HDO in the Mesosphere: Observation and Modeling of [HDO]/[H2O] Variability

We present 1992-2002 measurements of HDO and H₂O at 50-70 km altitude, derived from ground-based μwave observations at 32°N, 112°W. Observed HDO/H₂O ratios show HDO depletions relative to Standard Mean Ocean Water (SMOW) of δ D=-58% to -3%. Large variability of mesospheric δ D is in surprising contrast to stratospheric observations of δ D=-65% to -50% [Moyer et al., 1996], and indicates sensitivity to processes other than the CH₄ oxidation that pertains in the stratosphere. We observe anticorrelation of δ D and H₂O abundance, contrary to the stratospheric pattern. Observations are of the 225 GHz HDO and 203 GHz H_2¹⁸O lines, corresponding to rotational transitions of these molecules. The 12-meter radio telescope and T=4^oK receivers at Kitt Peak, AZ are used in frequency switching mode. HDO and H₂O altitude profiles are derived from sensitivity of the line shape to pressure. H_2¹⁸O is used as a proxy for H_2¹⁶O based on theory and available measurements [Kaye, 1987; Rinsland et al., 1991] of [H_2¹⁸O/H_2¹⁶O] = SMOW+/- 5% in the upper stratosphere. Standard theory holds that HDO is preferentially removed from water transported upward through the tropopause due to isotope-dependent freezing [Kaye, 1987], leading to lower stratosphere HDO depletions -65+/-10% [balloon- Rinsland et al., 1991; ATMOS- Moyer et al. 1996]. Depletion is less extreme in the upper stratosphere owing to conversion of CH₄ (D/H ~ SMOW) to H₂O. We model the photochemistry and conclude the observed mesospheric HDO behavior is driven by differing photolysis rates for HDO and H₂O at λ > 175 nm.