Amanda Grannas

Processes and properties of snow–air transfer in the high Arctic with application to interstitial ozone at Alert, Canada

Abstract Recent measurements of reactive chemical species in snow and firn at polar sites have served to underscore the importance of air–snow transfer processes in understanding changes in atmospheric chemistry. In this paper we present the first quantitative assessment of the impact of physical processes in the snow on air–snow chemical exchange of ozone. Measurements of snow properties, interstitial ozone concentrations, and an ozone kinetic depletion experiment results are presented along with two-dimensional model results of the diffusion and ventilation processes affecting gas exchange at Alert, Nunavut, Canada. The Arctic snowpack at Alert will allow rapid exchange of gases with the atmosphere. Even under natural ventilation conditions with moderate winds, the entire pack is exposed to air movement and therefore available for chemical exchange processes. Both measurements and model results indicate that ozone undergoes rapid depletion in the top centimeters of the snow—approximately within the top 5xa0cm under diffusion alone, and in the top 10xa0cm or less during ventilation. Due to the higher permeability of the snowpack on the sea ice site as compared to the terrestrial site, it is possible that chemical exchange processes could be even more rapid over the sea ice in the greater Arctic than at the terrestrial site. A quantitative discussion of complications that arise from current firn air sampling techniques is presented and possible improvements for future measurements are suggested.

Molecular halogens before and during ozone depletion events in the Arctic at polar sunrise: concentrations and sources

Abstract The molecular halogens Br2, BrCl and Cl2 were monitored from 9 February to 13 March 2000 as part of the ALERT 2000 campaign to investigate the causes of ozone depletion at polar sunrise. The measurements were performed over the transition period from winter to spring in the high Arctic, at Alert, on northern Ellesmere Island in Nunavut, Canada. The measurement campaign for these species covered the period from 24-h darkness, at the beginning of the campaign, to several hours of direct sunlight per day at the end of the campaign. The halogen measurements were made by atmospheric pressure ionization tandem mass spectrometry, using multiple isotopes for each species, and reporting a 20-s average for each species every 2xa0min. Bromine was observed above the 0.2xa0ppt detection limit throughout the campaign at mixing ratios up to 27xa0ppt. BrCl was not observed above its 2xa0ppt detection limit until mid-way through the campaign, but was present almost continuously thereafter, and reached levels of 35xa0ppt. Molecular chlorine was not observed above its 2xa0ppt detection limit. During periods of ozone depletion, there was a very strong inverse relationship between O3 and Br2, and a moderately strong inverse relationship between O3 and BrCl. The slopes of linear regressions of Br2 and BrCl vs. O3 indicate ≈1xa0ppb decrease in O3 mixing ratio for every ppt of either of the molecular halogens. In some cases, O3 depletion events seemed to be triggered by bursts of the halogen species initiated by photochemical processes, even in very weak “twilight”. In other cases, ozone depletion observed at Alert appeared to result from transport of O3-depleted, halogen-enriched air from other locations.

Snowpack processing of acetaldehyde and acetone in the Arctic atmospheric boundary layer

Abstract Acetaldehyde (CH3CHO) and acetone (CH3C(O)CH3) concentrations in ambient air, in snowpack air, and bulk snow were determined at Alert, Nunavut, Canada, as a part of the Polar Sunrise Experiment (PSE): ALERT 2000. During the period of continuous sunlight, vertical profiles of ambient and snowpack air exhibited large concentration gradients through the top ∼10xa0cm of the snowpack, implying a flux of carbonyl compounds from the surface to the atmosphere. From vertical profile and eddy diffusivity measurements made simultaneously on 22 April, acetaldehyde and acetone fluxes of 4.2(±2.1)×108 and 6.2(±4.2)×108xa0moleculesxa0cm−2xa0s−1 were derived, respectively. For this day, the sources and sinks of CH3CHO from gas phase chemistry were estimated. The result showed that the snowpack flux of CH3CHO to the atmosphere was as large as the calculated CH3CHO loss rate from known atmospheric gas phase reactions, and at least 40 times larger (in the surface layer) than the volumetric rate of acetaldehyde produced from the assumed main atmospheric gas phase reaction, i.e. reaction of ethane with hydroxyl radicals. In addition, acetaldehyde bulk snow phase measurements showed that acetaldehyde was produced in or on the snow phase, likely from a photochemical origin. The time series for the observed CH3C(O)CH3, ozone (O3), and propane during PSE 1995, PSE 1998, and ALERT 2000 showed a consistent anti-correlation between acetone and O3 and between acetone and propane. However, our data and model simulations showed that the acetone increase during ozone depletion events cannot be explained by gas phase chemistry involving propane oxidation. These results suggest that the snowpack is a significant source of acetaldehyde and acetone to the Arctic boundary layer.

Atmospheric chemistry of formaldehyde in the Arctic troposphere at Polar Sunrise, and the influence of the snowpack

The role of formaldehyde in the atmospheric chemistry of the Arctic marine boundary layer has been studied during both polar day and night at Alert, Nunavut, Canada. Formaldehyde concentrations were determined during two separate field campaigns (PSE 1998 and ALERT2000) from polar night to the light period. The large differences in the predominant chemistry and transport issues in the dark and light periods are examined here. Formaldehyde concentrations during the dark period were found to be dependent on the transport of air masses to the Alert site. Three regimes were identified during the dark period, including background (free-tropospheric) air, transported polluted air from Eurasia, and halogen-processed air transported across the dark Arctic Ocean. In the light period, background formaldehyde levels were compared to a calculation of the steady-state formaldehyde concentrations under background and low-ozone conditions. We found that, for sunlit conditions, the ambient formaldehyde concentrations cannot be reproduced by known gas-phase chemistry. We suggest that snowpack photochemistry contributes to production and emission of formaldehyde in the light period, which could account for the high concentrations observed at Alert.

A study of photochemical and physical processes affecting carbonyl compounds in the Arctic atmospheric boundary layer

Abstract Experiments were conducted during the ALERT 2000 field campaign aimed at understanding the role of air–snow interactions in carbonyl compound chemistry and the associated ozone depletion in the atmospheric boundary layer. Under sunlit conditions, we find that formaldehyde, acetaldehyde and acetone exhibit a significant diel cycle with average ambient air concentrations of 166, 53 and 385xa0ppt, respectively. A box model of Arctic surface layer chemistry was used to understand the diel behavior of carbonyl compound concentrations at Alert, Nunavut, Canada, with a focus on the chemical and physical processes that affect carbonyl compounds. Results of the study showed that the measured carbonyl compound concentrations can only be simulated when a radiation-dependent snowpack source term (possibly photochemistry) and a temperature-dependent sink (physical uptake on snow grains) of carbonyl compounds were added to the model. We are able to simulate the concentration and amplitude of the observed diel cycle, but not the phase of the cycle. These results help confirm the importance of snowpack chemistry and physical processes with respect to carbonyl compound concentrations in the Arctic surface boundary layer, and reveal weakness in the details of our understanding.

Acetaldehyde and acetone in the Arctic snowpack during the ALERT2000 campaign. Snowpack composition, incorporation processes and atmospheric impact

Abstract Acetaldehyde and acetone were measured in the seasonal snowpack near Alert (Nunavut, Canadian Arctic) in February and April 2000. Acetaldehyde concentrations in fresh surface snow in February were about 2.5xa0ppbw, decreasing to 1xa0ppbw after several days, while gas-phase acetaldehyde was about 75xa0pptv. Values for aged layers were 1.3–2.6xa0ppbw. In April, values for fresh snow were 5–10xa0ppbw, decreasing to 1–4xa0ppbw after several days (gas-phase values were around 230xa0pptv). Values for aged layers were in the range 0.7–3xa0ppbw. Snow-phase acetaldehyde represented 67% of the (snow+atmospheric mixing layer) system in winter and 94% in spring. Preliminary acetone measurements yielded values 1.5–3xa0ppbw in surface snow several days after deposition. To understand the kinetics of exchange of acetaldehyde between the air and the snow, its mechanism of incorporation in snow was investigated. Surface incorporation by adsorption, and volume incorporation by dissolution were considered. Winter and spring measurements showed very different trends, and spring concentrations were higher than winter ones, which is contrary to thermodynamic expectations. The photolytical production of acetaldehyde in the snowpack is proposed as an explanation.