A42B-01
Halogen Activation from Heterogeneous Uptake of N2O5
The nitrate radical, NO3, formed from reaction of NO2 with O3, and dinitrogen pentoxide, N2O5, formed from subsequent reaction of NO3 with NO2, drive several important nocturnal chemical processes, including oxidation of VOCs (by NO3) and removal of NOx (= NO + NO2) by the heterogeneous reaction of N2O5 to form either nitric acid, HNO3, and/or aerosol nitrates. Laboratory studies have shown that heterogeneous uptake of N2O5 on sea salt and chloride containing aerosol produces nitryl chloride, ClNO2. After sunrise, photolysis of ClNO2 provides a source of Cl atoms, an important oxidant of VOCs in the marine boundary layer. In spite of the potential importance of this halogen activation mechanism, field investigations have yet to directly confirm it. We have measured NO3 and N2O5, by cavity ring-down spectroscopy, and ClNO2, by I- chemical ionization mass spectrometry, on board the NOAA research vessel Ronald H. Brown during the Texas Air Quality Study - Gulf of Mexico Atmospheric Composition and Climate Study (TexAQS/GoMACCS) 2006. In this presentation, a few aspects of the interaction between nitrogen oxide and halogen cycles in the subtropical marine boundary layer are highlighted, including efficient ClNO2 production from N2O5 uptake on mixed continental and marine aerosol.
A42B-02
Hg(II) Sources, Sinks, and Reactions with Halogens in the Remote Atmospheric Marine Boundary Layer
We investigate the chemistry and sinks of reactive gaseous mercury (RGM) in the remote marine boundary layer, using a box model to interpret observations from the Atlantic and Pacific Oceans and from a coastal site in Okinawa. The diurnal cycles of RGM at all sites have morning increases that we show are quantitatively consistent with RGM production primarily from the oxidation of Hg0 by atomic halogens. Atomic bromine may generate the midday peak in RGM and could be the major daytime oxidant of Hg0, while atomic Cl could play a secondary role in the morning. Because RGM concentrations fall rapidly in the afternoon but persist above detection limit at night, we posit a photosensitive sink for RGM, possibly involving reduced sulfur species in marine aerosol. Alternately or in addition, a nocturnal oxidant, such as NO3, could supply detectable amounts of RGM at night. We find much larger amplitude diurnal cycles over tropical and sub-tropical oceans compared with the extra-tropics. The tropical enhancement of RGM concentration occurs in areas with little chlorophyll A, but where previous studies found significant biogenic halocarbon emissions.
A42B-03
Iodine Chemistry at the California Coast
The chemistry of reactive iodine species (RIS) in clean coastal environments has received considerable attention over the past years. It is known that the presence of reactive iodine in the marine boundary layer (MBL) can catalytically destroy ozone and change the NO/NO2 and OH/HO2 ratios. Iodine oxides can also lead to aerosol nucleation in the MBL, potentially affecting the radiative balance of the atmosphere. The limited observational database of RIS, however, restricts our ability to fully understand iodine chemistry, especially in polluted coastal environments, and to assess the significance of reactive iodine on a global scale. Over the past three years, we performed observations of reactive iodine species in the polluted MBL of North America's Atlantic (Isles of Shoals, ME in July-August 2004) and Pacific (Malibu, CA, in March–May 2004 and in October 2006) coasts using concurrent measurements by long-path (LP) and a multi-axis (MAX) differential optical absorption spectroscopy (DOAS) instruments. In this presentation we will focus on the observations of reactive iodine species in Malibu, CA and their behavior with respect to various environmental variables, such as the tidal cycle, wind speed and direction, solar radiation, and NOx levels. The spatial distribution of the reactive iodine species in the coastal MBL is poorly understood, and will be discussed based on the MAX-DOAS measurements and radiative transfer calculations in Malibu and at the Isles of Shoals. A comparison of the RIS observations in Malibu with those at the Isles of Shoals sheds light onto the chemistry of RIS in the polluted coastal environment.
A42B-04
In situ IO Measurements in the Marine Boundary Layer Using Laser-Induced Fluorescence Spectroscopy
The iodine monoxide (IO) radical plays an important role in the chemistry of the marine boundary layer (MBL). It is formed by the reaction of ozone with iodine atoms generated by the photolysis of I2 and photo-labile iodocarbons and at night from the reaction of NO3 with I2 and O3. IO is implicated in ozone destruction, DMS oxidation and new particle formation. IO has been measured previously using LP-DOAS, with absorption paths of several kilometres, and MAX-DOAS both of which are associated with significant spatial averaging over the halogen source regions. An in situ laser-induced fluorescence (LIF) technique has been developed to detect {IO} radicals in the MBL (Whalley et. al. (2007), J. Atmos. Chem. 58: 19 - 39) and has been employed in two separate instruments. In both field instruments, an all solid-state pulsed laser operating at 445 nm is used to excite IO in the A 2 Π3/2 \ (ν'=2) \ ← \ X 2 Π3/2 \ (ν"=0) electronic transition, with off-resonant fluorescence detected at 520.3 nm in the (2,5) band. The sensitivity of each instrument is determined by the generation of known concentrations of IO (between 5 and 100 pptv) from the photolysis of N2O at 185 nm followed by the subsequent reaction of oxygen atoms with CF3I. The first instrument was deployed in Roscoff, North-western France during August/September 2006 as part of the RHaMBLe (Reactive Halogens in the Marine Boundary Layer) campaign. The instrument limit of detection was 0.4 pptv for a 5 min integration time, with an uncertainty of 23%. Located on a small jetty, the instrument measured significant levels of IO on 11 days, with up to 29 pptv observed (10 second average). IO displayed a clear diurnal profile with a maximum at low tide, and lower concentrations observed on some nights. This is compared with LP-DOAS (University of Leeds) measurements also made at the site. The second instrument was deployed twice in 2007. The instrumental limit of detection was also found to be 0.4 pptv for a 5 minute integration time, with an overall uncertainty of 23%. The first deployment was aboard the RSS Discovery to measure open ocean IO over the Mauritanian upwelling during May/June 2007. The second deployment was to Mace Head, Western Ireland during August 2007. At Mace Head, IO was observed on all 9 measurement days, with a maximum of 28 pptv observed (1 min average) coinciding with low (spring) tide.
A42B-05
Observations of Halogens at Summit, Greenland
During May and June 2007 a field campaign was carried out at Summit, Greenland (altitude 3.2 km, latitude = 72.55°N) to investigate the importance of halogens in this remote environment. Instruments were deployed to obtain a large suite of observations that included: BrO, Hg°, reactive gaseous mercury (RGM) , OH, RO2, CO, NO, O3, HCl, HNO3, HO2NO2, SO2, soluble bromide, snow ionic composition, and J values. Significant levels of BrO (1-3 pptv) were often observed by both a differential optical absorption spectrometer (DOAS) and a chemical ionization mass spectrometer (CIMS). Both BrO and soluble bromide appear to be reasonably correlated with solar radiation. Depletion of elemental mercury and production of RGM was also frequently observed. These results indicate that halogen chemistry is active at Summit during summer. A summary of our preliminary observations as well as the impact of halogen chemistry on HOx radicals will be presented.
A42B-06
Global simulation of Polar Spring Bromine Explosion and ozone Loss
By considering sea salt production from sea ice, the bromine explosion events during polar spring time are well re-produced by the 3D global chemical model p-TOMCAT. The simulated tropospheric BrO in both polar spring time are successfully matching the satellite BrO distributions. We find the tropospheric BrO accounts for 30-40% of the total column BrO. The daily averaged BrO concentration in polar free troposphere is of 0.5-1 pptv. Model results also show that the heterogeneous reactivations on aerosols play a significant role in sustaining high BrO partitioning in total bromine. As to ozone loss, a zonal mean lose of 5-8% is simulated in lower troposphere with much higher loss locally. Although there is a significant bromine-induced ozone loss during the bromine explosion events, it is likely that much lower surface ozone concentrations must be a result of air mass isolated from free troposphere for sometime.
A42B-07
Is the Spring Maximum of MSA (Methane Sulfonic Acid) in the Arctic due to Oxidation of DMS (DiMethyl Sulphide) by BrO?
Long term measurements of MSA at Alert, NU have revealed a seasonal recurring pattern with two maxima: a first maximum occurs around April/May, while a second maximum is observed in summer (July/August). It has been shown that the origins of these maxima are different. The summer maximum is thought to reflect maximum DMS emission in Northern oceans off the coast of Norway and the northeast Pacific, while the spring maximum could be due to longer range transport of air originating in more Southerly regions of the north Atlantic where DMS emission would be occurring earlier in the season. The explanation for the spring maximum is debatable since by late April the Alert site is usually disconnected from such long range transport, although year-to-year variations in the MSA record appear related to sea surface temperature anomalies in the North Atlantic. It is well known that frequent ozone depletion episodes are observed in the spring at Alert, due to chemistry involving the BrO molecule. Chemical kinetic data show that the BrO molecule is also a very effective oxidant for DMS, and that this oxidation preferentially leads to the formation of MSA, rather than sulphate. In order to investigate whether the spring peak of MSA at Alert could be related to oxidation of DMS by BrO we have collected daily HiVol filters at Alert during the height of the ozone depletion season (late April- mid May 2007). We will report the results of this study and explore correlations with ozone and other chemical parameters, as well as the origin of the air masses sampled using 10-day back trajectories.
A42B-08
Simulating Arctic bromine explosion and surface ozone depletion with the GEM-AQ model
Episodes of bromine explosion and surface ozone depletion during the Arctic Springtime (for the years of 2000 and 2001) are simulated with an online air-quality transport model, GEM-AQ, and compared with in-situ measurements of ozone, temperature and wind speed/direction at the ground level and their profiles at selected stations as well as satellite BrO column measurements. GEM-AQ is implemented with tropospheric oxidant chemistry, gas-phase and heterogeneous bromine chemistry, and multi-component aerosol chemistry and microphysics. At this point model runs are performed with global variable-resolution horizontal grids having a high-resolution Arctic core at approximately 100 km resolution. A series of 30 hours meteorological forecasts are carried out, of which the first 6 hours are spin-up from objective analysis at different resolution and the last 24 hours are used for chemical transport simulation over months. Dry deposition velocities for HOBr, BrNO3 and HBr on the snow/ice pack are calculated online and their deposition is assumed to be converted to Br2 emission back to the atmosphere. Fresh, first-year sea ice is assumed to release Br2 as fresh emission and be more efficient at releasing Br2 than old, multi-year sea ice. We use the QuickSCAT satellite data of multi-year seaice fraction and Canadian Meteorological Centre's surface analysis of total seaice fraction for estimating the first-year fraction. Fair agreement is obtained between observed and simulated variabilities in surface ozone mixing ratios at Alert, Barrow and Zeppelinfjell on the timescale of several days, indicating the source region of bromine and synoptic- scale mass transport are simulated well. However, occasional, abrupt recovery of surface ozone at Alert is not captured by the model. Comparison with observed temperature and winds demonstrates that the model at the current resolution does not resolve abrupt wind episodes mediated by local mountain topography (e.g. Alert, Eureka, Zeppelinfjell); otherwise, simulated wind speed/direction agree well with the data from stations within relatively smooth topography (e.g. Barrow, Resolute, Inuvik) and from an ice floe buoy near the north pole.