The ¹⁸O isotope exchange rate between firn air CO₂ and the firn matrix at three Antarctic sites

Abstract

[1] Having measured d 18 O of firn air CO 2 for three Antarctic sites (Dronning Maud Land, Dome Concordia, and Berkner Island), we established that at depth the d 18 O(CO 2 ) values are systematically shifted away from atmospheric values toward isotope equilibrium with H 2 O from precipitation (firn matrix). The d 18 O(CO 2 ) shifts increase with depth and thus with CO 2 age. The equation of a first-order reaction is used to describe and calculate CO 2 oxygen isotope exchange rates. The calculated exchange rates clearly show a correlation with firn temperatures. Assuming that the same mechanism operates at different sites and at different temperatures, a pseudoactivation energy of 7.8 ± 0.8 kcal/mol is estimated. The occurrence of exchange means that firn is not chemically inert in that sense that some form of CO 2 hydration takes place. Possible mechanisms of CO 2 oxygen isotope exchange are discussed and liquid-like layers on firn grains are the most likely interface where this exchange takes place. Other gases (e.g., COS) may likewise be sensitive to hydration or related reactions and this may reduce their integrity in firn. A formal translation of the reaction rates of CO 2 oxygen isotope exchange in firn to a COS hydration rate appears to indicate COS preservation on the scale of tens of thousands of years. We suggest that the value of CO 2 oxygen exchange rate in firn, expressed as half-reaction time T 1 /2, may be used as a new firn characteristic, giving a measure of hydration activity in firn, an indirect measure of firn metamorphism and be correlated with other firn processes near a pore close-off-depth. T 1 /2 values may possibly correlate with the process of He and Ne diffusion/expulsion during the close-off of pores in firn, as well as the release of 14 CO produced in situ in the firn matrix.Citation: Assonov, S. S., C. A. M. Brenninkmeijer, and P. Jöckel (2005), The 18 O isotope exchange rate between firn air CO 2 and the firn matrix at three

High-precision dual-inlet IRMS measurements of the stable isotopes of CO<sub>2</sub> and the N<sub>2</sub>O / CO<sub>2</sub> ratio from polar ice core samples

“…Observations of ice core δ 18 O-CO 2 have previously been reported and discussed from low-resolution measurements from ice cores at Siple Dome, South Pole, and Byrd hereafter Siegenthaler) and high-precision measurements from firn sampling campaigns at Dome C, Dronning Maud Land, and Berkner Island (Assonov et al, 2005); hereafter ABJ). Siegenthaler observed δ 18 O-CO 2 values about 20 to 30 ‰ more depleted than typical atmospheric values and correlated to the variations of δ 18 O in the surrounding ice matrix.…”

A revised 1000 year atmospheric δ13C‐CO2 record from Law Dome and South Pole, Antarctica

“…[124] The range of ı 46 CO 2 of air in ice cores and firn is due to modification with depth from O-exchange with H 2 O [Assonov et al, 2005], which makes the ı 46 CO 2 of air in ice significantly more negative than that of atmospheric air (depending on the age/depth of a sample). For our measurements of modern atmospheric air, firn air, and ice core air against the pure CO 2 reference gas, ı 45 CO 2 varies between -2 (modern air) and 0 (ice) while ı 46 CO 2 varies from +14 (modern air) to -14 (ice).…”

CO2 isotopes as tracers of firn air diffusion and age in an Arctic ice cap with summer melting, Devon Island, Canada

“…With the exception of the anomalous samples at 15.95, 30.33, and 32.85 m, discussed above, the observed depletion trend with depth can be attributed to exchange with the ice matrix (À27.0 ± 1.7% VSMOW). This progressive exchange has been observed by Assonov et al [2005] in firn air from three Antarctic sites, and shown to be a first-order reaction.…”

“…14 CO 2 has diffusive properties that differ only marginally from those of CO 2 owing to its greater mass, and otherwise behaves the same as CO 2 in firn air. However, thermonuclear bomb tests from 1952 to [Assonov et al, 2005]. For Devon, error bars are for l CO 2 -H 2 O = 0.03276 (+0.0062/À0.0078) on the basis of the uncertainty of the depth of the 1963 thermonuclear peak used to calibrate the exchange rate.…”

“…If nonnegligible, this effect would act to reduce the calculated rate constant (discussed below) from that expected for the in situ temperature of Devon firn. However, good comparison with rate constants for three sites with different temperatures in Antarctica [Assonov et al, 2005] suggests that such an effect is negligible for Devon Island (Figure 6). …”