5.5. Firn Air Measurements

Past success in analyzing for halocarbons in air samples collected from firn (unconsolidated snow) at the South Pole [Elkins et al., 1996a] prompted us to continue these investigations in Greenland and Antarctica. Our initial measurements of South Pole firn air were from low-pressure glass flasks used in collecting air for carbon-cycle gases. Although these measurements seemed reasonable and the samples generally uncontaminated, they could only be run on two of our instruments; GC MS measurements were precluded for the South Pole samples because of a lack of available air. In April 1996 we assisted in the collection of samples from two holes at Tunu, Greenland (78°01’N, 33°59’E) and filled our usual 2.5-L stainless steel flasks to 375 kPa (yielding about 9 L of air) in addition to the glass flasks (150 kPa, yielding only 3.8- L air, most of which is used in analyses for carbon-cycle gases by the CMDL Carbon Cycle Group). Samples were returned to Boulder for analysis on all four instruments used in flask analyses (Table 5.2). Both glass flasks and steel flasks were run on the two GC/ECD systems and steel flasks were run on the two GC/MS systems. Similarly we obtained samples in both steel and glass flasks from deep and shallow holes drilled at Siple Dome, Antarctica (81°40’S, 148°49’W), in December 1996.

Data from these sites showed that CFCs, CH₃CCl₃, CCl₄, SF₆, halons, and HCFCs were essentially absent in the early 20^th century atmosphere (Figures 5.38, 5.39, and 5.40). This is not surprising information, but these samples, which, unlike the initial South Pole samples, were collected in such a way as to avoid low-level contamination of halocarbons, are the first verification of levels of CFCs and the major chlorocarbons that do not differ significantly from zero. The data demonstrate that if natural sources of these gases do exist, they are insignificant.

Fig. 5.38. “Conservative” gases of anthropogenic origin in Tunu firn air. (a) CFC’s, (b) halons, (c) SF₆. CO₂ in air at the bottom of the Tunu, Greenland, profile corresponds to atmospheric CO₂ in the early 1930s.

Fig. 5.39. Chlorocarbons in Siple Dome firn air. (a) CH₃CCl₃ and CCl₄, (b) CH₃Cl. CO₂ in air at the bottom of the Siple Dome profile corresponds to atmospheric CO₂ in the early 1950s.

Fig. 5.40. Replacement compounds for CFCs in Tunu firn air.

Results for methyl halides were more ambiguous, although it still may be possible to derive 20^th century atmospheric histories for them (Figure 5.39b and Figure 5.41). Methyl chloride in the diffusive zone is about 10% higher than it is at the bottom of the profiles suggesting that activities over the past century have elevated the mixing ratio of this gas in the atmosphere by about 50 ppt. Also, in the upper 10-12 m of the firn, CH₃Cl concentrations actually decrease toward the surface by about 30 ppt. This is consistent with seasonal cycles of CH₃Cl which are associated with photochemical cycles of tropospheric OH.

Fig. 5.41. Methyl bromide in firn air. (a) Antarctica, (b) Tunu, Greenland.

Data for CH₃Br from Siple Dome agreed well with those from the South Pole, both suggesting that atmospheric CH₃Br in the earlier part of this century was about 25% lower than it is today (Figure 5.41a). However, at Tunu, Greenland, a warmer and more coastally influenced site, CH₃Br was high near the bottom of the profile, reaching mixing ratios of nearly 50 ppt at the firn-ice transition (Figure 5.41b). Tests confirm that this elevation in concentration at depth is not an artifact of sample collection, storage, or analysis. This leads us to believe that the observed high values for CH₃Br in the firn at Tunu are real, although not necessarily of atmospheric origin. (This feature was also observed at Tunu for at least one other marine biogenic gas, CH₃I, and it was observed for CHBr₃ at Siple Dome where CH₃Br and CH₃I do not appear to have been produced.) A model of the firn profiles strongly suggests that CH₃Br is produced near the firn-ice transition, a process that could not have been happening at South Pole or Siple Dome and yield the observed profiles.

Processes in the upper 10 m of firn also may affect CH₃Br concentrations. At each antarctic site, the CH₃Br concentration was elevated by as much as 1 ppt (10-15%) in samples just below the surface. This feature was consistent, though at present unexplained. Such variations did not appear throughout the profiles which suggests this may be a phenomenon limited to the upper 10 m. How this affects the overall profile is difficult to ascertain without first understanding the process. All firn samplings were conducted during the summer months thus precluding any evaluation of seasonal effects in the surface. In the fall of 1997, flasks were sent to SPO for sampling from a 15 m deep hole in winter and summer of 1998.

[BACK] [CONTENTS] [NEXT]