A53B-1140
Nitrogen and triple oxygen isotopic composition of nitrate in surface snow in Antarctica.
Interactions between the chemical state of the atmosphere and climate are poorly documented mainly due to the lack of a suitable set of proxies for the past atmospheric oxidation capacity, which controls the chemical lifetime of greenhouse gases such as methane or ozone. For example, to date no quantitative relationship exists between nitrate (NO3-) preserved in polar snow, firn and ice and atmospheric nitrogen oxides (NOx = NO + NO2), which are key in modulating the burden of atmospheric oxidants. This is in part due to post- depositional processing taking place in the upper snow pack altering nitrate concentrations significantly. However, the stable isotopic composition of NO3- potentially provides a powerful tool to not only gain insight into post-depositional processes but also to assess past ozone chemistry and major NO3- sinks and sources. Here we report results from the first simultaneous determination of the nitrate 15N/14N, 18O/16O and 17O/16O isotopic ratios in Antarctic surface snow samples collected on a summer traverse between Dome Concordia (75°S) and Dumont d'Urville (DDU) (66°S). Reproducible isotopic ratios were achieved using the bacterial denitrifier method and an automated on-line GC- IRMS system. The observed mean NO3- oxygen isotopic anomaly (Δ17O = δ17O - 0.52 δ18O) of 28.8±2.2 ‰ is consistent with the expected anomaly transfer from atmospheric ozone through HOx-NOx gas phase chemistry. The spatial variability of the δ15N isotopic signature allows a clear distinction between near coastal (-8.6 to -5.6 ‰) and continental high elevation sites (31-47 ‰). Inland δ15N enrichment and depleted levels near the coast support the idea of significant continental snow reemissions of nitrate and transport towards lower latitudes.
A53B-1141
NOx and HONO Emissions from Natural Snow Surfaces in a Temperature Controlled Laboratory
Photochemical production of NOx and HONO from surface snow can significantly impact the NOx, OH, and O3 budgets in the overlying atmosphere. NOx production is initiated by the solar photolysis of NO3- within or at the surface of snowpacks. HONO is deemed to be a secondary species whose formation likely involves H-atom transfer between NO2 and natural humic substances. Climate and Global change can potentially affect the rate and efficiency of the NO3- (atm) -> NO3- (snow) -> NOx/HONO (atm) cycle via altered solar UV radiation spectrum and/or fluxes, concentration of pollutants, temperature and snow cover. As part of the international multidisciplinary OASIS (Ocean-Atmosphere-Sea Ice-Snowpack) program, emission fluxes of NO, NO2, and HONO from mid-latitude natural surface snow slabs were measured under controlled temperature, ozone mixing ratio and actinic flux conditions. Peak emission fluxes reached into the 0.5 to 1.5×1012 [molecules cm-2 sec-1] range for the three species, or ~ 103 times larger than those typically reported above Polar snowpacks. We present and analyze experimental results on the effects of temperature, actinic flux and the presence of humic acids on NO, NO2, and HONO emissions.
A53B-1142
Three Events of Nitrogen Emission From Snow Surfaces at Ny-Ålesund, Svalbard (Arctic)
Measurements of HONO, HNO3, NO and NO2 were made at two heights above the snow surface near the Italian Station "Dirigibile Italia" at the Kongsfjorden International Research Base at Ny-Ålesund, Svalbard (79°N), between February 20, 2006 and April 20, 2006. Median mixing ratios were 13.2 pmol mol-1, for HONO, and 19.8 pmol mol-1 for HNO3, measured during this period. Visible HONO gradients (25 pmol mol-1) were only observed twice during the period of 24-h sunlight; both times after snowfalls that deposited large amounts of HNO3 (-50 pmol mol-1) on to the snow surface. Three different periods of nitrogen emissions were studied during the campaign. The emission of NO, NO2 and HONO were correlated with fresh snowfall events. HONO emissions were also correlated with the presence of ammonium ions, which we take as tracers for organic compounds in marine aerosols. HONO was likely formed on the surface of ice crystals by the heterogeneous reaction of NO2 involving humic acids or other organic compounds.
A53B-1143
1-D Air-snowpack Modeling of Atmospheric Nitrous Acid (HONO) during ANTCI 2003
HONO measurements in ambient air during ANTCI (Antarctic troposphere chemistry investigation) 2003 exceed the pure gas phase model predictions by a factor of 1.92(plus/minus 0.67), which implies snow emission of HONO. A 1D air-snowpack model of HONO was developed and constrained by observed chemistry and meteorology data. The 1D model includes pure gas phase chemical mechanisms, molecular diffusion and mechanical dispersion, windpumping in snow, gas phase to quasi-liquid layer phase HONO transfer and quasi- liquid layer (QLL) nitrate photolysis. Dominant snowpack HONO sources include QLL nitrite in equilibrium with firn air deep below the snow surface and snowpack nitrate photolysis. The high concentration of HONO in the firn air is subsequently transported above the snowpack by diffusion and windpumping. The primary model uncertainties include the pH and volume of the QLL and snowpack nitrite concentration profiles.
A53B-1144
Coupled HOx, NOx and Halogen chemistry in the Antarctic Boundary Layer
The Chemistry of the Antarctic Boundary Layer and the Interface with Snow (CHABLIS) campaign took place at Halley Base in coastal Antarctica from January 2004 until February 2005. The campaign included a summer intensive focussing upon oxidant chemistry, featuring measurements of OH/HO2 radicals, total peroxy radicals and the halogen oxides IO and BrO in addition to long-term observations of NOx, VOCs, peroxides, HONO, HCHO and CO, together with radiation and meteorological parameters. Measurements of reactive species, especially HOx (OH and HO2) obtained during the summer period of the CHABLIS campaign have been analysed using box model simulations. The model, based upon the Master Chemical Mechanism, was constrained to observed meteorological parameters, photolysis rates, and concentrations of long-lived species such as VOCs and O3. The basic simulations overestimate the observed OH and HO2 levels and fail to replicate the diurnal NOx cycle. Addition of a strong halogen source, and attendant XONO2 hydrolysis sink for NOx, improves the model performance, however either additional radical sinks, or changes to the (highly uncertain) iodine oxide – HOx chemistry, are required to achieve good agreement. Overall the model simulations show the coastal Antarctic boundary layer to be an environment in which rapid radical cycling occurs, but driven by the halogen oxides IO and BrO rather than by NOx, and with attendant ozone destruction rather than production. A mechanism for the efficient recycling of halogen species, through the condensed phase, is required to explain the observed concentrations.
A53B-1145
Ozone Fluxes over Snow-Covered Environments
Until very recently ozone uptake to snow-covered environments was believed to be low and of secondary importance for surface-level ozone concentrations. Our recent studies from polar and midlatitude sites have resulted in contrary new insights into the interchanges of ozone with snow and shown the following: 1. Due to weak atmospheric ozone production and losses, ozone surface fluxes over snow, in particular in the polar regions, often have a determining influence on surface ozone levels. 2. Ozone fluxes in polar regions have significant diurnal and seasonal dependencies, with solar irradiance being a major driver of the ozone fluxes. 3. The ozone exchange is closely linked to photochemical production and exchange of nitrogen oxides, and under certain conditions increased levels of nitrogen oxides can result in photochemical ozone production and upwards ozone fluxes from the snow surface. 4. Gas exchange through the snow varies significantly depending on the gas permeability of the snowpack. 5. The substrate underneath the snow (e.g. glacial ice, sea ice, frozen (permafrost) soil, "warm" midlatitude soil) influences the process-level controls, magnitude and direction of ozone fluxes. These new findings call for the need to develop more detailed parameterizations of ozone exchanges over snow for use in atmospheric chemistry and transport models.
A53B-1146
Singlet Molecular Oxygen on Ice: Rates of Formation and Steady State Concentrations
Singlet molecular oxygen (1O2*), the first electronically excited state of molecular oxygen, reacts rapidly with certain types of environmental pollutants such as furans, phenols, and polycyclic aromatic hydrocarbons (PAHs). Its formation requires the absorption of light by a chromophore (a.k.a. sensitizer), which subsequently transfers energy to ground state molecular oxygen. In the environment, 1O2* chemistry has been studied primarily in the aqueous phase, such as in surface waters or cloud and fog drops. In this work, we expand our current understanding by investigating the rate of formation (Rf) and steady state concentration ([1O2*]) of 1O2* on ice. To investigate 1O2* kinetics, we use a chemical probe technique in which photoformed 1O2* reacts with furfuryl alcohol (FFA). To generate 1O2*, we illuminated frozen samples containing a sensitizer (Rose Bengal, RB) at 549 nm. The concentration of total solutes in each sample was controlled using sodium sulfate (Na2SO4). Following illumination, the decay of FFA was measured using high performance liquid chromatography (HPLC). Ice tests were conducted at 253, 263, and 268 K. Liquid tests for comparison were conducted at 278 K. Results showed dramatically faster (~104) FFA decay on ice than in liquid samples prepared from the same solutions, in agreement with the calculated solute concentration factor in the quasi-liquid layer (QLL) on ice compared to bulk solution. Varying the concentration of RB resulted in similar changes in both Rf and [1O2*], with magnitudes of change close to those expected. Changing temperature and total solutes, both of which control the volume of the QLL on ice, revealed two model regimes: FFA as a major (1) or minor (2) sink of 1O2*. Experimental results from the former regime show good agreement with expected values for both Rf and [1O2*]. Experiments in the later regime are currently in progress. We will also discuss the potential implications of 1O2* to the chemistry of naturally occurring snow and ice.
A53B-1147
Attenuated Total Reflection Infrared Spectroscopy of Ice Films
Ice near its melting point displays a mobile, dynamic layer where adsorption of gases and chemical reactions are most likely to occur. Studies of ice at these temperatures are lacking. A custom-made attenuated total reflection infrared spectroscopy (ATR-IR) apparatus was built to study ice films at temperatures relevant to the boundary layer and high latitudes. ATR-IR spectra of water and ice films over a range of temperature (-15 ° C to 25 ° C) will be presented. Probe molecules (e.g., acetic acid) were used to investigate the proton-donating ability of the bulk ice and ice surface. Spectral changes of the probe molecules with pH in liquid solutions as well as frozen films will be analyzed.
A53B-1148
A Solid-Phase Chemical Actinometer for Snowpack Solar Irradiation Measurements
Research over the past decade has established the importance of polar snow as a medium for chemical processing of atmospheric species; recent laboratory studies have demonstrated the photochemical transformation of anthropogenic organic contaminants in frozen water matrices. In order to investigate the role of snow photochemistry under environmental conditions, field researchers need methods for determining the intensity of light available at the surface and also penetrating the snowpack as a function of depth. While theoretical models exist for light penetration, they rely on a number of parameters relating to the formation and history of the snowpack that field researchers are unlikely to have knowledge of; this makes direct in-situ measurements desirable. Conventional liquid-phase actinometers are of limited use in polar environments where low ambient temperatures may interfere with diffusion-controlled processes - the freezing of aqueous solutions being an extreme case - resulting in varying quantum yields. A solid-phase actinometer system based on the well-studied photoisomerization of o-nitrobenzaldehyde (oNB) has been developed as an alternative device for the measurement of solar UV light intensity. Sample films consisting of oNB incorporated in a polymethylmethacrylate matrix are exposed to light and the extent of oNB loss from the film's surface is determined by ATR-IR spectroscopy. Films are rugged enough to allow deployment in environments that are impractical to access with bulkier electronic instruments, and small enough to allow precise placement and positioning without disturbing the bulk of the snowpack under consideration. Trial measurements made in late- thaw snowpacks near Barrow, Alaska in June 2007 demonstrate the usefulness of this technique for determining relative light intensity as a function of snowpack depth. The technique could readily be extended to other environments where solar irradiance cannot readily be measured by instrumental approaches.
A53B-1149
Occurrence of Organohalides in Snow and Air in Alert, Nunavut (2006)
We investigated concentrations of organohalides in Alert during a 3 week sampling campaign in spring 2006. In order to investigate the occurrence of species in the atmosphere and the snow pack, we have collected airsamples in electropolished canisters and snow samples by grab sampling including several depth profiles. Model compounds analysed on site from snow samples using gas chromatography and flame ionisation detection system (SPME-GC/FID) included chloroform, trichloroethene, dibromomethane, 1-chloro-2- bromopropane, dibromochloroethane, chlorobenzene, bromoform, 1,2,4-trichlorobenzene and 1,2- dichlorobenzene. Sample preparation was carried out using solid-phase microextraction (SPME). Air samples were analysed in the lab in Montreal using a cold-trap system for analyte preconcentration and a GC/FID system for analysis. We would like to discuss the implication of these results for snow-atmosphere exchange processes.
A53B-1150
Salt Distributions in the Sea-Ice Snowpack and Implications for Arctic Halogen Activation
Reactive halogens, derived from sea salt, are important oxidizers affecting arctic atmospheric chemistry. Bromine, and to a lesser extent chlorine and iodine, are periodically sourced, particularly during springtime, into the arctic atmosphere in the atomic or halogen oxide radical forms. These species subsequently cause depletions of surface ozone, deposit atmospheric mercury into the snowpack, and significantly change the overall oxidizing capacity and pattern of oxidation within the atmosphere. The process by which reactive halogens are released into the atmosphere is not understood, but has been linked to heterogeneous reactions on frozen surfaces that contain salts. Aerosols, frost-flowers, and open water leads have all been proposed as possible atmospheric halogen sources, but recent evidence suggests that snow on sea-ice might be the most important contributor. To better understand the impact of snow on arctic halogen chemistry, atmospherically accessible snow on various types of sea-ice were sampled from the University of Washington Applied Physics Laboratory Ice Station (APLIS 07) Beaufort Sea ice camp. Salt distributions were measured by means of bulk conductivity and ion chromatography. The high bulk salinities found support the idea of snow that has wicked up salts from newly formed sea-ice as a significant atmospheric halogen source. The resulting distributions show distinct differences in snowpack salinities on the different types of underlying sea-ice. The heterogeneous chemistry of halogen activation is likely to not depend upon the bulk salinity but instead on the surface salinity of the snow. From laboratory studies and theoretical considerations, there appears to be a saturation behavior where highly saline samples react at similar rates to less saline samples. Therefore, one needs to understand distributions of salinity and ion contents in snow to predict reactivity with respect to halogen activation. These distributions could be used with surface reactivity data and remotely sensed sea-ice data to model activated halogen inputs into the arctic atmosphere.
A53B-1151
Snow – Air Exchange Processes in Alert, Nunavut
The exchange of volatile organic compounds (VOC) between the snow pack and the atmosphere is not very well investigated. Little is known about the contributions of the snow pack to atmospheric processes, how much material is exchanged and what kind of reactions are taking place at the surface that the snow pack provides. In this study we have determined a wide range of VOC with different chemical functionalties (halogenated, aromatic, ketones) and observed their changing concentrations in snow and air over a period of time. With meteorological data collected on site, we would like to investigate potential causes of concentration changes and determine, how atmospheric and snow pack concentrations are linked.
A53B-1152
GEOSummit Baseline Measurements: Results and Interpretations of Surface Snow Elemental Concentrations
Long term measurements of the Arctic atmosphere and surface snow provide insight to the links between aerosol and snow chemical compositions. Current research activities at the Summit Greenland Environmental Observatory (GEOSummit) include high temporal resolution year-round measurements of DRUM aerosol size and S-XRF elemental composition, ICP-MS trace element measurements of surface snow and snow pits, snow accumulation and spatial variability, and other meteorological and snow properties. Year round surface snow samples allow for better understanding of the magnitude and timing of seasonal cycles in aerosol elemental concentrations deposited from the atmosphere to the surface snow. Several elements exhibit distinct seasonal timing of maximum concentrations found in surface snow samples (e.g., sea salts are largely deposited in the winter with dust predominantly deposited in the spring). In addition, snow accumulation rates were measured over the surface snow sampling period, thus aiding the evaluation of accumulation influences as well as wet and dry deposition. Due to the high temporal sample resolution, unique events that transport dust or pollution from North America and/or Asia can be readily identified. The source regions of these unique events are identified using the Lagrangian Particle Dispersion Model (LPDM) FLEXPART. Preliminary comparisons between surface snow and continuous ice core measurements indicate that the seasonal cycle of many of the elements are well preserved in ice cores. This correlation allows for better understanding and interpretation of the elemental records in ice cores, as well as reconstructing past atmospheric conditions from the elemental records.
A53B-1153
Recent Deposition of Trace Metals to Central (Summit) Greenland as Recorded in 3-Meter Snow Pits
During the summer 2005 and 2006 field seasons at Summit (3270 m) Greenland we collected snow core samples for comprehensive geochemical characterization. This sampling effort was one facet of our larger program with the overall objective of improving our understanding of the sourcing and post depositional diagenesis of organic carbon depositing on the Greenland ice sheet. From snow pits of 3-meter depth, representing ~4 years of recent accumulation, detailed profiles of a suite of chemical variables were obtained, including: total and water soluble organic carbon, particulate organic and elemental carbon, inorganic ions, and comprehensive elemental and isotopic analysis. The elemental characterization supports our source reconciliation efforts in providing sub-seasonal data on aerosol particulate matter chemistry from which sourcing vectors can be inferred. Elemental and isotopic analyses on the melted snow cores were carried-out using high-resolution (sector-field) ICP-MS (Finnegan Element 2). A large suite of elements were quantified, including: the major/crustal elements (Al, Ca, K, Fe. Na, Mg, Si), minor crustal elements (Ba, Cs, Li, Rb, Sc, Sr, Ti) light transition metals (Co, Cr, Cu, Mn, Ni, Zn), heavy transition metals (Ag, Cd, Hg, Pb, Tl, W), oxyanion metals (As, Mo, U, V), platinum group metals (Rh, Pd, Pt), rare earths (Ce, Er, Eu, La, Nd, Sm, Y, Yb), as well as, Be, Sb, Sn, sulfur and phosphorus. Very large (>30x) temporal variation in snow core concentrations were measured for Al, Ba, Cr, Cu, Fe, Mg, P, Rb, Sr, Ti, U, Zn and all the rare earths, while low variation (~5x) is observed for the elements As, Cd, Hg, Mo, S and Sn. The later group is representative of the more mobile, anthropogenically dominated/sourced trace metals. Principal crustal elements (Al, Ca, Fe, K, Mg, Na) and sulfate (S) present similar profiles, with significant burial peaks in spring. Major burial peaks are relatively uniformly spaced (~70 cm apart), indicating some consistency in net snow accumulation rates and transport vectors. A large suite of trace elements (Cd, Mn, Ni, P, Pb, Pt, Ti, U) and most rare earths exhibit deposition patterns similar to that of the crustals and S. However, the burial patterns of several elements (Cu, Hg, Sn, Zn) are unique. The Hg profile exhibits summer peaks and is anti-correlated with most other elements, but is correlated with TOC. TOC is, in general though, poorly correlated with most elements, indicating that post-depositional diagenesis may be significant for carbon.
A53B-1154
Atmospheric Deposition of Iron and Phosphorus to Greenland over the 20th- Century
The atmospheric supply of iron (Fe) and phosphorus (P) to nitrogen rich aquatic and terrestrial ecosystems may limit primary production and carbon cycling. During recent centuries, climate change, changes in land use, and industrial activities are thought to have altered the natural variability and atmospheric chemistry of these elements. In order to investigate the sources and perturbation of the atmospheric Fe and P over the 20th century, we have determined Fe and P in ice-cores and snow from the Greenland ice cap. Two ice –cores, D4 (71.4N, 44W) and, Basin 9 (65N, 44.9 W) and snow samples from the Greenland summit were analyzed by high resolution inductively coupled plasma mass spectrometry. Mean iron and phosphorus concentrations over the 20th century were found to be ~ 1 and 0.25 ppb respectively. Over the past 50 years, snow iron concentrations at the ice core sites displayed no significant trend, while phosphorus increased dramatically from 1990 to 2005. Comparison with black carbon records from the same ice cores suggest that boreal forest fires may be the source of the increase in phosphorus.
A53B-1155
Molecular transport and phase transition of polycrystalline ice doped with HCl and SO2 near its melting point
HCl and SO2 are major trace gases in atmosphere, which greatly affects chemical properties of atmospheric ice particles. A particular interest to atmospheric science is the effects of impurities on molecular transport and phase transition at grain boundaries in polycrystalline ice. Effects of doped HCl and SO2 on transport and phase transition at grain boundary of 2-4 micrometer polycrystalline ice were studied using a novel technique - Fast Thermal Desorption Spectroscopy (FTDS) [1] in the temperature range from -2 to -20 deg. C. In these experiments, H2O/D2O/H2O sandwich-like polycrystalline ice films doped with HCl and SO2 were vapor- deposited on the surface of a thin filament positioned in a vacuum chamber and maintained initially at cryogenic temperatures. After the deposition, the temperature of the filament was rapidly raised to a value near ice melting point, thus, initiating rapids H/D exchange reaction at the interface of H2O and D2O layers. Diffusion controlled rate of isotopic exchange in the desorbing film was monitored with a sensitive mass-spectrometer making it possible to gain quantitative information of the extent of diffusion of chemical species along the grain boundaries in polycrystalline ice samples. Comparisons of the experimental results in pure and doped polycrystalline ice show that water self diffusivity at the grain boundaries is significantly enhanced (by at least an order of magnitude) in the presence of HCl or SO2. The strong, non- Arrhenius, dependence of the water self-diffusivity on temperature indicates that this is due to grain boundary premelting [2], which may occur at temperatures as low as 10 deg. below ice melting point. We will discuss implication of this and other results of our FTDS experiments to various environmental phenomena. References: 1. Haiping Lu, Stephanie A. McCartney, M. Chonde, D. Smyla, and Vlad Sadtchenko, Fast thermal desorption spectroscopy study of morphology and vaporization kinetics of polycrystalline ice films, J. Chem. Phys., 125, 044709, 2006. 2. Dash, J.G., A.W. Rempel and J.S. Wettlaufer, The physics of premelted ice and its geophysical consequences, Rev. Mod. Phys., 78, 695, 2006. http://home.gwu.edu/~vlad/
A53B-1156
Investigation of the Factors Influencing Volatile Chemical Fate During Steady-state Accretion on Wet-growing Hail
Phase partitioning during freezing affects the transport and distribution of volatile chemical species in convective clouds. This consequently can have impacts on tropospheric chemistry, air quality, pollutant deposition, and climate change. Here, we discuss the development, evaluation, and application of a mechanistic model for the study and prediction of volatile chemical partitioning during steady-state hailstone growth. The model estimates the fraction of a chemical species retained in a two-phase freezing hailstone. It is based upon mass rate balances over water and solute for accretion under wet-growth conditions. Expressions for the calculation of model components, including the rates of super-cooled drop collection, shedding, evaporation, and hail growth were developed and implemented based on available cloud microphysics literature. Solute fate calculations assume equilibrium partitioning at air-liquid and liquid-ice interfaces. Currently, we are testing the model by performing mass balance calculations, sensitivity analyses, and comparison to available experimental data. Application of the model will improve understanding of the effects of cloud conditions and chemical properties on the fate of dissolved chemical species during hail growth.
A53B-1157
A Method for the Analysis of Ultra-trace Levels of Semi-volatile and Non-volatile Organic Compounds in Snow and Application to a Greenland Snow Pit
The remote polar regions of the earth yield a valuable resource as an historical record of past atmospheric climate conditions. In addition to climatological conditions, natural and anthropogenic pollution records are archived in the ice and snowpack of these areas. Understanding the depositional patterns and post-depositional processes of the various types of compounds preserved in the snow can help us more accurately interpret long- term data extracted from ice cores, as well as observed short-term changes. Organic compound tracers are useful tools for identifying the sources of aerosol contamination, and, while a significant amount of research has been done to identify these tracers in an aerosol format, their presence and fate in environmental media such as snow is less well documented. To gain a better understanding of the compounds reaching the Greenland ice sheet due to long-range transport, as well as their deposition in the snowpack, a method was developed to quantify a suite of organic compounds from snow meltwater by GC/MS. With this method we are investigating post-depositional changes, including photochemical reactions that may impact classes of organic compounds within the snowpack. A 3 meter deep snow pit was sampled at Summit, Greenland during the summer of 2005. From these samples, profiles representing 3-4 years of deposition were developed for a number of organic compound classes including alkanes, alkanoic acids, polycyclic aromatic hydrocarbons, and hopanes. In addition, previously established methods were used to obtain profiles for total organic carbon, metals, and low molecular weight organic acids.
A53B-1158
Photochemistry of PAHs on Arctic Snow Grains
Photochemical reactions can greatly influence the chemistry of surface snow and transform many of the organic pollutants in snowpacks. Little is known about the rates and mechanisms of these photochemical processes, but they likely play a large role in determining lifetimes of many pollutants in snow. Our goal here is to understand the fate of one class of organic pollutants, polycyclic aromatic hydrocarbons (PAH's), in snow and ice. PAH's are ubiquitous in the environment and have well-studied toxicities. We have focused on the degradation kinetics of phenanthrene, pyrene, and fluoranthene on ice, as these are the most abundant PAH's found in Arctic snow. Laboratory photochemistry experiments were conducted using frozen solutions of PAH's illuminated with simulated solar light. Direct photolysis rates were measured using frozen solutions of the PAH's in Milli-Q with added solutes, whereas indirect photo-oxidation reactions with hydroxyl radical were measured in similar frozen solutions with added hydrogen peroxide as a source of ·OH. Using data from our group's previous and ongoing research on the levels of hydroxyl radical in Summit snow, we have been able to begin to assess the relative significance of indirect photo-oxidation and direct photolysis of PAH's as mechanisms of degradation for these compounds on snow and ice. Summit field data indicate that direct photolysis is the main mechanism of phenanthrene degradation, and laboratory experiments confirm this finding for all three PAH compounds. Data extrapolated to Summit conditions suggests that indirect photo-oxidation via ·OH accounts for only a small fraction (about 2 to 4%) of PAH degradation on ice. Our direct photolysis data suggest that PAH lifetimes on snow and ice under Arctic conditions are on the order of 1 to 7 hours during mid-summer. These results are based on photolysis experiments conducted on ice pellets made from PAH solutions. Actual snowpack lifetimes are likely to be longer, given that PAH's are most likely to be associated with particulate matter (e.g, deposited atmospheric aerosols) in the snow.
A53B-1159
C2-C4 alkanes measured in a South Pole ice core: Are atmospheric histories of light hydrocarbons preserved in Antarctic ice?
Non-methane hydrocarbons play a significant role in global atmospheric photochemical system, but we have little knowledge about their atmospheric variability on long time scales. In this study, we analyze ethane (C2H6), propane (C3H8) and n-butane (C4H10) in a South Pole ice core, to examine the potential for using polar ice cores to reconstruct atmospheric histories of these gases. Air was dry- extracted from 124 ice core samples from the SPRESSO core, a 295 m core drilled in 2002 near South Pole as part of the ITASE campaign. The mean gas ages estimated for these samples range from 150 B.C.E. to 1720 C.E. The average mixing ratios for ethane, propane, and n-butane are 798±417 ppt, 234±89 ppt and 117±48 ppt, respectively. Point to point variability generally lies within the estimate of analytical uncertainty. These measurements demonstrate that 1) recoverable amounts of light hydrocarbons exist in polar ice, and 2) there is no evidence of down-core trends to suggest systematic loss or production with time. However, the hydrocarbon levels in this ice core are considerably higher than those in modern air over Antarctica (by factors of 2, 7, and 10 respectively for ethane, propane, and n-butane). The ice core data are not normally distributed, but appear to have a lower limit with superimposed variability. These "baseline levels" are roughly 300 ppt (C2H6), 80 ppt (C3H8), and 40 ppt (C4H10) and are consistent with modern ambient air and firn air levels measured at South Pole. A working hypothesis to explain these results is that alkane levels in ice reflect a combination of two components: 1) entrapped air possibly recording the atmospheric histories of these gases, and 2) some source of alkane contamination that is generated at or near bubble close-off, but does not continue at depth. It is unlikely that the elevated alkane levels reflect contamination during storage, extraction, or analysis.
A53B-1160
A 70 ka Record of Methyl Chloride From a Siple Dome, West Antarctica Ice Core
Methyl chloride (CH3Cl) is most abundant naturally occurring halocarbon in the atmosphere, with a modern mean mixing ratio of 550 parts per trillion (ppt). Prior studies in Antarctic ice and firn suggest that reconstruction of a paleoatmospheric record may be possible (Williams et al., 2007), but there is also some evidence for in situ production in glacial polar ice with high dust content (Saito et al., 2007). In this study, CH3Cl was extracted and analyzed in 42 samples from the Siple Dome, West Antarctic ice core covering the past 70 ka. Over the whole record, CH3Cl mixing ratios ranged from 407 to 819 ppt. Glacial core samples from 30-70 ka had a mean of 455∓46 ppt, significantly lower than the Holocene mean of 532∓61 ppt in this core. There are no data from the period between ~25-33 ka. Four data points around the last glacial maximum between 21-25 ka are higher, with a mean level of 625∓141 ppt. It is unclear whether these higher CH3Cl mixing ratios reflect higher atmospheric levels or in situ production within the ice. The observation of lower CH3Cl levels in glacial ice compared to interglacial ice is in apparent contradiction with earlier measurements from the Dome Fuji, East Antarctica ice core, which showed increased CH3Cl levels in glacial ice compared to interglacial ice (Saito et al., 2007). That core showed a strong correlation between CH3Cl and Ca+2 levels, implicating in situ production related to glacial dust. At Siple Dome, Ca+2 levels are generally lower, and there is no correlation with CH3Cl in the data set as a whole. However, the increased levels of CH3Cl at the last glacial maximum may indicate in situ production in samples with the highest dust loads.
A53B-1161
Ultrafast Thermal Desorption Spectroscopy and Microcalorimetry- New Tools for Aqueous Environmental Chemistry
In order to uncover the fundamentals of environmentally important reactions in ice, an experimental approach is necessary that simultaneously possess two important capabilities. First of all, it must allow control and characterization of aqueous phase structure, and, second, it must have sufficient sensitivity and time resolution in order to detect chemical products and intermediates produced the in the course of these reactions. In the past, these requirement were often satisfied in experiments were reactivity, surface properties, and phase transitions in thin ice films have been studied at cryogenic temperatures with standard surface-science techniques under high vacuum conditions. Unfortunately, these experiments were necessarily confined to temperatures below -100 deg. C due to ice volatility. With the objective of extending applicability of surface science techniques to studies of environmentally relevant processes in ice, we developed recently a new experimental approach that combines fast microcalorimetry with such typical high vacuum analytical techniques as thermal desorption spectroscopy (TDS) and time-of-flight mass spectrometry (TOF MS) while making it possible to study reactions and dynamics in thin polycrystalline ice films at temperatures as high as 0 deg. C. We will discuss the details of our approach, recent examples of the conducted experiments, as well as potential application of our novel technique in the field of Atmospheric Chemistry.
A53B-1162
Probing Chemical Properties of Interstitial Micro-fluids in Ice
Liquid is present as microscopic channels in polycrystalline ice at sub-freezing and even sub-eutectic temperatures. Not only do chemicals tend to concentrate substantially in this microscopic liquid phase, but local physicochemical properties may also differ widely from the bulk counterparts, therefore critically affecting the thermodynamics and kinetics of chemical processes occurring in frozen media such as snow, frost, and frost- flowers. This phenomenon has important implications in atmospheric chemistry such as affecting the composition of the atmospheric boundary layer in snow-covered regions. A method using con-focal laser scanning microscope equipped with a cryostat has been developed to measure physicochemical properties of the microscopic liquid phase in ice that are not readily extrapolated from the bulk data. The experimental setup allows for monitoring the freezing process of an aqueous solution with a sub- second time resolution and a submicron 3D spatial resolution. The physicochemical properties (e.g. viscosity, polarity, and acidity) can, in theory, be deduced from features of the fluorescence spectra of particular fluorescent indicators. For example, the acidity change during the freezing and melting process of electrolyte solutions has been monitored in real time by a pH-dependent dual emission fluorescent probe C-SNARF-1. The effects of temperature, freezing rate, and added electrolytes such as ammonium sulfate, sodium chloride and zwitterions are also examined. The findings complement the theory and previous experimental evidence of freezing hydrolysis.
A53B-1163
Microstructural approach towards prediction of air permeability and gas diffusivity in polar firn
Recent investigations of the air permeability and gas diffusivity of polar firn have shown that these proporties are not solely dependent on firn density, and that prediction of these properties can be improved by considering the effect of firn microstructural geometry. Co-registered measurements of air permeability, gas diffusivity, density and microstructure were made at 10 cm resolution on cores from East Anarctica, in a megadunes area, and West Antarctica, Hercules Dome. In addition, three-dimensional imaging of the firn microstructure usng micro-CT scans of selected firn cores was done. The 3D reconstructions of firn geometry were used in lattice-Boltzmann modeling of air flow through the samples. The measurements and modeling of air flow both indicate that the firn geometry can significantly influence air permeability.
A53B-1164
Methane isotope records from Antarctic firn air
The earth's climate is directly influenced by changes in the atmospheric loading of trace "greenhouse" gases. Methane is an important greenhouse gas whose atmospheric concentration has doubled over the last century as a result of various anthropogenic activities. Understanding the methane cycle in the recent past is a compelling scientific problem because increasing methane levels in the modern atmosphere will contribute to predicted future warming. The only source of air that allows us to study the long-term cycles of these gases is found in the trapped gases in ice cores and the interstitial pore spaces in firn from the central portions of Greenland and Antarctica. Firn air samples, retrieved from the South Pole in 1995 and 2001, were analyzed for δ13CH4. In December of 2005, another suite of firn air samples was recovered from the new US deep coring site located along the west Antarctic divide (WAIS D) which were analyzed for both δ13C and δ D of CH4. In general, replicate flasks were sampled from each depth from two separate drillings each year. The precision of the δ13C and δ D analyses is generally better than ± 0.2‰ and 2‰, respectively, with surface values in good agreement with NOAA (INSTAAR) biweekly flask measurements from South Pole. In all cases, δ13C and δD values decrease gradually with depth below the convective zone. However, these isotopes and CH4 mixing ratios decline sharply below the "lock-in" depth. The measured firn air profiles are primarily the result of three factors. First, tropospheric changes drive compositional changes in the near-surface convective layer that are subsequently mixed downward due to bulk air-movement within the shallow firn. Secondly, gases are mixed below the convective zone solely by molecular diffusion, leading to the gradual downward propagation of the compositional changes in the convective layer. Finally, differential gravitational settling causes the heavier atmospheric constituents (both elemental and isotopic) to accumulate deeper in the firn relative to the lighter gases. We use δ15N of N2 measurements to identify the depth of the convective zone and to correct for gravitational settling. We developed a physically based air transport model to account for the diffusive properties of the firn that allows us to test various historical records spanning the 20th century in a forward modeling exercise. Preliminary results suggest atmospheric δ13C and δ D values near the start of the 20th century were 1.8‰ and 11‰ lower than present day values, respectively. The increasing isotope trends throughout the last century are primarily the result of increased CH4 emissions with elevated isotope signatures (e.g. fossil fuel and biomass burning).
A53B-1165
Methods for radiocarbon measurements of ultra-small samples of paleoatmospheric methane
We present methods developed to perform the first measurements of 14C of methane (14CH4) on air extracted from ancient glacial ice. Air samples containing ~20 μg of CH4 carbon and dating to the last glacial termination were obtained by field melt-extraction from glacial ice outcropping at the Pakitsoq ablation margin in West Greenland. A low methane blank of < 4 parts per billion (ppb) for the field extraction was achieved by applying a chemical polishing treatment to the extraction vessel. CH4 was separated from the air by conversion to CO2 and subsequent cryogenic trapping. The amount of extraneous carbon added in this processing step was reduced to 0.26 ± 0.16 μg through the construction of a new small-volume CH4 conversion line that utilized platinized quartz wool for CH4 combustion. The resulting CO2 was subsequently graphitized and the 14C was measured by accelerator mass spectrometry. The amount of modern (100 pMC) carbon added during the graphitization step was reduced to 0.03 μg through the use of an ultra high-purity iron catalyst. Duration of the graphitization reactions for small (<25 μg C) samples was greatly reduced and reaction yields improved through more efficient water vapor trapping and the use of a new iron catalyst with higher surface area. The overall procedural 14C blank for all stages of sample handling was 0.75 ± 0.39 pMC for ~20 μg, 14C-free air samples with CH4 concentration of 500 ppb, which amounts to ~2.5% of the 14C content of a typical sample. The resulting overall 14CH4 uncertainties for the ancient air samples were ~ 1.0 pMC, or ~ 3% of measured values.
A53B-1166
Kinetic Gas Fractionation by Air Convection in Polar Firn
Recent firn air studies at the Megadunes site in central Antarctica (80.78°S, 124.5°E) show the existence of a 20-m deep convective zone. Large cracks or fissures in the firn likely enable this deep convection (Courville et al., 2007). The existence of fissures appears to be due to near-zero accumulation rate, intense weathering, thermal-contraction cracking, and sublimation. Measured xenon and krypton isotopes are less gravitationally enriched than argon and nitrogen isotopes, and the xenon-nitrogen isotope difference has been explored as a possible paleo-convective zone thickness indicator for use in ice cores (Kawamura et al., 2006). Here we show using first-principles arguments that this phenomenon may be viewed as a special case of a more general type of kinetic gas fractionation. In concept, convective mixing continually disturbs gravitational equilibrium, and gases diffuse back toward equilibrium, with heavier gases diffusing more slowly. Thus a steady state persists in which heavy gases are depleted relative to diffusive equilibrium conditions. The prerequisite for this mode of fractionation is that convective (turbulent) mixing and gravitational unmixing are in competition with each other, such that the effective eddy diffusivity and molecular diffusivity are similar (Prandtl number near 1). The effect on deep firn gases (and hence bubble air) increases with the thickness of the layer in which the Prandtl number is near 1. At Megadunes the magnitude of the effect is 10 per meg for 40Ar/36Ar. The effect may be safely neglected in ice core studies for most biogeochemical purposes.
A53B-1167
Firn-air Properties and Influences at the West Antarctic Ice Sheet Divide
In December 2005, we collected samples of firn air from a pair of dedicated boreholes drilled at the West Antarctic Ice Sheet Divide (WAIS-D), immediately adjacent to the WAIS-D deep ice coring effort currently underway at 79° 28'S, 112° 7'W at an elevation of ~1800m. The site is characterized by moderate temperatures (annual mean of -31°C) and moderate accumulation (24 cm/yr ice-equivalent). These samples were analyzed for a wide variety of atmospheric species by laboratories at the Scripps Institution of Oceanography, NOAA-ESRL, University of Colorado/INSTAAR, UC Irvine and Penn. State University. In this presentation, we focus on general properties of the firn air at this site and the influences on its composition, as inferred from concentration data for CO2, CH4, and a range of halogenated species, as well as the stable isotope ratios of N2 and several noble gases. Preliminary analyses indicate the presence of a shallow convective zone (a few meters or less), a diffusive region extending down to roughly 65m and a lock-in zone from 65m to the firn-ice transition at 76.5m. There is also evidence of a thermally-driven seasonal cycle in composition in the upper 25m of the firn. Modeling studies indicate that the accumulation rate at this site is low enough that the downward advection of air accompanying firn compression has a very small influence on the firn air profile. Air at the bottom of the diffusive column has a CO2-based age of 10-15 years (depending on the definition of "mean age"), while the air at the firn-ice transition is ~38 years old. Concentrations of halogenated species in the samples collected imply atmospheric histories that are generally consistent with those derived from direct atmospheric measurements and from firn air collected at other sites. Additional properties of the air, and their controlling processes will also be presented.