5.6. MEASUREMENT OF AIR FROM SOUTH POLE
As part of a cooperative venture with scientists from the University of Rhode Island, Pennsylvania State University, and CCG, NOAH analyzed the contents of flasks filled with firn air from the SPO in early 1995 [Battle et al., 1996].
The sampling system was designed so that large quantities of air could be pulled from discrete depths in the firn down to the firn-ice transition depth of 122 m. Because it was possible to obtain large amounts of air, flasks could be flushed adequately and filled for nearly routine air analyses in Boulder and elsewhere. Air at the bottom of the firn had a CO2 age of about 100 years [Battle et al., 1996].
N2O in these samples analyzed by NOAH forms a bridge between ice-core
data, which typically are much less precise owing to sample handling procedures
and small samples, and real-time, present-day measurements (Figure 5.32). These
results suggest that preindustrial levels of N2O in the atmosphere
had to be about 280 ppb and that N2O was increasing steadily through
the latter part of the 20th century. The growth rate of atmospheric N2O
from 1904 through 1958 was 0.06 0.01% yr-1 (95% confidence level);
thereafter, it has increased at a rate of 0.22 0.02% yr-1 (95% C.L.).
N2O covaried well with CO2 throughout the profile, although
the smoothness of the fit could be attributable to subsurface diffusion of the
gases. Nevertheless, the overall trend of N2O as a function of CO2
was 0.50 0.03 ppb N2O ppm-1 CO2-1
(95% confidence level, r2 = 0.98).
Fig. 5.32. History
of atmospheric N2O over the past century, derived from antarctic
ice-core measurements [Machida et al., 1995], real-time air measurements
in the southern hemisphere (NOAH), and analyses of South Pole firn air (NOAH).
Firn air ages are determined from correlation of CO2 in the samples
with the atmospheric CO2 history of Etheridge et al. .
Diffusivities of N2O and CO2 are assumed to be identical
and ice core data of Machida et al.  have been lowered by 1 ppb
to conform to the CMDL scale.
Surprisingly these flasks, which were sealed with Teflon o-rings, did not
cause significant contamination of most halocarbons. Consequently, depth profiles
were obtained of CFCs, chlorocarbons, and bromocarbons representing air as far
back as the late 19th century. (Dates assigned to halocarbons will be older
than CO2 in the same bolus of air owing to their slower rates of
diffusion.) As shown in Figure 5.33, which is a close-up of the lower portion
of the CFC-11 profile, the sampling and analytical precisions are on the order
of tenths of a ppt. Small amounts of contamination are suggested in that the
lowest values were still 2 ppt (<1% of today's atmosphere) and that two pairs
of flasks showed higher levels of CFC-11 in some of the deepest firn. This latter
contamination was probably caused by stress on the pump near the firn-ice transition
zone where less air was available to pull, thus increasing the probability of
sucking in unrepresentative air. This feature showed up in all of the gases,
further suggesting contamination with modern air. Nevertheless, this level of
contamination is not representative of the rest of the profile. Thus, we were
able to obtain precise, but probably accurate, measurements at sub-ppt levels
throughout most of the profile.
Fig. 5.33. CFC-11 in the lower portion of the firn at the South Pole. The
high degree of precision is shown in the actual measurements (triangles), where
two flasks were collected at each depth and plotted separately. Only at 108
m depth are these symbols distinguishable and there only because two pairs of
flasks from this depth in two separate holes were analyzed. Flasks at 120 and
122 m were subjected to some contamination with modern air during sampling,
probably owing to stress on the pump as the firn layers began to turn to ice.
Also shown here is the effect of the gravitational correction for settling of
CFC-11, which is a gas heavier than air. Again, the error is small.
These results yield entire atmospheric histories for CFCs, halons, and other
halocarbons of purely anthropogenic origin (Figures 5.33 through 5.37). They
also showed atmospheric trends for gases of both natural and anthropogenic origin,
such as CH3Br, during a time when the human population, its agricultural output,
and its industrial activity increased dramatically. These data demonstrate that
natural sources of CFCs and halons are minimal at best and most likely nonexistent.
Models of anthropogenic CFC emissions are supported by these findings, confirming
the predominance of anthropogenic activity in the atmospheric budget of refractory,
organic chlorine. The data for CFCs and the longer-lived organic chlorocarbons
ultimately will be useful in dating oceanic water masses and isolated ground
waters, providing atmospheric data where none existed before. Atmospheric CH3Br
in the southern hemisphere appears to have been about 25% lower at the turn
of the century than it is today (Figure 5.36).
Fig. 5.34. Depth profiles of CFC -11 and -12 in South Pole firn air. Mole
fractions of both gases near the bottom of the firn are less than 1% of the
present day values, suggesting that natural sources are minimal or non-existent.
Fig. 5.35. Depth profiles of CCl4 and CH3CCl3
in South Pole firn. There is some evidence of contamination of a few ppt in
the CH3CCl3 data, although mole fractions of this compound
came very close to zero near the bottom of the profile. Mole fractions of CCl4
never fell below 10 ppt, suggesting either significant, specific contamination
of this compound, a very early history of significant anthropogenic release,
or a natural source.
Fig. 5.36. Brominated gases in South Pole firn air. The anthropogenic halon
mole fractions both drop to zero early in the profile. These gases were not
introduced into the atmosphere in significant amounts until the 1970s. CH3Br
is about 6.5 ppt in air nominally dating back to about 1880.
Fig. 5.37. SF6 in South Pole firn air. Because of its small analytical peak
on the GC, the detection limit for SF6 in real air is around a few tenths of
a ppt. Like the halons, this gas was not introduced into the atmosphere until
the 1970s [Maiss et al., 1996].