Articles | Volume 12, issue 10
Research article
24 May 2012
Research article |  | 24 May 2012

The stable isotopic composition of molecular hydrogen in the tropopause region probed by the CARIBIC aircraft

A. M. Batenburg, T. J. Schuck, A. K. Baker, A. Zahn, C. A. M. Brenninkmeijer, and T. Röckmann

Abstract. More than 450 air samples that were collected in the upper troposphere – lower stratosphere (UTLS) region by the CARIBIC aircraft (Civil Aircraft for the Regular Investigation of the atmosphere Based on an Instrument Container) have been analyzed for molecular hydrogen (H2) mixing ratios (χ(H2)) and H2 isotopic composition (deuterium content, δD).

More than 120 of the analyzed samples contained air from the lowermost stratosphere (LMS). These show that χ(H2) does not vary appreciably with O3-derived height above the thermal tropopause (TP), whereas δD does increase with height. The isotope enrichment is caused by H2 production and destruction processes that enrich the stratospheric H2 reservoir in deuterium (D); the exact shapes of the profiles are mainly determined by mixing of stratospheric with tropospheric air. Tight negative correlations are found between δD and the mixing ratios of methane (χ(CH4)) and nitrous oxide (χ(N2O)), as a result of the relatively long lifetimes of these three species. The correlations are described by δD[‰]=−0.35 · χ(CH4)[ppb]+768 and δD[‰]=−1.90· χ(N2O)[ppb]+745. These correlations are similar to previously published results and likely hold globally for the LMS.

Samples that were collected from the Indian subcontinent up to 40° N before, during and after the summer monsoon season show no significant seasonal change in χ(H2), but δD is up to 12.3‰ lower in the July, August and September monsoon samples. This δD decrease is correlated with the χ(CH4) increase in these samples. The significant correlation with χ(CH4) and the absence of a perceptible χ(H2) increase that accompanies the δD decrease indicates that microbial production of very D-depleted H2 in the wet season may contribute to this phenomenon.

Some of the samples have very high χ(H2) and very low δD values, which indicates a pollution effect. Aircraft engine exhaust plumes are a suspected cause, since the effect mostly occurs in samples collected close to airports, but no similar signals are found in other chemical tracers to support this. The isotopic source signature of the H2 pollution seems to be on the low end of the signature for fossil fuel burning.

Final-revised paper