An innovative cryogen-free concentrator system for measurement of atmospheric trace gases at the parts per trillion level has been developed with detection by routinely used gas chromatographic methods. The first-generation system was capable of reaching a trapping temperature of -186 degrees C, while the current version can reach -195 degrees C. A Kleemenko cooler is used to create liquid nitrogen equivalent trapping conditions and eliminate the use of solid absorbents, a potential source of artifacts. The method utilizes dual-stage trapping with individual cold regions. The two stages are cooled to -20 and -175 degrees C for water management and sample enrichment, respectively. Both stages house a Silonite-coated stainless steel sample loop; the second stage loop is filled with 1-mm-diameter glass beads, which provide an inert surface area for analyte concentration. In our application, the complete system employed four channels utilizing two flame ionization detectors, one electron capture detector, and a mass spectrometer. The system was automated for unattended operation and was deployed off the New England east coast on Appledore Island to measure a suite of ambient non-methane hydrocarbons, halocarbons, alkyl nitrates, and oxygenated volatile organic compounds during the International Consortium for Atmospheric Research on Transport and Transformation field campaign in summer 2004. This robust system quantified 98 ambient volatile organic compounds with precisions ranging from 0.3 to 15%.

[1] The temporal and spatial variations of ozone (O 3 ) in polluted continental outflow over the North Atlantic were investigated during the International Consortium for Atmospheric Research on Transport and Transformation (ICARTT) field campaign in July-August 2004. Our analysis utilized measurements of O 3 from three smart balloons traveling at 0.5-3 km altitude in combination with simulations using the MM5/SMOKE/CMAQ air quality modeling system. Model results for over and within 300 km off North America were corroborated by comparison to a suite of measurements from ground stations, ozonesondes, and the NOAA ship Ronald H. Brown cruising in the Gulf of Maine. A prominent feature of the O 3 distribution was the high mixing ratios over the North Atlantic, reaching a peak value of 171 ppbv, compared to the northeastern United States (<$100 ppbv). The enhanced O 3 levels over ocean, mostly observed at night, appeared to be the result of four factors: (1) a supply of precursors in prevailing flow off the polluted U.S. east coast, (2) significant daytime in situ chemical production, (3) minimal depositional loss to the ocean at the balloon altitudes, and (4) small nighttime chemical loss. An important implication is that quantification of O 3 export from the United States must include estimation of downwind chemical processing in polluted air masses. Balloons 3 and 4 were launched within 18 hours of each other, and their tracks allowed examination of horizontal gradients in O 3 across distances varying from 200 to 400 km. In air masses influenced by recent outflow (<2 days) the O 3 gradient was À0.2 to 0.2 ppbv km À1 , while by distant source regions (>2 days) it exhibited only À0.05-0.05 ppbv km À1 . These same two balloons encountered Hurricane Alex at different times, but both measured O 3 mixing ratios >100 ppbv. Our model results show clearly that polluted air from the mid-Atlantic states was channeled directly into Alex's inflow region. Overall, variations in O 3 on timescales of tens of minutes to hours are attributed to its highly heterogeneous distribution in urban plumes, with variations over hours to days caused by changing source regions related to cyclonic activity.