A22C-01 INVITED 10:20h
An assessment of the HOx budget constrained by Summit 2003 field observations
The first observations of HOx at Summit, Greenland have shown elevated summertime values of OH and HO2 which are significantly higher than those measured in the South Pole (SP). The high levels of SP OH are believed to be the effect of the enhanced levels of NO, averaging from 100 to 200 pptv, from snow emissions. In the case of Summit, however, the NO averages are about 5 to 10 times lower those of SP. To better understand the sources and sinks HOx in snow covered polar region, the Summit HOx budget is compared to that of SP. The Summit HOx budget analysis is based on model calculations constrained by in-situ observations of precursors made in early July of 2003. This is a period, characterized by relatively calm wind, when model predicted HO2 is in close agreement with observations. Model estimated OH levels, however, are about 2 times lower than the observations. Summit summertime (early July) ambient temperature and dew point are, on average, about 10 and 20 degrees higher those of SP (December averages), respectively. As a result of these differences, the reaction of O(1D) + H2O is estimated to be the largest HOx source, instead of CH4 oxidation as concluded in SP studies. Contributions from snow emissions of HOx precursors, especially HONO, are significant but relatively less important for Summit than for SP. Calculations are carried out to evaluate the sensitivities of the total HOx source and sink to each of the snow emitted precursors. As for the HOx losses, the reaction of OH + NO2, like in SP, remains to be the major loss pathway. By contrast, loss through formation of HO2NO2 is much less significant in Summit due to the higher ambient temperature. The potential impact of halogen chemistry on Summit HOx budget is also evaluated. Finally, the consequence of Summit HOx chemistry is evaluated in terms of its impact on NOx lifetime and ozone levels.
A22C-02 INVITED 10:35h
Measurements of OH and HO2 + RO2 at Summit Greenland
A photochemistry experiment was conducted at Summit, Greenland during summer 2003 . Measurements of OH, HO2 + RO2, NO, O3, H2O, HONO, CH2O, HOOH, j values, soluble bromide, and a large suite of hydrocarbons were performed a few meters above the snowpack. Noon time values of OH and HO2 + RO2 were observed to range from 5-20 E6 and 2-5 x E8 molec. cm-3, respectively. For most conditions the HO2 + RO2 values were within 20% of calculations from a photochemical box model. However, the OH levels were consistently higher than predicted especially during times of high winds and blowing snow. These results suggest that there is a an significant mechanism for converting HO2 to OH in this environment that is unaccounted for at the present. We hypothesize that halogen chemistry may help account for this and other observations in the Summit environment.
A22C-03 10:50h
Measurements of snow grain hydroxyl radical at Summit, Greenland
Sunlit snowpacks release a number of volatile organic compounds (VOCs) such as formaldehyde and other carbonyls, carboxylic acids, alkenes, and alkyl halides. It has been hypothesized that this flux of VOCs to the overlying atmosphere is in part due to reactions of hydroxyl radical (OH) with snowgrain organic matter. Recent laboratory measurements by Grannas et al. support this idea by showing that the photolysis of polar snow releases formaldehyde, and that this release is enhanced by the addition of nitrate, a photochemical source of OH. In addition to its effects on organic chemistry, OH is probably also important in other snowpack reactions such as the oxidation of halides to form volatile, reactive gaseous halogens. However, the possible role of OH in these reactions has not been quantified. To begin to address the importance of OH in snowpack chemistry, we have measured the photochemical formation of hydroxyl radicals on snow grains at Summit, Greenland during the spring and summer. Measurements were made using a chemical probe technique where benzoate is added to the snow sample in order to scavenge OH and convert it into p-hydroxybenzoate, which is measured by HPLC. We found that OH is formed on snow grains during both seasons and that the rate of formation in the summer was more than an order of magnitude greater than the typical springtime value. Expressed on a bulk (melted) snow volume basis, the average summer value was approximately 200 nM/hr. Assuming that this reactivity occurs within a snowgrain "quasi-liquid layer" (QLL) that represents approximately 0.001% of the bulk liquid volume, rates of OH photoformation in the QLL are on the order of 10 mM/hr. The possible implications of this enormous rate of OH formation for snowpack chemistry (e.g., for VOC release) will be discussed. We have also examined the relative importance of nitrate and hydrogen peroxide as sources of photoformed OH on snow grains at Summit. Based on quantum yields determined in the laboratory, and measurements of hydrogen peroxide and nitrate in snow pits at Summit, we calculate that hydrogen peroxide is a much greater source of photoformed OH, accounting for approximately 40 times more OH than nitrate. Finally, we will discuss the lifetime of OH on snow grains and the calculated steady-state concentrations of OH in the quasi-liquid layer.
A22C-04 11:05h
Where Does Firn Air Come From?
Understanding linked physical and photochemical interactions in the near-surface snow is required both to assess the impact of the reactions on the composition of the atmosphere above the snow and to understand post-depositional processes for ice core interpretation. Field campaigns in 2003 and 2004 at Summit, Greenland have focused on simultaneous measurements of many chemical species, actinic flux, and temperature in and above the snow pack. The measured concentrations depend on the snow and interstitial air chemical content, and the volume from which samples are collected depends on the stratigraphy and microphysical characteristics of the snow pack as well as flow characteristics induced by the sampler. The layered nature of snow and firn has a large impact on chemical interactions. A critical element of understanding snow pack photochemistry is an understanding of airflow through the snow pack, under both natural conditions and the perturbed state during sampling. In situ experiments were conducted using SF6 as an inert tracer gas to verify flow paths in the near-surface snow. To gather non-reactive scalar data, vertical arrays of fine-gauge thermocouples were installed to record firn temperatures, both for model verification and for examining temperature versus photochemical effects. Finite element calculations using measured firn properties and boundary conditions were conducted for computation of flow field and resulting advective-diffusive temperature profiles a variety of conditions corresponding to the conditions of the group sampling. Measurements and modeling are shown for several scenarios, including group shading experiments and experiments conducted over long times in the near surface as well as deeper in the firn. The model results compare well with measured temperatures. Possible effects of stratigraphy, flow paths and flow rates on chemical measurements are discussed.
A22C-05 11:20h
Photochemical HCHO and H$_{2}$O$_{2}$ Processing in Snow at Summit, Greenland, and at South Pole
Heterogeneous photochemistry and temperature-driven recycling of formaldehyde (HCHO) and hydrogen peroxide (H$_{2}$O$_{2}$) in snow can significantly alter the composition of both the snow and the overlying atmospheric boundary layer. This has important consequences for the interpretation of ice-core records to understand the past oxidizing capacity of the atmosphere, as well as for the basic understanding of gas-phase photochemistry above snowpacks. Previous field and laboratory experiments combined with physically based air-snow transfer modeling showed that temperature-driven uptake and release explains at least 75% if not all of the observed net snow-air fluxes of both species at Summit, Greenland and South Pole, Antarctica. Other studies indicated that, at least for HCHO, some of the observed net flux is due to photochemical production in the snow. However, its actual contribution to the total observed flux from snowpacks into the boundary layer has not yet been quantified. Here we present new HCHO and H$_2$O$_2$ data from 2 field campaigns at Summit, Greenland, in summer 2003 and spring 2004 and from the ANTCI field campaign at South Pole in December 2003. Both species were measured in ambient air, and in firn air drawn from various depths in the snowpack while a large area of the snowpack (4m$^{2}$) was intermittently shaded. Filters of various transmittances of UV radiation were used during the shading experiments in order to separate the impact of changing temperature and radiation on the firn-air mixing ratios. In addition, profiles of temperature, radiation, and concentrations of HCHO and H$_{2}$O$_{2}$ in the snow phase were measured. Combined with new modeling studies, these data are used to quantify the relative importance of both, UV and temperature on firn air mixing ratios and snow-air fluxes for the wide range of conditions encountered at both locations and during different times of the year.
A22C-06 11:35h
Measurements of Short Chain (C$_{1}$-C$_{5}$) Alkyl Mononitrates and Related Compounds in North GRIP Firn air.
The paleoclimatic history of tropospheric ozone and nitrogen oxides are not well known and cannot be measured in polar firn air. However, alkyl nitrates can be used as tracers of tropospheric ozone and nitrogen oxide (NO$_{x}$) levels as they are formed in a minor channel during the oxidation of NO to NO$_{2}$ by peroxy radicals. Ozone plays a critical role in the atmosphere, and is present in the troposphere through transport from the stratosphere and as a result of in situ photochemistry involving the oxides of nitrogen (NO and NO$_{2}$). It is important for both atmospheric photochemistry and the global radiation balance through its roles as a precursor for the hydroxyl radical and as a strong infrared absorber. The anthropogenic perturbation of these species is critical to understanding contemporary changes occurring within the climate system. We present short chain alkyl mononitrate (C$_{1}$-C$_{5}$) measurements from firn air recovered from North GRIP ($75.1\deg$N, $42.4\deg$W) in central Greenland and compare to previous measurements made at two Antarctic sites, Dome Concordia ($75.1\deg$S, $123.4\deg$E) and Dronning Maud Land ($75.0\deg$S, $65.0\deg$E). The difference between the two hemispheres is distinct and pronounced with the northern hemisphere showing the greatest variations. In the northern hemisphere, the butyl and pentyl nitrates (C$_{4}$-C$_{5}$) show interesting reconstructed paleoclimatic trends that illustrate steep growth from the beginning of the last century before levelling off in more recent times. We consider the implications of this data for the composition and photochemistry of the troposphere over the last century.
A22C-07 11:50h
Mercury Concentrations in Coastal and Inland Snow In Arctic Alaska
Elevated mercury concentrations in arctic coastal and inland snow have been reported at remote locations. This mercury is attributed to mercury depletion events (MDEs) where gas-phase mercury is depleted due to photochemical conversion to reactive gaseous mercury that is subsequently deposited to snow and ice surfaces. The high mercury concentrations are generally observed near the coast, suggesting that sea ice or leads play a role in MDEs. We sampled the snow pack during 250-km over snow traverses from the central Brooks Range north to the sea ice off of Barrow, Alaska in the spring of 2002 and in the spring of 2004. Our goal was to gain a better understanding of the spatial pattern of mercury concentrations in surface snow. Mercury concentrations in surface snow more than 200 km inland from the Arctic Ocean coast ranged from 5 to 25 ng/L. Samples of a layer of recently deposited snow between 75 km inland from Barrow and Barrow yielded mercury concentrations that ranged from 25 to 147 ng/L and were not correlated with distance from the coast. Snow samples obtained from locations on the sea ice yielded far greater mercury concentrations than terrestrial snow. Mercury concentrations of as high as 820 ng/L were measured in surface hoar snow forming at the edge of a 200 meter wide open lead. Recently formed frost flowers on nilas ice at the edge of a refreezing lead contained mercury in concentrations ranging from 150 to 185 ng/L. Our work yields two important results. First, the relationship between mercury concentrations in snow and distance from the coast is complex and non-monotonic. Second, large leads, those with convective plumes that drive vapor transport, most likely play an important but unresolved role in lower tropospheric MDE chemistry near the Arctic Ocean coast. The existence of highly elevated mercury concentrations near leads suggests the need for a more detailed investigation of the chemical and physical regime near leads.
A22C-08 12:05h
Controls on the isotopic composition of reactive nitrogen species
The importance of sunlit snow surfaces in influencing reactive nitrogen chemistry is now well known. However, quantitatively separating the competing roles of transport, air-to snow transfer, and snow-to-air transfer remains challenging. Isotopic measurements represent a powerful new tool for addressing this problem. Recent innovations in analysis techniques - notably, the use of bacterial denitrification to convert nitrate to $N_{2}O$ for mass spectrometry - allow for the measurement of isotope ratios in snow and in gas-phase and aerosol species. To illustrate the potential utility of such measurements, we present a simple model of the controls on the nitrogen and oxygen isotopic composition of nitrate in remote areas such as central Greenland and west Antarctica. Given a source of $NO_{x}$ (e.g. PAN thermal decomposition) with a particular isotopic composition, isotopic variability in gas and aerosol phase $HNO_{3}$ is determined by the balance of "day" vs. "night" nitrogen oxide chemistry. With photochemical steady state, high local OH concentrations and low local $NO_{2}/NO_{x}$ ratios in summer, high $^{15}N/^{14}N$ ratios are found in $HNO_{3}$ due to the preference for the heavier isotope in $NO_{2}$. Under the same conditions, $^{18}O/^{16}O$ and $^{17}O/^{16}O$ ratios are low, due to low ratios in OH. In winter, $^{15}N/^{15}N$ ratios are low, but $^{18}O/^{16}O$ and $^{17}O/^{16}O$ high, due to the very high ratios characteristic of $O_{3}$, which is the source of virtually all $HNO_{3}$ oxygen during this season. Using output from the GEOS-CHEM global chemical transport model to estimate $NO_{x}$, $NO_{y}$, OH and $O_{3}$ concentrations, and published estimates of the isotopic composition of these compounds, we compare our results with measured seasonal profiles of nitrate in snow pits from Summit, Greenland, and Byrd Station, West Antarctica. Our model accommodates all of the observed variability in oxygen isotope ratios at both sites, but roughly 50% of the observed nitrogen isotope variability is unexplained and must reflect changes in the source values. These results suggest that measurements of both nitrogen and oxygen isotope ratios can potentially be used to separate variability in the sources of $HNO_{3}$ from local photochemistry.