ocean at the time of polar sunrise (Oltmans, 1981; Barrie et al., 1988). This lead to a series of increasingly intensive ground based campaigns in which the importance of halogen chemistry and the link to mercury depletion events were discovered (e.g., Schroeder et al., 1998; Bottenheim et al., 2002). The TOPSE airborne sampling campaign in 2000 helped to establish the vertical and geographic extent of ODEs in the Canadian Arctic, though the primary focus of this mission was to establish the cause of the springtime maximum in ozone in the Arctic mid troposphere (Atlas et al., 2003). The discovery of the stratospheric ozone hole over Antarctica led to a series of airborne campaigns probing also the Arctic stratosphere between 1989 and 2005. Unfortunately these missions devoted very few flight hours to sampling in the Arctic troposphere. The ABLE 3 summer campaigns in 1988 and 1990 did sample the North American free troposphere, but platform limitations and a focus on constraining surface fluxes resulted in much of the sampling at low levels (Harriss et al, 1992, 1994).

Recent years have seen tremendous advances in in-situ measurement capability and satellite observations of tropospheric composition. IPY offers a unique and timely opportunity for a coordinated and integrated international experiment to explore the chemistry of the entire arctic troposphere and its impacts on global chemistry and climate. Several unique phenomena have been identified in the arctic troposphere that need further systematic and coordinated investigation. Salient among these are: (1) causes of surface ozone and mercury depletion events; (2) the likely presence and role of halogen free radicals; (3) presence of atmospheric reservoirs of reactive nitrogen and their influence on ozone chemistry; (4) emissions of OVOC and NO_x from ice surfaces; (5) influences of stratospheric intrusions; and (6) investigations of glacial ice cores to understand past atmospheric composition and recent human impacts.

Surface ozone and mercury depletion . As noted above, the discovery of ODEs in the Arctic atmospheric boundary layer near the time of polar sunrise (Oltmans, 1981, Barrie et al., 1988) sparked one of the waves of interest in Arctic tropospheric chemistry. An equally surprising discovery was that gaseous elementary mercury (Hg) appeared to undergo depletion in concert with ozone/HO_x/NO_x raising the specter of a potential major contamination of the Arctic biosphere (Schroeder et al., 1998). Figure 5 and Figure 6 show examples of such coincident depletions and their vertical extent.

It has been postulated that this phenomenon is a result of the following gas-phase bromine atom chain reactions:

A mechanism suggested to cause the observed sudden BrO enhancements in the marine boundary layer is the autocatalytic release of BrO involving heterogeneous reactions on sea-salt surfaces (Tang and McConnel, 1996; Vogt et al., 1996). Substantial concentrations of BrO have been observed in both the Arctic and the Antarctic (Wagner and Platt, 1998, Frieß et al., 2004). The efficiency of this cycle is limited by conversion of Br to the non-radical reservoir species. Models have been developed to explain the role of bromine and iodine chemistry in ozone and Hg depletions (Calvert and Lindberg, 2003).

More recently, studies have pointed out the importance of polar snow as a source of ambient nitrogen oxides (NO_x) (Honrath et al., 1999) and of precursors of hydrogen oxide radicals (HO_x) such as HCHO during spring (Sumner and Shepson, 1999). Snowpack photochemistry has been identified as a likely cause of large interstitial-air and ambient HCHO concentrations, and was shown to result in NO_x concentrations in interstitial-air up to an order of magnitude larger than ambient levels, consistent with the presence of an unexpected diurnal cycle in ambient-air NO_x.

In short, the Arctic boundary layer in spring time is influenced by industrial pollution, halogen chemistry, and ice driven NO_x and OVOC intrusions at the same time. To date, attempts to develop a coherent mechanistic explanation for these depletion processes on the basis of known gas-phase chemistry has not been successful and heterogeneous processes involving snow and aerosol particles are likely implicated. Observational data are critically needed to analyze the coupled evolution of BrO_x/ClO_x/IO-NO_x-HO_x-O₃ chemistry during O₃ and Hg depletion events in the Arctic spring.

Halogens in the free troposphere . Little attention has been paid to the possibility that reactive halogens may have a significant impact in the Arctic free troposphere. Recent field campaigns at Summit , Greenland (72ºN, 38ºW, 3.1 km asl) suggest that this may indeed be the case. High levels of peroxy radicals (HO₂+RO₂ ) were measured consistent with photochemical theory given observed mixing ratios of precursors. However, OH levels were significantly elevated compared to steady state model simulations and previous measurements at South Pole (Huey et al., 2004a, 004b; Sjosted et al., 2005). Observed values of the (HO₂+RO₂)/OH ratio were generally 4 – 5 times lower than expected from theory. This disagreement was greatly accentuated during periods of high wind, when observed values of the (HO₂+RO₂)/OH ratio were more than an order of magnitude lower than model estimates. These observations lead to the hypothesis that halogen chemistry may be responsible for much of the observed disturbance in HO_x partitioning at Summit .

Satellite observations (GOME and SCHIMACHY) suggest that the atmospheric column of BrO above central Greenland during summer is often on the order of 3-5 x 10¹³ mol cm^-2 ( Figure 7) which would yield mixing ratios near 20 ppt if most of the BrO were in a 1 km deep boundary layer above the ice sheet (Richter et al., 1998; Wagner and Platt, 1998). Although substantial concentrations of BrO may be near the surface there is reason to believe that this BrO is distributed throughout the Arctic troposphere. Measurements of large perturbations in hydrocarbon ratios, enhanced soluble gas phase bromine (Dibb, unpublished data; Evans et al., 2003; Ridley et al., 2003) and ozone depletion (Peterson and Honrath, 2001; Helmig et al., 2005) in air filling the pore spaces of the snowpack indicate that halogen activation may proceed via heterogeneous reactions on ice crystal surfaces similar to those observed during polar sunrise at lower elevations in the Arctic. A small reactive halogen flux from the snow pack into overlying air could account for persistently elevated OH throughout summer. The impact on photochemistry in the free troposphere above sunlit snow may be significant if snow-impacted boundary-layer air is vertically mixed upward.

Reactive nitrogen, hydrogen, and ozone in the free troposphere. As has been stated above, the composition of the Arctic free troposphere and its linkages with the surface below and the stratosphere above have not been extensively studied. The ABLE-3A and 3B (Harris et al., 1992; 1994) campaigns did study O₃ chemistry but were restricted to middle troposphere altitudes due to platform limitations and were performed at a time when suitable instrumentation to measure many key species (e. g. free radicals) was unavailable. As noted earlier, one of the primary ABLE 3 objectives was to constrain biosphere/atmosphere exchange, hence there was a lower troposphere focus. A more recent effort was in TOPSE during Feb-March 2000 with an altitude limitation similar to the ABLE 3 experiments but improved instrumentation (Atlas et al., 2003). In this campaign, determining the impact of stratosphere-to-troposphere transport (STT) on the oxidative capacity of the troposphere was one of the key objectives. It should be noted that STT was found to be significant during both the summertime and the spring. However, model calculations constrained by the ABLE 3 and TOPSE data sets found that photochemical production appeared to be the dominant source of O₃ in the Arctic troposphere from late spring into summer.

The free troposphere of the Arctic is greatly influenced by fires in summer and Eurasian outflow in spring. Figure 8 shows an example of the role of PAN in the free troposphere where 80% of all reactive nitrogen is tied in this form. This is a unique feature of a very cold atmosphere where the organic forms of reactive nitrogen are highly stabilized (Singh et al., 1992). However, PAN is easily decomposed to produce NO_x and influence O₃ chemistry in other regions of the troposphere. There is a clear need to perform field missions that can cover the entire Arctic troposphere in at least two seasons with a capability to measure the O₃-HO_x-NO_x-halogen cycle including precursors and aerosols.

POLARCAT objectives related to tropospheric composition and chemistry.

The objective is to investigate the composition and chemistry of the entire Arctic troposphere in two seasons (spring and summer) with the goal of understanding the reactive nitrogen, reactive hydrogen, reactive halogen, and ozone cycles. An integrated approach that links surface, free tropospheric, and satellite observations with models of chemistry and climate is envisioned. The intensive sampling will quantify the relative importance of transport to and from the free troposphere, impact of halogen/OVOC/NO_x formation in the snowpack on the free troposphere, and the role of Arctic reservoir species on the global troposphere. Specific objectives are: