The focus of this work is on understanding the behavior of O3 and CO2, as two of the most important greenhouse gases that are as yet poorly understood. We also  measure bromine oxide (BrO), a key species responsible for the extraordinary polar springtime O3 and Hg atmospheric depletion , both of which have strong consequences for human and ecosystem health in the Arctic region. There are a variety of potential connections between these three species. O3 is a critical and central molecule in the troposphere. It indirectly but profoundly affects the abundance of most trace gases, including many greenhouse gases. Lower atmosphere depletions in O3 seen in the Arctic (Bottenheim et al., 2002) during springtime are highly correlated to deposition of Hg (Lu et al., 2001; Lindberg et al., 2002), a toxic and bio-accumulative pollutant; thus it also indicates and possibly determines Hg behavior. This coupled Hg and O3 chemistry is driven by bromine radicals (Br and BrO). Therefore, observations of BrO directly quantify a key catalyst in Arctic atmospheric chemistry and provide a link to the cycling of O3 and Hg in this region. A long standing hypothesis is that both micro and macro-algae emit organo-halogen species that may be initiators of the halogen chemistry that destroys O3, producing the intermediate BrO (Bottenheim et al., 1990). The fluxes of organo-halogen compounds are tied to primary productivity and thus to some extent to CO2 fluxes. Fluxes of halogen precursors and CO2 may or may not be tied to the presence of open leads. CO2 is exchanged by biological processes and thus is an indicator of biological activity, coupled to the physical characteristics of the open ocean and sea ice surfaces. While CO2 is readily exchanged between the atmosphere and the sea water, the role of sea ice as a barrier to, or integral player in, this process is only poorly understood. In spite of their importance, very little data exists on O3, BrO and CO2 concentrations in the Arctic Ocean region, and very little indeed over the Arctic Ocean surface. However, the scarcity of data for these key atmospheric species in the Arctic Ocean region itself is primarily due to the lack of capability.

An intense, short term high O3, low BrO depletion event on the ocean off Alert with simultaneous O3 enhancement, sampled with sensors similar to those proposed herein (2B and MAX-DOAS) during the OOTI 2004 experiment at Alert, NU. Adapted from Morin et al., 2005.
O3 is an enigmatic constituent of the natural atmosphere.  It is an essential component of the natural functioning of biogeochemical cycles, and the establishment of the “cleansing power” of the atmosphere, as O3 photolysis produces OH radicals. OH is the dominant oxidizer in the Earth’s atmosphere, responsible for the conversion of toxic non-water soluble pollutants into water soluble chemical species that are readily removed. On the other hand, O3 is a regulated air pollutant, toxic to both vegetation (Pell et al., 1997) and to the human respiratory system (Tilton et al., 1989).  It is also an important greenhouse gas, which has contributed ~15% of the global warming to date.

In 1986, independent reports of surface O3 depletion were published from the Barrow and the Alert Global Atmospheric Watch stations
(Bottenheim et al., 1986; Oltmans et al., 1986).  Pursuit of a causal relationship in photochemical O3 destruction led to the discovery by Barrie et al. (1988) that O3 depletion occurred in concert with large increases in particle-phase bromine, leading the authors to hypothesize the following chain reaction:
Br     +    O3            →     BrO          +    O2        (1)
BrO  +    BrO    →   2Br   +  O2                          (2)

That hypothesis is believed confirmed
through the observation of tropospheric BrO radicals during O3 depletion episodes (Hausmann and Platt, 1994), as reaction [1] is the only important source of BrO radicals.  BrO total atmospheric columns as well as tropospheric columns have now been measured from the ground and from satellite . In polar spring, areas exceeding 10 million square kilometers in sea-ice covered polar regions show enhanced BrO columns.
Surface layer O3 depletion in the Arctic often leads to near zero O3 concentrations, from the surface to as high as 1-2 km. Bottenheim et al. (2002) observed a period of several days of almost completely O3-depeleted air at the surface. Simultaneous BrO measurements by Multi-Axis Differential Optical Absorption Spectroscopy (MAX-DOAS) showed that enhanced BrO of up to 30ppt in the surface layer up to 1-2km altitude correlated to O3 depletion (Hönninger and Platt, 2002).



Ozone data from Ice Camp Narwhal (adapted Hopper et al., 1998).

However, those measurements were made at Alert, a coastal site. There are, to date, only a few sets of O33 data from ice camps: Ice Camp Swan (Hopper et al., 1994) and the Narwhal Ice Camp. From the fact that O was essentially depleted for an ~8 day period at the ice camp, with air motion from multiple directions, it was concluded that O3 depletion was a widespread persistent phenomenon over the Arctic Ocean surface at this time; the depletion only “breaks” by mixing from aloft, e.g. during storms. The difference in the data from the coastal sites and the Arctic Ocean surface is striking – O3 is often much higher at the coastal sites because of downward mixing related to orography and drainage flows. Recent data from the OOTI (“Out On The Ice”) experiment off Alert, Nu dramatically show how occasional drainage flow can lead to a sudden change from the common low O3, high BrO conditions on the frozen ocean to a high O3, low BrO episode (Morin et al., 2005). Thus, these data make clear the importance of continuous, real-time measurements from the Arctic Ocean surface.

If the chain reactions 1-2 are indeed responsible for the O3 depletion
, then a source of active halogen is required and sea salt is believed to be the major source. Activation of unreactive halogen ions into reactive halogen radicals (Br and BrO) is therefore needed. A major lack of understanding in the study of atmospheric chemistry during the Arctic springtime is the role of various frozen surfaces in halogen activation and O3 depletion (Simpson et al., 2005). In particular, the role of highly saline surfaces, such as brine and frost flowers on newly formed sea ice, is still poorly quantified (Kaleschke et al., 2004). Snow and/or sea ice appear to be critical components for halogen activation. O3 depletion events, as detected at coastal stations, cease at precisely the time of year of snowmelt. Satellite-based observations of BrO also show that enhanced BrO column densities, typical of O3 depletion events, cease upon snowmelt. It is likely that the high surface area and permeability of snow, as compared to liquid water or ice, plays an important role in the halogen activation process, despite the highly saline surface needed for halogen activation. Snowmelt on sea ice behaves differently from melt on land; the timing is delayed on sea ice, and some of the saline melt water pools longer on sea ice surfaces, for the duration of the melt season. We have no observations of what happens to O3 and CO2 during this transition in the Arctic Ocean. It is precisely this time of the year when the transparency of the snow/ice over the ocean increases and the biological systems on the base of the sea ice, leads and open waters grow. It should be recognized that the atmosphere also becomes unstable at this time, so vertical mixing could also account for the disappearance of the halogen chemistry. Observations of these three key species from autonomous buoys can fill in these missing observations, and will improve our understanding of sources, transformations and sinks of reactive bromine species in the troposphere as well as their impact on tropospheric chemistry and the Arctic ecosystem.

The concentrations of CO2 in the atmosphere are at the highest levels of the past 25 million years
. Current levels of CO2 have increased by 30% from 280 ppm in pre-industrial times to ~380 ppm today, and they continue to rise. For the decade of the 1990s, an average of about 6.3 Pg C per year as CO2 was released to the atmosphere from the burning of fossil fuels, and it is estimated that an average of 1.5-2.5 Pg C per year was emitted due to deforestation and land-use change during the same interval. Only half, on average, of the CO2 from anthropogenic emissions has remained until now in the atmosphere. Analyses of the decreasing 13C/12C and O2/N2 ratios in the atmosphere have shown that the land and oceans have sequestered the other half, in approximately equal proportions but with temporal and spatial variations. The Arctic Ocean is usually not included in these calculations as models presume a sea-ice capped region without much ocean-sea ice-atmosphere exchange. Because global climate models show large deviations in their simulations of current conditions in the Arctic region, the effect of changing ice cover (at ~7% decrease/decade; Comiso et al., 2002) and thickness on pCO2 fluxes in the Arctic Ocean is not clear. Climate models predict a predominantly ice-free Arctic Ocean in summer by the end of the century, which implies a change in the sea-air fluxes of CO2 due to either temporally and spatially enhanced photosynthesis or lowered productivity with nutrient depletion (Smetacek and Nicol, 2005). If the Arctic Ocean becomes ice-free, then the CO2 uptake due to ocean circulation coupled with the ocean solubility pump may increase and could be quantified by physico-chemical processes. However, the role of marine biota and changes in that role are less understood and hence, less predictable. Finally, the role of sea ice as a barrier to, or an integral player of, CO2 air-sea and/or air-ice fluxes is least understood, with both the direction and amount of CO2 transfer between air and sea/ice varying in the thaw/freeze and open water seasons due to sea-ice melt ponds, open brine channels, leads and photosynthesis. Independent information on spatial and temporal patterns of CO2 sources and sinks in the Arctic Ocean is of extraordinary value for challenging process-based terrestrial and oceanic carbon cycle flux models as well as atmospheric transport models, and thus improving our ability to predict future regional and global CO2 fluxes.