The evolution of our solar system remains one of the most fascinating questions since ancient astronomers first realized that the "stars" which do not twinkle are planets. In the modern era, the future evolution of the Earth has become a topic that may arise even in lay conversations. For example, the effects of changes in the earth's climate and the ozone layer are debated in policy and economic discussions as well as in scientific forums. One method for assessing our understanding of the Earth's atmosphere is to examine the atmospheres of the other terrestrial planets. The atmospheres of Venus and Mars share some common features with that of the Earth, but there are important differences. These differences are important because they force us to examine the often implicit assumptions that lie within the tunable parameters that exist in all models. Consequently, studies of the atmospheric chemistry on Venus and Mars can provide a test of the degree to which our understanding of the fundamental physical and chemical principles that govern the Earth's atmosphere is correct.
The chemistry of the Martian atmosphere has been "solved" to the extent that modern photochemical models [e.g., Nair et al. 1994] can reproduce the primary observable characteristics to within a factor of 2 - 3 using the results from laboratory studies of photochemical reaction rates. The chemistry of the Venus atmosphere, however, remains an "unsolved" problem in that the differences between the most recent generation of models [Yung and DeMore 1982] and some of the existing observations are greater than a factor of 10. The first part of this dissertation (Chapters 1, 2, and 3) combines several lines of research that focused on how the discrepancy between models and observations of the Venus atmosphere can be reduced. (The appendices for Chapter 2 are in Chapter 5.) The results provide a better understanding of the current photochemical processes in the Venus atmosphere and can serve as a baseline for future studies of its evolution.
The intervening chapter reports observations of Europa and Callisto in the 8-l3μm wavelength region. Those observations provide constraints on the composition and/ or physical state of the surficial material on the two satellites.
Part I (Observations and Photochemical Modeling of the Venus Middle Atmosphere)
The primary photochemical cycle of the Venus middle atmosphere (within and above the upper cloud layer) is the photolysis of C02 to form CO and oxygen atoms on the dayside followed by the reformation of C02 from CO and oxygen, primarily via catalytic reactions. Previous photochemical models-using ClOx, SOx, NOx, and HOx radicals to catalyze the reformation of C02-could qualitatively explain the stability of Venus' C02 atmosphere. Despite the powerful catalytic reactions introduced, however, none of the previous models could quantitatively explain either the low column abundance of molecular oxygen or the intense nightside air glow in the 02(a1Δ) band. The most comprehensive of the previous models, that by Yung and DeMore [1982], predicted a column abundance of molecular oxygen that was a factor of ~30 larger than the upper limit obtained the following year [Trauger and Lunine 1983], and it predicted a nightside airglow that was about three-quarters of the observed intensity. These discrepancies suggested that significant gaps remained in our understanding of the dominant chemical processes in the Venus middle atmosphere.
In the fifteen years since these studies, new observations have provided further insight into the current state of the Venus atmosphere, and our understanding of chlorine and sulfur chemistry has improved as a result of laboratory and field studies related to the terrestrial stratosphere. One idea that had been proposed is that the abundance of 02 in the Venus atmosphere might vary by as much as a factor of 10 over time scales of years. This proposal was based on observations of S02 and SO at the top of the Venus clouds which showed a monotonic decrease in the abundances of both species from 1980 to 1995 with S02 declining by a factor of 10 and SO by a factor of 5. An alternate proposal was that rapid reactions on aerosol particles, such as
HOCl + HCl +aerosol-+ H20 + Cl2 + aerosol
which converts Cl from a relatively nonreactive form, HCl, to a potentially highly reactive form, Cl2, could greatly enhance the effectiveness of the chlorine catalytic reactions that oxidize CO to C02.
This project has attempted to reconcile the differences between observations and photochemical models with regard to the oxygen budget of the Venus middle atmosphere. To examine the potential temporal variability of the 02 abundance, we attempted to detect molecular oxygen on the dayside of Venus using the Ultra-High Resolution Facility at the Anglo-Australian Telescope on 23 January 1995. We observed the 763.6325 and 763.2165 nm 02 lines at a spectral resolving power, λ⁄⁄λ, of 600000 with a square, 1.5 arc-second field of view. These observations reveal no evidence for 02 above the Venus cloud tops within our detection limit which is equivalent to a uniform vertical mixing ratio of 3 ppm. Our upper limit on the abundance of 02 in the Venus middle atmosphere is comparable to that obtained by Traub and Carlton [1974] and is a factor of ten larger (less stringent) than that obtained by Trauger and Lunine [1983] (2σ = 0.3 ppm). Within the obvious limitations imposed by the relative sensitivities of the non-detections, we find no evidence for an increase in the abundance of 02 in association with the observed decreases in S02 and SO abundances.
The parallel branch to new observations was development of a new photochemical model for the Venus middle atmosphere that incorporates the most recent chemical kinetic rates and photoabsorption cross sections. Our one-dimensional, steady-state photochemical model can reproduce (within measurement uncertainty and temporal⁄spatial variability) the retrieved SO profile [Na et al. 1994], the retrieved S02 abundance and scale height at the cloud top [Na et al. 1994], the retrieved CO profile [Clancy and Muhleman 1991], and the observed "global average" 02 (a1Δ) airglow [Crisp et al. 1996] using only gas-phase chemistry if we adjust key reaction rates within their assessed one standard deviation uncertainties. Our predicted column abundance for 02 is a factor of 10 smaller than than from previous models. Our results suggest SClx and ClSOx compounds may be important in the chemistry of the upper cloud layer, and we find that Reaction G10 is not an important source of Cl2 above the cloud top.
The current observational upper limit can be reconciled with the predictions from our photochemical model if one examines the absorption produced by our predicted 02 distribution instead of the column abundance. Our model predicts that most of the 02 in the Venus atmosphere will be located between 85 and 95 km altitude, several scale heights above the cloud top (~70-75 km altitude) and, thus, above the region in which multiple scattering is important. Radiative transfer model calculations indicate that the absorption due to our predicted 02 distribution is equivalent to that produced for a uniform vertical distribution of 02 with mixing ratio of 0.3 ppm, the two standard deviation upper limit from the observations. If the equilibrium constant for ClCO is adjusted by two standard deviations and the temperatures at 85 - 95 km altitude are decreased by ~15-20 K (two standard deviations), the predicted 02 absorption will lie below the observed one standard deviation upper limit.
Part II (Thermal Infrared Spectroscopy of Europa and Callisto)
The trailing hemispheres of Europa and Callisto were observed and a 9- 13 μm spectrum of Europa with better spectral resolution and better signal-to-noise than was previously possible has been derived. The ratio spectrum of the two satellites has a signal-to-noise ratio of approximately 30 and spectral resolving power of approximately 50. The ratio spectrum was combined with the average Voyager 1 spectrum of Callisto from Spencer [1987a] to obtain a 9- 13 μm spectrum of Europa with signal-to-noise that is a factor of 10 better than that in the average Voyager spectrum of Europa in Spencer [1987a]. No emissivity features due to water ice are apparent at the 3% level in our Europa spectrum. The disk-integrated, effective color temperature ratio for the two satellites is consistent with previous ground-based, broadband, thermal infrared photometry. One possible explanation for the absence of features in the thermal infrared spectra of Europa (at the 3% level) and Callisto (at the 1% level) is if the surfaces of both satellites have significant abundances of small particles (≲50 micrometer in size). This explanation is consistent with most of the published observations by Galileo.
