Temporal Behavior of Stratospheric Ammonia Abundance and Temperature Following the SL9 Impacts

Abstract Infrared emission lines of stratospheric ammonia (NH 3 ) were observed following the collisions of the fragments of Comet Shoemaker–Levy 9 with Jupiter in July of 1994 at the impact sites of fragments G and K. Infrared heterodyne spectra near 10.7 μm were obtained by A. Betz et al. (in Abstracts for Special Sessions on Comet Shoemaker-Levy 9, The 26th Meeting of the Division for Planetary Sciences, Washington DC, 31 Oct.–4 Nov. 1994 , p. 25) using one of the Infrared Spatial Interferometer telescope systems on Mount Wilson. Lineshapes of up to three different NH 3 emission lines were measured at a resolving power of ∼10 7 at multiple times following the impacts. We present here our radiative transfer analysis of the fully resolved spectral lineshapes of the multiple rovibrational lines. This analysis provides information on temperature structure and NH 3 abundance distributions and their temporal changes up to 18 days after impact. These results are compared to photochemical models to determine the role of photochemistry and other mechanisms in the destruction and dilution of NH 3 in the jovian stratosphere after the SL9 impacts. One day following the G impact, the inferred temperature above 0.001 mbar altitude is 283±13 K, consistent with a recent plume splashback model. Cooling of the upper stratosphere to 204 K by the fourth day and to quiescence after a week is consistent with a simple gray atmosphere radiative flux calculation and mixing with cold jovian air. During the first 4 days after impact, NH 3 was present primarily at altitudes above 1 mbar with a column density of (7.7±1.6)×10 17 cm −2 after 1 day and (3.7±0.8)×10 17 cm −2 after 4 days. (Errors represent precision.) We obtained >2.5 times more NH 3 than can be supplied by nitrogen from a large cometary fragment, suggesting a primarily jovian source for the NH 3 . By 18 days postimpact, a return to quiescent upper stratospheric temperature is retrieved for the G region, with an NH 3 column density of 7.3×10 17 cm −2 or more in the lower stratosphere, possibly supplied by NH 3 upwelling across an impact-heated and turbulent tropopause, which may have been masked by initial dust and haze. Above the 1-mbar level, the maximum retrieved column density decreased to 6.5×10 16 cm −2 . Comparison to photochemical models indicates that photolysis alone is not sufficient to account for the loss of NH 3 above 1 mbar by that time, even when chemical reformation of NH 3 is ignored. We speculate that the dispersion of plume material at high altitudes (above 1 mbar) is responsible for the change in the spectra observed a few days postimpact. Data on the K impact region provide qualitatively consistent results.