Monday, October 18, 1993 JUPITER I 8:30 - 10:00 AM South Ballroom Chair(s): P. Yanamandra-Fisher D. Gautier West R. A.* Karkoschka E. Tomasko M. G. Cunningham C. Wide-Field Camera Observations of Jupiter's Stratospheric Haze and Ammonia Cloud We observed Jupiter in early March of 1992 with the Wide Field Camera on the Hubble Space Telescope. We obtained images at UV wavelengths (284, 336, and 368 nm), blue (413 nm), red continuum (656 nm), and in the 889-nm methane band. The findings thus far indicate (1) jovian stratospheric haze extinction optical depth increases by about an order of magnitude from a minimum near latitude -40 degrees to maxima in polar regions, (2) there is a hemispheric asymmetry in the haze distribution such that the increase is most rapid between -50 degrees and -70 degrees, and between +30 degrees and +70 degrees, (3) stratospheric haze particles are strongly absorbing in the UV at high latitudes but not at low latitudes, (4) the wavelength dependence of the optical depth is similar to that of 0.16 micron radius spheres at high latitudes, but implies larger particles at low latitudes, (5) extinction optical depth of the underlying tropospheric ammonia cloud is approximately constant between latitudes -60 degrees and +40 degrees, but drops rapidly to near 0 poleward of those latitudes, (6) the north and south polar hood regions seen in the 889-nm images have distinctly different appearances, with considerable longitudinal structure seen only in the north. The loss of ammonia cloud opacity at high latitude correlates well with the magnitude of downwelling at the 270-mbar level indicated by the stratospheric circulation model of West et al. (Icarus 100, 245-259, 1992). Trafton L. M.* West R. A. Atreya S. K. Beebe R. F. Caldwell J. J. Owen T. C. HST Observations of the Spatial Variation of Gases and Aerosols in Jupiter's Atmosphere As part of a coordinated effort to obtain integrated dynamical and spectroscopic observations of Jupiter and Saturn, we have obtained spatially resolved spectra of Jupiter using the HST Faint Object Spectrograph with 0.5" and 1" circular apertures to sample selected regions along the central meridian and near the limb. The spectra span the wavelength range 1800-3200 Angstroms, covered by two gratings, G270H and G19OH, with resolutions from 1.8 -4.6 Angstroms. Various features were targeted in the blind pointing mode (1" pointing accuracy). The mostly northern hemisphere spectra, reduced to local reflectivities, are compared and contrasted. We report interim results for the 1992 apparition, taken in May. Evidence of a stratospheric absorber at 6 N includes limb darkening, which becomes more pronounced at shorter wavelengths. The spectra show a surprisingly marked variation of atmospheric haze content and haze properties with latitude. At low latitudes, a dramatic change is observed over a fairly small latitude interval. The presence of such gradients sets limits on the strength of the meridional circulation. At least in the northern hemisphere, Jupiter's upper atmospheric NH3 appears to be concentrated at lower latitudes. We investigate this in the context of a meridional circulation model. Cunningham C. C.* Caldwell J. J. HST Spectra of the Outer Planets I. The Grating Scatter Problem in the FOS Ultraviolet spectra of red objects, including planets, obtained with the HST Faint Object Spectrograph (FOS) are significantly affected below 2100 Angstroms wavelength by grating scatter of visible wavelength photons. Below 1800 Angstroms, the grating scatter dominates the signal. The HST Goddard High Resolution Spectrograph (GHRS) includes a low-resolution mode (G140L) that complements the FOS capability. Since the GHRS has "solar-blind" blind detectors, that would be the preferred mode for many planetary spectroscopy programs with the HST at short wavelengths. However, a partial electronic failure within the GHRS rendered the G140L inoperative soon after launch, so that the characteristics of the FOS below 2000 Angstroms have become a critical issue. Systematic ways to compensate for the effects of its grating scatter are required. Fortunately, the problem can be quantified. In May 1991, before the GHRS failure, the G2V solar analog star 16 Cygni B was observed with both the FOS ("blue" side) and the GHRS, throughout the wavelength range 1600-3300 Angstroms. As expected, the GHRS spectra of the solar analog star agreed well with independent observations of the Sun (Van Hoosier et al., Astron. Lett. Commun., 27, 163-168, 1988) throughout the entire spectral range, proving the reliability of the GHRS for this type of observation. The FOS observations have subsequently been used to develop techniques for correctly removing the instrumentally scattered, long-wavelength light. Additional FOS observations, using both its "blue" and "red" sides, of other stars previously observed with the GHRS are now scheduled. Repairing the GHRS is one of the items to be attempted during an STS maintenance visit to the HST, currently scheduled for launch on December 2, 1993. Barnet C. D.* Cunningham C. C. Caldwell J. J. HST Spectra of the Outer Planets II. Calibration and Other Problems Observations of Jupiter (May 1992-July 1993), Saturn (December 1992), Uranus (June 1992), and Neptune (August 1992) were made by the HST Faint Object Spectrograph (FOS) (1800-3300 Angstroms) and the Goddard High Resolution Spectrograph (GHRS) (1600-1850 Angstroms). The Jupiter and Saturn spectra sample the equatorial central meridian, equatorial limb, and the north pole (Saturn only). Spectra of Uranus and Neptune were only obtained at the center- of-disk. The FOS spectra were systematically processed to remove background grating scatter (see previous abstract) and to account properly for effects due to spherical aberration in the HST primary optics. Regions of overlap between FOS and GHRS spectra for Jupiter and Saturn indicate some absolute calibration problems. The calibration of planetary spectra is complicated by spherical aberration, since planetary targets overfill the spectrograph apertures. The standard data processing from the Space Telescope Science Institute applies a correction for aberration that assumes a point source is being observed. An additional systematic effect occurs when spectra from different gratings (FOS) or from different settings of the same grating (GHRS) are combined to produce an extended spectral range. The effect is consistent with a slight misalignment of the dispersion direction of the gratings with respect to the detector diode array, such that some light from the overfilled apertures is lost from diodes at the ends of the array. The effect is not seen in stellar spectra. Further calibration uncertainties are suggested by comparison with earlier spectra from the International Ultraviolet Explorer. Comparisons of the HST observations over the spectral range of 2100 Angstroms to 3300 Angstroms with earlier IUE measurements will be discussed. Subtle differences between the HST and IUE reflectivities exist. However, no significant changes in the reflectivity have occurred within the last decade with the exception of Saturn. This will be discussed in the context of the recent equatorial disturbance (Fall 1990). Caldwell J. J.* Cunningham C. C. Barnet C. D. HST Spectra of the Outer Planets III. Interpretation The HST spectra described above were ratioed both to HST spectra of the solar analog, 16 Cyg B, and to SUSIM/SpaceLab 2 solar data to derive albedo curves over the spectral range of 1600 Angstroms to 3300 Angstroms. In general, 16 Cyg B provided better cancellation of solar Fraunhofer features above 2000 Angstroms than did SUSIM, but SUSIM was better at shorter wavelengths. The reflectivity ratios are compared to vertically inhomogeneous models of the planetary atmospheres to determine abundances of minor and trace species therein. The models are sensitive to aerosol concentrations. All four giant planets show the influence of acetylene absorption. In particular, Uranus and Neptune have C2H2 abundances comparable to those previously determined from IUE observations. The HST spectra confirm that the abundances of C2H2 are very similar on Uranus and Neptune. Only Jupiter shows features due to NH3. The absence of NH3 on Saturn is in agreement with earlier TDla observations, and is a consequence of Saturn having lower tropospheric temperatures than Jupiter. No other species have been positively identified in the spectra above 1800 Angstroms, although some interesting possibilities are being considered. Attempts have been made to extend the spectra of Jupiter and Saturn to shorter wavelengths using a moderate resolution grating that is still operative on the GHRS. For Saturn, this was only marginally successful. For Jupiter, observations that have the potential to detect stratospheric CO have been made and will be discussed. Raman scattering within the solar Fraunhofer features at 2800 Angstroms (MgII) and 2850 Angstroms (MgI) are apparent in all spectra. Reuter D. C.* Deming D. Fast K. Thermal Wave Structure on Jupiter: Io Tidal Forcing As part of our continuing investigation of thermal wave structures on Jupiter we have obtained a series of 20-micrometer images of the planet using the cryogenically cooled grating spectrometer in the Goddard postdisperser in a stand-alone configuration. The observations were made from 3 to 7 May, 1993 using a 10 (spectral) x 50 (spatial) element BIB array at a spectral resolving power of about 1000 and a spatial resolution of 2 to 4 arcsec. Most of the observations were obtained in the equatorial region; however, occasional full- disk scans were performed. In addition to thermal wave structures of the type previously observed [e.g., 1] these data provide a sensitive probe of the thermal signature of tidal forcing due to Jupiter's moon Io. The strongest indicator of tidal motion is expected to be a wavenumber 2 mode that is locked to Io's orbital motion. Under the assumption that the deep atmosphere is neutrally buoyant only a weak response is expected to tidal forcing; however, it has recently been suggested [2,3] that even for small values of stratification the response is greatly increased. Thus measures of the amplitude of the tidal motion or sensitive upper limits to that amplitude, provide probes of the planetary structure. [1] Deming D. et al. (1989) Ap. J., 343, 456. [2] Ioannou P. J. and Lindzen R. S. (1993) Ap. J., 406, 252. [3] Ioannou P. J. and Lindzen R. S. (1993) Ap. J., 406, 266. . Bjoraker G. L.* Jennings D. Reuter D. McCabe G. Temperature Retrievals for Jupiter's South Pole Using H2 Quadrupole Emission at 17 Micrometers We observed Jupiter at 17 micrometers from the IRTF telescope on Mauna Kea using a new grating spectrometer developed at Goddard, known as CELESTE. We measured the 587.03 cm^-1S(1) quadrupole line of H2 in emission at high spectral resolution (Delta nu=0.08 cm^-1) for a region near Jupiter's South Pole (60 degrees South, 110 degrees System III longitude) on 7 March 1993, UT. At line center most of the emission originates from Jupiter's lower stratosphere between 10 and 60 mbars. The continuum produced by pressure- induced absorption by H2 sounds the tropopause at 100 mbar. These observations are useful because spatial variations in H2 emission from Jupiter are due solely to temperature changes rather than composition gradients. In addition, the H2 quadrupole line sounds a poorly understood part of the atmosphere. Voyager IRIS spectra of Jupiter sampled the 1 mbar region (using CH4) and 100- 200 mbar (using pressure-induced H2); however, temperature retrievals between 10 and 100 mbars are quite uncertain except at the spatial locations sampled by the radio occultation experiment. We retrieved a temperature profile intermediate between the Voyager 1 ingress and egress profiles. The ingress and egress temperatures at 30 mbar, which pertain to Jupiter's equator in 1979, were 135 and 115 K, respectively. Our south polar profile is consistent with -125 K at 30 mbar, using the Goorvitch & Chackerian S(1) line strength (Icarus 32, pp. 348-361, 1977). We have laboratory measurements of H2 absorption at high resolution and for long path lengths. Analysis is underway to improve estimates of the S(1) line strength. Beebe R.* Sada P. Conrath B. Observational Constraints for Great Red Spot Models Analysis of the highest-resolution Voyager imaging (ISS) and infrared (IRIS) data has been carried out, and improved navigational information has been utilized in a correlative study. These combined data indicate that the organized azimuthal flow is constrained in a sharply delineated collar 2300- 4500 km in width around the perimeter of the Great Red Spot (GRS). Lower- resolution Voyager observations indicate that the collar vorticity is replenished by the absorption of small westward translating eddies that are deflected around the spot. The central portion of the GRS, extending 12,000 km in longitude and 4500 km in latitude, contains smaller-scale structures with typical velocities of 5 m/sec that are suggestive of 2-D turbulence. Derived quantities from the IRIS data reveal a reduced H2 para fraction and enhanced ammonia gas and ammonia ice above the center of the spot. A temperature difference of approximately 10 K between the center of the GRS and its surroundings is indicated. These observations, along with the color ratio of the spot to its surroundings in groundbased observation at 889/7l5 nm, are consistent with a model of the GRS with a central region of elevated clouds surrounded by a constrained collar about the flanks of the raised center. If adiabatic heating is balanced by radiative cooling, the mean net upward velociy is approximately 5 x 10^-3 cm/sec. If the outward flow extends over a depth of one pressure scale height, then the divergent velocity is aproximately 1 cm/sec. The implications of the observations within the framework of current GRS models will be discussed. Hockey T.* Borucki W. Motion of the S-shaped Features Associated with Jovian Lightning Activity Studies of Voyager spacecraft images of Jupiter have revealed only one lightning storm that corresponds in latitude and longitude to a recognizable daytime feature: a pair of distinctive S-shaped clouds in the "disturbed region" of the planets northern hemisphere. If it is possible for a storm occurring at the water cloud level to affect the overlying ammonia cloud layer, then understanding these features may lend insight into the mechanism that controls the location of the lightning activity. We discuss the morpholoyg, dynamics, and development of the S-shaped feature. The data were culled from the collection of Voyager 2 images made before and at closest encounter. Images were selected that showed the features on each rotation of the planet while they were visible to the Voyager cameras. Discrete morphologies within the features are identified. Latitude and longitude measurements were made of each of these points for each image in which they appear. These measurements form the basis of a model of the motion of the features. The work of the first author is supported in part by a NASA JOVE Program fellowhsip. POSTERS Cochran W. D. Baines K. H. A High Resolution Spectral Atlas of Jupiter: 6000-7500 Angstroms We have obtained a high spectral resolution (R = 60000) map of selected regions of Jupiter in the wavelength region 6000-7500 Angstroms. Data were obtained using the Sandiford Cassegrain Echelle spectrograph on the McDonald Observatory 2.1m telescope on 1-3 May 1993 UT. Spatially resolved spectra of belts and zones were obtained along the central meridian. In addition, spectra were obtained of the South Equatorial Belt disturbance, the Great Red Spot, and dark spots in the North Equatorial Belt. Selected samples of the 160 spectra obtained will be shown. The data are of sufficient spectral resolution and signal/noise to allow isolation of individual H2, CH4, and NH3 rotational features for analysis. The distribution of the equivalent widths of several of these lines over the planet will be presented. These data will allow the pressure levels of the ammonia cloudtop near 600 mbars and the deepest tropospheric cloudtop near 3 bars to be obtained unambiguously for all of the re Borunov S. Dorofeeva V. Khodakovsky I. Drossart P. Lellouch E. Encrenaz Th. Phosphorus Compounds in the Atmosphere of Jupiter The only phophorus compound that has been detected in the jovian atmosphere from groundbased, airborne, and spacecraft observations is PH3. Abundance of PH3 is much larger than thermochemical equilibrium abundance at the levels of detection and it is possible to use these measurements for determination of the strength of convective mixing in the deep atmosphere of Jupiter. Here results of thermochemical equilibrium and chemical kinetic calculations are presented, including a sensitivity study to variations of elemental abundances of phophorus and oxygen. In contrast to what was found by Fegley and Prinn (1985, Ap.J., 299, 1067) in all the cases that are analyzed it is found that PH3 is mixed up from levels where the temperature is significantly lower than 1240 K. Clapp M. L. Miller R. E. FTIR Spectroscopy of Ammonia and Hydrazine Aerosols: Implications for Planetary Atmospheres We have recently obtained infrared extinction measurements for ammonia and hydrazine aerosols formed by homogeneous nucleation under conditions approximating the atmospheres of Jupiter and Saturn. The resulting extinction is very sensitive to particle size, phase (solid, liquid, etc.), and in some cases, shape. Indeed, comparison of Mie theory and Discrete Dipole Approximation calculations with observed spectra reveal a shape effect in the 9.4 micrometer region of the ammonia aerosol spectrum. This shape effect leads to an overall decrease in the peak absorption and may be responsible for the lack of observed condensed ammonia features in these atmospheres. We have also studied hydrazine and mixed ammonia/hydrazine aerosols using this technique. Both ammonia and hydrazine exhibit supercooled liquid phases at higher temperatures and are highly crystalline at lower temperatures when nucleated independently. When mixed aerosols are formed at lower temperatures, the spectra are perturbed, indicating the production of an amorphous solid aerosol. We have also observed the freezing of supercooled ammonia onto a hydrazine core as well as the coating of hydrazine particles by ammonia vapor. Representative spectra as well as their interpretation will be discussed. Huber L. F. Beebe R. F. Simon A. A. Danielson G. E. Limaye S. Zonal Wind Measurements Utilizing Hubble Space Telescope Wide Field/Planetary Camera Images Hubble Space Telescope Wide Field/Planetary Camera observations of the equatorial region of Jupiter in the Planetary Camera mode in broadband filters and in a narrow band filter centered on the deep methane absorption band at 889 nm have been deconvolved using both Lucy and Wiener algorithms. Pairs of images separated by 20 hours have been map projected and zonal winds have been derived for +20 to -30 degrees planetographic latitude. A discrepancy in the magnitude of the zonal winds near -7 degrees latitude between the measurements in the broadband F718M red filter and the narrow band FX89N methane filter is detected. The magnitude of this difference is approximately 30 +- 12 ms^-1. Comparison of the cloud morphology indicates that the clouds in the 718 nm image are strongly sheared near -7 degrees where the winds compare with those measured in low resolution Voyager images. In contrast the 889 nm image reveals chevron structures that extend +-2 degrees in latitude relative to the 718 nm shear zone. No discrepancy is seen at the corresponding northern latitude. Color vs. latitude across the equatorial zone indicate that the ammonia cloud deck is elevated and the two filters probe to the same depth at +7 degrees while the nonsimilar cloud pattern and color ratio near -7 degrees indicate that the cloud tracers in the two filters occur at different altitudes. This work is based on observations with the NASA/ESA Hubble Space Telescope, obtained at the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS5-26555. Support was provided by NASA through grant number GO- 2560.01-87A from Space Telescope Science Institute. Carlson B. E. Lacis A. A. Rossow W. B. Thermal Structure, Cloud Structure, and Gas Composition of the Great Red Spot We have analyzed the Voyager IRIS observations of the Great Red Spot (GRS) in order to gain a better understanding of its thermal structure, cloud structure, and gas composition. As a starting point for this investigation, we have defined a GRS spectral ensemble and have compared the average GRS spectrum with the average spectra corresponding to the belt and zone spectral ensembles that we have previously analyzed (e.g., Carlson et al., 1992, Astrophys. J. 388, 648). These comparisons reveal large differences in the spectra in the region of the S(sub)l(O) hydrogen line (near 602 cm^-l). This region of the spectrum is sensitive to pressures near 0.15 bar. We can further refine the location of these differences by using the emission angle variability of the IRIS measurements to improve our height discrimination. Comparison of spectral ensembles at higher emission angles reveals smaller and structureless differences in the region of the S(sub)1(O) hydrogen line, suggesting that the differences are confined to lower altitudes (i.e., upper tropospheric). Using the radiative transfer model described in Carlson et al. (1993, JGR, 98, 5251) we have analyzed the GRS spectra. Our thermal inversion procedure is similar to the standard inversion procedures described in Conrath et al. (1970); however, we have replaced weighting functions with normalized contribution functions, which more accurately depict the peak emission level. Consistent with previous investigations, we find that the upper troposphere is colder in the GRS. Similarly, consistent with the results of Griffith et al. (1992, Icarus, 98, 82) we find that the abundance of NH3 above the 0.3-bar level is depleted relative tothat found in the STrZ, while the PH3 abundance is comparable. In contrast with previous results, we find some spectral indications for a slight enhancement of NH3 in the 1-2 bar region relative to that in the surrounding STrZ. The dynamical, microphysical, and chemical implications of these results are discussed. Drossart P. Lecacheux J. Measurement of the Differential Rotation of Jupiter at Several Wavelengths in the Visible Images of Jupiter were acquired in the visible through interferential filters, at 8300 Angstroms(continuum), 8900 Angstroms(strong CH4 band) and 7250 A(weak CH4 band) from the 1 m reflector at Pic-du-Midi observatory. Mosaics within cylindrical projection, covering about 270 degrees of longitude, are obtained from images recorded during the nights of 1993 February 18 and 20, corresponding to a 50 hours time lag. The comparison between the two projections allows us to measure the horizontal wind velocities to an accuracy of about 10 m/s, between 30 degrees S and 30 degrees N latitudes. Wind velocities are obtained from several methods, following Limaye (Icarus, 1986), in particular by correlations between the two projections. At first order, the wind velocities are very similar to the Voyager wind profile, despite the very different aspect of the SEB at time of observations. Comparing the observations at two different wavelengths (in particular 8900 and 8300 Angstroms) shows no significant differences, despite the different aspect of the Jupiter images at these wavelengths. A comparison of images at 8900 Angstroms and 8300 Angstroms after processing with a contour enhancement numerical filter shows also that, despite the very different contrasts at both wavelengths, the contours of the cloud structures are nevertheless very similar. We conclude that the features responsible of the correlations in the wind profiles observed in the visible are the same, and that the spatial distribution of the high altitude clouds, predominent in the images at 8900 Angstroms, does not alter the appearance of the underlying cloud structure. Guillot T. Morel P. Gautier D. Chabrier G. Non-Adiabatic Models of Jupiter and Saturn Contrary to the usual assumption that the fluid envelope of the giant planets are fully convective, new calculations have shown that the radiative opacity in the deep atmosphere of Jupiter, Saturn, and Uranus (around the 1000-3000 K temperature levels), are small enough to permit the transfer of internal energy by radiation rather than by convection. We present here a numerical code developed to study the internal structure of Jupiter and Saturn, in which the Schwarzschild criterion determines whether the temperature gradient is radiative or adiabatic. The four differential equations of internal structure are solved by using an implicit collocation method based on a B-spline basis. The relative abundances of hydrogen and helium as well as the mass and the composition of the central core of rocks and ices are adjusted in order to reproduce the radius of the planet and the observed moments of the external gravitational field. The results are compared to current adiabatic models. We discuss the consequences of the occurrence of a radiative zone in the envelope of Jupiter and Saturn on their internal structure and their evolution. Conrath B. J. Gierasch P. J. Para Hydrogen Remote Sensing in the Atmospheres of the Outer Planets Knowledge of the para hydrogen fraction f(sub)p on the outer planets has important implications for atmospheric dynamics and energetics. In the present study, the problem of the retrieval of vertical structure of f(sub)p from Voyager IRIS data is addressed using direct inversion techniques. Contribution functions appropriate to the retrieval of the vertical structure of f(sub)p on Jupiter and Neptune have been calculated and are found to have half-widths of about 1.3 pressure scale heights. If the temperature profile and opacity sources are known a priori, then f(sub)p can be obtained between approximately 200 and 700 mbar on Jupiter and between 100 and 800 mbar on Neptune with vertical resolutions approaching 0.75 pressure scale height. However, simultaneous retrieval of temperature and cloud and gaseous opacities, along with f(sub)p, requires "redundant" spectral regions that sense the same atmospheric layer but possess differing sensitivities to the various parameters. This condition is poorly met at pressures greater than about 350 mbar on Jupiter and throughout most of the Neptunian upper troposphere, resulting in highly ill-posed retrieval problems with physically plausible but non-unique solutions. Between approximately 200 and 350 mbar on Jupiter, limited vertical structure information on f(sub)p can be unambiguously obtained, and results for this atmospheric region are presented. In some locations, such as the North Tropical Zone, correlation between large upper level values of f(sub)p and large 45-micrometer cloud optical depths is found, while in other areas such as the Great Red Spot, a strong anti- correlation between these parameters is observed. The dynamical implications of the results are discussed. Dowling T. E. An Explicit Planetary Isentropic-Coordinate Atmospheric Model for Jupiter, Saturn, Uranus, and Neptune We describe a new global, multilayer atmospheric model for the giant planets, called the EPIC model, and present applications to Jupiter, Saturn, Uranus, and Neptune that illustrate the model's strengths and weaknesses. The model uses finite differencing and an explicit timestep in order to run efficiently on massively parallel processors. We use an nCUBE 2 with 512 processors. The timestep is the 3rd Order AdamsBashforth timestep, which is appropriate for both the advective and diffusive terms, does not require restarting, and is more accurate than the leap-frog timestep. The horizontal differencing follows the Arakawa and Lamb scheme, which conserves total energy and total enstrophy. The vertical differencing follows the Hsu and Arakawa scheme, which uses potential temperature, a function of entropy and hence a conserved quantity for adiabatic motions, as the vertical coordinate. The result is one fewer prognostic equation, since the thermodynamic equation is incorporated into the coordinate system, and the elimination of vertical velocities except for diabatic motions, which improves accuracy. Giant planets benefit from the advantages of isentropic coordinates without suffering from the major disadvantage that is encountered with terrestrial planets, namely the intersection of the coordinate surfaces with the solid lower boundary. Instead, the convectively adjusted fluid interior of a giant planet makes the lower boundary a coordinate surface. Global models of Jupiter's atmosphere require approximately 10 x 10 more grid points than comparable models of Earth's atmosphere, because on both planets the deformation radius is 1000 km, but Jupiter's radius is 10 times bigger than Earth's. We typically run with 512 x 256 horizontal grid points and from 4 to 10 vertical layers. The model incorporates Voyager vertical temperature soundings and zonalwind profiles, and employs Newtonian cooling with realistic radiative cooling times. It simulates the long-term dynamical response of giantplanet atmospheres to events like comet impacts, and provides numerical weather prediction. It also provides a framework for testing the dynamical importance of proposed latent-heat sources, including ortho-para conversion and cumulus heating. The primary goal is to model the formation of zonal-wind profiles and long-lived vortices in a realistic manner. Lederer S. M. Marley M. S. Albedo Inhomogeneities and Jovian Seismology Mosser et al. (A & A (1993) 267, 604) have reported the detection of p-mode oscillations of Jupiter from Fourier Transform Spectrometer observations made at the CFHT. Detected modes had periods ranging from 8 to 17 minutes. However, their observations are also sensitive to flux variations produced by spatially periodic albedo features rotating with the planet. Such features, if present, could produce a spurious seismological signal. To investigate the magnitude of longitudinally periodic albedo features in wavelengths corresponding to methane absorption, we studied images of Jupiter taken through an 0.889-micrometer methane absorption filter at the Tortugas Mountain Observatory in Las Cruces, NM. Images from three consecutive nights within the 1991 observing run of Mosser et al. were analyzed for periodic albedo features. Average dN values within a circular region (analogous to the aperture employed by Mosser et al.) 15 degrees in diameter and centered on 7.5 degrees N latitude were calculated as a function of longitude and converted to the time of the meridian crossing, assuming solid-body rotation of Jupiter. Based on a periodogram analysis of this dataset, we see no albedo features that could produce a spurious seismological signal. We do, however, see a possible periodicity in one set of data corresponding to a 60-minute period. This may be related to the visible plume structures. Periodic albedo features with amplitudes >~1.5% would be detected in our analysis. To more fully investigate periodic jovian albedo inhomogeneities, we are now employing a dataset better suited for a time series analysis from the 1993 Jupiter opposition. Limits on the magnitude of albedo features that may hinder methane-band seismological observations of Jupiter will be presented. Schinder P. J. Flasar F. M. A Reexamination of the Radio Occultation Technique for Determining the Structure of Planetary Atmospheres We have performed a major reexamination of the radio occultation technique for determining the structure of planetary atmospheres. We derive from relativistic first principles the equations describing the propagation of the radio signal from the spacecraft to the Earth, both inside and outside the target atmosphere. In particular, we use the relativistically covariant eikonal equation (to order (v/c) in a frame comoving with a chosen point in the atmosphere) to determine the path of the ray through the differentially moving atmosphere. The atmosphere is assumed to be axisymmetric but not necessarily spherical. We examine the propagation of systematic errors in the retrieval of vertical profiles of refractivity, pressure, and temperature. Such systematic errors include uncertainties in the geopotential and isopycnal surfaces arising from uncertainties in the zonal winds, in the gravitational field, and in the position of the rotational pole. We focus in particular on the atmospheres of Titan, Saturn, and Neptune. We will discuss potential errors in the shape of the retrieved T-P profiles, which for the latter two can affect the atmospheric helium abundance determination obtained from the combined analysis of thermal infrared and radio occultation data. Monday, October 18, 1993 MARS SURFACE I 8:30 - 10:00 AM North Ballroom Chair(s): J. F. Bell III G. A. Swayze Lee S. W.* Clancy R. T. James P. B. Monitoring Regional Albedo Features on Mars: Results from the Hubble Space Telescope The Pioneer Venus Ultraviolet Spectrometer obtained extensive series of images of Venus in the 4th Positive system of carbon monoxide. The (0,1) band at 160 nm is excited by ionospheric processes and by resonance fluorescence of sunlight, while the (14,4) and (14,3) bands at 139 nm and 136 nm are produced by an accidental resonance between the (14,0) band and solar Lyman-alpha (the 136-nm feature is also contaminated by the photoelectron-excited 3P-5S multiplet of atomic oxygen). The 160-nm feature is emitted near 150 km, whereas the 139-nm and 136-nm features are emitted two or three scale heights lower. Thus the morphology of the images depends on the horizontal distribution of carbon monoxide in the lower and middle thermosphere. An earlier study (Alexander et al., 1993) of images in the 3P-3S resonance triplet of atomic oxygen at 130 nm showed that in the upper thermosphere, oxygen at high latitudes is more abundant in the afternoon than in the morning, but that on the equator this effect is less pronounced. Their interpretation invoked the dissipation of gravity waves generated in the middle atmosphere, a mechanism proposed by Alexander (1992) to explain the thermospheric superrotation. A preliminary inspection of the carbon monoxide images at all three wavelengths shows similar trends. Since the variations of the densities of CO and O across the sunlit disk are dynamically controlled, we will examine the images for evidence of changes in the dynamical regime with position and altitude. Referemces: Alexander M. J. (1992) Geophys. Res. Lett., 19, 2207. Alexander M. J. et al. (1993) JGR, 98, 10849. Singer R. B.* Miller J. S. Merenyi E. Evidence for a Global Compositional Dichotomy on Mars The north-south morphologic and age dichotomy is a profound aspect of Mars geology. Yet there has been limited evidence for compositional variations that correlate with this geomorphic division. This has been interpreted by many to indicate that current surface materials on Mars are globally distributed, largely homogenized, and generally unrelated to the underlying crust. Recent evidence, however, has shown that this is not the case. There is significant regional and local compositional variation around the planet (e.g. Bell et al. JGR 95 1990; Singer et al. Bull. AAS 22 1990; Singer and Miller LPI 92?2 1991; Mustard et al. JGR 98 1993; Murchie et al. LPSC XXIV 1993; Geissler et al. Icarus in press 1993; Merenyi et al., Icarus submitted, 1993). The manifestations of these variations in remotely sensed data are often subtle. We conclude, however, that a global compositional pattern is emerging. This result is based on the synthesis of various recent results by us and others, with emphasis on a series of visible and near-IR (0.44-1.02 micrometers) spectral images obtained by us during the 1988 apparition (Singer et al. LPSC XXI 1990). In summary, both high and low albedo materials currently exposed in the ancient highlands show more bulk crystalline hematite, implying weathering or other alteration that proceeded further toward equilibrium than in most of the lowlands. This may be a consequence of exposure age, but might also be related to differences in geologic processes and/or climate. Northern plains low-albedo regions, such as Acidalia Planitia and Oxia Palus are also different in their mafic composition than observed low-albedo regions on older terrains. Near-IR Fe^2+ pyroxene bands are observed centered between 0.92 and 0.99 micrometers for highland exposures, while Acidalia, a large rocky plain in the northern lowlands, does not show a band minimum shortward of 1 micrometer (where our data end). Various interpretations are possible for Acidalia, but we cannot yet choose among these with available data. What is clear, however, is that there is now evidence for compositional differences that correlate with the long-known global morphologic dichotomy on Mars. We can also conclude that (a) the surface of Mars has not been completely resurfaced and homogenized by aeolian action; (b) not all heavily weathered deposits are the same as each other or the global dust. Arabia, in particular, appears compositionally distinct from other large northern bright regions; (c) the composition of at least some of the basaltic lavas in the northern plains differs markedly from those to the south. Bell J. F. III* Pollack J. B. Crisp D. Geballe T. R. Cruikshank D. P. High Resolution K-Band Spectroscopy of Mars During 1990 and 1993 High spectral resolution telescopic observations of Mars were obtained during the 1990 and 1993 oppositions from Mauna Kea Observatory. During 1990, the IRTF CGAS spectrometer was used to observe ten 500-km regions at a spectral resolution of 1200-1500 from 2.10 to 2.47 micrometers [1]. These data were obtained simultaneously with near-IR imaging spectroscopic measurements discussed elsewhere [2]. During 1993, the UKIRT CGS4 spectrometer was used to obtain multi-pixel slit images of Mars at a spectral resolution of 350 from 2.04 to 2.46 micrometers. The 1993 data range from 500-1000 km spatial resolution and encompass about 70 spectra in total. Flux calibration (to within +-20%) of both of these datasets has been performed using IR standard stars. Many interesting absorption features can be seen in these data sets. Most prominent are the strong CO2 R-branch band near 2.05 micrometers and a weaker CO2 band near 2.15 micrometers. Also, a broad but weak absorption in the 2.3 to 2.4 micrometer region can be seen and is attributed to atmospheric CO. The presence of these features in our flux calibrated data will allow us to accurately model the relative contributions of the martian atmosphere and surface. As well, the presence of both strong and weak atmospheric bands may allow us to quantitatively derive the dust opacity during both oppositions [2,3]. Less prominent absorption bands are also seen in the data, and these features are tentatively assigned to mineralogic absorptions either on the surface or in the airborne dust. These bands are indicated in Table 1, along with the so- called "scapolite" bands observed by Clark et al. during the 1988 opposition [4]. There is enough overlap among the datasets from different oppositions to indicate that the features observed just outside of the martian atmospheric CO region are probably real and are probably due to martian bisulfate- and/or bicarbonate-bearing minerals, as proposed in [4]. Features observed inside the CO region require additional modeling to adequately separate the gaseous vs. mineralogic components [i.e., 5]. There is evidence in a few of the spectra for weak features near 2.20 and 2.25 micrometers, consistent with previous interpretations of weak phyllosilicate and/or bisulfate/bicarbonate absorptions in these regions [2,4,6]. More specific mineralogic inferences from all of these data will be possible only after more detailed modeling and understanding (not "removal") of Mars atmosphenc absolptions. Table 1, which appears here in the hard copy, shows Mars K-band absorption features detected in 1988, 1990, and 1993. Referernces: [1] Bell J. F. III and Crisp D. (1992) MSA7T Kona Workshop, LPI Contribution 787, 1-3. [2] Bell J. F. III and Crisp D. (1993) Icarus, in press. [3] Crisp D. and Bell J. F. III (1990) LPS XXIV, 343-344.[4] Clark R. N. et al., JGR, 95, 14463-14480. [5] Encrenaz T. and Lellouch E. (1990) JGR, 95, 14589-14593. [6] Murchie S. et al., (1993) Icarus, in press. Merenyi E.* Edgett K. S. Singer R. B. A New Type of Weathered, Immobile Soil Unit on Mars In the past few years, extensive VIS-NIR spectral analysis work has been done to investigate the compositional variability of the martian soils. Recent research by Pinet and Chevrel (JGR, 95, 1990) from high-spatial-resolution CCD imagery at selected wavelengths, by Mustard et al. (JGR, 98, 1993), Murchie et al. (LPSC XXIV), from high-spatial- and spectral-resolution Phobos ISM data, by Singer et al. (LPSC XXI, 1990), Singer and Miller (MSATT, 1991), Miller and Singer (LPSC XXIV, 1993), Bell et al. (JGR, 95, 1990; LPSC XXI, 1990), Merenyi et al. (LPSC XXIII 1992; Icarus, submitted) from high-spectral-resolution telescope imaging has shown a considerable variation of surface materials among bright regions as well as across dark albedo units. It has generally been observed and argued that dark regions exhibit greater variability than bright regions. Dark region spectra show evidence for mafic mineralogy, bright units are more altered. There are also examples of areas where the spectral properties strongly correlate with the physical characteristics of the soil. Here we follow up on previous work regarding a medium albedo unit, Deucalionis Regio, centered at (245 degrees W, 15 degrees S), which seems to be covered with a previously unreported type of soil. It is distinguished from the surrounding classic regions of Arabia, Sinus Meridiani, Oxia, and Acidalia in the 0.4-1.0-micrometer range. The spectra suggest that this area contains more pyroxene than Arabia and less than Sinus Meridiani. The crystalline hematite content appears greater than either in Arabia or Meridiani. Yet, Deucalionis cannot be modeled as a simple mixture of the nearby dark and bright units. Its physical characteristics, as derived from Viking IRTM data, indicate an immobile soil, maybe cemented by salts. We expect to test the mineralogical distinction with Mars Observer Thermal Emission Spectrometer in the future. The geographic extent of the spectral and physical anomalies show strong spatial correlation, which suggests a possible genetic tie. Very similar spectral and physical properties seem to be spatially correlated over another large area, Noachis. The anomaly of Deucalionis Regio was brought to our attention by the analysis of a telescopic spectral image. Recent work on another, similar image cube confirms the previous results. Geissler P. E.* Singer R. B. Search for Small Scale Ferric Oxide Mineral Deposits on Mars In preparation for the arrival of Mars Observer, we have begun a program of systematic examination of the Viking Orbiter multispectral imaging data for evidence of color anomalies indicative of ferric oxide mineralization. The motivation for this study is the interpretation of color anomalies observed in the central Valles Marineris to be due to local concentrations of hematite (alphaFe2O3) [1-3]. The mineralization occurs in two small depressions on the margins of Hesperian-aged layered deposits in West Candor Chasma. The distinct coloration is caused by the absorption edge of Fe3+ in hematite at 0.53 micrometers, which coincides with the Viking green filter center wavelength [4,5]. On Earth, concentrations of iron oxides similar to the mineralized zone in West Candor Chasma are formed through a variety of geologic processes, many of which may have occurred on Mars as well. Hematite is a major component of many gossans (supergene ore deposits) produced by weathering of near-surface orebodies. Volcanism also frequently results in local concentrations of hematite, as oxidation rinds on basaltic rocks and cinders or as hematitic soils produced by a combination of water and heat in regions of hydrothermal alteration. Ferruginous sandstones and latentes are cemented by mixtures of poorly crystalline hematite, goethite (alphaFeOOH) and hydrated iron oxides resulting from oxidation and precipitation of dissolved iron by surface- or ground-water. Identification of OH- or H2O-bearing ferric oxides inferred from Vilking 3-color observations and data expected from Mars Observer would be especially significant as new evidence for the past existence of water on Mars. Geologic features small enough to have escaped detection by telescopic scrutiny --like the 20-km-long depressions in West Candor Chasma--may exist in many parts of Mars and hold important clues to the composition and chemical environment of its surface. References: [1] Geissler and Singer (1992) LPI Tech. Rept. 92-04, 12-14. [2] Singer et al. (1992) Bull. A.A.S., 24, 977. [3] Komatsu et al. (1993) JGR, 98, 1110511121. [4] Singer (1982) JGR, 87, 10159-10168. [5] Sherman et al. (1982) JGR, 87, 10169-10180. McLarty J. E.* Singer R. B. Are Antarctic Dry-Valley Altered Volcanics Spectrally Consistent with Mars? Weathered or otherwise altered volcanic materials from the dry valleys of Antarctica have often been considered as reasonable analogs to martian rocks and soils, based on similarities in temperature and humidity (Gibson, JGR, 28 1983; Berkley, PLPS, 12B, 1981). To our knowledge, however, little spectroscopic work has been done to compare these terrestrial materials with known spectral properties of the surface of Mars. For this reason we have undertaken a spectral study of Antarctic samples of a particular core previously analyzed by Berkley and Drake (Icarus, 45, 1981). These samples were recovered by the Dry Valley Drilling Program (DVDP) on Ross Island, Antarctica. They are olivine-bearing alkalic basaltic tuffs, a composition predicted by some for martian lavas. They are also of interest because the climate on Ross Island parallels permafrost terrains on Mars. The samples are from DVDP hole number 2 unit 11, and are numbered 25 through 33 (omitting 27), with depth increasing with sample number. Spectral reflectance (relative to halon) from 0.35 to l.lO micrometers was measured for powdered samples of these units using the new laboratory spectrometer at the Planetary Image Research Lab (PIRL). The sample spectra show a range in both Fe3+ abundance and mineralogy. The samples at the top of the core (26 and 27) show clear evidence of bulk crystalline hematite, with diagnostic bands at 530 and 860 nm. Spectral evidence for crystalline hematite decreases further down the core. Most (if not all) of these samples contain poorly crystalline or very finely crystalline (nanophase) ferric oxide, as demonstrated by steep but relatively featureless spectral slopes from the near-UV to the near-IR. The steepness of this slope increases toward the bottom of the core, indicating a somewhat higher overall Fe3+ content than upper units (but less crystalline hematite). This sequence indicates a range of temperature (and/or humidity conditions) under which alteration occurred, with the most mature alteration at the top of the sequence. Samples from units 28 and 29 are very similar spectrally to telescopic spectra of high-albedo regions on Mars, and to Hawaiian palagonitic Mars analogs. None of these Antarctic samples shows evidence of hydroxylated ferric oxides, implying alteration involving relatively little water. This is consistent with the findings of Berkley and Drake (1981) that clay minerals are not abundant in these samples, and also consistent with the apparent low abundance of crystalline clay on the surface of Mars. Moersch J.* Nicholson P. Squyres S. Van Cleve J. Lee P. Hayward T. Houck J. Miles J. Thermal Infrared Observations of Mars During the 1993 Opposition Using the newly-developed SpectroCam-10 instrument on the Hale 200" telescope at Palomar Observatory, we obtained images and spectra of Mars in the thermal infrared during the 1993 opposition. Images were taken using a set of six wide band filters centered at 7.9, 8.8, 9.8, 10.3, 11.7, and 12.5 micrometers at three times over the course of the night of 11 January. Spatially resolved slit spectra with a wavelength resolution of Lambda/Delta Lambda = 200 between 7.5 and 13.5 micrometers were also obtained covering the entire disk. Seeing on the night of the observations in this spectral region was about 0.75", corresponding to a resolution of 350 km on the surface of Mars at the sub- Earth point. To analyze the data, we have developed a thermal model that uses finite difference techniques to solve the time-dependent one-dimensional heat diffusion equation for surface temperatures and then blackbody radiances over the entire planet on a 2 x 2 degree grid. The inputs to the model are thermal inertia, albedo, and geometry of insolation at each grid point. Running the model with a spatially uniform thermal inertia of 6.5 x 10^-3 cal cm^-2 s^-1/2 K^-l and and an average albedo of 0.21, we produced synthetic "average" Mars images. The actual images were then divided by the "average" images to remove most of the center-to-limb brightness gradient and reveal more subtle surface features. Many well-known features are readily apparent in such ratioed images, including Syrtis Major, Hellas, Sinus Meridiani, Argyre, and Acidalia. Using the Viking-derived thermal inertias and albedos as inputs to the model, a second set of spatially inhomogeneous synthetic Mars images was created. When the observed images are divided by these inhomogeneous images, the results are mostly flat, but a few significant features remain. In at least one case, we have identified the cause of one of these features to be a change in the albedo of the surface since Viking. However, other features may be related to absorption in the martian atmosphere, or to non-unit emissivity of materials on the planet's surface. Because of our high spectral and spatial resolution, it may be possible to interpret any emissivity features that may be found in terms of surface geochemistry. We will discuss these possibilities using Planck curve normalized color ratio fits as well as analysis of the spectral data. Clark R. N.* Christensen P. Barbera P. Swayze G. Moore S. Betts D. Comparison of Mid-Infrared Emission Spectra with Spectra Computed from Reflectance: Verification of Kirkoff's Law Thirty-four mineral samples (phyllosilicates, micas, amphiboles, feldspars, pyroxenes, pyroxenoids, and olivines), have been measured in emittance and reflectance to test Kirkoff's Law: emittance = 1 - reflectance. Samples were measured in emission at the Arizona State U, Mars Observer Thermal Emission Spectrometer facility. Emission measurements, from 5 to 25 micrometers, were measured on cooled and heated samples in a dry nitrogen atmosphere. Black body reference and liquid nitrogen (zero level) background were also measured. The emissivity spectrum was then computed by fitting and ratioing a black body to the calibrated maximum radiance measured from the sample. The reflectance of the same sample set was measured from 1.8 to 25 micrometers in biconical reflectance at the USGS Denver Spectroscopy Lab. Ideally, the measurement would be done with an integrating sphere, but a sphere would limit the possible wavelength coverage. Our biconical measurement integrates about 50% of the hemisphere and should be a close approximation to that from a sphere. The samples were also measured on a near infrared spectrometer from 2.4 to 2.5 micrometers with an integrating sphere as a check on the reflectance levels from the midinfrared instrument. Emittance was calculated using Kirkoff's law. Comparison of the emittance spectra show good agreement within the sample set limitation. There are differences observed, but they can be explained by sample variability. The same sample was measured, removed from the sample cup, reloaded into the cup and measured again several times. In general, the differences observed by repeated measurements are greater than the disagreement between measured emittance and emittance calculated from reflectance. No artifacts similar to those reported by others are observed with Kirkoff's Law. Swayze G. A.* Clark R. N. King T. V. V. Gallagher A. Calvin W. M. The U.S. Geological Survey, Digital Spectral Library: Version 1: 0.2 to 3.0 Micrometers We have developed a digital reflectance spectral library, library management software, and spectral analysis software. The library includes 500 spectra of 447 samples (some samples include a grain size series) measured from approximately 0.2 to 3.0 micrometers. The spectral resolution (FWHM) of the reflectance data is <=4 nm in the visible (0.2-0.8 micrometers) and <=10 nm (0.8-2.35 micrometers) in the NIR. All spectra were corrected to absolute reflectance using an NBS Halon standard. Library software enables users to search on parameters (e.g., chemical formulae, chemical analyses, purity of samples, mineral groups, etc.) as well as spectral features. Minerals from sulfide, oxide, hydroxide, halide, carbonate, nitrate, borate, phosphate, and silicate groups are represented. X-ray and chemical analyses are tabulated for many of the entries, and all samples have been evaluated for spectral purity. The library also contains end and intermediate members for the olivine, garnet, scapolite, montmorillonite, muscovite, jarosite, and alunite solid-solution series. We have included representative spectra of H20 ice, kerogen, ammonium-bearing minerals, rare-earth oxides, desert varnish coatings, kaolinite crystallinity series, kaolinite/smectite series, zeolite series, and an extensive evaporite series. Because of the importance of vegetation to terrestrial studies we have include 17 spectra of tree leaves, bushes, and grasses. The library and software will be available as U.S.G.S. Open File reports. Software will allow PC users to convert the binary data to ASCII files. Additionally, an ftp binary file will be on line at the U.S.G.S. in Denver. Users may build customized libraries for their own instruments using the library software. We are currently extending spectral coverage out to 150 micrometers and will make this library available on CD-ROM. POSTERS Kerchner V. C. An Autonomous Software Agent for Feature Extraction from Mars Images Apple's new Digital Graduate Student(TM) (DGS) system for the Macintosh(R) is a modular construction set for building distributed scientific applications. At first glance, DGS resembles other component-based analysis systems such as IRIS Explorer(TM) or AVS(TM). However, DGS extends this metaphor to include a rich component user interface, the use of concurrently running applications as components, arbitrary levels of component hierarchy, distributed and remote processing, and network interfaces to Unix(R), VMS(R), and DOS(R). Many data import/export types are supported, such as HDF and FITS. DGS applications may include "hot links" to other scientific applications such as Mathematica(TM) and Spyglass(TM). DGS is currently under development, and is expected to be released in May of 1994. As a test case for DGS, an autonomous agent capable of recognizing and counting arbitrary features (e.g., craters) in spacecraft photographs was constructed. Neural net technology was used to construct a programmable scanner that could be trained to recognize arbitrary bitmaps in images. Training sets of artificial, random craters with adjustable parameters were input to the net until it was capable of recognizing features in simulated images with acceptable accuracy. A complete autonomous agent capable of reading multiple images, extracting features, and compiling coordinates and statistics in background mode was then constructed and used to locate actual features in Viking and Mariner images of Mars. This agent can be trained to recognize arbitrary features in any image, and can be used "as is" or modified as desired. Hansen G. B. The Spectral Absorption of CO2 Ice in the Thermal Infrared Accurate absorption coefficients of CO2 ice in the thermal infrared are needed for enhancing the analysis of Mars Observer measurements of the surface deposits and clouds in the martian polar regions by the TES and PMIRR instruments. The spectral absorption of CO2 ice has been previously measured in the infrared wavelengths 2.5-26 micrometers by Ditteon and Kieffer (JGR, 84, 8294, 1979), but only two strong absorptions at 90 and 150 micrometers have been measured at wavelengths longer than 26 micrometers (Kuan, Ph.D. Thesis, U. Southern Cal., 1969). I am measuring transmission through long (2-100 mm) crystals of CO2 in the wavelength region 20 to 50 micrometers, using a Fourier Transform Spectrometer. My previous work (Hansen, Bull. A.A.S., 24, 978, 1992) revealed a large amount of spectral detail and many weak, narrow lines in the "transparent" wavelength regions of the near infrared. To resolve similar structure in the thermal infrared, I am using a maximum spectral resolution of 0.2 wavenumbers. I am able to measure absorption only in spectral regions where CO2 ice is relatively transparent (absorption coefficient in the range 0.001-30 cm^-l). These regions are significant in the study of radiative transfer in dense scattering media, such as the permanent and seasonal polar deposits on Mars, in that low absorption implies low emissivity (Warren, et al., JGR, 95, 14717, 1990). Head J. N. Dark Materials in the Cerberus Region, Mars: Local Sources? Previously, we have successfully modeled one scene within the Cerberus region as a variable areal mixture of Bright and Dark materials (Head et al. 1992 LPSC 23 509-510). The Dark material was interpreted largely as a sand sheet overlying the bright plains material, consistent with the most recent geologic maps (Greeley and Guest 1987, Geologic Map of the Eastern Equatorial Region of Mars). In principle, the Dark material could be either locally or externally derived-many dark streaks emanate from scarps and crater splotches, indicating the former. However, it was unclear whether these sources represented currently eroding, primary bedrock (cf Veverka et al. 1976 Icarus 27 241-253, Geissler et al. 1990, JGR, 95, 14399-14413), or merely the latest episode of transport of externally derived aelion material (cf. Lee et al. 1982, JGR, 87, 10025-10041). We now have evidence that the local sources may represent outcrops of prirnary bedrock. The best-studied example is in the splotch crater Mj (12N, 201W), which has been examined using Viking apoasis multispectral images from orbits 506 and 609. A well-defined, roughly rectangular area covering ~80km^2 has the following properties: it is distinctly darker in the violet filter images than the adjacent Dark materials; it has the lowest Red/Violet ratio of any unit in the scene (RIV = 1.93 for i = 16, e = 14, phase = 29); it requires a relatively much higher Shade fraction in linear mixture models constructed from imaging from both orbits, and its photometric response is distinguishable from that of adjacent material. Three much smaller (~1km^2) similar areas occur within large (>20km) crater splotches in the 609 series images, although the small size precludes any judgements about photometric response. It is possible that these effects are due to induration. However, the scarps (Cerberus Rupes), which are morphologically not depositional, exhibit characteristics similar to the "smaller" areas (including the difficulty in determining photometric response). We have identified several large crater splotches with darker (violet filter) interior areas for which we have repeat coverage. These will be examined to determine if they also have these linear mixing model and photometric properties. Klassen D. R. Howell R. R. Bell J. F. III Spectral Imaging of Mars in the 2-micron Region We obtained near infrared images of Mars on 26-28 January 1993 and 12-13 February 1993 using a near-infrared camera at the Wyoming Infrared Observatory. These images cover most of the 2.0- to 2.5-micrometer range, with a resolution of approximately 1.3%. The images from some nights achieve subarcsecond spatial resolution, corresponding to better than 400 km. Approximately 50% of the planet was imaged, over the longitude range 170-270 degrees and 50-135 degrees. The 2-micrometer region is known to contain features primarily due to carbon dioxide gas and ice, as well as water ice and carbon monoxide. Our observing program was the first to use this particular camera with its circular- variable-filter, and this configuration introduced some instrumental problems that we are currently trying to calibrate and correct. However, the preliminary reduction of the images show considerable detail. The classical albedo features and the polar cap are very obvious, showing high contrast. The longitudes observed complement those obtained earlier by Bell and Crisp 1993 (Icarus, in press). We hope to use these images to extend that earlier study of the surface and atmosphere. Monday, October 18, 1993 JUPITER II 10:30 - 12:00 NOON South Ballroom Chair(s): C. D. Barnet G. L. Bjoraker Orton G.* Friedson J. Yanamandra-Fisher P. Baines K. Kaminski C. Momary T. Infrared Characterization of the Early Stages of the 1993 SEB Revival: Temperature Structure and Vertical Cloud Distribution Images of Jupiter were made between 1.6 and 18 micrometers from the NASA/IRTF to map the spatial distribution and time dependence of temperatures at 20 and 250 mbar and the vertical distribution and thickness of clouds between 0.01 and 5 bars. In the current apparition, observations were obtained in 1992 October and are scheduled through 1993 August. Observations on April 24-27 supported the World Astronomy Day #2 (Jovian Atmospheric Reference) campaign and characterized the early stages of the violent South Equatorial Belt Disturbance and the SEB revival, which began April 7. The 250-mbar temperature of the initial outbreak location was about 0.5 K lower than its surroundings (at 4500 km resolution), as was a broader region some 15 degrees eastward. This upwelling had transported particles detectable near 2 micrometers about the 100-mbar level, but not up to the 10-mbar level. Both regions had cloud reflectivity and 5-micrometer radiance indistinguishable from the surrounding zone-like SEB. To the immediate west of each, however, were regions of darker cloud albedo and brighter 5-micrometer radiance. This coincidence implies that areas immediately west of the upwelling material were reduced of particles with bright albedos and providing significant 5-micrometer opacity, most probably at the 600-mbar level. It is the reduction of this cloud material that was one key to "reviving" the SEB to its more usual appearance in the ensuing months. Yanamandra-Fisher P. A.* Hoffmann W. Friedson A. J. Orton G. S. Duetch L. K. Hora J. Fazio G. G. Shivanandan K. 1993 SEB Revival: Expansion Phase Post-SEB revival images of Jupiter were recorded at 4.6, 8.56, and 13.0 micrometers at Steward Observatory, Using the Mid-Infrared Array Camera (MIRAC) on 3-6 June 1993 to map the longitudinal regeneration of the South Equatorial Belt (SEB) at the 250- and 600-mbar and as deep as 5 bar levels. These observations were recorded almost 2 months after the initial violent and dramatic onset of the SEB Revival, which occurred on 7 April 1993. Longitudinal recovery of the SEB is observed at the 13.0 micrometer level (near 500-mbar level) and the two equatorial belts (NEB and SEB) appear comparable. The 8.56 micrometer data, sensitive to the upper NH3 ice cloud deck, reveal that the SEB is comparable to the NEB throughout the hemisphere preceding of the Great Red Spot (GRS), but has not recovered fully at other longitudes. However, the 4.6 micrometer data, sensitive to the radiances from the deeper levels of the atmosphere and cloud tops, show that the SEB is reviving as two distinct components--northern (SEBn) and southern (SEBs). The SEBn is evident at all longitudes, albeit very faintly and with a hint of structure, but the SEBs exhibits longitudinal inhomogeneity. These data imply that the SEB revival occurs on different timescales at various pressure levels and is still ongoing. Halthore R. N.* Allen J. E. Jr. DeCola P. L. On the Nondetection of the V3 Band of Methane on Jupiter High-resolution spectra of Jupiter in the 3.3-micrometer region have failed to reveal either the continuum or the line emissions that can unambiguously be attributed to the V3 band of methane (Drossart et al., 1993; Kim et al., 1991). Broadband V4 emission, on the other hand, is easily observable either from space or from the ground. With the help of two simple nonLTE models--a two-level model and a three-level model--we explore factors such as collisional relaxation due to hydrogen and solar pumping of the V3 levels that affect the emitted line intensities. Both models use previously measured experimental values of relaxation coefficients. Predicted V3 line intensities are shown to be 1 to 3 orders of magnitude below the 3-sigma noise level of previous observations for 2- and 3-level models respectively. For a methane mixing ratio of 2 x 10^-3, applicable to levels below the homopause, a value of 320 K is derived as an upper limit to the temperature. Lee Y. T.* Allen M. Yung Y. L. Solar System Hydrocarbon Photochemistry: Impact of New CH4 Ly-Alpha Photodissociation Quantum Yields The observed abundances of C2H6 and C2H2 in the atmospheres of the outer planets are simulated by current photochemical models with various degrees of success. Since the ultimate source of C2 hydrocarbons is methane, which is the most abundant carbon species in these atmospheres, different photolysis pathways of CH4 may result in different distributions of C2 hydrocarbons. Recent measurements of the primary product channels in CH4 photodissociation at Lyman-alpha (Mordaunt et al., J. Chem. Phys., 98, 2054 (1993)) have been used in new calculations of C2 species. The preliminary results for the atmosphere of Jupiter indicate that the recently reported CH4 photodissociation quantum yields lead to a reduction in C2 abundances. The C2H6/C2H2 ratios are mainly affected by the chemical mechanism of hydrogenation from C2H2 to C2H6 and are less sensitive to the specific choice of CH4 dissociation pathways. Calculations of C2 abundances for other solar system atmospheres will also be discussed. Gautier D.* LeCluse C. Robert F. Laboratory Determination of Deuterium Exchange Rates Between CH4 and H2: Application to Giant Planets In the gaseous envelopes of the giant planets, hydrogen is mainly in the form of H2 and deuterium in form of HD. However, because of the difficulty to properly detect HD spectral lines, the D/H ratio in hydrogen (noted (D/H)H2) is inferred from the CH3D/CH4 ratio through the relation: (D/H)H2 = [1 / (4 f) ] CH3D/CH4 (1) where f is the kinetic fractionation factor. f is usually estimated from theoretical calculations (Beer & Taylor, 1972; Fegley & Prinn, 1988). Since pressure and temperature vary rapidly in the tropospheres of these planets, f varies with the atmospheric depth z and depends upon the vertical atmospheric velocity v. We report here on laboratory measurements of the isotropic exchange rate constants between CD4 and H2. They permit us to evaluate the rate of the reaction: HD + CH4---->k(T) H2 + CH3D (2) where k(T) stands is the isotopic exchange rate constant for 1 atom. The variation between k(T) and the temperature has been determined experimentally: k(T) = 6.1 10^-25 exp(4.4 10^3/T) cm^3 sec^- (3) Using the formula (4) derived from the standard kinetic isotope exchange equation: d(f) / dz = P(z) . k(T(z)) [alpha(T(z)) - f] . v^-1 (4) (where alpha is the isotopic fractionation factor under thermodynamical equilibrium and z the atmospheric depth) we have calculated numerically the profile of f(z). Accordingly, the f values to be used in equation (1) are: 1.157, 1.288, 1.441, and 1.355 for Jupiter, Saturn, Uranus, and Neptune, respectively. The (D/H)H2 values in these planets are then derived from the most recent determination of CH3D/CH4 ratios. It appears that, even in Uranus and Neptune, (D/H)H2 seems to be only moderately enriched relative to the Protosolar value, recently revised by Geiss (1993) who proposes: (D/H)Protosolar = (2-6 +- 1) x 10^-5. References: [1] Geiss (1993) pre-print. [2] Beer & Taylor (1973) Astrophys. J., 179, 309-327; Fegley & Pnnn (1988) Asrophys. J. 326, 490-508. Varanasi P.* High-resolution Laboratory Data on the Infrared Spectral Lines of Methane and Other Hydrocarbons Needed in Studies of the Jovian Atmosphere Observed infrared spectra of the Jovian atmosphere contain a wealth of information on the spectral lines of CH4 and of several hydrocarbons that are created by the photodissociation of CH4. Accurate retrieval of atmospheric properties is dependent upon the availability of reliable laboratory data on the spectral lines of interest at the appropriate temperatures. Here we present absolute intensities, collision-broadened half-widths, and pressure- induced shifts on several lines in the fundamental bands of ^12CH4, ^13CH4, ^12CH3D, ^12C2H2, ^12C^13CH2, and ^12C2H4. Using a tunable diode laser spectrometer as well as a state-of-the-art Fourier transform spectrometer and He and H2 as broadening gases, the data have been obtained at several temperatures appropriate to the Jovian atmosphere. The data are useful in the spectroscopic determination of H:D and ^13C:^12C in the planetary atmosphere. Supported by the Planetary Atmospheres Branch of the Solar System Exploration Division of NASA under Grant-in-Aid No. NAGW-1894. Spilker T. R.* Hydrogen Broadening of Ammonia's Inversion Spectrum Microwave spectral measurements have been made on ammonia's inversion spectrum, broadened by hydrogen, at temperatures of 203 and 298 K and pressures from 0.5 to 4.5 atm. Over the range of conditions covered, the data most closely agree with predictions of the absorptivity formalism by Spilker [1990] as compared with predictions of the Van Vleck/Weisskopf [1945] and the Berge and Gulkis [1976] formalisms. Small systematic variations between the data and Spilker formalism predictions suggest minor adjustments to that formalism, which would increase its accuracy. This would bring the accuracy of our knowledge of the microwave absorptive behavior of ammonia more in line with the best opacity measurement accuracies for giant planet atmospheres. Results from this program will find immediate application in the interpretation (or reinterpretation) of radio astronomical and radio occultation data from observations of the giant planets, especially Jupiter and Saturn. They will also play an important role in the design of radio experiments for future missions to the giant planets. This research was done while the author held a NASA/National Research Council Research Assistantship at the Jet Propulsion Laboratory. Baguhl M.* Grun E. Linkert G. Linkert D. Siddique N. Identification of Small Dust Impacts in the Ulysses Dust Detector Data: Relevance to Jupiter Dust Streams Since October 1990 the dust detector onboard the Ulysses spacecraft recorded impacts of cosmic dust particles. A very rigid scheme has been applied so far to identify dust impacts from the noise background. These data with large signal amplitudes previously led to the identification of interstellar dust and of dust streams from Jupiter (Grun et al., Nature, 362, 428, 1993). Additional data with small signal amplitudes could be extracted from the noise background using coincidence arguments. In the time period from October 28, 1990 to December 31, 1992, 637 additional impacts have been identified. The total amount of reliably identified dust impact data increases by a factor of about 3. This increase permits a better statistical analysis especially of the Jupiter dust streams, which consist mostly of small and fast particles. Additional dust streams have been identified between the already known streams before and after Jupiter flyby. The dependence of the deflection from the Jupiter direction, the stream intensity and width on Jupiter distance support the assertion that they have been emitted from the jovian system. Io and Jupiter's gossamer ring has been proposed previously as a source of the dust. The relevance of the newly identified streams for the models is discussed. Eshleman V. R.* Gurrola E. M. Analytic Power-Law, Exponential, and Isothermal Models of Planetary Atmospheres The benchmark of analytic models of planetary atmospheres assumes isothermal ideal gas with constant scale height, so that both pressure and density decrease exponentially with radius. These combined exponential-isothermal attributes are well known to be incompatibile for real atmospheres since they require the strength of gravity to be constant with radius. For more realistic inverse-square gravity, unique analytic models are tractable if pressure is exponential, if density is exponential, or if the atmosphere is isothermal, but no two of these properties can apply to the same model. Although the above models can have useful appplications, all four leave much to be desired in problems related to occultation observations of tenuous planetary atmospheres. For example, the observable phase and frequency in radio occultations and the refractivity-controlled signal intensity in both stellar and radio studies of such atmospheres, can all be related to S/V, S'/V', and S"/V". Here S and V represent the density integrated along infinite tangential and semi-infinite veltical paths through the atmosphere, respectively, and the primes denote differentiation with respect to their common minimum radius. For the above four models, these three ratios involve complicated and different mixes of terms such as modified Bessel, modified Struve, exponential, and power functions of radius or its inverse. While remaining analytic, these results are not convenient to use. However, the complexities are unnecessarily obfuscating since there is a singular model with inverse-square gravity where S/V = S'/V' = S"/V" = a constant. It is neither isothermal nor exponential but pressure, density, temperature, pressure scale height, density scale height, S and its derivatives, and V and its derivatives all have the same functional form as powers of radius. We are particularly interested in this model as the fundamental unit for developing a superposition-ofmodels technique that can provide analytic representations of complex atmospheric structures and corresponding features of remote-sensing measurements. POSTERS Chanover N. J. Beebe R. F. Kuehn D. M. Temporal Development of the SEB Disturbance in Jupiter's Atmosphere Low-resolution Voyager 2 mosaics constructed by C. Avis and A. Collins at JPL have been used to characterize typical convective activity within Jupiters South Equatorial Belt (SEB). Between May 27 and June 8, 1979, three bright convective bubbles emerged 83 degrees, 104 degrees, and 106 degrees west of the Great Red Spot (GRS) respectively. The average north-south expansion rate of these bubbles was roughly 3 m/s, although at their onset they underwent a much more violent expansion, when they expanded roughly two degrees of latitude in two rotations of the planet. The resulting white clouds dispersed to the extent that they would not be visible at groundbased resolution in a time span of roughly eight days. Within that time they drifted eastward and expanded in longitude. An SEB Disturbance was observed on April 9, 1993, with the Tortugas Mountain Observatory at NMSU. An eight-color series of images was taken with a 24-inch telescope and full longitudinal coverage was obtained. This observation was followed by systematic monitoring of the disturbance over a period of 65 days. By May 7, 1993, the disturbance had developed to the point that five distinct bright clouds were visible to the east of the GRS. The extent to which the development of the 1993 cloud pattern is consistent with the Voyager convection is examined in detail. Center-to-limb reflectivity of the bright clouds is interpreted as vertical structure using the multilayered multiple scattering atmosphere model of Kuehn (Kuehn and Beebe, Icarus, 101, 2, 282). Sanchez-Lavega A. Gomez J. M. Miyazaki I. Lecacheux J. Colas F. Laques P. The Life Cycles of the Jovian SEB and New Phenomena Observed During the April 1993 Outburst The historical records and our own observations of Jupiter during the last twenty years are analyzed together to characterize the life cycles of the South Equatorial Belt (~ -20 degrees to -l0 degrees) in the 0.35-1.0 micron range. During the last century the SEB has suffered regular but aperiodic changes in its structure. We present a comparative study of their morphology, motions and characteristic times for their evolution patterns. Four major sequential phases can be distinguished: (a) Zone-like ("fade") status usually accompanied by a darkening of the EZs (from -10 degrees to 0 degrees) and by a uniform, very red and contrasted GRS; (b1) Outbreak of a bright white spot and development of a planetary-scale disturbance (SEBD); (b2) Disturbance interaction with the GRS and subsequent propagation of the activity to adjacent latitudes (EZs and STRZ Disturbances); (c) Quiescent belt-like status, with minor outbreaks of white cloud activity. Phase (b2) is not always observed. We also describe and document new phenomena observed during the 1993 April outburst phase: (1) A large low albedo uniform veil westward of GRS suggesting dispersion of cloud elements when interacting with the GRS and injection of particles to high altitudes; (2) A series of diagonal streaks crossing the equatorial region (a manifestation of waves triggered by the outburst ?), and (3) The retard in the development of the SEBD southern branch with the formation of a very dark spot to the northwest of the GRS. All these new patterns are probably related to the unusual proximity of the outbreak site, of only 20 degrees, to the east edge of the GRS. Molina A. Moreno F. Ortiz J. L. CCD Observations of Jupiter and Saturn During 1992-1993 Several sets of Jovian and Saturnian images covering the 0.3-1 micrometer spectral range are shown. The observations were carried out during 1992 and 1993 using the CCD camera of the 1.5m telescope at Calar Alto and that of the 2.5m Nordic Optical Telescope at La Palma (Spain). The high quality of some of these images allow us to study different atmospheric parameters. Specially interesting are the March to June 1993 Jovian images, which show the development of several features relationated with the 1993 South Equatorial Belt revival. Momary T. W. Baines K. H. Orton G. S. Jupiter's Polar Hoods and Prominent Temperate Latitude Features: An Assessment of Temporal Variability Between the 1992 and 1993 Apparitions Near-infrared imagery acquired at wavelengths within deep H2 and CH4 absorption bands reveal prominent polar hoods and, at temperate latitudes in both the northern and southern hemisphere, anomously bright, localized features ("spots"; each is 3-5 times brighter than their surroundings). We have observed such features at sub-arcsec resolution during the past two apparitions (23 Feb 1992 and 25 April 1993) with the ProtoCAM near-ir camera at the NASA/IRTF. We find that both polar hoods grew both significantly brighter and more extensive in area during this period. The northern hood is approximately 60% brighter in 1993 compared to 1992 at 2.14 micrometers, indicating perhaps a significant increase in the vertical column abundance of stratospheric aerosols. A distinctive asymmetrical shape is apparent in data acquired in both years, with increased asyrmmetry in 1993. Specifically, the hood boundary extends southward to 51 degrees lat at 319 degrees lon (System III), but only to 63 degrees lat at 184 degrees lon. When viewed from intermediate longitudes, the north polar hood appears "cocked", much as reported from 0.89-micrometer observations by W. Pryor (personnel communication). In contrast, the southern hood does not appear "cocked" in either year. One bright, localized spot was observed at northern temperate latitudes each year. In 1992, the northern spot was located 35.6 N lat and spanned 4*2 degrees in latitude and longitude. The 1993 spot was at 34.7 N lat, but was significantlylarger (8*6 degrees). In 1992, two southern spots were observed, co-located at 27.5 degrees S. lat. Each was significantly larger than their northern counterpart (~9*4 degrees). In 1993, a southern spot was located at 29.1 degrees S and extended 5.3 * 7.0 degrees. Finally, a brightening in the region of the SEB revival was centered at 10.7 degrees S lat, 322.4 degrees lon on 25 April 1993, spanning 5.0*6.3 degrees. Monday, October 18, 1993 MARS SURFACE II 10:30 - 12:00 NOON North Ballroom Chair(s): J. F. Bell III G. A. Swayze Miller J. S.* Singer R. B. Laboratory Reflectance of Ferric Mineral Mixtures: Application to Regional Mapping of the Martian Surface We are conducting a suite of laboratory reflectance measurements of ferric mineral mixtures to support analysis of our martian spectral imaging dataset (Singer et al., LPSC XXI, 1164, 1990). Our dataset was obtained during the 1988 and 1990 oppositions, and covers substantially all of Mars south of 40 degrees N with spectral coverage from 0.44 to 1.0 micrometers. The lab measurements will cover this range, focusing on the spectral regions near 0.53 micrometers and 0.86 micrometers, the location of absorption bands characteristic of Fe^3+ in hematite. Understanding the behavior of these spectral features when hematite is mixed with spectral dilutents of high and low albedo, and with martian surface analog materials will be useful for interpreting visible and near-infrared spectroscopy of the martian surface. Past investigations have revealed the detailed shape of crystalline hematite in ratio spectra of Meridiani Sinus to Acidalia Planitia (Singer and Miller, BAAS 23, 1174,1991). The characteristic hematite features are subtle in the original spectra and differences in strength are hard to see by overlaying. The ratio technique has proved useful in spectral comparisons for highlighting differences in spectral contribution from specific minerals. We have attempted to quantify the strength and map the distribution of hernatite spectral features on the martian surface to help in understanding regional geology and weathering history. These efforts have been complicated by the variation over the martian surface in relative strength of the two main hematite absorption features centered near 0.53 micrometers and 0.86 micrometers, and the strong positive (red) slope of martian reflectance spectra in the visible. These factors motivate us to gain a better understanding of the bchavior of Fe^3+ in spectral mixtures through laboratory spectral analyses. Objectives of the lab measurements include determining the behavior of the 0.53 micrometer and 0.86 micrometer features in mixtures of varying dilutions and in ratios of mixture spectra. Mixture components will include hematite, montmorillionite clay, oxidized and unoxidized basalt, and amorphous or nanocrystalline Fe^3+-containing palagonite. The shape and strength of the hematite Fe^3+ features will be observed in mixtures with high and low albedo spectral dilutents, as well as with martian spectral analog materials. These determinations will support a better quantification of the hematite spectral features, which will be applied to improve the quality and interpretability of regional mapping in our telescopic data set, and should be applicable more generally to analysis of visible and near-IR spectroscopy of Mars. Lee P.* Thomas P. C. Veverka J. Calvo S. Discovery of Longitudinal Dunes on Mars The features are resolved in three adjacent high-resolution (30 m/pxl) Viking Orbiter frames (525B11, B12, B13) and occur over an area of at least 900 km^2 in the northern circumpolar erg. The dunes appear as closely-spaced (~100 m) sets of subparallel linear ridges in the vicinity of a small (D = 6.5 km) impact crater. They reach several kilometers in length and occasionally display kinks and Yshaped junctions. As evidenced by a morphological continuum between the longitudinal dunes and the (transverse) barchane dunes present in the same area, the former appear to originate from the latter by modification of the barchane horns in a bimodal wind regime. While longitudinal dunes account for more than half of all the dunes in the sand seas on Earth, early studies noted that they were strikingly absent on Mars (Breed et al. 1979; Tsoar et al. 1979, JGR, 84). Because longitudinal dunes represent masses of passing sand in transit across a desert surface and are themselves usually transitory bedforms, their virtual absence on Mars implied that much of the available saltating material had already been stabilized in sites of accumulation. However, the localized occurrence of longitudinal dunes reported here suggests that aeolian processes involving saltation likely remain active in at least restricted areas of the northern circumpolar erg. The identification of these features as longitudinal dunes will be confirmed using very high resolution (2 m/pxl) Mars Observer images. Clifford S. M.* Mars: Thermal and Hydraulic Considerations Regarding the Fate of Water Discharged by the Outflow Channels to the Northern Plains The identification of possible shorelines in the martian northern plains suggests that the water discharged by the circum-Chryse outflow channels may have led to the formation of transient seas, or possible even an ocean, covering as much as a third of the planet. Speculations regarding the possible fate of this water have included local ponding and re-infiltration into the crust; freezing, sublimation, and eventual cold-trapping at higher latitudes; or the in situ survival of this frozen water to the present day--perhaps aided by burial beneath a protective cover of eolian sediment or lavas. However, consideration of the thermal behavior of water above and beneath the martian surface, and the kilometer or more hydraulic head difference between the water discharged from the channel source region and confined beneath the frozen crust of the northern plains, precludes any local mechanism for the significant reassimilation of flood water by either the frozen crust or the underlying regional aquifer under climatic conditions resembling those of today. On the other hand, neither cold-trapping at higher latitudes nor the subsequent freezing and burial of flood waters can be ruled out. It is important to note, however, that in the case of burial, thermal processes will rapidly redistribute any H2O present at depth into the overlying mantle--a process that will continue until the pore volume of the depositional layer is saturated throughout. These results further constrain the identification of geologically reasonable sinks for the water discharged by the outflow channels and may provide further insight to the subsequent geomorphic evolution of the northern plains. Hartmann W. K.* Survey of Lunar and Martian Cratering Record: Modifying the Current Paradigm Lunar and martian cratering data were reviewed in preparation for Mars Observer data analysis. Two standard format graphs of crater number/km2 vs. diameter have been developed, which allow quick and easy comparison of cratering data from different planetary geological provinces and also avoid the relatively illegible steep slope published by some workers. As I reported earlier, the lunar survey leads to significant modification of the paradigm that was used by the Voyager team to interpret planetary and satellite crater diameter distributions. The Voyager team reported that on the "relative plot" or "R-plot" format, lunar uplands generally are represented by what I call a V-shaped curve. They also found this shape of curve on some outer planet satellites, and interpreted it as the signature of a unique (ancient?) population of projectiles. This is probably incorrect. The V shape does not represent the lunar uplands generally. It is associated with regions that have ancient intercrater plains. Spectral evidence (including Galileo data) indicate these plains were mostly formed by flooding by ancient basaltic lavas, now covered by a veneer of highland dust. Models show how flooding can produce the V shape. These results will affect future interpretations and prepare the way for Mars Observer studies of martian cratering records from old and young martian regions. Mariner era interpretations stressed cratering evidence for an early erosive era followed by fairly abrupt climate change to current conditions. More recent work posits other interpretations. The small crater population, to be resolved by MO, will be especially informative in clarifying the issue, in dating channels, and perhaps in permitting searches for effects of a possible ancient denser atmosphere. Barlow N. G.* Martian Impact Craters as Indicators of the Subsurface Distribution of Volatiles The fluidized ejecta morphology surrounding the majority of fresh impact craters on Mars has been attributed to either impact into near-surface volatiles (Carr et al., 1977, JGR, 82) or by interaction of the ejecta with the thin martian atmosphere (Schultz, 1992, JGR, 97). A multifaceted study has been undertaken to determine the distribution and quantifiable qualities of the various ejecta morphologies in an attempt to constrain their manner of formation. Evidence to date supports the theory that near-surface volatiles (most likely H2O) are the primary cause of the fluidized ejecta morphologies observed on the planet. Five major ejecta morphologies are considered: single-lobe rampart (SL), double-lobe rampart (DL), multiple-lobe rampart (ML), radial (Rd), and diverse (Di). The three rampart morphologies are of the fluidized type, the radial morphology is more similar to typical lunar-type ejecta patterns, and the diverse morphology is a combination of fluidized and radial morphologies. The sizes and distributions of 3819 craters >=8 km diameter reveal a number of correlations. SL morphologies are primarily associated with craters in the 8- to 20-km diameter range in the equatorial regions, but cover a larger diameter range (up to 65 km) at higher latitudes. ML morphologies are found surrounding craters in the 1645-km-diameter range in the equatorial region, but are rare to nonexistent at higher latitudes. DL craters cover the 8-50-km-diameter range and are primarily found in the 30 degrees-60 degrees N latitude range. Rd craters are found globally for craters >60 km. Di morphologies are very localized in extent and cover the diameter range between ML and Rd morphologies. Few correlations between terrain type and ejecta morphology were discovered. Studies to determine variations in ejecta morphology with elevation are in progress. In addition, sinuosities of the fluidized morphologies were computed and ML morphologies are more sinuous than SL or DL morphologies (~1.2 for ML compared to 1.1 for SL or DL, where 1.0 is a circular morphology). Inner lobes of DL and ML craters are generally less sinuous than outer lobes. When depth-diameter relationships are computed, the observed diameter-latitude trend for SL and ML craters correlates well with the proposed areal and vertical distribution of subsurface H2O on Mars. According to this scenario, SL morphologies result from impact into ice whereas ML morphologies indicate influence from liquid water reservoirs at depth. The higher sinuosities for ML craters can be explained by an decrease in the clast/volatile ratio in the martian equatorial substrate. The observations are more consistent with the buried volatile model than with the ejecta-atmosphere interaction model. de Grenier M.* Pinet P. C. Chevrel S. Visible and Near-Infrared Reflectance Geometric Corrections of the Martian Surface and Implications for Spectral and Spectro-Imaging Mappings A nearly global visible (VIS) and near infrared (NIR) coverage of the martian surface has been achieved during the last 1988 and 1990 oppositions, by spectro-imaging telescopic means implemented at the 2-m telescope of Pic du Midi (France). The spectral observation focused on the 0.56-1.05-micrometer domain with a spectral resolution (R = lOO) and spatial resolution (100-300 km). Taking advantage of the substantial spatial overlapping existing within our data and resulting, for a given wavelength dataset, in a 15 to 25 degree martian longitude interval between the corresponding subterrestrial points, we propose a well-constrained estimate of the Minnaert exponent k for different wavelengths. This estimate is produced by minimizing the brightness difference rms residual of about 400,000 martian surface elements, observed under different geometric conditions and distributed across the available martian surface mapping in several separate overlapping frames. The phase angle is almost constant in these observations and ranges below 5 degrees. The incidence and emergence angle values of the dataset, due to the observation geometry, range from 0 to 90 degrees in a coupled behavior, and are predominantly distributed toward the low angle values (0-45 degrees). The results indicate that the mean value of exponent k varies from 0.59 to 0.68, with the rms residual varying from 4 to 11% when the wavelength is successively 0.56, 0.73, 0.91, 0.98 micrometers. The 4% residual means nearly all of the data spectral variance is explained and is consistent with the predominant effect of atmospheric attenuation known to occur in the VIS domain. The important NIR residual suggests that in addition to the geometric limb-darkening attenuation effect, there is a surface material contribution. Indeed, when considering the 0.98-micrometer wavelength, the k mean va1ue over the eastern hemisphere varies from 0.64 to 0.72 in relation to the proportion of dark and bright regions. This trend is consistent with a more precise analysis revealing a systematic k linear increase from low to high albedo areas. A linear regression fit of the dataset permits us to establish the following albedo (A)-dependent laws on the Minnaert coefficient, for the three wavelengths: 0.73 micrometers, k = (0.631 + 0.003).A + (0.500 + 0.003), 0.15 < A < 0.38 0.91 micrometers, k = (0.920 + 0.001).A + (0.397 + 0.002), 0.13 < A < 0.41 0.98 micrometers, k = (1.074 + O.OOl).A + (0.356 + 0.002), 0.15 < A < 0.41 These results allow presently to correct reflectance geometric variations and to produce VIS and NIR geometric albedo extended mosaics of the surface within a 3% to 8 % consistency. It also shows that one has to take into account the wavelength and albedo-dependent behavior of the Minnaert coefficient in the use of the spectroscopic and imaging data before any interpretation, a first estimate of the associated discrepancy between corrected and uncorrected data amounting up to 5 to 10% under certain conditions. Hayashi J. N.* Jakosky B. M. Haberle R. M. Atmospheric Effects on the Thermally Derived Albedo of Mars Haberle and Jakosky (1991) investigated the effect on thermal inertia of including a dusty C02 atmosphere and sensible heat exchange with the surface as opposed to the previously assumed constant atmospheric contribution equal to 2% of the maximum solar insolation (Kieffer et al. 1973, 1977; Palluconi and Kieffer 1981). We have utilized the Haberle and Jakosky model to investigate the effects on the thermally derived albedo. The thermally derived albedo together with the thermal inertia provides the best match for the observed temperatures. At low dust opacities, the 2% model derived albedo underestimates the Haberle and Jakosky model-derived surface albedo. This is most pronounced at low albedos and is attributed to the effects of CO2 in the atmosphere. For a constant low surface albedo and increasing dust opacity in the Haberle and Jakosky model, the thermal behavior is matched by an increasing albedo in the 2% model. This effect is attributed to scattering by dust in the atmosphere. For a constant high surface albedo and increasing dust opacity in the Haberle and Jakosky model, the thermal behavior is matched by a decreasing albedo in the 2% model. This effect is attributed to heating of the atmosphere by the dust. At low dust opacities, the 2% albedo adequately estimates the Haberle and Jakosky model mean planetary albedo. However, increasing the dust opacity increases the slope of the Haberle and Jakosky planetary albedo vs. 2% albedo to greater than 1. This change in slope is contrary to that of the observed albedo vs. the 2% albedo, which has an approximate slope less than 1. All of our model runs had a ratio of the dust infrared to visible optical depth equal to 0.5, consistent with Haberle and Jakosky (1991). We are currently exploring the effect of changing the dust properties on the slope of the Haberle and Jakosky planetary albedo vs. the 2% albedo. Zent A. P.* Quinn R. Laboratory Measurement of Co-Adsorption of H2O and CO2 Under Mars-like Conditions We are completing the first laboratory measurements of the simultaneous adsorption of CO2 and H2O on Mars-analog materials at temperatures and panial pressures appropriate to Mars. In laboratory experiments reported in the chemical engineering literature, H2O displaces CO2 from adsorption sites very effectively at temperatures and partial pressures higher than appropriate for Mars (e.g., Carter and Husain, Chem. Eng. Sci., 29, 267-273). While both species are present in the martian atmosphere and pore gases, climate models, which invoke regolith adsorbate as a volatile reservoir are based on laboratory experiments in which only one adsorbate is used. Future models must be augmented in order to account for the competition for a finite number of adsorption sites. At temperatures below the frost point however, H2O adsorbate may be segregated into ice, allowing more effective competition by CO2. In addition, the seasonal thermal wave passing through the shallow regolith may force in situ phase changes in H2O, and change the number of adsorption sites available for visitation by CO2. We find that CO2 is displaced from adsorption sites by the presence of H2O under Mars-like temperatures as long as H2O ice is absent. The implications for models of the martian cliate will be discussed. In general, high atmospheric pressure regimes during periods of high obliquity must be considered less likely. The inventory of CO2 adsorbed in the regolith must also be revised downward. We will present new estimates of the total exchangeable CO2 in the atmosphere-cap-regolith system. Wisdom J.* Touma J. The Chaotic Obliquity of Mars Numerical integration of the rotation of Mars with the chaotically evolving planetary system over 100 m.y. shows large chaotic variations in the obliquity of Mars. In our integrations the obliquity of Mars varies from about 10 degrees to 50 degrees on timescales of several million years. The chaos is associated with repeated separatrix crossings of a chaotically pulsating secular spin-orbit resonance. POSTERS Pierazzo E. Singer R. B. Britt D. T. Doose L. R. Smith P. H. Tomasko M. G. Keller H. U. Knudsen J. M. Soderblom L. A. Geologic Remote Sensing Science on Mars from MESUR Pathfinder The Imager for MESUR Pathfinder (IMP) is an innovative stereoscopic multispectral camera system based on a single CCD detector (see Britt et al., this volume). The CCD has a standard silicon spectral response, dropping off sharply below wavelengths of 0.4 micrometers and above about 1.0 micrometers. Fortunately this spectral region is well suited to studying alteration products and processes (including weathering) in dust, soil, and rocks, based on visible and near-IR spectral characteristics related to Fe^3+ abundance and mineralogy. At longer wavelengths (~0.8 to 1.0 micrometers) the system is sensitive to spectral features of Fe^2+ in mafic materials such as pyroxenes and basalts, and is capable of discriminating among various compositions and mineralogies. Mission constraints on cost, weight, and data rate required us to keep the design simple and limit the number of bandpasses for geologic remote sensing to 8. Optimization of location and width of these bandpasses has therefore been crucial for spectra discrimination of minerals, rocks, and soils known (or suspected) to occur on the surface of Mars. The science capabilities of this camera system are therefore excellent, both to guide the rovermounted Alpha/Proton/X-ray spectrometer to specific targets of interest and to remotely investigate and map the terrain surrounding the landing site. Initial bandpass positions were chosen based on prior knowledge, then refined analytically through modeling with laboratory and telescopic spectra. The first three bands are at 0.45, 0.53, 0.60 micrometers, for two reasons. First, the band centers are similar to Viking Orbiter and wlll facilitate comparisons at different spatial scales. Second, as proven with Viking images and telescopic spectroscopy, this combination is excellent for detecting subtle differences among ferric oxides and oxyhydroxides, and is especially sensitive to buLlc crystalline hematite. The fourth band, at 0.75 micrometers, corresponds to a reflectance maximum for many Fe3+ minerals. The fifth band (0.86 micrometers) is located at an absorption in crystalline hematite, while the sixth band (0.90 micrometers) is closer to the absorption for hydroxylated ferric oxides. The seventh and eighth bands (0.95 and 1.0 micrometer) help define the long-wavelength shoulder of the near-IR Fe^3+ bands. The fifth through eighth bands (0.86 to 1.0 micrometer) are also designed to characterize Fe^2+ absorptions diagnostic of pyroxenes and other mafic minerals. For low- to moderate-Ca pyroxenes the absorption minimum location can be estimated. For materials with minima at 1 micrometer and longer, such as high-Ca pyroxenes and olivine, we can characterize the short wavelength side of the absorption. Band widths vary from 30 to 50 nm, chosen to minimize exposure times in all channels. Modeling of bandpass width shows little difference in interpretive capability between narrow filters (10 nm) and the broader filters in IMP. Hauber E. Neukum G. Age Determinations of Valles Marineris Interior Deposits, Mars The large canyon system of the Valles Marineris offers a unique view into the upper part of the martian crust. Canyon walls, canyon floor, and various interior deposits like landslides or layered deposits represent a complex assemblance. Knowledge about the time and the mechanisms of emplacement of these materials is important in order to understand the sequence of geologic processes, which has led to the present state of the region. Of special interest are the absolute and relative ages of the different geologic units. While the surrounding plains are sufficiently large to yield well based crater retention ages, the small size of the interior deposits makes it difficult to get reliable crater counts. In detail, these tiny and sometimes not clearly separated units that formed subsequently to the ongin of the canyon itself consist of layered deposits, landslides, irregular floor deposits, fractured floor material, and surficial deposits (Lucchitta et al., in Mars, Univ. of Ariz. Press, 1992). Yet, only landslides have been dated by crater counting in an early work by Lucchitta (JGR 84, No. B14, 1979). In most cases, however, relative ages have been determined on the basis of stratigraphic positions. Our goal has been to define crater retention ages in the western part of the Valles Marineris, i.e., in Ius and Tithonium Chasmata. To this purpose, about 100 high-resolution Viking-Orbiter images have been map-projected and compiled to image mosaics covering the entire western Valles Marineris and parts of Noctis Labyrinthus. Using this image base as well as low-resolution color data, geologic units have been mapped and craters have been counted on these units. The results have been evaluated through application of the cratering chronology model of Neukum and Hiller (JGR 86, No. B4, 1981). Butler B. J. Muhleman D. O. Slade M. A. Results from 1992 and 1993 VLA/Goldstone 3.5 cm Radar Data In an attempt to further understand the surface and subsurface of Mars, radar experiments were undertaken during the most recent opposition. The motivation was threefold: first, we wanted to compare the radar return from the residual north polar ice cap (RNPIC) to that of the residual south cap (RSPIC). Second, we wanted to obtain data for those longitudes that we did not sample in our experiment in 1988. And finally, we wanted to verify results from our 1988 experiment (the 1988 experiment is explained in Muhleman, et. al., Science, 253, 1508-1513,1991). The experimental setup was similar to that described in Muhleman, et. al., with the exception of the transmitted power from the Goldstone antenna, which has been increased from 350 kW to 460 kW. We performed experiments on 3 dates: December 9, 1992 (subearth latitude, phi,~12 degrees), December 29, 1992 (phi ~9 degrees), and January 12, 1993 (phi ~6 degrees). Only the SS echo (or so- called "depolarized") data have been fully reduced to this point, and discussion will be limited to that data. Also, the December 9 data were compromised by transmission problems, and the only result from that date was an indication that the RNPIC was not nearly as efficient a backscatterer as the RSPIC. As far as verification of the 1988 results, we found that "Stealth" was still present, and the volcanic regions (Tharsis and Elysium) were still the most highly reflective regions on the planet, excluding the RSPIC. Global fits of the SS backscatter data have been made, incorporating all of the data taken thus far (including the 1988 data). Significant results are: "Stealth" is the least reflective region on the planet, but there are other regions which exhibit depressed reflectivities. The RSPIC is the most reflective region on the planet. The RNPIC has a slight enhancement in reflectivity in selected regions, but reflectivities are much less than those for the RSPIC. The reduced reflectivity can be explained by assuming that the ice is contaminated by silicate inclusions, with a volume fraction of ~50%. Other highly reflective regions are mostly contained in the Tharsis and Elysium volcanic regions. New global SS radar reflectivity maps will be shown. Lucey P. G. Williams T. Horton K. Rafert J. B. SMIFTS: A Cryogenically Cooled, Spatially Modulated, Imaging, Fourier Transform Spectrometer We describe a novel cryogenically cooled, spatially modulated, imaging, Fourier transform interferometer spectrometer for spectral measurement in the 1-5 micrometer range. Using spatial modulation and a detector array to sample the interferogram, the instrument employs no moving parts to obtain spectra. It is extremely robust and potentially more reliable than other interferometers in addition to taking advantage of the multiplexing afforded by array detectors. The instrument technology possesses a unique combination of characteristics that form a niche for spectral measurement not commonly recognized but of great potential value. These characteristics include broad wavelength range, wide field of view, simultaneous measurement of all spectral channels, compactness, no moving parts, and moderate resolution (lambda/Delta lambda = 100-1000). We will present test data derived from the instrument. The spatially modulated FTS spectral measurement technology has been described by several authors including Yoshihara and Kitade (1967), Okamoto et al. (1984), Barnes (1985), and Smith and Schempp (1991). The spectrometer requires no moving parts, is compact, and enjoys a number of advantages over other spectral data collection technologies such as gratings and tunable filters. The unique combination of characteristics define an important niche for remote sensing spectral data acquisition. In the simplest mode of operation, the spectral resolution is determined by the number of elements in the detector array and the cutoff frequency. Our implementation of the instrument employs an Amber Engineering 256 by 256 InSb element detector array, which enables a 10,000 wavenumber (1 micrometer) cutoff frequency and therefore a spectral resolution of 76 cm^-1. We will list and discuss the characteristics of the SMIFTS for spectral measurement, describe the interferometer technology upon which the instrument is based, and other optics essential to the instrument performance, and describe the cryogenically cooled implementation of this technology. Jakosky B. M. Henderson B. G. Mellon M. T. Chaotic Obliquity and the Nature of the Martian Climate Recent analysis of the martian obliquity shows that it is chaotic on timescales longer than 10^7 years. This suggests that the obliquity might have been as high as 60 degrees within the last 10^7 years and that it would almost certainly have been this high within the last 10^8 years. Using our polar-cap thermal models (Icarus, 102, 286,1993), we have calculated polar temperatures and water-ice sublimation at obliquities up to 60 degrees. At the highest obliquities, up to a meter of water ice could sublime each martian year, and the entire thickness of the polar deposits could sublime in under 10^4 years. This ice would presumably condense where it would be most stable-at middle and equatorial latitudes. Additionally, CO2-H2O clathrate hydrate is currently stable in the polar deposits at depths below a few meters. If the caps actually contain clathrate, the CO2 would be released as the caps sublime; if the caps are predominantly clathrate, then they could contain the equivalent of one bar of CO2. This CO2 would then be released at the highest obliquities, where it would remain predominantly atmospheric. This CO2 (i) could provide a sink for the putative early CO2 atmosphere, which provided some greenhouse warming and (ii) could be responsible for occasional warmer climates during recent epochs. Clearly, these results are very speculative, as conditions would be so different from those input into the model that the polar-cap temperatures, the water (and CO2) sublimation, and the climate would be very different from what we can predict. Stephens S. K. Stevenson D. J. Rossman G. R. Keyser L. F. Carbonate Formation on Mars: Latest Experiments and Models Laboratory simulations and computer models of martian CO2 storage may answer some fundamental questions about martian climate history and surface weathe?g. Could carbonate formation have reduced CO2 pressure from a hypothetical >1 bar to the present 7 mb in 3-4 b.y.? We address this and other problems with on- going pressure-drop (PD) experiments, coupled with various analytical techniques and modeling. Recent PD experiments exposed silicate powder (surface area ~3 m2/g) to 1 bar CO2 at ~20 degrees C for periods of 101-2 days. "Damp" (1 ml(sub)H2O for each 5 g(sub)silicate) results fall into 2 groups: [1] diopside (~CaMgSi2O6) and olivine ((Mg,Fe)2SiO4), [2] quartz (SiO2) and plagioclase (~CaAl2Si2O8). Group [1] shows rapid short-timescale (<3 days) CO2 pressure drop (Delta P) and slow long-tau Delta P, while group [2] gives Delta P ~ 0. "Dry" (no H2O), "vapor" (4 micro l), "damp," and "wet" (5 ml) diopside runs yield generally increasing Delta P for all taus. Results suggest adsorption (and other physical storage) and irreversible reaction (with CO2 + H2O); any reaction beyond a surface layer has not yet been investigated; and differences between [1] and [2] point to a thermodynamic or crystal structure role. Applying long-tau rates of <2.7 x 10^-5 g(sub)CO2/g(sub)silicate/day h(sub)regolith ~100 m, a limit of <75 mb/yr can be placed on CO2 uptake for a warm, wet Mars. We will use SEM to test for possible crack diffusion, and further PD runs (e.g., T < 0 degrees C, P < 1 bar, or using igneous glass) to characterize reaction mechanisms. Additional results involve an X-ray photoelectron spectroscopy (XPS) experiment and Fourier transform infrared spectroscopy (FTIR) analysis. Using XPS, a pristine diopside cleavage surface exposed to conditions similar to a "vapor" PD run shows no added carbonate, and FTIR spectra after the PD run show a change due to <1 wt% of an unidentified carbonate phase. These findings are interesting; we plan further analysis. Models of Martian CO2 history using experimental rates will address several issues: What is the effect of other sources or sinks on weathering timescales? Is Delta P affected more by gradual or catastrophic processes? Does carbonate decomposition (from rapid burial and shallow isotherms) stop early due to waning thermal flux? Where are martian carbonates? TITLE-ONLY PRESENTATIONS Britt D. T. Doose L. R. Singer R. B. Smith P. H. Tomasko M. G. Keller H. U. Knudsen J. M. Soderblom L. A. The Imager for MESUR Pathfinder (IMP) During the winter and spring 1993 the IMP team under Peter H. Smith (PI) competed in and won the JPL-supervised competition to provide the imager for the MESUR Pathfinder Mars lander. The IMP instrument is a binocular CCD-based imager with a 12-position filter wheel and full azimuth/elevation pointablity. The camera has a 12-cm stereo separation between the "eyes" and the optics are set (f/18, focal length 23 mm) so that the camera does not require active focusing (depth of field 0.65 m to infinity). The CCD uses frame transfer to shutter images electronically, eliminating another moving part. The active sensing area is 256 x 256 pixels in each "eye". The field of view for each eye is 14.4 degrees square and the IFOV is 1.0 mrad. The maximum resolution on the ground will be approximately 0.6 mm. The camera will sit on a deployable "jack-in-the-box" mast that can extend as much as a meter above the lander. Since the camera has full functionality in the stowed as well as the deployed configuration, it is possible to greatly expand the stereo baseline by taking panoramas both before and after mast deployment. The 12-position filter wheel is only on the right "eye" and contains eight filters optimized for Mars surface geology, three filters for atmospheric water vapor and dust measurements, and one broadband filter for stereo imaging with the other eye. Science experiments planned for the IMP include filter-wheel spectral mapping of the landing site to determine compositional variation and identify mineralogical units as targets for further investigation using the rover-based Alpha/Proton/X-ray spectrometer. Spectral mapping will also study weathering processes and products in the dust, soil, and rock. Included in the instrument package is J.M. Knudsen's magnetic properties experiment that will use up to five magnets of differing strengths to identify magnetic minerals in the martian dust. Stereo imaging and panoramas will study the large- and small-scale structure of the landing site, rock and dune features, and any erosional features. Additional images will be taken to study time-variable phenomena such as frost, dune formation, and seasonal changes. The instrument is also designed for a number of significant atmospheric studies. These include measurements of atmospheric water vapor, atmospheric dust, optical depth, and the size, shape, absorption spectra, and vertical distribution of dust in the atmosphere. Additional measurements include imaging a "wind sock" to estimate wind direction and imaging cloud patterns. Monday, October 18, 1993 INVITED TALK 1:30 - 2:20 PM Grand Ballroom Chair(s): G. Orton Moreno F.* The World Astronomy Days #2 Project: Preliminary Results on Groundbased Multispactral Imaging Analysis of Recent Changes in the Jovian Atmosphere The project #2 of the so-called World Astronomy Days of the IAU concerns the creation of a physical map through remote sounding of its atmospbere during the 1993 apparition. This apparition has been characterized by rapid changes in various jovian regions, the most remarkable being the South Equatorial Belt disturbance (SEBD). In this paper a preliminary analysis of this and other active regions is made from the dynamical evolution and photometric properties of features mostly based on 0.37-0.95 microns groundbased sub-arcsec imagery acquired using a large format CCD camera at the 1.52m spanish telescope at Calar Alto Observatory at various dates from March to June, 1993. Monday, October 18, 1993 INVITED TALK 2:20 - 3:00 PM Grand Ballroom Chair(s): G. Orton Shoemaker E. M.* Shoemaker C. S. Levy D. H. Collision of P/Shoemaker-Levy 9 with Jupiter The discovery of an impending collision of a comet with a planet is a unique event in the annals of astronomy. Detection of P/Shoemaker-Levy 9 in orbit around Jupiter and of two other comets (P/Gehrels 3 and P/Helin-Roman- Crockett) shortly after their escape from jovicentric orbit suggests that the steady population of comets brighter than H(sub)1 = 7 that are captured in orbit around Jupiter is about one. Collision of such comets with Jupiter probably occurs about once per century. If the pre-1992 precursor of P/SL 9 was substantially larger than the nuclei of P/G 3 and P/HRC, as seems likely, impact of an object with the energy of P/SL 9 may occur with a frequency of about once per millenium. Low-velocity encounters with the Jovian system evidently occur about an order of magnitude more frequently than calculated by the linearized two-body method of Opik. Collisions with Jupiter and the Galilean satellites are correspondingly more frequent. Collisions of comets on free heliocentric orbits are several times more frequent than collisions of comets captured as jovian satellites. Dimensions of the nuclei of P/SL 9 are not yet well established. If the largest objects are about 5 km in diameter, as much as 10 megatons total energy may be delivered to Jupiter's atmosphere. If the nuclei are smaller, however, the total energy could be one or two orders of magnitude lower. It is anticipated that impact of some of the largest nuclei, in July 1994, will be detectable from the Voyager and Galileo spacecraft. A campaign to study the nuclei is needed in order to predict observable effects of the collisions. The train of nuclei appears to be oriented at an angle to the orbit of the comet, and some objects may escape impact. Monday, October 18, 1993 Comet P/Shoemaker-Levy 9 and It's Impact on the Jovian System 3:30 - 5:30 PM Grand Ballroom Chair(s): E. M. Shoemaker, C. S. Shoemaker, D. H. Levy Jewitt D.* Luu J. Chen J. Physical Properties of Split Comet Shoemaker-Levy 9 We describe an ongoing program of observations of the split comet Shoemaker- Levy 9, designed to elucidate its basic physical and dynamical properties prior to the 1994 impact with Jupiter. The observations are taken under subarcsecond seeing conditions from Mauna Kea, and include both charge-coupled device images and spectra. To date, at least 21 separate nuclei have been identified (see below). Each nucleus supports a weak coma, so that photometry is able to yield only upper limits to the sizes of the underlying nuclei. CCD spectra show no evidence for molecular emission features due to CN, C2, or C3. The photometric and kinematic evolution of the fragments will be discussed, based on observations from the 1993 March to August observing window. Weaver H. A.* Feldman P. D. A'Hearn M. F. Arpigny C. Brown R. A. Helin E. F. Levy D. H. Marsden B. G. Meech K. J. Larson S. M. Noll K. S. Scotti J. V. Sekanina Z. Shoemaker C. S. Shoemaker E. M. Smith T. E. Storrs A. D. Yeomans D. K. Zellner B. HST Observations of Comet Shoemaker-Levy (1993e) Comet Shoemaker-Levy (1993e) was observed by the Hubble Space Telescope (HST) during July, 1993. The first observation executed successfully on July 1, and the second is scheduled for July 28. During the observation on July 1 four images were obtained with the PC using the F555W filter (similar to V-band) and exposure times of 100, 400, 700, and 700 s. Each nucleus is surrounded by a (roughly) spherical coma. The spatial brightness distribution of the coma is significantly flatter than the rho^-1 profile that is typically observed for comets having r~1 AU, so the coma is not due to the steady-state production of dust flowing outward from the nucleus. In order to estimate the magnitudes of the individual nuclei, the contribution from the coma must first be subtracted. Our initial attempts to subtract the coma result in the following preliminary estimates for the observed nucleus V magnitudes for the six brightest nuclei (starting from the southwest end and moving along the train to the northeast end): 23.8, 23.2, 23.5, 23.5, 24.0, and 23.5 with a relative uncertainty of ~0.1. The absolute value is dependent on the coma subtraction, and we believe that the numbers quoted are conservative lower limits to the true values. Assuming that the nuclei have a geometric albedo and color equal to that of the nucleus of P/Halley, and that the phase law is asteroidal, then an upper limit on the diameter of the largest nucleus is ~5 km. A 16 min. FOS spectrum covering the spectral region from 2223-3278 Angstroms was also obtained in order to search for gaseous emission from OH. Only scattered solar continuum radiation was detected, and the 3 sigma limit on the OH emission corresponds to an upper limit on the water production rate of about 3 x 10^27 s^-1. Chodas P. W.* Yeomans D. K. The Upcoming Collision of Comet Shoemaker-Levy 9 with Jupiter In late July 1994, the more than a dozen pieces of comet ShoemakerLevy 9 will likely collide with Jupiter. Using observations of the comet's central region over the interval from March 17 through June 26, preliminary orbital calculations have been carried out in a heliocentric frame and a backward integration of the comet's motion indicates that on July 8, 1992, the comet passed to within 1.6 Jupiter radii north of the planet's center, a distance that is well within this planet's Roche limit. The resultant tidal forces are the likely cause of the comet's splitting into multiple fragments. Prior to this Jupiter close approach, the comet was in a highly eccentric orbit about Jupiter itself. A forward extrapolation of this comet's motion and position uncertainties suggests that the center of the comet's train of particles has a high probability of colliding with Jupiter on July 20-21, 1994. When viewed in a heliocentric frame, the orbit of this comet resembles that of Jupiter. However, the comet is currently in a very eccentric (e > 0.99) and highly inclined (I = 83 degrees) Jupiter orbit with an apojove of 0.33 AU. Assuming a Jupiter radius of 69,700 km, Jupiter entry occurs on about July 20.5, 1994 at a relative velocity of 60 km/s. At entry, the Sun-Comet-Jupiter angle will be 64 degrees (night side) at a Jovicentric latitude of -45 degrees. As additional astrometric data become available, the comet's orbit will be improved and the rather uncertain Jupiter collision circumstances updated. Zahnle K.* MacLow M.-M. Chyba C. F. Some Consequences of the Collision of a Comet and Jupiter At the time we write this, comet Shoemaker-Levy IX (1993e) is expected to strike Jupiter on July 21 1994 (IAU circular 5800; D. K. Yeomans, pers. comm.). Based on the assumption that the fragments are as dark as P/Halley, the largest may be 10 km across (G. Shoemaker, pers. comm.). If so, the collisions will release some 10^31 ergs, which is comparable to the energy released by the Cretaceous-Tertiary (K-T) impact to the detriment of the dinosaurs et al. The energy to be released scales with the size of the fragments; a 1-km fragment would release only 10^28 ergs. Here we present model results of the catastrophic disruption and consequent explosion of cometary bodies hitting Jupiter. The model is directly based on our models of similar phenomena occurring on Earth. In these models we integrate equations describing aerodynamic drag and ablation of a spreading, fragmenting impactor. Ten kilometer comets explode at about the 100 bar level. One kilometer comets explode at about the 5 bar level. In either case the resulting fireball will rise bouyantly out of the atmosphere on a time scale of a minute. The fireball will lift with it water and other less volatile volatiles that are presently trapped below the visible cloud cover. The predicted Jovian impact qualifies as a serendipitous possible test of these models, although it may have more primal resonances as well. Marley M. S.* The Seismological Impact of Comet Shoemaker-Levy The impact of Comet Shoemaker-Levy (1993e) into Jupiter's atmosphere will release energies of l0^3l to 10^28 ergs (assuming impactor diameters of 1 to 10 km). A portion of this energy will be carried away as acoustic waves. A shock wave with an initial characteristic period of about 1 to 3 min (Zahnle et al., submitted) is generated by the explosive destruction of the comet. With distance from the explosion the shock dissipates into a pressure wave and the wave period increases. Acoustic waves will also be excited by the rapidly rising fireball resulting from the explosion. Downward-propagating components of the wave front will refract upward at various depths in the planet. Upon reaching the upper troposphere, those waves with periods longer than the acoustic cutoff period (about 4 min for Jupiter) will reflect downward and become trapped in the planet. These waves will excite the acoustic oscillation modes of the planet. Observed Jovian acoustic modes (Mosser et al. (1993) A & A 267:604) have reported periods of 8 to 17 minutes and energies of about 10^28 ergs/mode. Since the comet-produced acoustic waves will have predominantly shorter periods and comparable or smaller energies, the influence of the comet impact on the observable p-mode spectra may be small. However, those waves with periods less than the acoustic cut-off period will, after refraction in the interior, propagate upward past the tropopause. These waves will break, delivering energy and momentum into the stratosphere. The locations of stratospheric perturbations on the opposite side of the planet will reflect the paths taken by these waves. An annular "shadow zone" with no short-period wave paths connected to the impact point will be produced by the lower sound speed in Jupiter's core. Time- and spatially resolved observations of stratospheric temperature following the impact may thus provide a diagnostic of Jupiter's interior structure. Ahrens T. J.* Takata T. O'Keefe J. D. Comet Shoemaker-Levy 9 Impact on Jovian Atmosphere Dispersed fragments of Shoemaker-Levy 9 are expected to impact Jupiter in July 1994. Impact is expected because the predicted orbit intersects ~4 x 10^4 km from Jupiter's center. The largest of some 10 dispersed fragments are reported to have a diameter of 10 km. These fragments are expected to penetrate the jovian atmosphere with the initial velocity of 60 km/s at an angle of 40 degrees from the zenith. We have conducted three-dimensional numerical simulations of cometary penetration into the jovian atmosphere using the Smoothed Particle Hydrodynamics (SPH) method. The SPH method is a fully Lagrangian method and suitable for describing the motion of a highly distorted impactor. Hydrogen and an equation of state for ice are used to describe the jovian atmosphere and the icy comet respectively. We predict the energetics of energy transfer to the jovian atmosphere, the amount and distribution of cometary materials inserted into Jupiter's atmosphere, and the degree of atmospheric ionization, its radiative signature as a function of time, versus atmospheric height. We predict the light available, which could be reflected by the Galilean satellites and hence be immediately visible to the Earth upon impact. Postimpact events, which include focused free oscillations, may be observable from the Earth. A 10-km-diameter cometary fragment suffers ablation and vaporization during penetration. Transfer of ~50% of the kinetic energy to the jovian atmosphere occurs within a few hundred kilometers below the cloud tops and the object totally disperses before it reaches a depth of ~500 km. Harrington J.* LeBeau R. P. Jr. Backes K. A. Dowling T. E. Dynamical Response of Jupiter's Atmosphere to its Collision with Comet P/Shoemaker-Levy 9 The predicted collisions of pieces of comet P/Shoemaker-Levy 9 with Jupiter (I.A.U. Circ. 5800) would deposit 10^18-10^24 J of energy in the atmosphere for impacts of pieces 0.1-10 km in diameter respectively. Using the Explicit Planetary Isentropic-Coordinate atmospheric model for Jupiter (Dowling, this conference), we predict observable effects resulting from meteorological adjustment of the atmosphere for several months after the collision. We simulate the atmosphere vertically to a depth of 10 bar (stratosphere and troposphere) in 4-10 layers, and horizontally resolve the first baroclinic deformation radius. Since the model assumes a hydrostatic atmosphere, we do not model the initial nonhydrostatic phase of an impact event; rather, we add energy as heat and model its effect on the weather. The simulations cover the varying sizes of the cometary pieces, the effects of impacts in and near existing spots and shear zones, and a range of deposited energies and energy release depths spanning the work of Sekanina (1993, Science, submitted) and Zahnle et al. (1993, Nature, submitted). Dessler A. J.* Sandel B. R. Effects of Comet Shoemaker-Levy on the Jovian Magnetosphere Comet Shoemaker-Levy will undoubtedly either modify the existing workings of the jovian magnetosphere, or it will produce some new, hitherto unobserved magnetospheric phenomena, or it will do both. What will be the nature of these comet-induced phenomena, and will they be observable? The value of Q, the total outgassing rate of the comet and its accompanying cloud of dust and debris, is an important variable needed to answer the above question. When the cometary gas cloud crosses Jupiter's solar-wind bow shock and then enters the magnetosphere, it will encounter accelerated electrons that are not present in the solar wind. At this point, electron impact will start dissociating and ionizing the gas. The line-of-sight column density of un-ionized gas is approximately N ~ Q/vr, where v is the speed of gas escaping from the comet, and r is the distance from the comet train. If we assume Q is large enough to be interesting, one observable phenomenon would be the magnetosheath and magnetopause becoming invisible as seen in the glow from dissociation and ionization products. The effect on radio emissions ought to be profound. From about 6 R(sub)J outward, the jovian magnetosphere is filled to the bursting point with plasma from Io. The torus weighs about 10^9 kg, and the injection rate from Io is about 10^3 kg/sec The deposition of more plasma than this ought to cause an explosive disruption of the magnetosphere. If the comet does not add enough plasma on its inbound leg, the splash from its atmospheric impact will surely do the job. Finally, there is a faint possibility of an effect of the magnetosphere on the comet nuclei. The energetic electrons in the inner magnetosphere can penetrate about a meter into the comet, and their flux is sufficient to charge this subsurface layer at a rate of about 10^-11 Coulombs/cm2-sec. This rapid charging raises the possibility of causing the comet to shed meter-sized pieces from its surface. Also, dust and ice particles in the range 1-100 micrometers can be disrupted and vaporized by becoming electrically charged by magnetospheric plasma. These last two effects will increase the mass loading rate. Romig J.* Shoemaker-Levy 9: Radio Emission from the Terminal Encounter The center of the swarm of objects that comprise Comet Shoemaker-Levy 9 will enter Jupiter's atmosphere on about July 21, 1994. There are at least 17 objects in the swarm and individual entries will be spread over several days. The entry velocity will be 60 km/sec and is likely to make an angle of 45 degrees with the local normal. The entry point is on the night side at 45 degrees south latitude and 26 degrees behind the dawn terminator (information courtesy of D. K. Yeomans, JPL.) Major uncertainties exist concerning the size, composition, and strength of the individual objects. The upper limit on size is believed to be 10 km.(G. Shoemaker, PC.) An object of this size having the density of water would deliver approximately 1.0E + 31 ergs to Jupiter's atmosphere--an energy comparable to that in a typical solar flare. The entry phenomena, terminal flare, ascending fireball, and attending shocks generated by objects having a range of sizes, compositions, and strengths believed to be representative of the swarm have been modeled by Z. Sekanina, and by K. Zahnle, M-M. MacLow, and C.Chyba. These models lead us to anticipate robust radio emission during the several-minute interval encompassing the entry and demise of the larger objects of the swarm. Since the entry point is on the night side of Jupiter, none of this emission can reach Earth. All of this emission, however, will reach Voyagers 1 and 2 and may be detectable by the onboard Planetary Radio Astronomy (PRA) instruments, both of which are still active. The Voyager PRA Team is evaluating the type and strength of the anticipated radio emission and devising appropriate observing strategies. This talk will summarize our analysis and conclusions. Johnson T. V.* Shoemaker/Levy Observations from Galileo At the time of the predicted impact of Comet Shoemaker/ Levy 9 into Jupiter, the Galileo Spacecraft will be in a position to view Jupiter at a phase angle of ~50 degrees. Current estimates of the Comets trajectory indicate that the mean impact point will be visible on the limb of Jupiter as seen from Galileo. Preliminary plans and capabilities for Galileo observations will be discussed. This work was carried out at Jet Propulsion Laboratory/California Institute of Technology under a contract with NASA. Horanyi M.* New Rings for Jupiter? The fragments of the recently discovered Comet Shoemaker-Levy 9 is expected to return to the close vicinity of Jupiter on July 25, 1994 (Marsden, IAUC #5800). The parent nucleus is thought to have broken up during its previous close encounter around July 9, 1992 and the fragments now orbit Jupiter on a loosely bound high eccentricity orbit. The fragments are slowly spreading apart and some of them may hit Jupiter. Fragments that hit Jupiter will burn up, perhaps resulting in a spectacular event that could be observed from Earth. Pieces of the original nucleus, due to the freshly exposed volatiles, seem very active and during the next (final ?) approach they will deliver dust in situ to the jovian magnetosphere. Dust grains, once exposed to the plasma environment, will collect electrostatic charges and interact with the magnetic and electric fields of the magnetosphere. For larger grains, radius a >= 10 micrometers, magnetospheric perturbations are negligible and they will faithfully follow their source and crash into or escape Jupiter. Very small grains, a <= 0.1 micrometers, can rapidly gain (loose) energy outside (inside) synchronous orbit and escape (crash) in a short time (Horanyi, Morfill, and Grun, Nature, 1993). However, dust grains in between these size limits will get captured in the jovian magnetosphere. These grains will get dispersed or focused depending on their size and location in the plasma environment. Using a detailed description of the particles and fields environment, I modeled the orbital evolution of dust particles that will be produced as the fragmented nucleus of Comet Shoemaker-Levy 9 traverse the jovian magnetosphere. I will give estimates on the size distribution of the grains, the location, and the optical depth of the expected new rings. Schenk P.* Melosh H. J. Split Comets and Crater Chains on Callisto and Ganymede Voyager 1 returned, among other things, images of several prominent crater chains on the surface of Callisto. These impressively straight chains, or catena, consist of between 6 and 30 craters with diameters up to 30 km. The largest catena stretches some 700 km across the surface. A total of 18 such chains have been identified on Callisto, and at least 4 crater chains have been identified on the younger surface of Ganymede. Ejecta deposits and central peaks indicate these craters are of impact origin. They resemble some secondary chains of large lunar basins, but a search for source basins for the Callisto chains has been unsuccessful. The recent observations of comet Shoemaker-Levy 9, which split into a line of ~20 fragments during a pass inside the Roche limit of Jupiter, suggest to us that split cometary nuclei may be responsible for at least some of the crater chains on Callisto. We have modeled the tidal splitting of a cometary nucleus following encounter with Jupiter. Lengths of comet chains are a function of comet size, distance of closest approach to Jupiter, and distance from Jupiter at impact (radius of satellite orbit). We find that the observed lengths of crater chains on Ganymede and Callisto are similar to the lengths of comet fragment chains predicted by our model. The simple theory of splitting we employ predicts a general correlation between crater chain length and the average crater size within each chain, with scatter depending on satellite encounter geometry. A rough correlation between crater size and chain length is observed. The preference of crater chains for the Jupiter-facing hemispheres of both Ganymede and Callisto is consistent with tidal breakup of comets as a source for some of the crater chains on Ganymede and Callisto. From the number of crater chains on Callisto, we estimate a frequency of fragmentation of between 50 and 100 years, consistent with observations. This apparent fossil record of cometary fragmentation observed on Ganymede and Callisto may ultimately allow us to derive limits on the material strengths of comets. POSTERS Chen J. Jewitt D. On the Rate at Which Comets Split Time-resolved charge-coupled device (CCD) images are used to assess the frequency of splitting of comets. We compare four R-band images of each comet to search for co-moving companions. When blinked on a computer, the CCD images provide relatively high and uniform sensitivity to secondary nuclei. In a sample of 50 comets, 3 are found to be split. The observable lifetime of the secondary nuclei is estimated to be ~6 yrs. Hence, we estimate the lower limit to the cometary splitting rate to be S ~ 0.01 per year. This large rate suggests that 1) nuclear splitting is common in comets, 2) the parent nuclei have a tendency to "peel off" rather than break up, and 3) splitting may be an important destructive process for the cometary nuclei. Hicks M. Fink U. Wisniewski W. P/Shoemaker-Levy 1993e: A Time-dependent Coma Development Model Images of Shoemaker-Levy 1993e (SL92e) were obtained on March 28, April 24, and June 20, 1993 at the 2.3 meter telescope on Kitt Peak and the 1.5 meter telescope on Mt. Bigelow in order to follow the development of this singular ob3cct. We wish to determine if the dust coma has its origins in the traditional schema of hydrodynamic drag acting upon entrenched dust grains or if it is due to some other mechanism acting upon small-scale particles produced in the tidal breakup. Preliminary results show that the developing coma can be fitted with a time dependent model of coma formation with the outgassing of the individual nuclei in the comet train starting at the time of initial disruption. Our model computes the column density of the dust as a function of distance perpendicular to the comet train and as a function of time. For a steady state coma emitted from a point source in the absence of radiation pressure the column density falls off as 1/R. However, in the case of SL93e, the coma behaves more like a line source near the center of the train with a pronounced flat region extending some finite distance. By observing how this region develops with time and the application of our model we are able to deduce an outflow velocity for the dust on the order of 1-3 m/s. This model assumes no initial coma before breakup, which is supported by the absence of discernible coma in the prediscovery plates. Through the application of a hydrodynamic gas drag model (Probstein 1968) our dust outflow velocity is consistent with a gas production rate of ~10^14 particles cm^-2 sec^-1. This production rate can be supported by the sublimation of H2O, which at the heliocentnc distance of SL93e, has a an energy-limited maximum production rate of 3.4 x 1015 particles cm^-2 sec^-1. It is unnecessary to call upon a more volatile icy component. The maximum particle size lofted from the nuclei should be on the order of 5 micrometers assuming silicate dust and a nuclear size of ~2 km. The above discussion neglects the effects of radiation pressure. We shall present results from a model, which implicitly incorporates this mechanism. The well defined leading edge of the coma of SL93e in the sunward direction should provide an independent check of the maximum particle size and outflow velocity. Given the segregating action of the radiation pressure upon partides of various sizes, information on particde size distributions shall be obtained. Friedson A. J. Orton G. S. Effect of the Collision of Comet Shoemaker-Levy 9 on the Appearance of Jupiter in the Near and Thermal IR Current estimates for the trajectory of comet Shoemaker-Levy 9 predict that the string of fragments will most likely strike the night side of Jupiter. The target region will not rotate into view until some 1-5 hours after the impact of comet material. Hence, observations of the effects of large impacts on the neutral atmosphere will be limited to examination of residual phenomena that linger hours to days after any dramatic fireball associated with a large airburst has come and gone. Here we attempt to estimate the extent and nature of any alteration to the near and thermal infrared appearance of the planet from its precollision state. Emission from a spatially resolved dust cloud of optical depth ~0.05, deposited in a layer near the 0.1-mbar pressure level and radiating with the ambient atmospheric temperature, would be easily detected in the 8-micrometer spectral region. A cloud with at least this optical depth could be produced if ~10^l4 g of condensible material with ~0.5-micrometer- radius particles were spread over an area of radius 10^4 km in the upper stratosphere, not an unreasonable result in the aftermath of a number of large impacts. The same cloud would also be seen in the near infrared by observing in strong methane and hydrogen absorption bands such as those between 1.7 and 2.3 micrometers. Large convergence of horizontal winds in the rebound following the rise of a large fireball would be expected to generate vorticity in the region of the impact. Rossby waves or closed vortices produced in the troposphere would probably be detectable by their thermal signature at 18 micrometers. Boslough M. B. Crawford D. A. Kipp M. E. Trucano T. G. McGlaun J. M. Simulations of Cometary Interaction with Jupiter: Preliminary 2D and 3D Calculations of the Sensitivity of the Event to Bolide Representation and Angle of Incidence Simulations of the predicted hypervelocity interactions of Comet Shoemaker- Levy 1993e fragments with Jupiter will provide insights into the observable results of the events, allowing estimates of disturbance amplitudes and providing guidance for observations. Preliminary numerical simulations of the early phase of the interaction events will be reported, with a focus on the generation of atmospheric disturbances by the impacting bodies and on energy coupling to the planet. Two-dimensional calculations will indicate the influence of gravity and upper atmospheric density gradients on the interaction. Three-dimensional simulations will deterrnine the extent to which the impact obliquity is important and how the angle of incidence may influence fragmentation of the objects. Three-dimensional calculations will also determine the spatial dispersion of the projectiles, as well as trajectory alterations driven by pressure on the objects during entry. These numerical simulations make use of a strong shock, large deformation, Eulerian code, which has been experimentally validated over a wide range of conditions. It is capable of modeling high-temperature equations of state for multiple phases, including states ranging from condensed matter through plasma. This work performed at Sandia National Laboratories supported by the U.S. Department of Energy under contract DE-AC04-76DP00789. Chhabildas L. C. Boslough M. B. Crawford D. A. Furnish M. D. Grady D. E. Applications of Experimental Hypervelocity Impact Techniques to Studies of Cometary-Planet Impacts Sandia has developed experimental techniques and capabilities to investigate impact effects at hypervelocities in excess of 12 km/s. These techniques can now be used to investlgate interactions of bolides with planetary bodies. We will present the techniques that have been developed and how they can be used to study high-energy shock physics, impact-induced light emission (impact flash), fracture, fragmentation, shock-induced vaporization, and other phenomena associated with hypervelocity interactions. Diagnostic techniques include flash X-ray radiography, pulsed-laser holography and photography, fast framing and streak photography, time-resolved visible and ultraviolet spectroscopy, time-resolved velocity interferometry (VISAR), and time-resolved piezofilm (PVDF) stress measurements. These capabilities can be used to experimentally benchmark processes expected to occur when Comet Shoemaker-Levy 1993e fragments collide with Jupiter. This work performed at Sandia National Laboratories supported by the U.S. Department of Energy under contract DE-AC04-76DP00789. Kim Y. H. Kim S. J. A'Hearn M F. Ionospheric Effects of the Collision Between Comet P/Shoemaker-Levy 9 (1993e) and Jupiter The anticipated collision of Comet P/Shoemaker-Levy 9 (1993e) with Jupiter in July 1994 is an invaluable cosmic experiment that can provide information on violent changes in cometary and Jupiter's atmospheres as they interact during the entry phase. Although the main bodies of cometary fragments may explode below the Jovian homopause (Sekanina, private communication 1993), a significant amount of gas and dust from the comae and the blazing cometary surface will be injected into the thermosphere and ionosphere. We attempt to model the ionosphere disturbed by the huge amount of cometary gases, mainly H2O, OH, CO, and CO2. The dominant abundance of H2O in the comet will cause charge transfer reactions of H+ and H3 with H2O, resulting in a decrease of H3 and electron densities in the ionosphere. The injected cometary gases will spread away from the comet's collision path by diffusion and thermospheric wind. Information on thermospheric wind may be derived from imaging observations of infrared H3 emission at 3.5 microns since the pattern of H3 deficiency relative to the surrounding undisturbed ionosphere will reflect the spreading of the cometary H2O gas in the thermosphere. The Jovian thermospheric wind in the equatorial region is totally unknown yet, but a possible thermospheric wind in the north polar region was reported by Kim et al. (a paper presented at the 18th meeting of the European Geophysical Society held in Wiesbaden, Germany, May 1993) who presented an apparent movement of a small patch of H3 emission. Kim S. J. Kim Y. H. A'Hearn M. F. Cheng A. F. Collision Between Comet P/Shoemaker-Levy 9 (1993e) and Jupiter's Magnetosphere: Its Influence on the Auroral Activities of Jupiter Although the exact location of the collision with Jupiter is not absolutely yet definite, it is certain that there must be collisions of the cometary tails, comae, and nuclei with the magnetosphere of Jupiter. In order to understand this gigantic cosmic experiment, we attempt to make models such as expected collisional, radiative, chemical, and MHD processes in the magnetosphere. In particular we will focus on: (1) the formation of bow shocks of cometary nuclei inside the magnetosphere; (2) the energization of ionized cometary constituents in the magnetosphere; (3) the diffusion of ionized particles in the magnetosphere; and (4) the eventual precipitation of the energized particles on the auroral regions of Jupiter. We will discuss any significant changes in the flux and spectrum of precipitating particles, and the spectral characteristics of the auroral regions. We will also discuss possible changes in the intensity and morphology of H3+ emission in the auroral regions. Garcia L. Reyes F. Carr T. D. Observations from the UFRO of the Decametric Emission of Jupiter During the Collision of Comet Shoemaker-Levy 1993e The decametric radio emission of the planet Jupiter has been monitored at the Univ. of Florida Radio Observatory (UFRO) since the year after their discovery in 1955. The main telescopes used at the UFRO to monitor the emission are the 640-dipole large array at 26.3 MHz, a broadband array of 4 right hand circularly polarized logspiral antennas (TP) having a frequency range of 12 to 40 MHz, and several Yagi antennas at the frequencies of 15, 18, 20, 22, and 26.3 MHz. The instrumentation attached to the telescopes are an Acousto Optical Radio Spectrograph (AORS) having a frequency range of 24 to 39 MHz connected to the TP array, a high resolution high sensitivity time expansion spectrograph based on an instrumentation tape recorder and a 200- channel real time spectrum analyzer connected to the 26.3 MHz large array. We expect to have a second array of 4 left hand circularly polarized log spiral antennas operational for the observation of the collision. During the collision the UFRO will observe the planet as part of the IJW Decametric Network. The network will include several observatories providing world-wide coverage of this unique event. From a few weeks before, during and a few weeks after the collision we will monitor the emission from the planet. We will obtain daily plots of intensity vs. time for each discrete frequency channel. The two spectrographs will provide dynamic spectra of moderate (200 kHz and 15 ms) and high (3.3 kHz and 0.3 ms) resolution of selected portions of the storms. The dynamic spectra will allow the detection of changes in intensity, duration, and frequency coverage of the emission . The entry of the comet into the Jovian magnetosphere will provide ionized material that may trigger, absorb, or modify the decametric emission. If the electron density around the pieces is large enough, a current could flow between the pieces and the Jovian magnetosphere, cretaing a temporary 'comet flux tube'. Some of these phenomena are similar in nature and magnitude to those caused by Io, which has a well known effect on the decametric emission. The orientation of the orbit of the comet may provide a unique opportunity to test some of the emission models. Chapman C. R. Galileo Imaging Team Plans to Observe the Jupiter Impact Point of Comet 1993e Using the Galileo SSI Camera The Galileo camera may provide a unique opportunity to obtain spatially resolved observations of the impact of Comet Shoemaker-Levy into Jupiter next year. The currently projected impact of the center of the train of comet fragments (~22 July 1994) occurs on the backside of Jupiter as viewed from Earth, but approximately at the limb (perhaps slightly on the frontside/ darkside beyond morning terminator) as seen from Galileo. The spacecraft will be ~1.5 AU from Jupiter, with Jupiter at a phase angle of ~50 degrees and ~60 pixels in diameter. The Galileo camera would then be the only instrument anywhere in the solar system capable of directly observing the impact point, and related darkside phenomena, with spatial resolution. Playback of Ida encounter data will have been completed and the spacecraft will have a 10 bps capability to play back portions of comet impact pictures recorded on tape. We will describe our preliminary plans to use the SSI camera for this purpose, including considerations of timing, exposures, filters, and playback strategy. The experiment will clearly be aided by the best possible predictions of the times of individual fragment impacts and by after-the-fact determinations of those times from Earth-based observations of indirect phenomena. Our data can provide unique documentation of these extraordinarily rare large "cratering" impacts into Jupiter's atmosphere, and we can search for lingering luminous phenomena as the impact points rotate toward the morning terminator. TITLE-ONLY PRESENTATIONS West R. A. Friedson A. J. Cometary Particles as a Tracer of Jupiter's Stratospheric Circulation The impact of fragments of comet Shoemaker-Levy 9 on Jupiter's atmosphere in July, 1994 may provide an unprecedented opportunity to study Jupiter's stratospheric circulation. Recent calculations (Z. Sekanina, paper submitted to Science, 1993) predict that much of the comet material will be deposited in Jupiter's stratosphere. If so, and if the material is deposited in a confined region (10,000 km or less, horizontally) we can expect a situation analogous to an El Chichon or Pinatubo event for the terrestrial stratosphere. Initially the volatile material will be vaporized and will rapidly recondense. The large ice crystals and dust particles will rain out and be lost to the troposphere. The cloud of small particles that remain may have settling times of more than a year. These sub-micron to micron particles would probably be easily seen in methane filter images in the near-IR, and possibly in the ultraviolet. An observational program to monitor the dispersal of this cloud or clouds would reveal much about the nature of the circulation. Some predictions about the meridional evolution of the clouds can be made already, based on the meridional circulation model of West et al. (Icarus 100, 245-259, 1992) unless the impact itself significantly disrupts the annual average circulation well after the initial transients die away. Tuesday, October 19, 1993 COMETS I 8:30 - 10:00 AM South Ballroom Chair(s): H. Campins N. Samarasinha Schleicher D. G.* Eby R. M. Alciatore L. A. The Photometric Behavior of Comet P/Halley in 1985 As has been widely reported, large day-to-day variations were observed in Comet Halley's level of activity throughout its recent apparition. Most previous efforts at constructing a detailed lightcurve have concentrated on the March through June 1986 timeframe. During this interval, Halley clearly exhibited a slowly evolving lightcurve with a period of approximately 7.4 days (cf. Schleicher et al., 1990, Astron. J., 10, 896). However, a good lightcurve for the months preceding perihelion could not be constructed due to the comet's smaller amplitude of variation and lower mean brightness, and to the less complete observational coverage from the four sites included in the Schleicher et al. study. In order to facilitate the interpretation of a variety of nonphotometric observations of Halley, as well as to further constrain the rotational state of Halley's nucleus, we have attempted to create the most complete lightcurve possible for 1985 by incorporating all narrowband photometry archived by the International Halley Watch (IHW). Preliminary analysis indicates that the data are consistent with a 7.4-day period, but that more rapid evolution of the lightcurve may have occurred in the fall of 1985 than took place in 1986. The IHW data also clearly show a strong phase effect for the dust particles as Halley varied in phase from 25 degrees in October to <2 degrees in mid-November, and back to >50 degrees in December. This effect is similar to that reported by Meech and Jewitt (1987, Astron. Astrophys., 187, 585) from spectroscopic data obtained at phase angles less than 10 degrees. We will present final results of both the lightcurve investigation and the phase analysis of the dust. This research was supported by NASA grant NAGW-2366. Klavetter J. J.* A'Hearn M. F. An Extended Source for CN Jets in Comet P/Halley--Models and Results Various groundbased and spacecraft observations indicated a possible extended source for some species in comet P/Halley (e.g., A'Hearn et al. 1986. Nature. 324:649 and Eberhardt et al. 1987. Astron. Astrophys. 187:481). We used CN images to determine that the gaseous jets have an extended source origin but that the diffuse component of the coma does not (Klavetter and A'Hearn 1991. BAAS. 23:1167 and Klavetter and A'Hearn 1993. Submitted to Icarus). We present our methods and quantitative examples showing the extended source for the jets and contrast this with the diffuse CN and the continuum. In addition, we present jet and diffuse models showing basic agreement with the data. We also show that as much as 50% of the diffuse CN component originated in the jets. Moreels G.* Clairemidi J. Rousselot P. Goidet B. Detection of Aromatic Compounds in the UV Spectrum of Comet P/Halley Emission spectra of comet P/Halley in the 275-710 nm range were obtained with the three-channel spectrometer of Vega 2 during the encounter session on March 9, 1986. The instrument scanned a region in the inner coma at cometocentric distances rho = 400-40000 km. The spectra show the molecular emissions of OH, NH, CN, CH, C2, and NH2 superposed on a dust-solar scattered continuum that is very intense at rho < 3000 km. Very close to the nucleus, i.e. when the projected distance rho is smaller than 1000 km, new bands appear between 310 and 380 nm with an intensity increasing as 1/rho. A broad band extending from 342 to 375 nm showing four peaks at 347, 356, 364, and 374 nm is identified as due to the fluorescence of phenanthrene, on the basis of a comparison with a laboratory spectrum obtained under jet-cooled conditions and laser excitation at 282.6 nm. A second intense band, which arises at 323 nm close to the nucleus is identified as due to naphthalene, according to a comparison with a laboratory fluorescence spectrum obtained in exciting naphthalene in the gas phase with a laser at 289 nm. The detection of two polycyclic aromatic molecules through their UV electronic S(sub)1 --> S(sub)O transition is in good agreement with the suggestion that part of the 3.2-3.5 micrometer X-CH vibrational band detected in the IR is due to aromatics. It has two consequences. First, molecules with many C atoms such as naphthalene or phenanthrene can produce, as a result of photolytic processes, C2, C3, and CH molecules. This might provide a solution to the problem of the parents of C2 and C3. Second, the detection of aromatics in a comet clearly demonstrates that interstellar and cometary materials present a similarity of chemical composition. This is a strong argument in favor of the hypothesis that comets are built from interstellar material. Storrs A. D.* Morphology of UV Images of OH in P/Halley Recent reanalysis of images of the OH (0-0) and (1-1) band emission of comet are described. Observations with exceptionally high Q(sub)OH (immediately post-perihelion) produce a "shadow cone" in the antisunward direction, but no other spatial structure is seen. Azimuthally averaged brightness profiles cannot be uniquely fit with the Festou vectorial model. This is probably due to the high variability of the OH production rate, the limited size of the images, and uncertainty in the amount of sky and continuum contamination. Crovisier J.* Bockelee-Morvan D. Colom P. Gerard E. Jorda L. Lecacheux J. Colas F. Davies J. K. Despois D. Padman R. Paubert G. Observations of Dust and Gas Jets in Comet P/Swift-Tuttle 1992t by Visible Imaging and Radio Spectroscopy Nearly simultaneous observations of comet P/Swift-Tuttle 1992t were undertaken in November-December 1992: visible imaging with the 1-m telescope at the Pic du Midi Observatory; observations of the 18-cm lines of OH with the Nancay radio telescope; millimeter observations of the HCN, CH3OH, H2S, and H2CO lines with the IRAM 30-m telescope; submillimeter observations of the CH3OH, HCN, and H2CO lines with the JCMT 15-m telescope. Visible imaging shows a pattern of strong dust jets varying from day to day, from which we derive a rotation period of 2.8 days and we deduce the spin axis. All radio lines are asymmetric, with a cusp at negative velocities indicating a bulk gas motion towards the observer of about 0.8 km s^-1. These observations suggest that half of the dust and half of the gas are ejected in a single jet, the other halves being roughly isotropic. The velocity of the gas jet is 0.8-1.0 km s^- 1. From this scheme, we infer corrections to the molecular production rates previously estimated from a spherical symmetric model. We estimate the non- gravitational forces resulting from this anisotropic outgassing and discuss their influence on the comet orbit and spin axis. Schloerb F. P.* Lis D. Schilke P. Sanders D. Deane J. Ziurys L. Observations of Parent Molecules in Comet Swift-Tuttle Millimeter and sub-millimeter wavelength spectroscopy offers a very useful means to probe the composition of comets. We present observations of the parent molecules HCN, H2CO, and CO in Comet Swift-Tuttle, which have been obtained by this technique. The comet was observed using the Caltech Submillimeter Observatory 10m antenna and the Five College Radio Astronomy Observatory 14m antenna during November 1992. The HCN J=1-0, HCN J=3-2, HCN J=4-3, H2C0 515-414, and H2C0 312-211 transitions were detected; no detection was made of the CO J=3-2 line. The HCN production rate is about 2 x 10^26 s^- 1, which suggests a nominal abundance of 10^-3 for HCN in Comet Swift-Tuttle. The H2CO production rate is approximately 3 x 10^27s^-1 assuming that H2CO is distributed as a parent molecule in the coma. These preliminary production rate estimates are uncertain by up to, typically, 50% due to uncertainties in ephemeris used to track the position of the comet during observations. The shapes of the observed lines are all consistent with expansion velocities of about 1.0 km/sec in the coma. In addition, all detected lines show a strong blue shift of the velocity of the emission relative to the velocity of the nucleus. A simple model of uniform velocity outflow has been fit to the spectra with the best signal-to-noise ratio in order to determine the velocity of outflow and the fraction of the gas in the coma that originates from the day and night sides of the nucleus. The model fits suggest that about 80% of the observed molecules are produced on the day side. However, these simple models do not fully account for the shapes of the lines, and we suggest that further modeling including the effect on non-uniform expansion velocities in the coma will be required to match the data. Yau K.* Yeomans D. Weissman P. Comet P/Swift-Tuttle: Its Orbital Motion from 703 BC to AD 2392 We have performed an investigation of the orbit of P/Swift-Tuttle via a long- term integration forward to AD 2392 and backward to 703 BC. The initial conditions for the integration are determined from observations obtained at the 1992-1993,1862, and 1737 returns. Perturbations due to the nine planets have been fully taken into account. The coordinates of the planets are taken directly from the JPL long ephemeris DE102. The integrator itself is a predictor-corrector based on a Gauss-Jackson procedure. Our successful integration has enabled us to identify with certainty two of its previous apparitions in AD 188 and 69 BC from the ancient Chinese observations. No other early observations of P/Swift-Tuttle have been found. The unobserved returns between AD 188 and 1737 can easily be explained by the large minimum distance between the comet and the Earth, which prevented naked-eye visibility. We find that the comet must achieve a visual magnitude 3.4 to be discovered. The observing conditions at each return computed from the osculating orbital elements resulting from our integration suggest that the comet has maintained about the same intrinsic brightness for more than two millennia. Our forward integration puts the comet 0.153 AU away from the Earth at closest approach during its next return in 2126. Thus, there is no immediate chance of a collision with the Earth. Hoban S.* Mumma M. J. Reuter D. C. Davies J. K. High-Resolution Spectroscopy of Organic Species in Comet P/Swift-Tuttle Recently, the nu(sub)2 and nu(sub)9 bands of methanol (CH3OH) have been found to contribute up to half of the emission in the broad "cometary organic feature" centered near 3.4 micrometers (Reuter, 1992, Ap. J., 386, 330). Moderate-resolution infrared spectroscopy of Comet P/Swift-Tuttle shows evidence for structure within organic feature in the long-wavelength portion where the contribution from methanol is minimal (Davies et al., 1993, MNRAS, in press). We present high-resolution spectroscopy (lambda/delta lambda ~ 8000) of the feature centered near 3.425 micrometers in P/SwiftTuttle obtained with the CGS4 on the UKIRT on 1993 Nov. 26. Possible progenitors of this emission will be discussed. Highresolution spectra of the nu(sub)3 band of methanol centered at 3.516 micrometers in P/Swift-Tuttle from 1993 Nov. 25 will also be presented. The temporal evolution of the methanol production rate in P/Swift-Tuttle will be discussed. Yeomans D.* Yau K. Weissman P. Comet P/Swift-Tuttle: Constraints Upon its Nucleus For the same heliocentric distances, the gas production rates are comparable for comets Swift-Tuttle and Halley. Whereas outgassing thrusts (nongravitational effects) increase Halley's period by 4 days per return, nongravitational effects are not significant in the motion of SwiftTuttle over two millennia. One possible explanation would require that the comet's outgassing is directed radially toward the Sun and in a fashion that is symmetric with respect to perihelion; in addition the direction of the rotation axis, as well as the size and location of the nucleus' active areas could not have changed substantially over the entire two-thousand-year observed interval. A second, and perhaps more likely, possibility is that the nucleus of comet Swift-Tuttle is substantially more massive than Halley's so that its motion is relatively unaffected by nongravitational accelerations. If one assumes the lightcurve of Swift-Tuttle reflects the comet's outgassing behavior, then a 1992-93 lightcurve that is asymmetric with respect to perihelion would strongly favor the second possibility. POSTERS Huang B. Combi M. Cochran A. Fink U. Schulz R. Time-dependent Model Analysis of 8 Days of CN Spatial Profiles in P/Halley The Halley CN profiles were constructed from observations taken at three observatories during an eight-day period in April 1986 (Cochran and Barker, 1986, ESA, SP-250; Schulz and Schlosser, 1990, ESA, SP-315, 121; Schulz, 1992, Icarus, 96, 198-203; Fink et al., 1991, ApJ, 383, 360). These data provide a time series of CN spatial profiles spanning just over one 7.37-day period from April 7 to April 15, 1986, and sample distances from the nucleus from just over 10^3 km to 10^6 km. The effect of the 7.37-day periodic variation on the CN distribution in P/Halley has been examined by using the time-dependent Haser model of Combi and Fink (1993, ApJ, 409, 797). Because of the large spatial scale of the Schulz and Schlosser data on April 7, 8, & 9 (~10^6 km), information present in the spatial profiles regarding the gas production rate actually covers nearly two full periods. These spatially extended profiles clearly show the wavy structures outside 10^5 km predicted in the previous analysis that was based solely on profiles that extended only to distances less than 10^5 km. By using the same model as before with standard scale lengths and the same phase lag and amplitude correction for the variation in gas production rate, we are able to satisfactorily reproduce the highly variable Halley CN profiles over the eight days. Lutz B. L. Womack M. Wagner R. M. Pre- and Post-Perihelion Abundances of Gas and Dust in Comet Halley Photometrically calibrated spectra of comet P/Halley (1986 III) were recorded between 1985 September 12 and 1986 June 10 using the Ohio State University Image Dissector Scanner (IDS) on the Perkins 72-inch telescope at the Lowell Observatory. Column densities of CN, C3, CH, C2, and NH2 were calculated from measured fluxes in these spectra, and molecular lifetimes were deduced from the radial distribution of CN, C3, C2, and NH2. Production rates were computed using the new lifetimes and a vectorial model analysis. Continuum emission at 4260 Angstroms was used to derive gas-to-dust ratios. The data indicate that Comet Halley was ~2-5 times more abundant in gas and dust at postperihelion than preperihelion. The gas-to-dust ratios appeared to be subject to changes as a result of short-term outbursts, but otherwise did not exhibit any systematic dependence on heliocentric distance. Schulz R. A'Hearn M. F. Detection of Shells in the C2 Images of Comet P/Halley Two distinct features, namely jets and shells, have been detected in emission- band images of Comet P/Halley during its apparition in 1986. The existence of gaseous jets was first reported for the CN and later traced for several other coma species (C2, C3, [OI] + NH2). CN jets could be detected in CCD images of the inner coma (r < 10^-5 km) as well as in digitized large scale photographs (r > 5 x 10^5 km). The CN shells, however, were only found in the large-scale photographs where their day-to-day expansion could be followed with the naked eye. Subsequent extensive investigations showed that no shells are detectable in the inner coma (r < 7 x 10^4 km) and led to evidence for the evolution of CN jets into CN shells. If jets evolve into shells that expand thereafter with an expansion velocity of V(sub)Exp ~ 1 km/s, the shells with radii 7 x 10^4 km < r < 10^5 km should be detectable in the CCD images. We have therefore applied an image processing technique to the CCD images in CN, which enhances shell structures and found them indeed. For example, the expanding CN shell first visible in the digitized photographs on April 7th, 1986, could also be detected in the CCD image obtained approximately 12 hours earlier on April 6th, 1986. It already has an asymmetric shape that remains during its observable expansion up to April 9th, 1986. In the next step a search for shells was conducted in the the C2 images taken on those dates for which CN shells were detected. Shells also exist in the C2 images. The morphology of the shells in CN and C2 is essentially equal on the same observational date. Both have approximately the same radius and show almost identical asymmetries. Therefore, it is likely that the CN and the C2 in the shells originate from the same general source, which might be an organic macromolecule. Additionally to the shells, the jets in CN and C2 have been compared after the continuum that severely contaminates the C2 images and adulterates the appearance of the C2 jets was two-dimensionally subtracted. Tozzi G. P. Feldman P. D. Weaver H. A. The Spectrum of Comet P/Swift-Tuttle (1992t) in the Near Infrared Region Long slit spectroscopic observations of comet P/Swift-Tuttle in the near infrared region were obtained on the 12th and 14th of November 1992. The spectral range was about 9000-10,200 Angstroms with a slit length of 4 arc minutes on the sky. The FWHM of the instrumental profiles were ~5 Angstroms along the spectrum and 4.5 arc seconds along the slit. The raw spectra show an intense continuum, due to the solar scattering by the dust component of the coma, and emissions due to the (1-0) and (2-1) bands of the CN Red System, together with numerous overlapped atmospheric emission lines of OH. After a careful subtraction of sky lines and continuum, the data from the gas component have been analyzed. The spatial profiles of the CN bands show a marked asymmetry with the maximum shifted some thousands of km with respect to that of the continuum, probably due to the strong activity of this comet. The profiles have been studied with a vectorial model in order to determine the time dependence of the production rate. In addition to the CN bands, some OH lines remain in the cometary spectrum, the most intense being some transitions in the (3-0) band. Since these lines are predicted to have an intensity much lower than that measured, care was taken to verify that the residual OH emission was of cometary origin. The implication of this detection will be discussed. Larson S. M. Scotti J. Coma Morphology of P/Swift-Tuttle During October-December 1992, P/Swift-Tuttle exhibited complex dust coma morphology qualitatively consistent with its previous return in 1862 and which repeated close to the 2.8-day rotation period determined by Sekanina (A.J., 86, 1981). The dust jet structure indicates a well-developed crust with discrete emission sources similar to P/ Halley. Images of CN and C2 do not show the same structure as the dust jets, suggesting a source other than the dust jet sources. This work is supported in part by NASA grant NAGW-247. Womack M. McKeown S. M. Millimeter-Wavelength Spectra of H2CO and CH3OH in Comet Swift-Tuttle Spectra were obtained of the H2CO 3(sub)12-2(sub)11 transition at 226 GHz on UT 25 Nov 1992 using the CSO 10-m telescope and of the CH30H J = 3-2 transition at 145 GHz on UT 18 Nov 1992 using the NRAO 12-m telescope. Five rotational transitions of CH3OH were simultaneously observed from which a relatively low rotational temperature was directly determined. Spectra, column densities and production rates will be presented and implications for parent and/or extended source origins for formaldehyde and methanol will be discussed. Spinrad H. Brown M. E. Johns C. M. Kinematics of the Ion Tail of Comet P/Swift-Tuttle Using a series of high-resolution (lambda/delta lambda 33,000) very long slit (slit length ~ 6 arc minutes) of the 6199-Angstrom H20+ complex of Comet P/Swift-Tuttle obtained from the coude auxiliary telescope feeding the Hamilton echelle spectrograph at Lick Observatory, we have measured the velocity and velocity dispersion of water ions in the tail of the comet. We oriented the slit both along the sun-comet vector and perpendicular to that vector centered a distance down the tail. The outflow velocities measured from the tail-oriented spectra clearly confirm the observations by Rauer and Jockers and by Wycoff and Lindholm, which suggested that the tail motions are indeed true bulk flows in the antisolar direction. Out to about 3 x 10^5 km down the down, typical flow speeds are about 30 km s^-l. For the first time, velocities of ions picked up ahead of the nucleus are also observed, showing the deceleration of the solar wind as it encounters the comet. The velocity dispersion of the H20+ lines (sigma), a measure of the ion temperature of the comet, varies systematically down the tail of the comet; sigma is smallest at the position of the nucleus and increases slowly down the tail. On the sunward side of the comet, sigma increases much more quickly. On the spectra perpendicular to the tail axis, sigma is always lowest in the center of the tail and higher off-axis. The ion temperatures corresponding to the observed as range from about 10^5 to l0^6 K. We also note a clear anticorrelation between the H20+ line intensities (related to the number density) and the flow speed and dispersion; variations in the total mass loading from the comet or in the flow speed of the solar wind may be responsible. Finally, we discuss several approximate methods of inferring the local magnetic field induced by the interaction of the cometary ions with the solar wind. Typical estimates of the magnetic field yield values of about 40 microteslas. Tuesday, October 19, 1993 JUPITER AND SATURN 8:30 - 10:00 AM North Ballroom Chair(s): G. R. Gladstone J. Connerney Kostiuk T.* Livengood T. Espenak F. Buhl D. Fast K. Jupiter North/South Polar Hydrocarbon Emission, February 1993 Hydrocarbon emission from the polar regions of Jupiter was measured in February 1993 as part of a long-term study of the temporal and spatial variability of thermal infrared aurora. Nearly simultaneous line spectra were obtained from ethylene (10.5 micrometers) and ethane (12 micrometers) at a spectral resolution of 0.00083 cm^-1. Measurements were made using the Goddard Space Flight Center Infrared Heterodyne Spectrometer at the NASA IRTF on Mauna Kea, Hawaii. Ethylene lines showed a significant enhancement in the south polar region near 90 degrees longitude and 60 degrees S latitude (System III, 1965). At 60 degrees north latitude ethylene emission was prominent near the nominal 180 degree longitude "hot spot." Less prominent enhancements in ethane were observed at both poles. Retrieved thermal structure at the 10 microbar (from the ethylene lines) and 1 mbar (from the ethane lines) pressure regions of the polar stratosphere will be presented. These results will be the first for the south polar region. The north polar results will be compared to retrievals from 1989 measurements. The northsouth results will be compared and the relationship of the thermal structure, abundances, and variability will be discussed. Livengood T. A.* Buhl D. Deming L. D. Espenak F. Kostiuk T. Reuter D. Fast K. E. Location of Hydrocarbon Emission Near Jupiter's South Pole February 1993 and December 1989 observations of jovian south polar C2H4 (ethylene) and C2H6 (ethane) molecular line emission and April 1992 imaging observations of C2H6 band emission will be presented. Emission line spectra were acquired using the Goddard Infrared Heterodyne Spectrometer at NASA/IRTF, providing information on chemical abundance and temperature profile in the stratosphere. Images of ethane emission were acquired using the Goddard Infrared Postdisperser Camera at a spatial resolution of ~2.2". Evidence of an emission feature at ~60 degrees S latitude and ~90 degrees longitude was observed. Results from the emission line and imaging observations will be compared with previous observations, e.g., Orton et al. (BAAS, 23, 1131, 1991), which reports enhanced south polar hydrocarbon emission at 60 degrees +/- 5 degrees System III longitude; Caldwell et al. (Icarus 74, 331-339, 1988), which reports a "hot spot" of variable longitude; and Livengood et al. (Icarus, 97, 26-45, 1992), which reports enhanced ultraviolet emissions occurring within the range 24 degrees +/- 26 degrees System III longitude. We will discuss species-specific morphology and compare the location of IR and W maximum emission. Waite J. H. Jr.* Gladstone G. R. Na C. Franke K. Jovian Auroral Observations with HST The set of jovian auroral observations carried out by Hubble Space Telescope (HST) during the Jupiter/Ulysses encounter using several different filters are examined together. Recent HST GHRS and reanalysis of Voyager infrared results are used to define an auroral atmosphere model [Trafton et al., 1993]. This atmospheric model is then used to carry out a complete radiative transfer simulation of the HST observations including H2 band emissions and H Ly alpha. This allows polar plots of the energetic particle precipitation to be reconstructed for both the northern and southern auroral zones, showing the location of the auroral oval relative to the footprints of both the Io torus and the magnetotail. Turgeon B.* Caldwell J. J. Jupiter's Southern UV Aurora Imaged with the Hubble Space Telescope The southern ultraviolet aurora on Jupiter was imaged by the the Faint Object Camera (FOC) onboard the Hubble Space Telescope (HST) on 1993 July 9. The images reported here are part of an ongoing effort to obtain high-quality data on the aurora to gain a better understanding of its morphology and to determine the properties of its northern and southern components. Two auroral images were obtained on each of two consecutive HST orbits, using the filters F175W and F152M (effective wavelength ~160 nm). On the first orbit, the System III CML was 16 degrees (+/-6 degrees due to Jupiter's rotation) and on the second, the CML was 70 +/- 80 degrees. In addition, continuum images were acquired in order to navigate the images of the aurora. One pair of continuum images, using filters F175W and F170M, observed the north and south limbs of Jupiter near 170 nm. One other image, using filters F220W, F231M, and F2ND, measured the north polar limb near 230 nm. Preliminary inspection of the images reveals a well-defined auroral arc, of nonuniform brightness. Subsequent analysis will determine the position of the arc with respect to the jovian magnetic field. Scattergood T. W.* Clegg S. M. Photochemistry in Outer Planet Atmospheres: Effects of Hydrogen on the Photolysis of Methane and Acetylene Many simple hydrocarbons, including ethane (C2H6), ethylene (C2H4), acetylene (C2H2), and propane (C3H8), have been detected in the atmospheres of the outer planets. These compounds are produced from the photochemistry and energetic particle induced chemistry of methane (CH4) present in these atmospheres. However, the ratios of hydrogen (H2) to CH4 generally exceed 500:1 in these atmospheres, and vary in different atmospheres and in different regions within the atmosphere. The presence of H2 should effect the photochemistry of CH4 (and the other hydrocarbons), as noted by many investigators (see e.g., Strobel, Int. Rev. Phys. Chem. 3,145 [1983]). Recently, Allen et al. (Icarus 100, 527 [1992]) pointed out that the ratio of C2H6 to C2H2 in the atmosphere of Jupiter varies with altitude (at least in the 2 to 12 mbar region) and showed that H2 may be, in part, responsible for this variation. In order to experimentally study this, and to study the effects of H2 on the photochemistry of hydrocarbons in outer planet atmospheres, we have carried out a series of experiments in which different mixtures of CH4 and C2H2 were photolyzed at either 1849A or 1236A. The results of these experiments show that many hydrocarbon products, at least up to C4 compounds (butanes), are made even when the ratio of H2 to reactant gas was 1000. The ratio of C2H6 to C2H2 was found to increase with increasing H2, by about a factor of 4 with a 7-fold increase in the H2/CH4 ratio, consistent with the predictions of Allen et al. (1992). This change in C2H6/C2H2 was also dependent on the photolysis time, a parameter not yet explicitly dealt with in experimental simulations or, perhaps, in computational models. Finally, the results of experiments in which C2H2 was photolyzed at 1849 Angstroms with CH4 (but no H2) showed that the concentration of CH4 did not change with photolysis time, in spite of the loss of C2H2. This result suggests that equal amounts of CH4 are simultaneously produced and catalytically decomposed by the photolysis of C2H2. Prange R.* Dougherty M. K. Dunlop M. W. Balogh A. Field Aligned Current Driven Aurora in the Jovian Magnetosphere: A Possible Identification with Correlated HST and Ulysses Observations Ultraviolet images of the north Jovian aurora were taken with the Faint Object Camera onboard the Hubble Space Telescope on February 9, 1992 in conjunction with the close flyby of the Ulysses spacecraft with Jupiter. The images ohtained at Lyman alpha showed evidence of two major auroral structures at different latitudes, corotating with the planet. The higher latitude one consists of a narrow arc roughly highly aligned with the footprint of a magnetic shell connected to the middle magnetosphere. During the same period, field aligned current signatures were identified in the magnetic field data measured by the dual fluxgate/vector helium magnetometer onboard Ulysses. These type of field aligned currents can be generated as a result of departure from corotation of the plasma flow, or of transfer of momentum either from centritugally driven plasma outflow or from the solar wind. Using a three dimension field model we compare the projection of the field aligned signatures onto the surface of the planet and the locus of the auroral emission, and we discuss the possible association between these phenomena, identified for the first time on a planet other than the Earth. Gladstone G. R.* Waite J. H. Jr. Hall D. T. Moos H. W. Feldman P. D. Strobel D. F. McGrath M. A. Clarke J. T. Bagenal F. Schneider N. M. Shemansky D. E. EUVE Observations of Jupiter's Plasmasphere The first observation of Jupiter using the Medium Wavelength spectrometer of the Extreme Ultraviolet Explorer (EUVE) satellite shows a feature at 304 Angstroms originating from the vicinity of the planet and not the Io torus. We interpret this feature as being due to resonant scattering of solar He 304- Angstrom photons by He+ ions in the plasmasphere of Jupiter. The lack of other detectable EUV emission features from Jupiter and the large column of hydrogen in the jovian upper atmosphere argue against an auroral or dayglow origin for the 304-Angstrom feature. The observed brightness is 0.5 +/- 0.2 R, which corresponds to a helium ion density on the order of 30 cm^-3 near Jupiter. We will examine various models of the jovian plasmasphere to determine the most reasonable source for these He+ ions. Weisstein E. W.* Serabyn G. Detection of the 267 GHz PH3 line in Saturn and Implications for Its Mixing Ratio in Saturn's Upper Atmosphere A Fourier Transform Spectrometer has been constructed at the 10.4-m Caltech Submillimeter Observatory (CSO) for the purpose of studying planetary atmospheres at millimeter and submillimeter wavelengths. Initial measurements have concentrated on the jovian planets, where highly pressure-broadened rotational lines are expected to be resolved by the instrument, which has a resolution of ~200 MHz. Spectra were taken in all available submillimeter atmospheric windows between 150 and 1000 GHz. Jupiter and Saturn were observed at the CSO in December 1992 and July 1993. The deep and very wide (~12 GHz FWHM) J = 1-O rotational transition of PH3 (phosphine) at 266.945 GHz was unambiguously detected in Saturn on both occasions. Surprisingly, PH3 was not observed in Jupiter. The PH3 line in Saturn has been modeled using a radiative transfer model. The large width of the observed line implies a stratospheric mixing ratio of 8 ppm, significantly larger than the mixing ratios of 1-2 ppm previously obtained from IR measurements (Courtin et al., 1984; Noll et al., 1990). The large mixing ratio is also somewhat unexpected given the rapid photodissociative depletion predicted by photochemical models (Kay and Strobel, 1983). The absence of any observed PH3 in Jupiter indicates that it must be significantly less abundant in Jupiter than Saturn. IR measurements give a mixing ratio of 0.7 ppm for Jupiter, but our new data suggest that this value is too large. The observed difference in PH3 mixing ratios in Saturn and Jupiter may place dynamical constraints on the strength of convection in these two atmospheres, or may be related to differing NE3 abundances, since PH3 photodissociation is inhibited by the presence of NH3 (Strobel, 1977). Johnson R. E.* Pospieszalska M. K. Crosby C. K. Jurac S. Grosjean D. Shi M. Baragiola R. A. Sittler E. C. Morfill G. E. Neutral Torus of Sputtered Products from the Icy Satellites of Saturn: Revisited The observation of a much larger than expected OH density near Tethys lead Shemansky et al. (1993) to propose that an enhanced micrometeorite flux ejects water products from the icy satellites. Also, Morfill et al. (1993) suggested that the Ering particles are more efficiently sputtered since grain charging can increase the effective ion flux. Here we re-evaluate our earlier calculations of the plasma ion sputtering rates of the icy satellites (Johnson et al. 1989) and the E-ring particles (Johnson et al. 1984). We include recent measurements of the sputtering of ice vs. temperature for the ions, energies, and temperatures appropriate for the icy satellites and the E-ring grains, including charging. We combine this with an earlier calculation of the flux of ions to the surface of an icy satellite treating the ion gryromotion (Pospieszalska and Johnson 1989), and account for local pick-up ion formation. These effects combine to give an enhancement in the satellite sputter source rate that may account for the OH observation. The hydrogen torus and E-ring erosion rate are also considered. References: Johnson R. E. et al. (1984) Adv.Space Phys.4, 41. Johnson R. E. et al. (1989) Icarus, 77, 311. Morfill et al. G. E. (1993) JGR, 98, 11,285. Pospieszalska M. K. and Johnson R. E. (1989) Icarus, 78, 1. Shemansky D. E. et al. (1993) Nature, 363, 329. POSTERS McConnell J. C. Miller S. Prange R. H3+ Rotovibrational Temperatures at Mid-Latitudes on Jupiter Recently H+ emissions have been observed at mid-latitudes in Jupiter's thermosphere extending the probed region beyond the auroral regions. The main sources of the excited H3^+ states are thought to be thermal excitation and exitation by the reaction H2^+ + H2 --> H3^+ + H in which the H3^+ molecule contains about 1.5 eV distributed among rotovibrational levels. Energy transfer from vibrationally excited H2(v=1) may also be important. For the auroral regions it appears that the observed emissions may be due to thermal excitation although the calculations of Kim et al (1992) indicate that one can expect that the higher vibrational levels will not be in thermal equilibrium. At mid-latitudes, if electroglow sources of ionization are unimportant, then the main source of H3^+ should be photoionization of H2. However, our calculations show that the densities where the EUV photoionization energy is deposited are sufficiently low that the H3^+ may not be rotationally equilibrated, nevermind vibrational equilibrated. Thus if the observed H3^+ emission appears thermal it may be yielding information on alternate ionization sources. If the source of emission is thermal then it will be possible to monitor thermospheric temperatures in this region. Clearly it is important to resolve the question of the mid-latitude source in order to be confident in using the H3^+ emissions as a monitoring tool. In addition, it is important to understand if other ionization sources are playing a role. We will show the results of some preliminary calculations using both a simple rotational model, which indicates non-thermal rotational temperatures and a more complex model using the energy level information from Kao et al (1991). Griffioen E. McConnell J. C. Rego D. Prange R. Dols V. The Effects of a 2-D Radiative Transfer Spatial Broadening on Jovian Lyman Alpha Aurorae We have investigated the effects of multiple scattering of the 1216 A Lyman line on the spatial extent of Jovian auroral arcs. The temperature profile adopted for the calculations was based on a set of recent auroral observations in the IR discussed at the Annapolis workshop (1992). The H and minor species distributions were estimated using a 1-D chemical diffusive model. Several different values were assumed for the strength of the auroral H source and the magnitude of the eddy diffusion coefficient. We have computed the altitude distribution of the Lyman alpha excitation rate using the deposition code of Rego et al. (1993) in which we have assumed that electrons are the precipitating particles. Energies of 1, 10, 50 KeV have been considered to explore the effects of different altitude distributions. To calculate the spatial broadening effects due to multiple scattering we have used a new 2-D radiative code, (Griffioen, 1993) which uses a modified Lambda operator for speeding convergence in an optically thick medium such as that presented by H in the Jovian atmosphere. The results will be compared with FOC images of the Jovian aurora recorded using the Hubble Space Telescope at Lyman alpha and the H2 bands. Baron R. L. Owen T. Satoh T. Connerney J. Harrington J. The Jovian H3+ Aurora: Time Variability and Solar Wind Parameters The 3.4-micrometer aurora of Jupiter was observed dunng the months of January, February, and March 1992.We report on the photometric variability of the integrated auroral intensity at 3.4 micrometers of each polar region as a function of CML and as a function of days and as a function of selected solar wind parameters. We also describe the first results comparing tbe observed auroral structure with that predicted from the O6 magnetic field model. Auroral oval shapes and locations are also reported. Satoh T. Baron R. L. Connerney J. Detailed Structure of H3+ Aurora on Jupiter Detailed source modeling of Jupiter's H3^+ aurorae are discussed based on the analysis of the high spatial-resolution images taken at the NASA/IRTF on Mauna Kea, Hawaii. Sixteen 3.4 micrometer images acquired during the time between 5:53UT (when Jupiter's Central Meridian Longitude was 69.7deg) and 13:34UT (CML 348.5deg) on March 22, 1992 are used in this analysis. On that date the solar wind was near a minimum and the basic structure of the aurora oval was clearly observed. Models are constructed to represent the appearance of the aurora in individual images and the total flux variation curve as a function of the CML of Jupiter. Five parameters, the size of the aurora oval (whose shape is taken from the O(sub)6 model), its position on Jupiter's surface (longitude and latitude of the oval center) and the width and height of emitting area have been studied. Harris W. M. Clarke J. T. McGrath M. Analysis of Jovian Auroral Ly-Alpha Emission Variability with the IUE Satellite Archives A new technique has been developed to extract Ly-alpha emission from IUE low dispersion spectra of the Jovian Northern and Southern aurorae that were obtained during the period from 1980 to 1991. The extraction of Ly-alpha from these spectra is complicated by the variable level of sky background from the Earth's geocorona, and by non-linearities in the IUE camera response at this wavelength. Both difficulties are addressed in this work. The extracted Ly- alpha intensities have been analyzed for changes associated with longitudinal variability previously identified in the Jovian auroral system. The results of this analysis will be presented along with a comparison of changes in auroral Ly-alpha emission with changes in solar Ly-alpha intensity. Previous studies of these IUE spectra have focused on longitude and optical depth variability in the auroral H2 Lyman (B^1 sigma+ - X^1+ sigma(sub)g +) and Werner (C^1 II(sub)micro - X^1 sigma(sub)g +) band emissions. The extracted Ly-alpha data have been incorporated into this data base, and the results of a comparison of the relative intensities of H2 and Ly-alpha with Jovian longitude will be presented. The ratio of H2 band intensities at different wavelengths has been used previously to identify changes in the optical depth of auroral emission. A similar comparison with Ly-alpha and H2 emission will be presented, along with a discussion of the implications of the result regarding the observed line profile and possible generation mechanisms for the emission. Rego D. Prange R. Benjaffel L. Gerard J. C. Modeling the Emergent Profiles and Intensities of Lyman Alpha and H2 Bands Induced by Protons Precipitation in the Jovian Atmosphere As a final step towards the theoretical modeling of the HLy alpha/H2 hands ratio observed in the Jovian aurora, we have developed a model of energy deposition and optical lines excitation (HLy alpha and H2 bands) for protons precipitating in the Jovian atmosphere. In this model, we have focused on the effects of the (H+/H,H+/H2) charge exchange in various energy ranges and those of the secondary electron (e^-(sub)s) produed by ionization of the constituants of the atmosphere by proton impact. The contributions of the e^- (sub)s and the primary H/H+ beam to thc volume of ionization and excitation rates are calculated. We have modeled the Doppler-shifted HLy alpha emissions produced by the fast H, which precipitate with the protons of the beam. Finally, the volume excitation rates have been used as input for a radiative transfer code that we have developed in order to compute the emergent profiles a Feldman P. D. Bagenal F. Belton M. J. S. Broadfoot A. L. Clarke J. T. Delamere A. Lane A. L. Skillman D. R. Earth-Orbital UV Jovian Observer: A Discovery Mission We have begun a one-year definition study of an Earth-Orbital UV Jovian Observer as a mission within the Discovery Program of the Solar System Exploration Division of NASA. The goal is to study the long-term temporal behavior of both Jupiter (aurora; airglow; Lyman-alpha bulge) and the Io plasma torus from Earth-orbit through ultraviolet (550-1750 Angstroms) spectroscopy and EUV (700-1450 Angstroms) imaging. The payload consists of a 0.6-m telescope with a split focal plane to provide simultaneous spectroscopy and imaging. It would be placed into a high Earth orbit (perhaps the L1 Lagrangian point) by a Delta II expendable launch vehicle. The selected orbit will provide continuous viewing of Jupiter over an eight-month period and also greatly simplify ground operations. The performance requirements and the current optical design characteristics will be discussed in the context of the detailed scientific objectives. Karkoschka E. Tomasko M. G. Saturn's Aerosols 1991: Sizes and Distribution We used Hubble Space Telescope images of Saturn at wavelengths 0.30-0.89 micrometers to derive properties and distribution of Saturn's aerosols. Our key findings are (1) There are two distinct types of aerosols in Saturn's atmosphere. Small aerosols (~0.15 micrometer radius) populate the stratosphere, larger ones (~0.5-2 micrometer radius) populate the troposphere. Variation of tropospheric aerosol radii can explain some belt and zone features; (2) Our model is consistent with a vertically uniform mixture between gas and aerosols within each layer; (3) The latitudinal distribution of aerosols is very different in both layers. In the troposphere, aerosols are strongly concentrated towards the equator. In the stratosphere, they are concentrated at the North Polar Region while mid-northern latitudes are clear; (4) Both types of aerosols are much darker in the ultraviolet than in the infrared. For tropospheric aerosols, most of the variation of absorption with wavelength occurs in the visible, but less so in the near ultraviolet. Tejfel V. G. Saturn 1991: A Comparison of Zonal Photometry from Hubble Space Telescope and from Groundbase Observations The results of detailed photometry of Saturn from HST obtained in July 1991 (Westphal et al., Icarus, 1992, 100, No. 2) were compared with our photographic observations of Saturn in July 1991 made with a 1-m telescope and good image quality (Astron. Vestnik, 1992, 26, No.4). The latitudinal variations of the Minnaert limb-darkening coefficients derived from these observations are in very good agreement, as well as the relationship between normal albedo and limb darkening. Both cycles of the observations showed the decrease of limb darkening from the equator toward the planetographic latitude +40 degrees and the increase of limb darkening toward the north pole of Saturn. The latitudinal variations of zonal limb darkening in different spectral regions were analyzed on the mutual correlation and sensitivity to the Rayleigh optical depth of the atmosphere above the clouds. These variations may be connected with noticeable differences of the mean sizes of the cloud particles at low and high latitudes and partially or especially with the altitudinal differences of the upper effective cloud or haze boundary. McGhee C. A. Nicholson P. D. French R. G. Saturn's Central Flash from the 3 July 1989 Occultation of 28 Sgr at McDonald and Palomar Observatories During the 3 July 1989 occultation of 28 Sgr by Saturn, the central flash was recorded at both Palomar and McDonald Observatories using 58 x 62 InSb arrays. Individual images can be seen on Saturn's limb, which permits a detailed comparison of the observations with models. The star light was substantially attenuated by the ring system reducing the brightest observed flash to just 1% of the full stellar intensity. The central flash profile is sensitive to the shape of the planet's limb, so that the strong zonal winds on Saturn must be included to obtain the most accurate model. This figure includes the photometry of one of the central flash images from the Palomar data (shown in crosses) and the predicted light curves for this image according to two models for the shape of Saturn's limb. An equipotential suface of a rotating fluid with the effects of Saturn's zonal winds included ('winds') and without the winds included ('no winds'). Fig. 1, which appears here on the hard copy, shows how refined photometry of the McDonald observations will permit a more detailed comparison of these two limb shape models with the data. This work was supported in part by NASA Grants NAGW-544 and NAGW1368, and (for C. A. M.) a Hughes Fellowship. Weir A. L. Conrath B. J. Zonal Variation of the Relative Humidity of Ammonia on Saturn At Saturn direct retrieval of NH3 abundance from Voyager IRIS spectra is not possible. Instead, an approach is adopted wherein variations in ammonia absorption line strength of individual spectra about their zonally averaged mean can be explained in terms of variation of the NH3 relative humidity about some undetermined value. When this approach is carried out on spectra in the 10-30 degrees N latitude range, the result is a strong wavemode two variations in longitude. Interpreting regions of enhanced relative humidity as cooler than those of reduced relative humidity implies a thermal structure at about the 400 mb level that is 180 degrees out of phase with that reported at about the 100 mb level in the latitude range 30-40 degrees N (Achterberg and Flasar, Planetary Scale Thermal Waves in Saturn's Upper Troposphere, DPS 1993). The dynamical implications of this structure will be addressed. This work supported in part by the Planetary Data System. Kerola D. X. Larson H. P. Tomasko M. G. Near-IR Spectroscopic Studies of the Troposphere of Saturn High-resolution spectra from 1.7-3.3 micrometers acquired at the NASA Kuiper Airborne Observatory include 2 of Saturn's near-IR atmospheric transmission windows that are at least partially obscured by telluric H2O and CO2 absorptions at groundbased telescopes. We fitted this entire spectral region to an atmospheric model that included gaseous absorption by H2, CH4, NH3, and PH3 and the effects of multiple scattering by haze and clouds. We calculated H2 absorption coefficients using code furnished us by Borysow (1992, private communication) and we used the ESFT technique of Tomasko et al. (1989, BAAS, 21, 961) to synthesize the contribution of CH4 to Saturn's observed spectrum. For gaseous NH3 we used molecular parameters from Stout (1974, Ph.D. Dissertation) and for solid NH3 we used absorption coefficients in Sill et al. (1980, JOSA, 70, 724-739). Only for the prominent PH3 absorption feature at 2.9 micrometers did we have to use low temperature (200 K) laboratory comparison spectra because of the lack of molecular parameters. Our objectives were to determine accurate elemental abundance ratios (e.g., C/H, P/H, etc.) and to characterize the size, distribution, and composition of haze and cloud particles in Saturn's atmosphere. The results for C/H and P/H are 7.5 x 10^-4 and 5.0 x 10^-7, respectively. We found no evidence of gaseous NH3. Our upper limit to the NH3 mixing ratio at Saturn's radiative-convective boundary is ? 10^-9. Ammonia is decidedly undersaturated at atmospheric pressures lower than ~1 bar. Our upper limit to gaseous NH3 is extremely low compared to detected amounts derived from observations at visible, mid-IR, and microwave wavelengths. We are able to reconcile these differences on the basis of different mechanisms for spectral line formation in these disparate spectral regions. Our search for solid phase NH3 was also negative. From thermochemical arguments it has been widely assumed that NH3 ice crystals comprise the upper clouds on Saturn, although no incontrovertible spectroscopic proof has ever been presented. Strong bands of solid NH3 at 3 micrometers therefore offer an important test of this assumption. We placed Saturn's observed spectrum on an absolute reflectivity scale, which we then compared with synthesized spectra of candidate haze and cloud particles. Our calculations demonstrated that the reflectances of pure, polydisperse NH3 ice crystals with effective radii ranging from 0.1 to 2.25 micrometers are not compatible with Saturn's 3 micrometer spectrum. A reasonable fit to Saturn's continuum spectrum can only be achieved by using bright, micron-sized scattering haze particles mixed in with H2, CH4, and PH3 in Saturn's middle and upper troposphere. Other candidate haze materials such as hydrazine (N2H4) and acetylene (C2H2) can be tentatively rejected upon examining recent absorption spectra of cryogenically nucleated aerosols (Dunder, 1991, Ph.D. Dissertation). This research was supported by NASA grant NAG2-206 and GSRP grant NGT-50782. Achterberg R. K. Flasar F. M. Planetary-Scale Thermal Waves in Saturn's Upper Troposphere We have analyzed data from the Voyager IRIS (Infrared Interferometer Spectrometer) experiment to search for waves in the upper troposphere of Saturn. The data used are temperatures in layers about one scale height in thickness centered near pressures of 130 mbar and 270 mbar, from three global mapping sequences taken by Voyagers 1 and 2, which together provide coverage between 60 degrees S and 80 degrees N latitudes. The field of view of the instrument is about 8 degrees of a great-circle arc on the planet at the subspacecraft point. To analyze the data, zonal means were subtracted from each data point to give the deviations from the zonal mean, which could be caused by real temperature variations or instrument noise. The data were then broken down into ten degree bins in latitude. A periodogram analysis (in longitude) was then applied to each latitude bin to search for waves. Since the data for each map were taken over periods of 12-18 hours, the phase speed of the waves is ambiguous and must be assumed. The assumed phase speed can be varied to maximize the amplitude of the periodogram response at integer zonal wavenumbers, and thus estimate the phase velocity of the waves. The most significant result occurs in the 30 degrees N to 40 degrees N latitude bin in the Voyager 1 data at 130 mbar, where zonal wavenumber two is observed with a false alarm probability of 10^-5 at phase velocities near-570 ms^-1,-30 ms^-1 and 520 ms^-1 in System III. Zonal wavenumber two remains statistically significant at the 99% confidence level over the latitude range from 20 degrees N to 40 degrees N; the phase of the wave is also roughly constant over this latitude range. Modeling of this wavenumber two wave suggests that it is a planetary-scale Rossby mode confined in latitude by meridional variations of the zonal mean winds. Zonal wavenumbers one and four are also found in the 30 degrees N to 40 degrees N bin, and wavenumber four, and possibly nine, in the 35 degrees N to 45 degrees N bin of the Voyager 1 data at 130 mbar. None of these waves are seen in the 270 mbar data. Wavenumbers one, two, and four are also significant at the same latitudes in the Voyager 2 data; however, wavenumber one has become dominant, indicating that the waves are variable on the timescale between the Voyager 1 and Voyager 2 encounters. Miller S. Lam H. A. Trafton L. M. Geballe T. R. Ballester G. E. Tennyson J. An H3+ Database for the Jovian Ionosphere If predictions of the impending impact between Comet Shoemaker-Levy and Jupiter are correct, the energy released will cause a major disruption of the Jovian atmosphere. Even a glancing collision will almost certainly result in significant changes in the outermost layer, the ionosphere. Detailed knowledge of the pre-impact ionosphere will therefore be crucial in attempting to analyze the effects of the collision. A spectral image taken last year by Miller et al, using the CGS4 spectrometer on UKIRT at around 3.5 microns, showed that the H^3+ molecular ion is distributed across the Jovian disk, away from the bright auroral regions. Emissions were sufficiently bright to obtain temperatures of H^3+ in the ionosphere as a function of all latitudes. Further studies carried out during May 3-5 (U.T.) of this year have built up a large databank of spectral images showing up H^3+ emissions at all latitudes through a large range of central meridian longitudes. Work is in progress produce H^3+ temperatures and column densities. We believe the resultant database will not only provide vital input for current models, but serve as a crucial baseline for post-LS impact studies. Lam H. A. Miller S. Joseph R. D. Tennyson J. High Resolution Spectroscopy of H3+ on Jupiter High resolution infrared spectral images of Jupiter were taken on the night of April 18 and 19, 1992 (U.T.) using the NASA IRTF Cshell spectrometer at wavelengths around 4.05 microns. Analysis of the spectra showed that the brightest emissions were concentrated close to the poles; integration times used were too short to pick up emissions on the body of the planet. Due to the high spatial resolving power of the instrument (0.35") we were able to make out a narrow (<1") region of emission where the fitted rotational temperature exceeded 2000K. We believe that previous lower spatial resolution spectroscopic studies of H^3+ ions on Jupiter have 'smeared out' this feature across the auroral region, giving rise to a fairly consistent temperature profile around l000K. There exist some uncertainties in our fitted temperatures due to the rather small energy differences between the upper, emitting levels of the lines we have measured. Nevertheless we believe that the existence of the anomolous temperature region is real and that the matter is worth further investigation. Billebaud F. Drossart P. Vauglin I. Sibille F. Merlin P. 10-Micron Array Camera Observations of Hydrocarbon Emissions in the Atmosphere of Saturn We present here preliminary observations of C2H6 and C2H2 emissions in the atmosphere of Saturn, recorded on December 9, 1992, at the 3.6 meter Canada- France-Hawaii telescope, using the french national 10-micron array camera (C10 micrometers) and its Circular Variable Filter (Lambda/Delta Lambda = 50). The size of the planet was about 15 arcsec at this epoch and the spatial sampling was of 0.7 arcsec per pixel. The images were recorded at six wavelengths: 10.91, 11.69, 12.47, 13.09, 13.29, and 13.48 microns. The image of 12.47 microns highlights emission of C2H6. Two main features are observed: an excess of emission near the north pole and a bright emission in the north equatorial zone, with a strong east to west assymetry (the eastern part being brighter), previously detected by the 10-micrometer Goddard camera (Gezari et al., Nature 342, 777, 1989). The image at 13.29 microns, which is taken in the wing of the CH2 band presents similar structures but with a different relative contrast betweent the north equatorial zone and the pole and the eastern and western parts of the north equatorial zone. The most plausible interpretations is that we observe emissions of the high stratosphere, probably in relation with interactions with magnetospheric currents. Drossart et al. (1993, sumitted to JGR Planets) have shown for Jupiter that heating of the high stratosphere can produce important enhancements of hydrocarbon emissions without modifying the lower part of the stratosphere. This may be a similar phenomenon on Saturn. The diferences in emission intensities between C2H6 and C2H2 could then come from different vertical distributions of these constituents, due to the photochemistry. Tuesday, October 19, 1993 COMETS II 10:30 - 12:00 NOON South Ballroom Chair(s): D. Schleicher D. Boice Luu J. X.* Jewitt D. C. Cloutis E. Near-Infrared Spectra of Primitive Bodies Many molecular-vibrational features occur in the near-infrared spectral region and thus serve as useful diagnostic tools in analyzing reflection spectra. We have exploited this feature in obtaining moderate resolution near-infrared spectra of primitive solar system bodies with the Cooled Grating Spectrograph CGS4 of the 4-meter United Kingdom Infrared Telescope (UKIRT). The spectra cover the H (1.4-1.8 micrometers) and K (2.05-2.45 micrometers) wavelength regions at resolutions lambda/Delta lambda ~ 480 and lambda/Delta lambda ~ 600, respectively. Our sample includes 16 asteroids found in the Trojan clouds and in the outer main belt, as well as the unusual distant object 5145 Pholus and the distant comet Chiron. The Trojan and outer main belt asteroids consist mostly of P- and D-types and are recognized to be among the most primitive of the known asteroid types. Both Pholus and Chiron may be objects recently perturbed from the Kuiper belt beyond Pluto. The observations were searched for signs of absorption bands, particularly the C-N overtone at 2.24 micrometers reportedly present in some asteroid spectra. They were also compared with spectra of terrestrial organic materials that may be analogs of primitive solar system objects (e.g., tar sands, coal, oil shale). The spectra will be presented and the results discussed. Meech K. J.* Buie M. W. Mueller B. E. A. Belton M. J. S. The Coma and Light Curve of Chiron In support of our HST observations of Chiron obtained during 1993 February and March, we have obtained extensive groundbased broadband (BVRI) CCD data on this object. Thirty-six hours of data were obtained on 15 nights of observation between UT 1993 January 24 and 1993 March 31 using the UH 2.2-m telescope on Mauna Kea (MKO), the Lowell Observatory 42-inch and 72-inch telescopes, and the KPNO 2.2-m telescope. Chiron was at heliocentric distances between 9.35-9.25 AU and geocentric distances between 8.35-8.63 AU during the observations. The purpose of the MKO dataset was to ascertain the extent and color of the dust coma and the KPNO and Lowell data were for further refinement of the rotation period. The extended coma during January and February appeared to be less prominent than during January 1992, however, the rotational lightcurve was not detectable, indicating a larger amount of material in the coma near the nucleus. The extent of the coma and the total brightness behavior of Chiron from this observing season in combination with the 35 nights of data from Meech & Buie (in prep.) from 12/88-03/92 will be discussed in the context of the model of the bound dust atmosphere. Holman M. J.* Wisdom J. Dynamics of the Kuiper Belt In a study of the dynamics of small bodies in the Kuiper belt, 1500 test particles were placed in orbits in the Kuiper belt and evolved for 200 m.y. using the symplectic mapping method of Wisdom and Holman (1991). The four giant planets were included in the integration as perturbers. Numerous test particles with semimajor axes between 30 AU (Neptune) and 43 AU encountered Neptune during the course of the integration. Test particles that encounter Neptune late in the integration generally reach Neptune by a path that roughly preserves semimajor axis, similar to the delivery of meteorites from mean-motion resonances in the asteroid belt to Earth- and Mars-crossing orbits. The resulting distribution of encounter times suggests that the times of first encounter with Neptune can reach several billion years. Futhermore, the flux of new encounters decays slowly, roughly as the inverse of time. This gives new insight into the dynamics of the delivery of short-period comets from the Kuiper belt and permits a rough estimate of the mass of the Kuiper belt of 0.1 Earth masses. We compare results with the preliminary orbits of recently discovered Kuiper belt objects. Marcialis R. L.* Larson S. M. Deep CCD Imaging and Photometry of 2060 Chiron During the 1992-1993 Apparition Approximately 300 CCD images of 2060 Chiron were obtained with the Steward Observatory 2.3-m telescope and Larson IHW CCD camera during November and December of 1992, and on part of one night in January of 1993. The effective focal ratio of the system was /1.9, which allowed properly exposed frames to be read out less than 2 minutes apart. Most of the images were in the Cousins R passband, and were obtained with the goal of exploring the high-frequency ( ~10 min) "dropouts" in Chiron's rotational lightcurve (Marcialis and Buratti 1993, Icarus, in press), and to use these dropouts to determine whether the object rotates in a prograde or retrograde manner. The November observations were made within about two weeks of Chiron's pre-opposition stationary point, at a solar phase angle of nearly 6 degrees Co-addition of these images show a well-developed tail at least 50" long at position angle 288 degrees. Assuming the tail to be pointed in the anti-sunward direction gives it a length of at least 16.6 million km (0.11 AU). Photometric reductions of the observations are in progress, and results will be presented at the meeting. Cochran A. L.* Long Slit CCD Observations of Comet Schaumasse (1992x) We obtained long slit CCD spectroscopic observations of comet Schaumasse during 1992 and 1993 in order to study the distribution of the gas in the coma. All observations were obtained on the 2.7m telescope of McDonald Observatory at a resolution of 7 Angstroms with a spectral range from 3000- 5600 Angstroms. With this spectral range, we can observe emissions due to OH, NH, CN, C3, CH, C2, and CO^+. In this paper, we will discuss the distribution of the gas in the coma. The gas showed a slight asymmetry in the sunward/tailward sense but was resonably symmetric perpendicular to the sun/tail line. We will concentrate on the distribution of the OH gas. With these observations we will derive the H20 lifetime against photodissociation in a manner analagous to Cochran and Schleicher (Icarus (1993) in press). These observations are particularly complementary to that study since they were obtained at moderately low solar activity, as evidenced by the He 10830 Angstroms EW (solar minimum is defined as EW=48m Angstroms; solar maximum=75 m Angstroms). These are the first long slit observations we have obtained with EW< 63m Angstroms. With these data, we intend to continue our study of the trend of H20 lifetime with solar activity. Knopp G. P.* Meech K. J. The Fragmentation of Comet Wilson (1986l) Comet Wilson (19861) was observed on 23 nights from October 1986 to November 1989 at the UH 2.2-m telescope on Mauna Kea (MKO), the KPNO 2.1-m and 4-m telescopes and the CTIO 1.5-m and 0.9-m telescopes. During the observations, which bracket perihelion, Comet Wilson was at heliocentric distances of 1.2- 7.8 AU. Broadband imaging photometry was obtained (BVR), which was more extensive after the nucleus' splitting was observed in February 1988. This has allowed examination of the photometric and dynamical behavior of the fragment and parent body. The photometry demonstrates that both objects were fairly constant in their behavior, with highly isotropic comae. However, the fragment's coma does cease to be centrally condensed in November 1989, implying the suspension of detectable activity. The relative trajectories of the nuclei allow an estimation of the time of splitting and may show the presence of non-gravitational forces. The general photometric behavior will be discussed, as well as surface brightness profiles, in relation to the physical characterisitics of the fragment and parent bodies. Campins H.* Osip D. Gustafson B. A. S. Rieke G. Rieke M. Larson S. Schleicher D. A Multi-Wavelength Study of the Potentially Meteorite-producing Comet P/Wilson-Harrington (4015 1979VA) The identification of object 4015, 1979 VA, as Comet P/Wilson-Harrington, P/W- H, (IAUC 5585, 1992) supports the notion that extinct or dormant nuclei of periodic comets can have asteroidal appearance, and that some fraction of the Earth-crossing asteroids have a cometary origin. Furthermore, orbital characteristics make P/W-H particularly interesting even among the comet- asteroid transition objects. It has a low inclination orbit, perihelion just inside 1 AU, and aphelion in the middle of the asteroid belt. This orbit has the potential of delivering meteorites to Earth at relatively low velocities. The figure, which appears in the hard copy, plots the orbital eccentricity vs. reciprocal semimajor axis of meteorite-producing meteoroids (crosses) listed by Halliday et al. (J. R. Astron. Soc. Can., 83, 49-80, 1989) and of P/W-H (open circle). P/W-H plots close to several meteoroids; we are in the process of doing orbital integrations to determine under what circumstances the comet may have produced any of these objects. We also obtained time series CCD images (red filter lambda-C = 7000 Angsroms delta lambda = 2600 Angstroms) on UT 1992 Dec. 01 and 02 at Lowell Observatory and in the Cousins red bandpass on UT 1992 Dec. 21 and 1993 Jan. 26 at U. of Arizona. These images are being used to determine the rotational characteristics of the comet as well as to search for any detectable coma. Simultaneous reflected (1.26, 1.60, and 2.22 micrometers) and thermal (10.6 micrometers) infrared observations were obtained on UT 1992 Dec. 17 at U. of Arizona. Thermal modeling will provide estimates of the radius and albedo. However, at the time of this writing, the rotational characteristics and surface temperature are uncertain. If slow rotation is assumed (standard thermal model) a preliminary estimate of the radius is 1.3 km. Samarasinha N. H.* Evolution of Cometary Rotational States Under Sublimation Induced Torques Recently a number of authors (Julian 1990 and references therein) showed by means of numerical simulations that cometary nuclei can change their rotational states during a single apparition due to torques caused by localized outgassing. However, the behavior of the lightcurve of comet Halley during the last two apparitions (Millis and Schleicher 1986; Schleicher and Bus 1991) and the historical data over the last two millennia (Yeomans and Kiang 1981) indicate that the activity of the comet was nearly unchanged from one apparition to the next, suggesting a constant rotational state. This suggests that sublimation-induced torques are canceling out at least over each apparition for comet Halley. In addition, the complex rotational states of comet Halley, which are consistent with most observations, have near- commensurate periods for the components of rotation (Samarasinha and A'Hearn 1991; Belton et al. 1991). We will present the results of numerical simulations that are based on the integration of sublimation-induced torques over a number of apparitions. We will also address the near constancy and the commensurability of the rotational state of comet Halley and whether there are preferred rotational states for a given cometary nucleus. Mueller B. E. A.* Complex Spin States of Cometary Nuclei P/Halley is the only comet known with a well-established complex rotation [1]. Extensive lightcurve coverage of active comets far away from the Sun when their activity is not dominant has only been done for a few comets. Recent observation and periodicity analysis of P/Schwassmann-Wachmann 1 suggest that this comet might also be in an excited spin state [2]. P/Tempel 2 on the other hand shows no indication of complex motion at all. What about other comets? We investigated published lightcurve data of other comets with respect to their periodicity, with special emphasis on multiple periodicity analysis. Is the difference in rotational behavior due to different mechanisms of outgassing from the nucleus (e.g., due to structure of the crust, number of vents, dynamical age) or is it just due to the poor lightcurve coverage of comets and/or our methods of periodicity analysis? The result of this investigation will be presented. References: [1] Belton, M. J. S. et al. (1991) Icarus, 93, 183. [2] Meech, K. J. et al. (1993) A. J., in press. POSTERS Ballou R. G. Schleicher D. G. Whole Coma Narrowband Imaging of Comets In order to better understand physical and chemical processes taking place within cometary comae, it is desirable to know the detailed spatial distribution of both the dust and the various gas species in a comet. However, due to the large spatial scales involved, it is difficult to obtain the spatial distribution of individual species for the whole coma. For instance, photometry and long-slit spectrophotometry usually sample only a small fraction of the coma, while radio observations seldom have high spatial resolution. We obtained over a dozen nights of simultaneous wide-field CCD images and conventional aperture photometry of Comets P/Brorsen/Metcalf (1989 X) and Austin (1990 V) using a Takahashi f/4 telescope and the Lowell 42-in telescope respectively. The use of the wide-field telescope, having a field of 50 arcmin with 4 arcsec/pixel resolution when coupled to an 800 x 800 TI CCD, allows us to image the whole coma and make simultaneous sky measurements. Standard image-quality IWH filters (2 x 2 inch to eliminate vignetting) isolate the CN, C2, C3, CO+, and H2O+ bands, along with three continuum points. Results of this investigation, including derived scalelengths for CN, C2, and C3, and how the dust distribution compares to the expected 1/rho fall- off, will be presented. This research was supported by NASA grant NAGW-2366 and by the Research Corporation. Bus S. J. Wasserman L. H. Elliot J. L. Chiron Stellar Occultation Candidates: 1993-1996 The unusual nature of (2060) Chiron has been repeatedly discussed in the literature. Given the recent discoveries of other distant objects of size comparable to that of Chiron (5145 Pholus, 1992QBl and 1993FW), one might suppose that Chiron is representative of what may be a substantial population of large, distant solar system objects. Chiron's brightness and current approach toward perihelion provides an important opportunity to study the physical properties of such objects. A basic property of Chiron that still remains uncertain is its size, and its related geometric albedo and effective surface temperature. These parameters play a significant role in the ability to constrain both the volatile makeup of Chiron, and the development and evolution of the observed dust coma. As early as 1983, thermal infrared observations of Chiron were made from the ground, and a nominal diameter of 180 km was determined. Using the IRAS database, Sykes and WaLker (1991, Science, 251, 777) placed an upper limit on the diameter of 372 km. More recent observations in both the infrared and submillimeter spectral ranges again help place upper limits on Chiron's size, but still do not allow for useful constraints to be placed on the mechanisms underlying Chiron's cometlike activity. We hope to successfully predict and observe a stellar occultation by Chiron as a means of directly measuring its nuclear diameter and, if the dust coma is substantial at the time of the occultation, probe the optical depth and distribution of dust in the inner coma. An earlier search for candidate stars was carried out by Mink and Stem (1990, BAAS, 22, 1358) utilizing the HST Guide Star Catalog. With a more accurate ephemeris, we carried out an independent search for fainter occultation star candidates over the three Chiron apparitions from fall 1993 through summer 1996. Using films taken by E. M. and C. S. Shoemaker with the 46-cm Schmidt telescope at Palomar Mountain, 44 candidates were identifled to a limiting V magnitude of 16, and whose predicted minimum separation from Chiron is 2.5 arcsec or less. The three most promising occu1tation events occur during the 1993-1994 apparition; two in particular involve stars with V magnitudes of 11.9 and 10.7. We present the results from this candidate search, up-to-date improvements to the predictions (including estimated errors), and nominal observing circumstances for the more promising events. Churyumov K. I. Kleshchonok V. V. On the Spectrum of Comet P/Schaumasse (1992 X) Obtained with the 6-Meter Reflector on March 14-15, 1993 The spectrum of comet P/Schaumasse (1992X) was obtained with the 6-m reflectoq BTA of the special Asaophysical Observatory of Russia Academy of Sciences on March 14-15, 1993 (see Figure). Several gas production ratios, including Q(C2)/Q(CN) and Q(C3)/Q(CN), were determined. The Q(C2)/Q(CN) ratio of 1.35 corresponds to a "normal" ratio. The peculiarity of the comet is the anomalous Q(C3)/Q(CN) ratio, which is equal to 0.94. This is approximately four times more than that of the "normal" comets [1]. References: [1] Cochran A. (1989) ACM, III, 281-184. Fig. 1, which appears here in the hard copy, shows spectrum of comet P/Schaumasse (1992 X) on March 14/15, 1993 (dashed line is the solar spectrm). Spahr T. B. Hergenrother C. Larson S. M. High Ecliptic Latitude Asteroid and Comet Search Most asteroid and comet searches concentrate near the ecliptic plane around opposition. We have carried out a small pilot survey with the Catalina Station 0.4-m, f/3 Schmidt camera to search photographically for objects brighter than about Mv = 17 at high ecliptic latitudes. Although such a strategy is not expected to yield large numbers of new objects, it will be biased toward high inclination main belt asteroids, Earth approachers, and long-period comets. Being outside the region of the major search programs also has the effect of providing more complete sky coverage. Unhypersensitized Kodak TMAX400 films were exposed for 8 minutes twice for each 40 square degree field and scanned visually with a stereo viewer. A total of 14 nights in 1992 November, 1993 March, April, and May covered about 2000 square degrees and yielded 6 new or lost asteroids; 4 were main belt asteroids with inclinations 15-28 degrees and eccentricities 0.14-0.35, while two do not have reliable orbits at this time. These objects were near perihelion at discovery, and would have been too faint for detection when in the usual opposition fields of the other searches. No comets were discovered, but Comet Spacewatch (1992h) was observed. Although the sample is small, the results indicate that systematic searches at high ecliptic latitudes would be productive. Film was provided through a grant from the Eastman Kodak Co., Scientific Imaging Division. TITLE-ONLY PRESENTATIONS Churyumov K. I. Shabas N. L. Kravtsov F. I. Investigation of the Magnetic Field in the Plasma Tail of Comet Abe (1970 XV) Investigation of the physical conditions in cometary plasma tails is of great value because of such structural features as frequent disconnection events. It is important to determine the physical parameters in the plasma tails in order to elucidate the causes of these phenomena. The estimates of some physical values in the plasma tails of comets through comparison of observations with the diffusion model have been conducted by Churyumov and Tesselko [ll in comet P/Halley (1986 III) and in comet P/Churyumov-Gerasimenko (1982 VIII). The estimates of the magnetic field induction are in good agreement with the direct measurement results made onboard the Vega and Giotto spacecraft in P/Halley (B~50-75 nT) and the International Cometary Explorer in the plasma tail of comet P/Giacobini-Zinner (B~10-60 nT). We studied the plasma tail of comet Abe (1970 XV). The physical parameters of the magnetic field in the plasma tail of comet Abe were determined on the basis of photometric profiles from 24 of the comet's plasma tail images. The films were obtained by K. Churyumov with the 0.17-m Schmidt telescope at the Coronal station of the Astrophysical Institute of the Kazakh Acadamy of Sciences during September 30- October 7, 1970. Mean values of the magnetic induction, B, ranged from 35.2 nT to 54.0 nT dunng the period of observation. References: [1] Churyumov K. I. and Tesselko N. L. (1992) BAAS, 24, 1012. Moreels G. Rousselot P. Goidet B. Clairemidi J. The C2 (1,1)/(0,0) Intensity Ratio in Comets Schaumasse and Halley as a Tracer of Diffuse Source The C2 Swan bands generally appear as the most intense features in the visible part of cometary spectra. However, up to now, parent molecule(s) of C2 have not yet been identified. A new method for obtaining informtion about the lifetime and the ongin of C2 consists to measure and analyze the intensity ratio R(sub)O I(1,1)/I(0,0) at 512-516 nm. This ratio was measured in spectra of comet Halley and in spectra of Schaumasse taken on March 5, 1992 with the 193 cm telescope at Observatoire de Haute-Provence. Both observations show that the intensity ratio I(1,1)/I(0,0) increases from 0.75 to 1.1 when the cometocentric distance increases from 400 to 10000 km. The data on comet Halley are depicted in a 2D visualization of an angular sector originating at the nucleus where two gaseous jets are apparent. The ratio R(sub)O shows a high 1.1 value in a region at 20000-30000 km associated with an apparent diffusion of material issued from the jet. In an attempt to explain the variation of this ratio as a function of distance to the nucleus and to the jets, a new model, based on a Monte-Carlo computation was constructed to calculate the populations of the triplet d^3pi, c^3Sigma, b^3Sigma, a^3Pi and singlet A^1Pi and X^1Sigma states of C2. The model shows that the time necessary to reach fluorescence equilibrium is as long as 3000 s. Two initial population distributions are considered. Distribution I consists of low excitated C2, 75% in the ^1Sigma state and 25% in the 3Pi state. Distribution 2 consists of highly excitated C2, 100% in the 3Pi state. The observational data of comets Schaumasse and Halley are interpreted as follows. The C2 molecules responsible for the Swan bands are created as daughter or granddaughter products of parent molecules among which are polycyclic hydrocarbons. The main source is located at the nucleus. The C2 molecules are created with low initial excitation energy (R(sub)O ~= 0.75 to 0.90) that progressively increases with increasing cometocentric distance in the expansion process (R(sub)O ~= 0.95 to 1.05). A second source, diffuse, results from the fragmentation of dust particles carried out in jets to a distance of 20000-30000 km. The C2 molecules of this type have a higher excitation. Their parent probably originate from some of the CHON particles. Tuesday, October 19, 1993 MARS ATMOSPHERE I: OBSERVATIONS 10:30 - 12:00 NOON North Ballroom Chair(s): A. Grossman J. Pearl Lellouch E.* Goldstein J. J. Bougher S. Theodore B. Rosenqvist J. Mars' Middle Atmosphere Circulation Near Equinox from Microwave Observations On December 29, 1992-January 2, 1993, we observed several CO lines on Mars near opposition with the IRAM 30-m radiotelescope at Pico Veleta, Spain. At that time, shortly after northern spring equinox (L(sub)s = 19 degrees) Mars' apparrent size was slightly larger (15") than the telescope beam diameter at CO J=2-1 (12-13"), providing some modest spatial resolution of the martian disk. We observed the J=1-0 ^12CO and the J=2-1 ^12CO and ^13CO rotational lines, at 115, 230, and 220 GHz respectively, on several locations on the disk. Data were obtained at 1 MHz, 100 kHz, and 10 kHz resolution with high signal-to-noise. The ^12CO(2-1) spectra are used for direct absolute wind measurements near 47 km altitude, with an absolute 1 sigma accuracy provisionally estimated at about 5-10 m/s. After our 1988 observations, which pertained to solstice conditions (Lellouch et al., Ap.J., 383, 401, 1991), these data provide the first direct wind measurements in Mars' middle atmosphere near equinox. Surprisingly, results clearly indicate the presence of strong (about 120 m/s) retrograde zonal winds at most latitudes, whereas moderate prograde jets were expected to prevail at mid-latitudes. In relation with the dynamical picture, our data should also provide information about possible hemispheric variations of Mars' thermal structure. Finally, high signal-to-noise combined observations of ^12CO and ^13CO may also allow to retrieve the CO vertical profile at 0-70 km. Gurwell M. A.* Muhleman D. O. Berge G. L. Observations of Middle Atmospheric Winds on Mars Observations of the middle atmospheric circulation of Mars (45-100 km) are sparse. Direct measurements were obtained during the 1976 Viking Lander descents. More recently, single antenna observations of Doppler shifts in the CO J=2-1 line (Lellouch et al., 1991) and interferometric observations of Doppler shifts in the CO J=1-0 line (Berge and Muhleman, in prep) were made during the 1988 and 1990 oppositions, respectively. To improve and expand this database, observations of the CO J=1-O line (2.6 mm) on Mars were made in early February 1993 with the OVRO four element array (L, = 38 =mid-northern spring). The synthesized beam (3.5" FWHM) allowed good spatial resolution of the 12" disk of Mars; the subearth point was at 4 degrees north, allowing both hemispheres good coverage. The observations were made with 62.5 kHz channel resolution symmetrically covering the inner 8 MHz of the CO absorption line. Centroid fitting of the line core allows LOS velocity determination to high precision (order 10 m/s). The data are additionally improved with the application of self-calibration techniques to correct for spatial phase errors, as verified with broadband continuum channel maps. Synthetic CO line core weighting functions peak in the range of 40 to 60 km altitude (at the subearth point), showing that this dataset is sensitive to the middle atmosphere. The experiment, performed in 3 10-hour tracks over a week, is especially sensitive to zonal mean circulation. The 1988 IRAM and 1990 OVRO data clearly show strong low latitude zonal winds, and a strong southern zonal jet is predicted in GCM simulations for L(sub)s ~44 degrees (Pollack et al. 1990). We will explore this zonal structure with improved sensitivity and spatial resolution. In particular, we will be able to probe the zonal circulation in each hemisphere. In addition, we will examine our sensitivity to meridional flow in the polar regions. Grossman A. W.* Clancy R. T. Muhlemann D. O. Microwave Limb-Sounding of Martian Water Vapor We present preliminary results from a continuing campaign to monitor the vertical, latitudinal, and diurnal distribution of water vapor in the atmosphere of Mars, using a new technique of groundbased microwave spectroscopic limb sounding. At centimeter wavelengths spectroscopic transitions are isolated and highly resolved. Furthermore, the radiative transfer analysis is simplified by the linear dependence of the source function on temperature and the insignificance of atmospheric scattering. Finally, contamination by terrestrial water vapor is reduced by the fact that the martian line appears as a narrow feature on top of a broad continuum of terrestrial tropospheric emission. The technique was first demonstrated during the 1990 opposition (Clancy et al., Icarus, lOO, 1992), when Mars was mapped with the Very Large Array (VLA) interferometer at a rotational transition of H20 corresponding to a wavelength of 1.35 cm. Success with this technique requires the high spatial resolution available at the VLA, which yields a large contrast of line emission for the long atmospheric path-length at the limb of Mars. Latitudinal variations are derived from the spatial variations along the limb, while the difference between the morning and afternoon limbs yields a measure of the diurnal variation. The measured line shape returns information on the vertical distribution of water vapor. The 1990 observations, taken during the late northern winter season (Ls = 344 degrees), yielded a global average water column of 3.0 +/- 0.8 pr micrometer, which is less than that previously measured at this season in other years. These observations showed that water vapor was uniformly mixed to altitudes of 50 km, but decreased uniformly in abundance from equator to pole in both hemispheres. Observations were obtained during the 1992-1993 opposition at Ls = 349 degrees, 43 degrees, 76 degrees, and 99 degrees. Preliminary results indicate a global column abundance of 3 pr micrometer at Ls = 349 degrees increasing to 8 pr micrometer at Ls = 43 degrees. Furthermore, this increase appears primarily in the northern hemisphere and is confined to altitudes below 15 km, suggesting a colder atmosphere than expected from Viking measurements. Continuing observations of Mars H20 with this technique are expected to provide information on seasonal variability and interannual variations of water vapor. Clancy T.* Grossman A. Muhleman D. A Cold, Dry, Cloudy, Dust-free Mars Atmosphere in the 1990's We have employed the VLA for spectral line (22 GHz, 1.35 cm) mapping of the latitudinal and vertical distribution of Mars atmospheric water vapor for four observational periods in 1992-1993, corresponding to late northern winter- spring (Ls = 340-100 degrees) on Mars. A 225-GHz whole-disk spectrum of HDO was also observed in April 1993 (Ls = 68 degrees). Several aspects of the thermal state of the Mars atmosphere and its seasonal variation are apparent from preliminary analysis of these observations. The Mars atmosphere appears as dry in 1992 as it did in 1990 (Clancy et al., Icarus, 100, 1992) during the late northem winter season (Ls = 340-350 degrees, global average col ~3 pr micrometer). Furthermore, the water appears well mixed to high altitudes (>40 km), and equally distributed between the northern and southern hemispheres. In contrast the northern spring observations (Ls = 43,76 degrees) indicate a strong concentration of water vapor in the northern hemisphere and a global saturation altitude at or below 15 km. These measurements may also indicate a drier atmosphere (global col ~5-8 pr micrometer) than observed by the Viking spacecraft for the northern spring season in 1977-1978, although their analysis is very preliminary. The low altitude of saturation observed for northern spring season implies a 15 K colder atmosphere than observed by Viking for the same season. Such a change is consistent with the overall colder atmospheric temperatures obtained from groundbased microwave CO observations (Clancy et al., JGR, 95, 1990), which have also been observed over the 1990-1993 period. The general picture inferred for the Mars atmosphere from these microwave observations and HST ultraviolet and violet imaging (James et al., Icarus, 1993) is a colder, drier Mars atmosphere with global cloud formation during the coldest northern spring season (Ls = 20-100 degrees) and very low atmospheric dust loading for the period 1990-1993. Such a state is markedly different from the relatively warrn, dusty, wet, and less cloudy atmosphere observed by Viking in 1977-1978. James P. B.* Clancy T. R. Lee S. W. Martin L. J. Singer R. B. Synoptic Observations of Mars Using the Hubble Space Telescope: Second Year Hubble Space Telescope observations of Mars during 1993-1993 commenced on May 30, 1992 (L(sub)s = 259) and continued until June 16, 1993 (L(sub)s = 93). Most of the sequences were designed specifically to repeat coverage at the identical seasonal dates observed during the first year of HST observations (James et al., 1993) in order to document interannual variations in atmospheric and surface features. The period of observations includes the summer dust storm season in the south (reported at 1992 DPS), the winter north polar hood, and the north cap recession and orographic cloud development of northern spring. These phenomena will be discussed in the context of the 1992-1993 observations. Interannual variability will be investigated through comparison of these phenomena during the two years. The background dust opacity history, deduced from modeling with a radiative transfer code, will be compared to that previously determined for 1990-1991. We thank the Space Telescope Science Institute for support of this project through Grant #3763. Martin L. J.* James P. B. Clancy R. T. Lee S. W. Beish J. D. Parker D. C. Mars Apparition of 1992-93: CCD Imaging from HST and Ground-based Telescopes CCD cameras have become standard imaging tools for Mars observations, not only from the Hubble Space Telescope (HST), but also from many telescopes that remain Earthbound. These include some sophisticated, although moderately priced, CCD systems obtained by serious amateur observers. Although the results do not have resolutions that are comparable to Mars orbiting spacecraft, these data do afford whole-hemisphere formats during each single exposure, giving us complete views of major weather systems. These data can also provide monitoring of weather and albedo changes during periods, like the present, when no working spacecraft is at Mars. The HST images are superior to the best groundbased images, and the best groundbased images are obtained less than 1% of the available time. On the other hand, time on HST for Mars observations has naturally been limited. Earlier HST observations of Mars have shown that useful images can be aquired even when its apparent diameter is at its smallest size. The only physical limitation on observations is the HST restriction that Mars must be at least 50 degrees from the Sun. This covers roughly one-half of a martian year. At Lowell Observatory, a telescope and CCD system were dedicated to these Mars observations. Bad weather was a major problem, eliminating any chance for contiguous observations and allowing for only a handful of reasonably useful images. On nights that were clear, the atmospheric turbulence was usually extreme. During this period, Flagstaff experienced several times its normal precipitation for the season. In Miami, Beish and Parker were still recovering from Hurricane Andrew. Although their telescopes were not damaged, their "domes" required extensive rebuilding. Nevertheless, Parker intermittently obtained good Mars images, beginning in July 1992. Using the above sources, together with data gathered by Beish of the Intemational Mars Patrol (ALPO) and by R. McKim of the British Astronomical Association, we can piece together much of the actvity of the apparition. This is still not as complete as data from the International Planetary Patrol network of 1969 through 1978. This research is supported in part by NASA grants NAGW 2257 and NAGW 3311 and Space Telescope Science Institute grant GO 2379. Pearl J. C.* Conrath B. J. McMillan W. W. Christensen P. R. First Observations of the Martian Atmosphere from the Mars Observer Thermal Emission Spectrometer Experiment The Mars Observer Thermal Emission Spectrometer (TES) experiment will study the surface and atmosphere of Mars. The instrument has three 2 x 3 arrays of detectors: broadband solar (0.3-2.7 micrometers), broadband thermal (5.5-100 micrometers), and spectrometric (200-1600 cm^-1, at selectable 5 and 10 cm-^1 resolution). Full disk measurements of the planet will be made in early August, 1993. At this time L(sub)3=11.5 degrees, near the time of maximum water vapor abundance in the north polar region. The signatures of atmospheric H2O, CO2, and dust are expected to dominate the spectrum. The data will be discussed in the context of atmospheric science. McMillan W. W.* Pearl J. C. Conrath B. J. Mariner 9 IRIS Revisited: Mars Observer or Bust In preparation for the Mars Observer mission and the return of tens of thousands of infrared spectra per day from the Thermal Emission Spectrometer (TES), we have developed a fast inversion algorithm to retrieve temperature structure and aerosol opacity from martian infrared spectra. As a test of our algorithm we are undertaking a systematic analysis of the entire Mariner 9 Infrared Interferometric Spectrometer (IRIS) dataset of 21,000+ spectra. While portions of the IRIS dataset have been previously analyzed, the lack of a speedy and robust algorithm to invert the IRIS spectra to retrieve temperature profiles and aerosol opacities has been a major impediment to a truly comprehensive analysis. Independent temperature profiles are obtained by separately inverting the two sides of the 15-micrometer (667 cm^-1) band (P and R branches). In the absence of any other atmospheric opacity, these profiles should be identical to within the rms error resulting from measurement noise propagation (1-2 K) and uncertainties in the CO2 absorption properties. However, in practice, additional wavenumber dependent extinction due to absorption and scattering by atmospheric aerosols causes the retrieved P and R branch profiles to differ. These differences are exploited to assist in deriving information on opacity resulting from the presence of the particulates. Additionally, the simultaneous analysis of IRIS spectra of the same geographic location, but different viewing geometries, is used in the characterization of the optical properties of airborne particulates. Preliminary results for temperature profiles and aerosol opacities will be presented. Herkenhoff K. E.* Ground-based Monitoring of Martian Atmospheric Opacity The amount of dust in the martian atmosphere is variable in both space and time, and complicates quantitative analysis of martian surface properties. Dust storms have been observed telescopically for almost 200 years, and are known to have major effects upon the structure and circulation of the martian atmosphere. Great dust storms tend to occur during the southern spring and summer, and may be an important mechanism by which dust is transported into the polar regions. Zurek and Martin (JGR 98, 3247-3259, 1993) found that "planet-encircling dust storms do not occur every Mars year, and...that there may have been periods of several successive years without such storms." The clarity of Mars images taken during the 1988 and 1990 oppositions suggests that the martian atmosphere has been less dusty recently than in previous years. Hubble Space Telescope images of Mars show that the dust opacity was less than 0.06 in December, 1990 (R. T. Clancy, personal communication, 1992). Martian dust storms can be detected only when Mars is relatively close to Earth, so a complete seasonal or interannual history of dust storms is impossible to obtain from groundbased data alone. The optical depth of aerosols in the martian atmosphere between dust storms has been determined primarily from spacecraft data, but can also be inferred from groundbased observations. Groundbased images of Mars show that atmospheric dust opacity significantly affects the photometric behavior of the planet. Lumme (Icarus 29, 69-81, 1976) modeled martian limb brightening using high-quality visible- light photographs taken on 3 September 1973. The optical thickness (0.16) and single-scattering albedo (0.55) at 435 nm that he derived are consistent with more recent results using Viking Orbiter violet-filter television data, suggesting that groundbased data may be used to determine the scattering properties of the martian atmosphere. Models of limb profiles derived from calibrated photographs of Mars taken during the Viking Orbiter mission will be compared with Viking opacity measurements in order to assess the utility of groundbased data in determining the opacity of aerosols in the martian atmosphere between dust storms. POSTERS de Pater I. Mitchell D. L. BIMA Observations of Mars in the J= 1-0 CO Line During the 1988 Opposition We imaged Mars in the 3 mm thermal continuum and in the J = 1-0 ^12CO line, using the BIMA millimeter interferometer^1, which is located at the Hat Creek Radio Observatory. The observations were obtained during September and October of 1988 when Mars was near opposition and perihelion. In the 3 mm continuum one typically probes a few centimeters into the martian crust. Our observations reveal diurnal heating effects in these subsurface layers, and show that the south pole is somewhat cooler than surrounding areas. We also obtained spectra at all locations on Mars' disk, with an effective resolution of ~6". The spectra are seen to evolve in a systematic way from the morning to the evening, which we attribute primarily to variations of the surface temperature and in the atmospheric temperature- pressure profile as the martian day progresses. We will show several spectra, and compare these with model atmosphere calculations and with single-dish spectra obtained by other researchers (e.g., Clancy et al. and Lellouch et al.) during the same time period. ^1Berkeley-Illinois-Maryland Array, operated by the University of California at Berkeley, the University of Illinois, and the University of Maryland, with support from the National Science Foundation Cernicharo J. Paubert G. Encrenaz Th. Lellouch E. Gulkis S. A Tentative Detection of the 183 GHz Water Vapor Line in the Martian Atmosphere The 183 GHz water vapor line was tentatively observed on Mars with the IRAM 30m millimeter antenna at Pico Veleta, under extremely dry atmospheric conditions. The Doppler shift of Mars was -9 MHz. A spectrum of Saturn was used to remove the absorption component due to the terrestrial mesospheric water. The central absorption core of the martian line was marginally detected, with a depth of 4% and a FWHM of 2.5 MHz. Preliminary calculations indicate that this profile can be fit with a mean H2O abundance of 1 mu-ppt (H2O/CO2= 10-5), uniformly distributed. If confirmed, this value, which can be considered as an upper limit, would be smaller than the H2O abundance inferred from the VLA measurements at 22 GHz (Clancy et al., Icarus, 100, 48), recorded one month earlier. Alternatively, the absence of a strong absorption core could also be explained by a cut-off in the vertical H2O distribution; in this case, no constraint would be derived from our observation upon the H2O abundance in the lower martian atmosphere. DiSanti M. A. Mumma M. J. Bjoraker G. Global Atmospheric Studies of Mars Existing models developed for the odd-hydrogen chemical cycle on Mars have neglected the role of transport within the atmosphere. We are investigating this problem with the use of spatially resolved global maps of CO, H2O, HDO, and CO2 absorption lines, to assess transport and photochemistry in the martian atmosphere. Our observations were conducted UT 9-12 March 1993 at the IRTF, using the recently commissioned facility spectrometer (CSHELL) at resolving powers of (2-4) x 10^4. A principal advantage in using modern array detectors such as CSHELL is that they afford simultaneous spatial coverage for extended objects. In our case, we oriented the slit north-south, which allowed full latitudinal coverage over the disk of Mars. The odd-hydrogen photochemical cycle is investigated through observations of the H2O 111-000 band at 8820 cm^-1 and, when combined with observations of the HDO nu(sub)1 band at 2720 cm^-1, addresses the question of possible hemispheric asymmetry in D/H ratio contained in the polar ice reservoirs. Quasiconserved species such as CO can be used as tracers of atmospheric transport. Through observations of the 2-0 and 3-0 bands, plus coincident CO2 lines, we have obtained global maps of the CO/CO2 mixing ratio, at a spatial scale of a few hundred kilometers. This permits searches for latitudinal and longitudinal variations, as well as variations with local martian topography. Wells W. K. Grosse F. A. Hunten D. M. Rizk B. Sprague A. L. Hill R. Morris R. L. Observations of Mars H2O and O2 During 1993 Groundbased observations of the column abundance of H2O vapor in the martian atmosphere corresponding to L(sub)s = 50 and 75 degrees (late northern spring) are reported and results are described. Doppler-shifted lines of the 8200 Angstrom band were observed in absorption and observed to have significant latitudinal variation. Measurements of molecular O2 were also acquired. Bilal A. Tyler G. L. Systematic Errors in Atmospheric Profiles Obtained from Abelian Inversion of Radio Occultation Data: Implications for Mars Observer Reduction of radio occultation data to retrieve atmospheric profiles, T-p(h), requires knowledge or assumption of the horizontal structure of the atmosphere. In the case of terrestrial planets the atmosphere usually is assumed to be spherically symmetric. This assumption leads to an analytic expression relating refractivity vs. height to the bending angle and the ray impact parameter, both of which are directly obtainable from the frequency data and trajectory information (Fjeldbo et al., 1971, Astron. J. 76, 123-140.) Occultation studies of the giant planets have demonstrated that departures from spherical symmetry, if not accounted for, can result in serious errors (Eshleman, 1975, Science, 189, 876-878). In the present study, we have analyzed and quantified errors in temperature and pressure profiles due to large scale atmospheric "tilts" in an otherwise spherical atmosphere, including the role of occultation geometry in enhancing or suppressing this error. The effect of atmospheric tilt is to introduce an approximately constant bias in the temperature profile. For example, a horizontal tilt of 0.6 mrad in the martian atmosphere can bias temperature profiles by up to 0.3 K, depending on the occultation geometry and on the orientation of the tilt. A much more severe slope of 4.8 mrad, which might exist in the upper atmosphere at the boundary of the polar caps (Pollack et. al., 1990, JGR, 95, 1447-1473) and where most occultations of Mars Observer are expected to occur, could bias the temperature profiles by up to 3.2 K for an orbit opening of 30 degrees, if such effects were neglected. Our approach in modeling the tilts has been to assume the local atmosphere to be spherically symmetric but with a different center of symmetry. This approach is satisfactory for the case in which the ray periapsis does not vary too much in latitude and longitude during the course of an occultation. Results from the NASA-Ames GCM for Mars suggest that horizontal variations in pressure are much more severe across latitude than they are across longitude. An axially symmetric figure of revolution should be a better model for the shape of the martian polar atmosphere than is a large-scale tilt. The current approach incorporates errors due to imprecise knowledge about the location of the transmitter, the receiver, and the center of the occulting planet, and is not restricted to any specific occultation geometry. Limaye S. S. Sromovsky L. A. Krauss R. Wright E. Santek D. Fry P. Saunders R. S. McIDAS-eXplorer: A Vehicle for Analysis of Solar System Data McIDAS-eXplorer is a software environment being developed to provide access to and enable efficient analysis of geophysical data acquired about solar system objects. It is an enhancement of McIDAS-X (a toolkit directed at the terrestrial meteorological community for analysis of real-time weather satellite, conventional and forecast model output. McIDAS-eXplorer provides access to and enables investigation of data from the CD-ROM volumes published by the Planetary Data System from NASA's solar system missions and incorporates the SPICE subroutine library developed by the Navigation and Ancillary Information Facility (NAIF) at the Jet Propulsion Laboratory for the display, navigation, animation, and analysis of planetary data on most UNIX workstations with X-windows support. Primarily intended for analysis of image data, the user extensible environment provided by McIDAS-eXplorer allows analysis of a wide variety of data with minimal effort in a multiframe (with overlay graphics), multitasking environment. Currently, tools are available for calibration, navigation and analysis of Voyager images of the giant planets and their satellites, Magellan radar and altimeter data, Viking Orbiter, Mars Mosaicked Digital Image Model data, and Galileo images. Tools are being added for access to and analysis of atmospheric profiles as well as spectral data such as Mariner-9 IRIS and Voyager IRIS observations. This work is funded by Contract NAS5-31347 from NASA's Applied Information Systems Research Program. Tuesday, October 19, 1993 1992 UREY PRIZE LECTURE 1:30 - 2:20 PM Grand Ballroom Chair(s): C. Pieters Lissauer J. J.* On the Diversity of Plausible Planetary Systems No abstract available. Tuesday, October 19, 1993 INVITED TALK 2:20 - 3:00 PM Grand Ballroom Chair(s): C. Pieters de Bergh C.* Breakthroughs in Groundbased Infrared Spectroscopy of Planets Recent progress in infrared detector technology and instrumentation and the development of infrared arrays have strongly benefited the study of planetary atmospheres and surfaces using groundbased spectroscopy. We will review the most recent of these studies. Three important breakthroughs will be more particularly discussed: 1) the new study of the deep atmosphere of Venus by measuring near-infrared radiation of the dark side of the planet; 2) the monitoring of processes occurring in the ionosphere of Jupiter using H^+3 infrared emissions; 3) the detection of nitrogen ice at the surface of Pluto. Tuesday, October 19, 1993 COMETS III 3:30 - 5:30 PM South Ballroom Chair(s): S. Hoban J. Crovisier Weissman P. R.* No, We Are Not in a Cometary Shower The flux of long-period comets through the planetary region will vary as a result of the magnitude and rate of external perturbations on the Oort cloud. In extreme cases, star passages through the Oort cloud or encounters with GMCs can cause showers of 10^9 or more comets to enter the planetary system (Hills, J. G., Astron. J. 86, 1730,1981). In order to correctly estimate the population of comets in the Oort cloud, it is necessary to know whether or not the solar system is currently experiencing an enhanced cometary flux. In addition, cometary showers have been invoked to try and explain the enhanced cratering rate currently estimated for the Earth over the past 250 m.y., which is about twice that estimated for the Moon over the past 3 G.y. Using computer-based dynamical simulation models, it is shown that the current distributions of long-period comet orbital elements are inconsistent with a cometary shower. Two different dynamical tests are employed. First, it is shown that the predicted orbit element distributions from a cometary shower are highly nonrandom, in contrast with the random distributions of the long- period comets. Second, the 1/a(sub)o distribution for the long-period comets shows no evidence of a perturbation of the inner Oort cloud, as would be expected for a major cometary shower. Taken together, these two tests allow one to set an upper limit on the magnitude of any recent major perturbations on the Oort cloud. This work was supported by the NASA Planetary Geology and Geophysics Program. Stern S. A.* Stewart G. R. Collisional Evolution in the Kuiper Disk We have constructed a model that computes the collision rates between bodies of various sizes in the Kuiper Disk (KD). We follow the trajectory of test bodies orbiting in an azimuthally homogeneous but radially varying disk. The test particles orbit on inclined, elliptical orbits. They are followed using Kepler's Equation to compute their residence time in each radial bin. The model allows one to explore the collision rate as a function of the total mass and the statistical distribution of orbital (a,e,i) elements in the disk. These calculations demonstrate that significant collisional evolution can take place in the present-day Kuiper Disk, and that if the disk was initially massive (i.e., like the cores of Uranus and Neptune), then a particularly intense, early period of collisional evolution must have occured billions of years ago. The existence of the recently discovery Kuiper Disk objects 1992QB1 and 1993FW likely indicate that such a period produced accumulation-driven growth to sizes at least as large as hundreds of kilometers in diameter. Why much larger (e.g., Earth-mass or larger) bodies did not accumulate as well in the Kuiper Disk is a mystery, but may be related to either (i) the removal of gas from this system before sufficient time was available for growth to proceed to this degree, or (ii) the lack of sufficient material in this region to result in larger bodies over 4.5 x 10^9 years. Jackson A. A.* Zook H. A. Dust Orbital Evolution from Outside the Orbit of Jupiter We have made a numerical study of the orbital evolution of particles released from sources outside the orbit of Jupiter as a continuation of modeling dust orbits in the solar system. We first examined the motion of dust from periodic comet Schwassman-Wachmann I, which has its perihelion outside the orbit of Jupiter (q = 5.772, Q = 6.31, e = .045 and i = 9.4). These particles evolve in orbits under the influence of radiation pressure, Poynting-Robertson and solar wind drag ,and gravitational perturbations by the planets. The particles were taken to be ideal black bodies with a density of one gram per cubic centimeter. The grains are taken to leave the parent body with zero relative velocity at perihelion. Out of a group of 27 particles ranging in size from 2 to 50 microns in radius only 4 made it past Jupiter. All other particles were ejected on unbound orbits. During orbital evolution, most of the dust particles followed quite irregular orbits before being scattered onto unbound orbits. A two micron radius particle was even observed being lofted into a one-to-one resonance with Saturn. The evolution of low inclination dust grains past Jupiter led us to try a simulation of a hypothetical source object beyond Jupiter in a circular orbit but with inclination varying between 0 and 180 degrees. It was found that almost 50% of the prograde orbits were ejected from the solar system while only about ten percent of the retrograde orbits were so effected. This may have relevance to the observation by the Ulysses spacecraft [2] of retrograde interplanetary dust particles. Keller H. U.* Kuhrt E. On the Formation of Cometary Surface Crusts The largest part of a typical cometary nucleus surface is inactive. Jet activity is restricted to a few percent of the surface. Various perceptions have been developed to model the formation of dust mantles. Thin layers of dust will readily choke free sublimation of volatiles. Heat, however, is still conducted through the mantle and builds up vapor pressure at somewhat elevated temperatures of sublimation. Not later than during the next perihelion passage of the comet the dust mantle will be blown off if gravity alone is the retarding force. The results of numerical calculations are presented that describe the growth of depleted layers of refractory matrix material on the surface. Cohesive forces within the nuclear matrix are strong enough to withstand the pressure build up of the underlying ice-rich material over a wide range of realistic parameters of the cometary matrix material. Crusts of a few centimeters up to 10 meters of thickness slowly form depending on the parameters after tens of orbital revolutions. Water sublimation is quenched but not extinguished. Sublimation underneath and diffusion of vapor through the inert crust can provide substantial contributions to the total gas production after perihelion due to a strong phase lag. Boice D. C.* Huebner W. F. Inner Coma Models with Dust for Comets Halley, Giacobini-Zinner, and Grigg-Skjellerup We present multifluid, hydrodynamic simulations for the gas, dust, and plasma flow appropriate for the spacecraft encounters with comets Halley, Giacobini- Zinner, and Grigg-Skjellerup. The simulations are based on a spherically symmetric neutral coma model with detailed photo and gas-phase chemistry and dust entrainment by the gas. The model includes a separate energy balance for the electrons, separate flow of the neutral gas, fast neutral atomic and molecular hydrogen, dust fragmentation, and distributed coma sources of gas- phase species related to the dust. The results from the ICE intercept of comet Giacobini-Zinner suggest a complex energy distribution for the cometary electrons. The electron energy budget in the model has been extended and the effects of hot electrons on excitation, dissociation, and ionization are found to be comparable to the corresponding photo processes, except in the innermost coma where collisions with water cool the electrons. These simulations allow a study of how certain coma features, e.g., spatial distributions of electrons and ions within the diamagnetic cavity, change with cometary gas production and dust-to-gas production rate. In comparison with observations, the simulations can give insights into the similarities and differences of these three comets. This research was supported by the NASA Planetary Atmospheres Program. Levasseur-Regourd A. C.* Renard J. B. Goidet B. Hadamcik E. Polarimetric Properties of Cometary Dust Grains and Asteroidal Surfaces Extended databases, together with new observations (1992-1993) from Pic-du- Midi telescope and from Giotto spacecraft, have allowed us to determine empirical curves for the polarization of light scattered by cometary dust grains or by asteroidal surfaces. Polarimetric data obtained from the same body at different phase angles can indeed be immediately used to draw phase curves, without a strong need for normalization. It has been noticed that all cometary data obtained over the whole coma with a large aperture compare quite well below approximately 45 degrees, and may be divided into two subsets corresponding to active comets and to less active comets above 45 degrees phase angle. Also, it has been found that the data corresponding to asteroids of the same taxonomic type compare remarkably well, and that measuring the polarization at a phase angle greater than 25 degrees may be used to suggest a taxonomic type. The evolution of the previous results with (1) wavelength in the visible domain, (2) spatial resolution for cometary observations, (3) temporal resolution for asteroidal observations is presented and discussed in terms of physical properties of small bodies in the solar system. Moore M. H.* Ferrante R. Nuth J. Donn B. Condensation of Crystalline Phase Ices on Silicate Smokes at T < 20 K The signature of silicate grains is often detected in comets and in a variety of interstellar objects. These grains may have complex compositions resulting from a variety of processes and often are sites for the condensation of volatile molecules at T < 80 K. We have examined the infrared spectra of several simple molecules condensed on laboratory-produced smokes. Detailed spectra of the OH stretch region of CH3OH, the nu(sub)1 of SO2, and the 45-micrometer band of H2O show that a dominant crystalline phase ice forms when these gases condense at T < 20 K on the silicate smokes (typically 0.1-0.5 mm thick). Similar spectra of CH3OH, SO2, and H2O condensed on aluminum substrates at T < 20 K reveal amorphous phase spectra; annealing to T ~ 130 K, 155 K, and 90 K, respectively, is required to convert these ices to a crystalline phase. Laboratory silicate smokes are made either by evaporation of SiO solid or by combustion of SiH4 followed by vapor phase nucleation and growth in a hyudrogen atmosphere. Smoke grains (typically tens of nanometers in size) contain unreacted sites and are amorphous in composition and in morphology. Since ices form nearly crystalline phases on these smokes, we suggest that the unreacted grain sites catalyze the phase transition during deposition. This is supported by our observation that amorphous phase deposits form on similar thicknesses of grains prepared by grinding commercial Si and SiO2 (a loose powder was formed on the aluminum surface). Amorphous water ice mixed with more volatile molecules has been suggested as a way to produce clathrates during warming in the outer solar nebula, which might explain the volatile content of comets and the icy satellites. The crystallization of amorphous water in comets is also one mechanism proposed to explain cometary outbursts. However, the assumption that amorphous phase deposits routinely form on grains at T < 80 K may meed to be reexamined based on these results. Brooke T. Y.* Crovisier J. Bockelee-Morvan D. Hanner M. S. Summary of the Possible Contribution of Aromatic Hydrocarbons to the Cometary 3.4 Micrometer Emission Feature In several comets, the broad 3.4-micrometer emission feature extends to the 3.23-3.31-micrometer region (3090-3020 cm^-l) where emission from methanol is not significant according to current model fits. A distinct feature centered at 3.28 micromters (3050 cm^-l) is seen in spectra of Halley and Levy 1990 XX. This frequency is highly suggestive of the C-H stretch in aromatic hydrocarbons (though other unsaturated hydrocarbons could contribute to the flux in the region). Following Baas et al. (Ap.J., 311, L97, 1986), we calculated the abundances of aromatic hydrocarbons in comets assuming that they emit in the infrared as a result of pumping at UV and visible wavelengths, as is believed to be the case in the interstellar medium. There is reasonably good evidence for comet-to-comet variability in the abundances. The comet aromatic signature is clearly different from that seen in reflection nebulae and HII regions. The implications for the composition and origin of cometary aromatics will be discussed. Constraints on the abundances of some saturated organic compounds with bands in the 3.2-3.6-micrometer region will also be presented. It is perhaps worth noting that some polycyclic aromatic hydrocarbons are among the most potent carcinogens known on Earth. Wyckoff S.* Kleine M. Wehinger P. Peterson B. Carbon Isotope Abundance Ratios in Comets An improved model of the excitation of CN in comets based on flourescence equilibrium has been developed for determining the carbon and nitrogen isotope abundance ratios from high resolution optical spectra. The fluorescence model includes the effects of collisions in the inner coma, the Swings effect, and improved radiative and molecular parameters of CN. All permitted transitions among the lowest electronic states, B^2Sigma^+, A^2II(sub)i, X^2Sigma^+, for the six lowest vibrational levels, and the 31 lowest rotational levels have been included in the model. Thus a total of 1452 individual energy levels were modeled for all stable isotopes of CN: ^12C^14N, ^13C^14N, ^12C^15N. The model is used to calculate the fluorescence efficiencies as a function of heliocentric velocity for all observed rotational lines of each isotopic species. These fluoresence efficiencies (g-factors), converted to isotope intensity ratios, can be used to derive the isotope abundance ratios of carbon and nitrogen in comets. The model has been used to determine the ^12C/^13C isotope abundance ratios in comets Halley (1986r), Levy (1990c), and Austin (1989c). We find that the carbon isotope ratios in these three comets are consistent with the bulk solar system value (90) within the observational errors. Limits on the nitrogen isotope ratios are also consistent with the nitrogen solar system value. Mumma M. J.* Hoban S. Reuter D. C. DiSanti M. Methanol in Recent Comets: Evidence for Two Distinct Cometary Populations The abundance of methanol has now been measured for seven comets. Two are dynamically new, one is a very long period comet, one is of intermediate period, and three are short-period comets. None are Jupiter-family comets. The sampled comets exhibit two distinct values for the relative abundance CH3OH/H2O. The simplest interpretation is that the two values are cosmogonic, i.e., that the comets represent two populations that formed in different regions of the solar nebula, or that accumulated at different times during nebular collapse. Two of the four comets with high (~5%) methanol abundance are dynamically new, thus we associate high methanol abundance (Population I comets) with formation in the Uranus-Neptune region of the solar nebula. The third is P/Swift-Tuttle, which is now a short-period comet. However, its methanol abundance suggests that it once was an Oort cloud comet. If so, it must have been captured long ago since it has resided in the inner solar system long enough to generate a meteor stream. The two comets with low (<-1%) methanol abundance (Population II) may have originated elsewhere, possibly in the Kuiper Belt. The observed compositional dichotomy is also consistent with a single formation region, if the abundance of methanol in pre-cometary ices varied strongly with time in the nebula. Such differences might reflect increased radiation processing of ices in relatively clear regions of the nebula. Then late-forming comets may contain less methanol than those formed earlier in the nebular collapse phase. Measurements of methanol and related species in additional comets would test the composition and relative numbers of type I and II comets, for comparison with radiation chemistry models and predicted orbital statistics [1,2]. It is of particular interest to extend these measurements to comets that are thought to have formed in the Kuiper belt, such as the Jupiter-family comets. References: [1] Duncan M. et al. (1988) Ap.J., 328:L69-L73. [2] Mumma M. J. et al. (1993) In Protostars and Planets III, 1177-1252. Budzien S. A.* Modeling the Quenched Cometary OH Observed in High Resolution IUE Spectra Solar resonance fluorescence can create a population inversion of the X2II lambda-states of cometary OH molecules that enables 18-cm radio observations and affects the ultraviolet fluorescence spectrum. However, the inversion is collisionally quenched by ions and neutrals in the inner coma, and this process changes both the OH ultraviolet fluorescence efficiency and the observed 18-cm flux. A model of the OH quenching applicable to both ultraviolet and radio observations is presented and compared to results from high-resolution IUE spectra observed in several comets from 1978-1990. The effects of solar variabilty, the Swings effect, a possible OH population inversion induced through the dissociation process, and fluorescent nonequilibrium, and the implications for ultraviolet and radio observations are discussed. Yi Y.* Brandt J. C. Global Simulation of a Comet Crossing the Heliospheric Sector Boundary: Disconnection Events The interaction between the solar wind and a comet can be simulated through a 3-dimensional magnetohydrodynamic (MHD) code. We adapted the code of Ogino, Walker, and Ashour-Abdalla to a differential grid MHD code and applied the revised code to the case of a comet crossing the heliospheric current sheet (the sector boundary) in the solar wind. Frontside magnetic reconnection occurs between the reversed interplanetary field lines draped around the comet in the cometosheath. Subsequent evolution, including the formation of symmetric ion tail rays (antisunward), the production of the disconnected tail, and the establishment of a new tail of opposite polarity, reproduces the basic observed morphology of disconnection events (DEs). These simulation results explain the strong association of DEs in Comet Halley 1985-86 with sector boundaries (Brandt et al. 1992) and, by implication, the association in general. Thus, the frontside magnetic reconnection model for DEs (Niedner and Brandt 1978) appears to be confirmed. Farnham T. L. Meech K. J. Finson-Probstein Modeling of the Dust Tail of Comet Torres 1987V We are developing a kinetic model of cometary dust grains, based on the Finson-Probstein (F-P) technique, to study the particle populations in cometary dust tails. The appearance and morphology of a comet's dust tail are the result of the opposing forces of gravity and solar radiation pressure that act upon individual grains. Using the dynamics of the system, the surface density along the tail is determined by summing up the scattering contributions from all particles. By matching models to observed images, various grain parameters can be determined. The main parameters are the particle size distribution, the dust production rates, and the escape velocities. Our model is similar to the original F-P work, however, we have made several improvements including incorporating non-spherical emissions and more realistic scattering functions into the model. In addition, we are working on a process to use a direct inversion of the image to obtain the grain parameters. Comet Torres (1987V) is a dynamically new comet that was not heavily observed during its apparition, although it exhibits an interesting and long-lived tail. We have images from 12 observing runs spanning the period May 1987 through January 1992 (heliocentric distances from 3.6-13.1 AU postperihelion). In the images out to at least 11 AU, there is a broad, straight dust tail whose basic appearance is remarkably similar from image to image, regardless of the geometry of the comet, Earth, and Sun. Near the nucleus, the tail shows a slight spiral structure such as that often seen in P/SW1, but preliminary indications suggest that this may be modeled by small particles. At 13 AU, the last remnants of coma were observed as a diffuse spherical cloud, and the overall brightness had faded substantially from previous runs. The application of several suitability tests indicates that Torres is a good candidate for modeling using the Finson-Probstein techniques, especially with the large time base that our images cover. A preliminary analysis of the tail indicates that it is composed of large particles (possibly up to centimeter- size) that were ejected near perihelion with very small velocities. Smaller particles (micrometer size), which would be expected for a dynamically new comet, could be a component of the tail in the earlier images, but they will dissipate too rapidly to contribute at the larger heliocentric distances. A detailed analysis of the tail of Comet Torres will be presented, including specifics on the particle parameters that are determined and the status of the inversion process. Verkhoglyadova O. P. Kotsarenko N. Ya. Churyumov K. I. Towards the Phenomenon of "Closing-in" Rays in Cometary Plasma Tails Filamentation instability study in terms of one-dimensional model proposed in [1] with taking into account weak plasma flow to tail axis far from the nuclei is undertaken. The data obtained ICE during it flight through Giacobini-Zinner tail are in favor of this model [2]. The system is assumed to be homogeneous and infinite. The aperiodic instability development results in exponentially exceeding in time and periodic in space disturbances of cometary plasma density, magnetic field etc., across the tail. Rays take the form of plasma fibers with thickness determined by spatial scales of the above disturbances. Instability increment and maximum ray thickness coincide approximately with ones obtained in [1]. However the small transverse velocity causes travelling wave of stratification, so the ray structure 'closes' to the axis with the mean velocity of transverse plasma flow. References: [l] Kotsarenko N. Ya., et al. (1992) BAAS, 24, 1024. [2] Verkhoglyadova O. P. et al. (1993) Russ. Astron. J. (in print). [2] McComas P. J. et al. (1987) JGR, 92, 1139. Combi M. R. The Fragmentation of Dust in the Inner Comae of Comets Cometary dust particles are entrained in the expanding gas flow created by the vaporization of ices. Traditional approaches to dustygas dynamics in the inner coma consider there to be an initial distribution of dust particle sizes that do not fragment or evaporate. The standard Finson-Probstein model (and later variations) yields a "1-to-1-to-1" correspondence between the size of a dust particle, its terminal velocity owing to gas drag, and its radiation pressure acceleration, which creates the notable cometary dust tail. The relative magnitudes for this correspondence depend on the variations of particle densities and light scattering properties with particle size. Our dust model can include realistic particle size distributions, radiation scattering properties for dark absorbing grains, a density distribution of porous aggregates, terminal velocities consistent with dusty-gas dynamics, radiation pressure acceleration, and true 3-D and time-dependent particle trajectories. Models of dust coma images show that the elongated shapes of dust-coma isophotes on the scale of 10^4-10^5 km from the nucleus requires that the "1-to-1-to-1" correspondence cannot in general be correct. Particles of a certain size (and radiation pressure) must have a broad range of velocities in order to explain the elongated shape. Such isophote shapes are also seen in the hydrogen coma. Although there is approximately a single value of the radiation pressure acceleration for all H atoms, there is a broad distribution in velocities caused by the combination of photochemistry and molecular collisions. A straightforward way to create such a distribution for dust is if fragmentation occurs within or even somewhat outside the dusty-gas dynamic acceleration region (~100several hundred km). In this way, initially large particles, which are accelerated to fairly slow velocities by gas-drag, fragment to form small particles that still move slowly, but are subject to large radiation pressure accelerations. Fragmentation has already been suggested as one possible interpretation for the shape of the spatial profiles of dust extracted from Giotto images of Comet Halley. The general elongated isophote shapes seen in groundbased images may represent a signature of fragmentation that would have to occur on scales of up to several hundred kilometers. Hanner M. S. Yanamandra-Fisher P. A. West R. A. Scattering Properties of Aggregate Particles We are engaged in modeling the scattering and emitting properties of irregular particles with the goal of understanding the composition and structure of cometary dust particles. Our numerical code is based on the discrete dipole approximation, in which a particle is modeled as an array of radiating dipoles on a three-dimensional lattice (Draine, Ap.J. 333, 848, 1988). Some of the optical properties of an aggregate particle are determined by the optical properties of the monomers (West Appl. Opt. 30, 5316, 1991). Spherical monomers do not reproduce the observed cometary features; moreover, spurious resonances are introduced. We are creating aggregates composed of non- spherical monomers of differing sizes and optical properties. In particular, tetrahedral monomers do give rise to the 11.25 micrometer olivine emission peak observed in comets, as well as the negative polarization at small phase angles for size parameter X >~ 3. Results on the single-scattering albedo, polarization, and shape of the 10-micrometer silicate emission feature for these aggregates will be presented. This research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under contract with NASA. Rettig T. W. Tegler S. C. Pasto D. J. Mumma M. J. Spontaneous Polymerization as the Trigger for Cometary Outbursts Cometary outbursts occur in both short-period and dynamically new comets with a wide range of parameters. A number of mechanisms have been suggested for outburst activity but none have been generally accepted. As a possible energy source for outbursts, we calculate the energy released by polymerization of a variety of simple molecules such as HCN. We suggest that polymerization in localized planetesimal-sized regions of a cometary nucleus is a possible source of energy for outbursts. We have chosen to investigate this mechanism because of the presence of HCN in comets and in the interstellar medium, the presence of possible polymeric dissociation products (NH2, CN) in cometary comae, and because of the energetics and reactivity of the molecular components. Also, HCN polymers have been postulated to exist on the dark surface of comets such as Halley. The calculation of HCN polymerization provides sufficient energy in a localized planetesimal sized region (20-100 m) of a comet to raise the temperature of the subsurface ices and produce sublimation. The mechanism requires a nonhomogeneous accretion of planetesimals. Only near or beyond the orbits of Uranus-Neptune are temperatures cold enough (around 40 K) to maintain the original content of the interstellar grains including HCN. With radial gravitational mixing of the planetesimals, the final accretion to comet-sized objects resulted from planetesimals from a variety of distances in the Jupiter to Neptune regions. The size calculation is in close accord with the accretion models predicting planetesimals 5-100 m in diameter. HCN polymers produce both carbon ladder polymers and to a lesser extent, reactive carbon-nitrogen ladder structures called polyamidines. H20 is a part of this matrix and may interact with the HCN polymers to release higher amounts of C02 and CO observed in outbursts. New results of recent experiments to determine fragmentation rates of various polymers will be presented. Schwehm G. H. Laurance R. J. Hechler M. Rosetta: The Planetary Cornerstone Mission in ESA's Long-Term Science Programme During the past year the Rosetta Mission scenario went through a considerable evolution. After the original mission objective the comet sample return could not be maintained, the mission had to be descoped to fit within the financial and technological envelope of ESA. The technological constraints being the use of solar arrays to provide power, the use of an ARIANE launch vehicle and a ground segment primarily based on ground stations from ESA member states. From a variety of mission scenarios that proved to be feasible within these constraints, a comet rendezvous mission with emphasis on insitu analysis of the cometary matter was selected as baseline mission. The mission scenario includes the deployment of a surface science station or several smaller stations and flybys at one or two asteroids en route to the comet. Although designed now as a 'European-only scenario' the mission is open for an international collaboration, which could improve the overall science return. On 5 November 1993 the Science Programme Committee of ESA will take a decision if Rosetta will be implemented as Cornerstone 3 with a potential launch date around 2003. Tuesday, October 19, 1993 MARS ATMOSPHERE II: MODELING 3:30 - 5:30 PM North Ballroom Chair(s): M. Allen J. Barnes Allen M.* Nair H. Yung Y. L. Anbar A. D. Clancy R. T. A Homogeneous Gas-Phase Photochemical Model of the Atmosphere of Mars The primary constituent of the martian atmosphere is C02. Solar irradiation should lead to significant amounts of CO and O2, in the ratio of 2:1. However measurements of the atmosphere have revealed a composition quite different from this initial prediction. For more than two decades, atmospheric modelers have sought to explain the observed composition, but good simulations have been achieved only with the adoption of key simplifying assumptions. We shall present a self-consistent model of martian atmospheric photochemistry that can reproduce modern measurements with a minimal number of imposed constraints on the calculated results. We have solved the one-dimensional continuity equation, in which chemistry and vertical diffusive transport are coupled, for 27 species containing carbon, nitrogen, and/or oxygen atoms. The model includes 102 homogeneous gas-phase reactions. Mixing ratios at the surface are prescribed for CO2 and H2O to account for atmosphere-surface mass transfer. Escape effusion velocities and/or fluxes are prescribed for H, H2, and O at 240 km altitude, the upper boundary. Zero fluxes through the top and bottom boundaries are assumed for all other species since these are thought not to have significant escape fluxes nor major interaction rates with surface materials. The model does not simulate in detail H2O condensation in the lower atmosphere, but instead we prescribe the vertical scale heights for H2O vapor to account for this process. When currently accepted photodissociation parameters and kinetic rate coefficients are used, the model abundance for CO is significantly smaller than observed values. However, we achieve satisfactory agreement between calculations and measurements for CO, O2 and O3 by revising the rate coefficients of a few key reactions within their published uncertainties. The adjustments in the parameters for OH + H2O, and CO + OH also would resolve discrepancies between models and observations of the terrestrial mesosphere. At this time, we feel there is no need to invoke significant rates for heterogeneous processes to bring models into better agreement with observations, as has been suggested by several recent investigations. Atreya S. K.* Gu Z. G. Stability of the Martian Atmosphere: Is Heterogeneous Catalysis Essential? The classical problem of the stability of carbon dioxide in the martian atmosphere was recognized nearly two decades ago. Briefly, sunlight below 2200 Angstroms would irreversibly convert C02 in the martian atmosphere to its products CO and O at a rate of (1-2) times 10^12 cm^-2 s^-l since the reverse reaction is spin forbidden. The implication is that all of the C02 in the martian aanosphere should have been destroyed in less than 10,000 years, and the present atmospheric levels of CO 8 times 10^-4) and O2 (~1.3 times 10^-3) should have been attained in 3-6 years. Yet, it is found that the level of CO2 in the atmosphere is stable except for the normal 30% seasonal change in surface pressure, and that the mole fractions of CO and O2 are also stable and do not continue to build up. McElroy and Donahue (1972)Science, 177, 986 and Parkinson and Hunten (1972)JAS, 29,1380, proposed recycling CO into C02, through OH catalysis. These pre-Viking models, however, assumed an unacceptably large value for the middle atmospheric eddy mixing coefficient (>=10^8 cm^2 s^-1) or an excessively high water vapor content (500-1000 ppm). To overcome these difficulties (Atreya and Blamont (1990) GRL, 17, 287) suggested heterogeneous surface or chemical catalysis of CO + O --> CO2 reaction on aerosols of dust or ice in the martian atmosphere. This talk presents results of a comprehensive photochemical model, which is a single unified neutral atmosphere-ionosphere model from the surface to about 250 km. It will be shown that with the latest available laboratory kinetics data on the critical source of OH (HO2 + O), the CO2 absorption cross-sections, especially their temperature dependence, and the martian atmospheric parameters, it is possible to explain the stable atmospheric levels of CO2, CO, and O2 almost entirely on the basis of homogeneous gas phase photochemistry alone. Heterogeneous processes may still be necessary; they will, however, be needed to suppress the production of CO2. Nair H.* Gerstell M. F. Yung Y. L. Photochemistry and Thermal Structure of the Early Martian Atmosphere Many investigators of the early martian climate have suggested that a dense CO2 atmosphere was present in order to warm the surface above the melting point of water (e.g., Pollack et al. 1987). However, Kasting (1991) recently pointed out that previous thermal models of the primitive martian atmosphere had not considered the condensation of CO2. When this effect was incorporated, Kasting found that a purely CO2 greenhouse is an inadequate mechanism to warm the surface. Early in martian history, the Sun was expected to have had a much greater output in the ultraviolet than it does today. We suggest that under the influence of the early solar spectrum, an initially pure CO2 atmosphere will rapidly be converted to a predominantly CO atmosphere, enhancing the greenhouse effect through pressure broadening of the CO2 absorption lines. We will investigate the climatology of an atmosphere where CO2 is a minor constituent but still the key radiative species. The thermal structure of the dust-free atmosphere is estimated by employing a simple radiative-convective model similar to that used by Gierasch and Goody (1968). Heating and cooling rates for a martian atmosphere containing O2, O3, H2O, CO, and CO2 are calculated using the quasi-random model described in Crisp (1986,1990). Colwell W. B.* Stewart A. I. F. The Mars Hot Oxygen Exosphere: A Two-Hemisphere Approach The existence of a hot oxygen exosphere on Mars was suggested by McElroy (Science, 175, 443, 1972). Recent energy spectra and altitude profiles for the case of a homogeneous exobase are calculated by Zhang (JGR, 98, 3311, 1993); see also Nagy et al. (Ann. Geophys., 8, 251, 1990). The immediate source of hot oxygen is the dissociative recombination of O2+, which is itself produced through photoionization of the ambient atmosphere. We adopt the branching ratios of Paxton (Thesis, Univ. of Colorado, 1983) for the five recombination channels. The dayside energy spectrum of hot atoms is derived at midsolar cycle, taking into account the escape and collisional thermalization processes. Exospheric densities are calculated using Liouville's theorem. A distribution of Keplerian trajectories through a test point in the exosphere are traced to the exobase. Conditions at their feet are considered when integrating their contributions to hot atom density over solid angle. This technique permits consideration of a nonhomogeneous exobase, specifically the absence of sources on the nightside. We present energy spectra and altitude profiles of hot oxygen for various solar zenith angles. Our values of the escape rate of hot oxygen are in accord with earlier calculations. We discuss the flow of dayside hot oxygen atoms and energy to the nightside. Of the 4.2 x 10^l4 erg/sec deposited into the exosphere of hot atoms by the recombination process; 7.7 x 10^13 erg/sec are carried to the nightside exobase. We also consider the effect of exospheric hot oxygen on thermal hydrogen and find that sputtering of hydrogen by hot oxygen is small in comparison to the dayside thermal loss rate, unless there is a substantial cold-trap enhancement of hydrogen abundances on the nightside. Bougher S. W.* Mars TGCM Calculations of the Upper Atmosphere: Predictions for Mars Observer Arrival The National Center for Atmospheric Research thermospheric general circulation model (TGCM) for the Earth has been modified to examine the three-dimensional structure and circulation of the upper atmosphere of Mars (MTGCM). The MTGCM presently solves primitive equations yielding global solutions for the zonal, meridional, and vertical velocities, total temperatures, geopotential heights, and important neutral (CO2, CO, O, N2, Ar, O2) and dayside ion (O+, CO+, O+, NO+) densities. The solar, orbital, seasonal, and upward propagating tidal responses of the atmosphere above 70 km have been studied for various combinations of input parameters. Calculations driven by the changing in situ solar forcing are generally successful in explaining existing Mars data. However, Mariner 9 UVS airglow measurements taken during a global dust storm provide evidence of large temperature and density variations uncorrelated with solar activity. Significant forcing of the upper mesosphere and thermosphere from below apparently occurs as a result of upward propagating gravity waves or tides generated by solar heating of airborne dust. The predicted effects are strongest and most likely during solar minimum conditions around southern summer solstice (Ls = 270). Similar solar, orbital, seasonal, and potential dust storm conditions are appropriate to the first 6 months of Mars Observer mapping activity. MTGCM predictions are presented for upper atmospheric fields corresponding to these Mars conditions. Temperatures over 70-90 km are examined with a view to comparison with PMIRR temperatures. Simulations with and without upward propagating thermally driven tides are presented to illustrate the range of temperatures possible. Future plans include the sequential coupling of this MTGCM model with a lower atmosphere GCM, thereby providing an improved way to study these tidal effects throughout the entire atmosphere. Krasnopolsky V. A.* Bowyer S. McDonald J. Gladstone G. R. Chakrabarti S. First Measurement of Helium on Mars: Implications for the Problem of Radiogenic Gases on the Terrestrial Planets A hundred photons of the martian He 584-Angstrom airglow detected by the Extreme Ultraviolet Explorer satellite during a two-day exposure (January 22-23, 1993) correspond to the effective disk average intensity of 43 +/- 10 Rayleigh. Radiative transfer calculations, using a model atmosphere appropriate to the conditions of the observation and having the exospheric temperature of 210 +/- 20 K, result in the He mixing ratio of 1.1 +/- 0.3 ppm in the lower atmosphere. Nonthermal escape of He is due to electron impact ionization and pickup of He+ by the solar wind, to collisions with hot oxygen atoms, and to charge exchange with molecular species (1.4 x 10^5, 3 x 10^4, and 7 x 10^3 cm^-2s^-1, respectively). The lifetime of He on Mars is equal to 5 x 10^4 yr, and the measurement reveals outgassing processes that appear to be rather strong on Mars. The He outgassing rate coupled with the 40Ar atmospheric abundance and with the K:U:Th ratio measured in the surface rocks are applied as the initial data of modeling that presumes loss of all Ar accumulated in the atmosphere during the first Gyr by large-scale impacts. The two-reservoir model results in the full mass ratios of K, 6 x 10^-6 g/g, of U, 1.9 x 10^-9 g/g, and of Th, 6.5 x 10^-9 g/g, which are smaller by factors of 45 for K and of 12 for U and Th than the values for the Earth. The radiogenic heat flux is equal to 1.7 erg/(cm^2 s). Similar modeling for Venus gives the mass ratios of K, 4 x 10^-5 g/g, of U, 6 x 10^-9 g/g, and of Th, 2 x 10^-8 g/g, and the radiogenic heat flux of 14 erg/(cm^2 s). Barnes J. R.* Walsh T. Murphy J. Transport and Mixing Processes in the Mars Atmosphere as Simulated by the NASA-Ames GCM Atmospheric transport is of considerable importance for the climate system of Mars, especially for the interrelated cycles of dust, water, and carbon dioxide. Transport and mixing is also of importance in relation to photochemical processes in the martian atmosphere. It is thus of interest to attempt to characterize some of the basic aspects of the atmospheric transport and mixing processes on Mars. Taking the atmospheric circulation simulated in the NASA-Ames Mars GCM, passive tracer transport experiments have been performed with an aerosol transport model (also developed at NASA-Ames) in order to quantitatively assess basic aspects of transport and mixing in the Mars atmosphere. By carrying out "twin" transport experiments employing very different initial states, the full eddy transport tensor can be determined from the model data. This then yields the eddy mixing or diffusion coefficient values, as well as the so-called mean transport circulation that acts to produce advective transpon in the latitude-height plane. The latter transport could be characterized, for example, in terms of a "turnover" time scale for a middle atmospheric region penetrated by the mean circulation. GCM experiments have indicated that such a time scale can be relatively short as a result of the very vigorous and deep mean meridional circulations in the Mars atmosphere. The latter (as well as the atmospheric eddy motions) have been found to change very substantially in the GCM simulations with season and atmospheric dust loading, so it is anticipated that the strength of the tracer transport should vary considerably as well. It is expected that the GCM transport measures should effectively represent lower limits (unless the mean circulations and large-scale eddies in the model are much stronger than on Mars), since the horizontal resolution in the GCM is too coarse to resolve most of the gravity wave motions generated by topography and other processes. There is increasing evidence from modeling and observations that breaking gravity waves may act to provide strong vertical mixing in the middle atmosphere of Mars (above 30-50 km). Hollingsworth J. L.* Haberle R. M. Houben H. C. Young R. E. Mars Climate Model Annual-Cycle Simulations A 3-D climate model based on the primitive equations in log-pressure coordinates is used to perform multi-annual simulations of Mars' atmospheric circulation. The model uses a spectral (spherical harmonic) representation in the horizontal and finite differences in the vertical. Simplified physics are applied: thermal forcing is expressed in terms of Newtonian cooling and momentum drag is specified in terms of Rayleigh friction near the model top to provide a "sponge" layer. In the multiyear simulations, the model is configured at 06Tr16L14, a relatively low spatial resolution corresponding to roughly 18 x 5 degrees longitude-latitude with 14 unequally spaced vertical levels. Results indicate significant interannual variability, with the greatest variations occurring in the southern hemisphere during late winter and spring seasons. During southern winter, transient eddies are much weaker than their counterparts during northern winter. However, eddy activity in the southern hemisphere rises dramatically during spring and is greater than that in the north at any season. The model's interannual variability is assessed by examination of hemispherically integrated zonal (ZKE) and eddy kinetic energies (EKE). In the southern hemisphere, EKE is a relative minimum near summer/winter solstice and maximum during fall/spring transitions. In the northern hemisphere, EKE shows a similar seasonal pattern, yet the fluctuations are much less pronounced. Since these results occur in a model with simple, but repeatable forcing, it is suggested that Mars' climate system has enough nonlinearity to produce substantial interannual variability. Lyons J. R.* A Comparison of Dust Raising on Earth and on Mars A review of terrestrial duststorm data reveals that significant dust raising only occurs during high surface wind conditions and usually over subsynoptic scales. Terrestrial duststorms are therefore classified as severe weather events. It has previously been suggested (Woiceshyn, 1977) that martian duststorms occur under conditions similar to terrestrial storms. Here I propose specific mechanisms for dust raising on Mars by analogy with terrestrial processes. Terrestrial dust raising can occur by several mechanisms: strong surface pressure gradients; transport of upper level momentum to the surface via an isentropic boundary layer; downslope winds on the lee side of a mountain range. Each of these processes can accompany cyclogenesis on Earth. When an upper level weather system forms a rapidly deepening surface storm circulation over an arid region, very severe soil erosion and duststorms can result. Such storms are well known in the Sahara, northern China and to a lesser extent the U.S. Great Plains. They occur most often in spring, but are also seen during winter and summer seasons, and exhibit a marked interannual variability. Fronts, such as those accompanying intrusions of cold air masses, and mid-tropospheric jets, caught in a deep rnixed layer, can also produce wind gusts at the surface (Prospero,1981). Downslope winds are thought to be associated with gravity waves (Durran,1990) and so occur in a stable surface layer. Observations of local dust clouds by Viking (Peterfreund and Kieffer, 1979), while not exhaustive suggest that there are two main classes of local storms: subtropical and polar cap edge storms. The former may be associated with momentum transport (e.g., descending subtropical jet) to the surface, particularly near subsolar latitudes close to perihelion. If the boundary layer is stable due to heating of a dust haze, then downslope winds may be important. The polar cap storms are likely to be associated with baroclinic disturbances accompanying the polar jet, perhaps enhanced by the sublimation flow from the cap. Low level cyclogenesis at midlatitudes may be responsible for generating large surface pressure gradients for periods exceeding a martian day, thereby raising substantial quantites of dust. The interannual variability of such processes may be an important component of the interannual variability of global dust storms, particularly if local dust storms are indeed a trigger for global events (Pollack et al, 1979). The observation of local dust storms by Mars Observer should yield important information about both synoptic weather and subsynoptic dust raising on Mars. Johnson B. R.* Atreya S. K. The Far Infrared Spectra of Martian Ices Aerosols, including dust and H2O and CO2 ices, are believed to play an important role in the composition, thermal structure and the stability of the martian atmosphere. The capability of far infrared spectral observations to determine planetary ice composition was investigated. Thin film transmission spectra of crystalline (Ic) and amorphous H2O ice and crystalline CO2 ice deposited on a silicon substrate were measured in the laboratory by a Fourier transform spectrometer and used to calculate their refractive indices at temperatures of 77 to 150 K over the range 50-500 cm^-1. The refractive index data were compared with previously published data and generally agreed well to within the 20% experimental error. The chief source of error was due to uncertainty in film thickness, determined by fitting channel spectra with a theoretical transmission model. The derived refractive indices were incorporated into a radiative transfer model used to study the far infrared properties of cloud and surface ices on Mars. Thin mid-latitude H2O ice clouds with typical integrated ice amounts of 0.5 pr micrometer have vertical far infrared optical depths of the order of 10^-4, making their detection difficult. However, cloud models based on Viking observations of the polar regions produced observable spectral features (Delta T <= 8K) in calculated spectra near 225 cm^-1. Inclusion of a CO2 ice haze in the calculation lowered the apparent surface brightness temperature by 10-20 K. Theoretical emission by CO2 frost shows strong spectral contrast of ~20-30 K near the lattice absorption bands in solid CO2. In the weakly absorbing inter-band region, CO2 frost emissivity is ~0.4 increasing to ~0.7 with the incorporation of small amounts (0.1-1.0%) of dust or water ice. H2O frost exhibits poor spectral contrast in this region with an emissivity close to unity. Forget F.* Pollack J. B. Impact of the CO2 and H2O Clouds of the Martian Polar Hood on the Polar Energy Balance Clouds covering extensive areas above the martian polar caps have regularly been observed during the fall and winter seasons of each hemisphere. In particular, the very cold temperatures observed during the polar night have been identified as CO2 clouds. Estimating the impact of the polar hood clouds on the energy balance of the polar regions is necessary to model the CO2 cycle, and address puzzling problems like the polar caps assymetry. For example, by altering the thermal radiation emitted to space, CO2 clouds alter the amount of CO2 that condenses during the fall and winter seasons. We have analyzed the complete set of Viking IRTM data to define the spatial and temporal properties of the polar hoods, and estimated how their presence affects the energy radiated by the atmosphere/caps system to space. To recover the integrated thermal fluxes from the IRTM data, we have developed a simple model of the polar hood, thought to consist of "warm" H20 clouds, lying above colder and opaque CO2 clouds. Such a model is based on the analysis of the IRIS spectra [1]. It is possible to fit most spectra available. The high thermal emissivity of the CO2 ice particles is possible because of contamination by dust and water [2]. Besides, we have found new evidence for such opaque CO2 clouds in the IRTM data. A specific treatment of the IRTM data has also allowed us to recover the "true" ground ice temperatures, which were difficult to measure directly because of the presence of clouds. These temperatures are colder than expected. Thus, we have been able to estimate the wavelength integrated thermal emission by the surface/atmosphere system at the top of the atmosphere, and to compare this with what would have been emitted by the bare polar caps. The polar clouds are found to have a very strong impact on the polar energy budget, especially in the northern hemisphere where CO2 clouds can be formed at most polar latitudes. This is consistant with what has been concluded from models trying to fit the Viking Lander pressure curves [3]. [1] Paige, D. A. (1990) Bull. of AAS, 22-23. [2] Warren et al. (1990) JGR, 95, 717-741. [3] See Hourdin et al., this issue, and Pollack et al. (1993) JGR Planets 98(E2), 3149. Talagrand O.* Hourdin F. Forget F. Numerical Simulation of the General Circulation of the Martian Atmosphere and the Viking Pressure Data On the basis of results of the martian atmospheric General Circulation Model (GCM) developed in the last four years at Laboratoire de Meteorologie Dynamique, we have shown that the seasonal variations of the surface pressure, as measured at a fixed point of the surface of the planet, can be decomposed into three components: (1) the time evolution of the global atmospheric mass, due to condensation and sublimation of carbon dioxide; (2) an orographic contribution due to the modulation of the large spatial pressure variations along slopes, by the seasonal evolution of temperature; and (3) a dynamical contribution due to the geostrophic balance between the near-surface winds and surface pressure. The decomposition can be justified mathematically based on the sigma-coordinate version of the classical geostrophic equation [1] and is accurately satisfied in the simulations. As a consequence, the condensation sublimation cycle cannot be accurately determined, from local pressure measurements, without information on the atmospheric dynamics, as produced by GCMs. Additional experiments have shown that the seasonal evolution of the global atmospheric mass is very sensitive to the visible albedo A(sub)ice and to the thermal emissivity epsilon(sub)ice of the polar ice, which are poorly known from observation. A best-fit simulation to the Viking pressure data was derived by allowing different values of both parameters for the two caps: A(sub)ice = 0.62 and epsilon(sub)ice = 0.6 for the north cap and A(sub)ice = 0.45 and epsilon(sub)ice = 0.76 for the south cap. As already mentioned by Pollack et al. [2], these "best-fit" values of the ice parameters must not be interpreted as real physical properties of ice materials since they also account for some model deficiencies, and in particular the nonrepresentation of CO2 ice clouds, created by atmospheric condensation during the cap formation. This may explain the very low value of the "best-fit" emissivity of the northern cap, which has been observed to be preferentially covered by polar haze [3]. The low emissivity will account for the trapping of thermal radiations by clouds during the formation phase. [1] Hourdin et al. (1993) J. Atmosph. Sci., in press. [2] Pollack et al. (1993) JGR Planets, 98(E2), 3149-3141. [3] Forget and Pollack (1993) this conference. POSTERS Banfield D. Toigo A. Ingersoll A. P. Martian Weather Correlation Length Scales from Viking IRTM T15 Mars Observer soon will be returning data on the state of Mars' atmosphere. Several groups have set out to assimilate these data into numerical models to track the weather systems of Mars. However, to properly assimilate the data into a model, the correlation length scale of weather patterns on Mars must be known. This quantity can be inferred from pressure-wind correlations of the Viking Lander data, estimated from cloud images, or directly measured from spatially resolved atmospheric temperature data. We are investigating the spatially resolved data from the 15-micrometer channel of the Viking IRTM instrument. This yielded atmospheric temperatures at about the 0.6-mbar level. Certain periods of the mission gave adequate coverage of the planet in latitude, longitude, and time of day for an analysis of the weather correlation. We are currently studying a subset of the data spanning 348 degrees < Ls < 20 degrees during the first martian year that Viking observed, just after the second large dust storm of 1977 cleared. However, other subsets will also be studied. In order to measure the weather correlation length scale, the systematic temperature variations must first be removed from the dataset. This "mean state" of the atmosphere is interesting in itself as it contains information about seasonal effects, dust storm effects, and thermal tides. It is primarily a function of season, latitude, and time of day, but effects such as look angle of the instrument and local topographic relief need be considered. This "mean state" can be determined by fitting the data to spherical harmonics in latitude and time of day at the mean altitude of the observations as well as fitting the vertical derivatives of these same functions. Subtracting these systematic temperature variations of the atmosphere leaves residuals that represent the transient weather deviations. We then correlate each residual with other nearly contemporaneous residuals as functions of the separation distance between the points. These correlations may also be functions of season, latitude, dustiness, and direction of the separation vector. The widths of these correlation functions define the weather correlation length scales of Mars. Murphy J. R. Pollack J. B. Bridger A. F. C. Large Scale Martian Atmospheric Response to Developing Global Dust Storms: Thermal Tides The thermal structure of the martian atmosphere is intimately coupled with the quantity and distribution of dust suspended within it. During the developing stage of a planetary-scale dust storm, the amount and spatial distribution of dust rapidly increase, thereby imparting a significant temporal variability to the atmospheric thermal forcing, which in turn induces an intense atmospheric dynamical response. This response has been observed to take the form of several phenomena, including a dramatic increase in the thermal tidal (pressure, wind) amplitudes as seen by the Viking landers, the generation of a winter polar warming and the presence of large amplitude waves at high northern latitudes, and modulation of baroclinic wave activity observed at the northern middle latitude Viking Lander 2 site. We will present results from three-dimensional numerical simulations of global dust storms, with emphasis upon the time variation of atmospheric tidal components. Comparisons will be made with available Viking data to show that the model response is similar to that observed. Heating fields that drive the tides will be isolated to compare the response to that expected from classical tidal theory. The influence of Mars' large amplitude topography will be explored via topography-less simulations in an attempt to isolate non-Sun synchronous tidal components that have been deduced from the Viking observations. Magalhaes J. A. Young R. E. Hydrostatic Mountain Waves in the Martian Atmosphere: An Analytical Study We have studied analytically two-dimensional finite amplitude hydrostatic mountain waves generated by localized topographic obstacles under martian conditions. A fully nonlinear lower boundary oondition is used, and a radiation upper boundary condition is assumed. The upstream atmospheric structure is idealized as two layers of constant Brunt-Vaisala frequency. For Mars, the lower layer represents the diurnal thermal boundary layer while the upper layer represents the remainder of the atmosphere. The solution of the model is completely determined by the shape of the obstacle, the nondimensionalized height of the obstacle (hN(sub)L/U(sub)O), and N(sub)U/N(sub)L, where h is the obstacle height, U(sub)O is the upstream surface wind, and N(sub)U and N(sub)L are the upstream Brunt frequencies in the upper and lower layers. We find that during the martian night when N(sub)L is large and much greater than N(sub)U, the vertically propagating waves can be partially reflected at the interface between the two layers and produce resonances that lead to important increases in downstream winds even behind quite small obstacle. In addition as the ratio N(sub)U/N(sub)L changes, the transmission of vertically propagating mountain waves is strongly affected. This "filtering" behavior of the lower portion of the martian atmosphere could be an important factor in regulating the dynamical coupling between the atmosphere and surface. This work has been supported by the Planetary Atmospheres Program of NASA. Houben H. Haberle R. M. Young R. E. Passive Tracer Transport in the Martian Atmosphere Atmospheric transport plays a vital role in the martian C02 and dust cycles, and an uncertain role in the water vapor cycle. In view of the extreme interannual variability of major atmospheric dust transport events (i.e., planet-encircling and global dust storms), it is probable that atmospheric water vapor transport also exhibits a high degree of interannual variability. We have therefore developed a Martian Climate Model (MCM) capable of fast, multiyear simulations of atmospheric dynamics. The MCM is coupled to the Aerosol Transport Model of Toon et al., allowing us to characterize the tracer transport properties of the martian atmosphere over long periods of time. We have completed a ten year simulation of martian atmospheric dynamics and have studied two components of the general circulation, which are important to tracer transport: the mean meridional circulation and traveling planetary waves. The mean meridional circulation varies significantly over the course of a artian year, from an equatorially symmetric Hadley circulation at the equinoxes to a single direct cell at the solstices. The intensity of this overturning circulation is a sensitive function of atmospheric dust load and thus will be highly variable from year to year. On the other hand, we find virtually no interannual variation in the mean meridional circulation when the dust load is fixed. This is not the case for the traveling waves. As expected, both hemispheres exhibit minima in traveling wave activity during the summer. The southern hemisphere also shows a minimum in wave activity during midwinter. Both hemispheres have strong wave activity in spring and fall. In fact, southern spring is the period of most intense wave activity in both hemispheres. This is also the season of greatest interannual variability in our simulations. That this is the peak dust storm season suggests that planetary waves may play a role in generating global storms. Martin T. Z. Lachin A. IR Opacity Maps of Martian Local Dust Storms Recent work to map the global 9 micrometer dust opacity of the Mars atmosphere using Viking IR Thermal Mapper data (Martin and Richardson, 1993; Martin and Lachin, 1992) has led to a new effort to study local dust events that are captured in individual IRTM mapping sequences. These observations provide true "snapshots" of dust behavior, unlike the time-averaged nature of the global map set. Opacity is derived using the approach described by Martin (1986). Samples in the 6-8 micrometer band are paired with nearest-neighbor 8-10 micrometer samples for this purpose, a necessity because the sampling in these bands was not cospatial (Kieffer et al., 1976). A variety of interesting dust events have been studied, including those described by Peterfreund and Kieffer (1979) and Peterfreund (1985). The total dust mass raised during a storm may be calculated from the area-integrated opacity, for the L(sub)s 226 degrees storm in 1977, a value of 1.6 x 10^8 kg is derived. References: [1] Martin T. Z. (1986) Thermal infrared opacity of the Mars atmosphere. Icarus 66, 2-21. [2] Martin T. Z. and Lachin A. (1992) Topographic control of martian dust opacity. Bull. AAS, 24, 1009. [3] Martin T. Z. and Richardson M. I. (1993) JGR, in press. [4] Peterfreund A. R. and Kieffer H. H. (1979) JGR, 84, 2853-2863. [5] Peterfreund A. R. (1985) Contemporary Aeolian Processes on Mars: Local Dust Storms. Ph.D. Thesis, Arizona State Univ. Ockert-Bell M. E. Pollack J. B. Arvidson R. Shepard M. Improved Radiative Properties of Martian Atmospheric Dust Particles One of the key radiative agents in the atmosphere of Mars are the suspended dust particles. We are carrying out a new analysis of Viking 1 and 2 images of the martian atmosphere to better evaluate the radiative properties of the atmospheric dust particles. The properties of interest are the size distribution information, the optical indices and other radiative properties, such as the single-scattering albedo and extinction coefficient. These properties will provide a good definition of the influence that the atmospheric dust has on heating of the atmosphere. Our analysis will represent a significant improvement on our past analysis [1,2] by using a better representation of scattering by nonspherical particles and by using data closer to the Sun. The improvements allow us to better define the diffraction peak and, hence, the size distribution of the particles. The radiative transfer method that we have used is based on the doubling method [3]. A semi-empirical method is used to model the scattering by nonspherical particles [4]. Hapke [5,6] theory is used to model the photometric properties of the surface. The computer code has been revised and tested extensively. The data used for this study includes data from Vikings 1 and 2. Images included those from the blue, green, red, ir1, and survey filters. The images were calibrated and a connection for vignetting was added. We used an iterative method to fit the parameters of interest to the observed data: small phase angles were used to find the size distribution information, phase angles of about 50 degrees were used to determine the imaginary index, and the data at larger phase angles determined the shape of the particles. We calculate the intensity expected in a given range due to variation of one parameter and do a chi-squared fit to the variance to find the best fit of the parameter in question. The resu1ting best fit is used as a set parameter while another is varied, etc. The results of this analysis using two size distributions (log-normal and the modified-gamma distribution [2]) will be presented. The radiative properties of the suspended dust particles will be given for each filter of the Viking cameras. References: [1] Pollack J. B. et al. (1977) JGR, 82, 4479-4496. [2] Pollack J. B. et al. (1979) JGR, 84, 2929-2945. [3] Hansen J. E. (1969) Astrophys. J., 155, 565, 1969. [4] Pollack J. B. and Cuzzi J. N., J. Atmos. Sci., 37, 868- 881. [5] Hapke B. (1981) JGR, 86, 3039-3054. [6] Hapke B. (1986) Icarus, 67, 264-280. Monday, October 18, 1993 LATE POSTERS Giuliatti-Winter S. Murray C. D. Observational Constraints on Dynamical Models of Saturn's F Ring Voyager 1 and 2 images of Saturn's F ring have revealed radial and azimuthal structure. Previous attempts at understanding these unusual features have concentrated on explanations for the azimuthal structure in the ring. For example, the perturbing effects of Prometheus, Pandora, and putative small satellites have been invoked to explain the wave-like or 'braided' appearance of the ring at some longitudes. We have analyzed Voyager 2 images of the F ring in an attempt to describe the ring's radial structure. Several images show isolated, non-intersecting strands of material with no evidence of wave phenomena. These results allow us to place additional constraints on plausible dynamical models of the F ring, especially those that involve the presence of small embedded or nearby satellites. We review the problem of the radial confinement of the F ring in the light of these observations and propose a model of the F ring that is consistent with all the imaging and charged particle observations. Coustenis A. Lellouch E. Maillard J. P. Monitoring of Titan's Near-Infrared (1-2.5 Micrometer) Spectrum We have observed Titan with the CFH Fourier Transform Spectrometer in Mauna Kea (Hawaii) on the 1, 2, 5, 6, 9, and 10 August 1993, UT. The orbital longitude of Titan on these dates was approximately 165, 188, 253, 275, 345, and 10 degrees respectively. Our data therefore include Titans inferior conjunction, western elongation, and superior conjunction. We have covered the four infrared windows between 1 and 2 microns (1.08, 1.28, 1.6, and 2.0 microns). More precisely, using filter K, H, and J we covered the spectral ranges: 390-5100, 5600-6650, and 7400-9650 cm^-1 with spectral resolutions of 5.2, 6.6, and 8.7 cm^-1 respectively. The signal-to-noise ratio is 50-120 for the 2-micron window and 100-150 for the others. The SAO 164861 star was used for calibration. A preliminary reduction of the data shows no significant change in the shape of the spectrum as a function of orbital position. Wednesday, October 20, 1993 IO I: SURFACE TO ATMOSPHERE 8:30 - 10:00 AM South Ballroom Chair(s): N. Schneider M. McGrath Matson D. L.* Veeder G. J. Blaney D. L. Johnson T. V. Goguen J. D. Brown R. H. Modeling Io's Thermal Emission Io's strange thermal emission signature at infrared wavelengths was recognized more than 20 years ago. The discovery of volcanic hot spots by Voyager (Hanel et al., 1979) explained many, but not all of the puzzling aspects of Io's spectrum. Attempts to model multiwavelength eclipse data (Sinton and Kaminsky, 1988) and radiometry (McEwen et al., 1988) with "standard" thermal models (applicable to asteroids and airless satellites) have failed. For Io's bolometric Bond albedo and heliocentric distance, the observed emission at 20 micrometers is too low for these models by about a factor of 2, even before including the additional emission from the hot spots. We have developed a new class of model for the background emission that results from the absorption of sunlight on Io's surface. The new background model includes (1) significant reradiation of absorbed sunlight on the night side due to a heat storage unit, and (2) the thermal pedestal effect due to the blue-shifted, reradiation of the insolation absorbed by the thermal anomalies themselves. This approach accounts for Io's low 20-micrometer emittance and in-and-out-of-eclipse radiometry at 4.8, 8.7, and 20 micrometers. It also requires large, low- temperature anomalies that radiate the major portion of Io's observed heat flow, in agreement with new analyses of Voyager IRIS data (McEwen et al., 1992). More sophisticated models of Io's surface state and thermal insulating properties are needed to explain the rate of eclipse cooling in the context of the pedestal effect. This work was done at JPL/Caltech, under contract to NASA. Dumas C.* Spencer J. R. Woodney L. Angara A. Clark B. Fourier Deconvolution of Io Infrared Images: A New Step in the Survey of Io's Volcanic Activity We have applied Fourier deconvolution by Wiener filtering to a large set of infrared images of Io obtained with the 1-5 micrometer infrared camera ProtoCAM at the IRTF. This allows us to approach the diffraction limit of the instrument in the L, L,' and M bands. On 30 occasions since 1989, we have observed the Jupiter-facing hemisphere of Io during eclipses by Jupiter, which occur every 1.8 days. In eclipse, only the volcanic thermal emission is visible in the 1.5-4.8 micrometer region. We take sequences of about 100 short-exposure images, select those with the highest peak flux, and shift-and-add them to increase the S/N. The same process is applied to a nearby bright star to get the point spread function (PSF) of the instrument and atmosphere. Using this PSF the image is deconvolved by a Wiener filter, resulting in a dramatic gain in spatial resolution. To locate individual hot spots, we fit the deconvolved images with a synthetic image containing up to six point sources with adjustable location and flux, using the Simplex algorithm. Similar techniques were applied to a smaller data set of sunlit images of Io by McLeod et al. (1991, A. J. 102, 1485). Reproductibility is excellent: the location we obtained for the persistent hot spot "Kanehekili" was 10.4 S,37 W on 1993/05/02 and 8.6 S, 38.5 W on 1993/05/09. Loki, the hot spot "Hiialul" on the leading hemisphere, and two other hotspots near Creidne Patera and Sengen Patera can also be seen and located in the eclipse images. Sunlit images at 4.8 micrometers taken on 1993/05/02 show Loki (then quiescent but still prominent at this wavelength) against the sunlit disk of Io: fits to these images give Loki's location as 12 N, 310 W, and Io's diameter as 1797 Icm, within 1.3% of the correct value. Lightcurves of Io obtained during its disappearance or reappearance behind Jupiter's limb provide additional constraints on spot positions and fluxes. These techniques improve our ability to monitor the location and time variability of each individual hotspot, and to prepare for the nighttime observations of the Jupiter facing hemisphere of Io by Galileo in December 1995, immediately after the Io flyby. Blaney D.* Hanner M. Russell R. Lynch D. Hackwell J. Infrared Spectrophotometry of Io Between 3-13 Micrometers Infrared observations of Io were collected at the NASA Infrared Telescope Facility, Mauna Kea Hawaii on February 7 and 8, 1993 using the Aerospace Corp. liquid-He-cooled spectrograph. The instrument used two 58-element Blocked Impurity Band (BIB) linear arrays to cover the 3-13 micrometer wavelength region with a resolving power of 30-120. All spectral elements were observed simultaneously. The central meridian of the disk of Io, ranged between 225-260 West longitude on February 7 (Trailing Hemisphere) and between 70-105 West longitude on February 8 (Leading Hemisphere). The broad spectral range covered contains reflected solar radiation, ernission from thermal anomalies (i.e., volcanic regions), and background emission from solar heating of the surface. Thermal emission from the leading hemisphere varied less than 2% on February 8. Emission from the trailing hemisphere on February 7, increased throughout the night by roughly 30%. The increase in ernission with increasing west longitude is consistent with that expected by the volcanic region Loki (309 West longitude, 10 North latitude) approaching the center of the observed disk. The emission levels measured, however, indicate that Loki was in a period of relatively low activity. The absorption feature due to SO2 frost is present at 4.08 micrometers. The band depth, measured using the values at 3.47 micrometers for the continuum and at 4.08 micrometers for the band minima, was roughly 50% on both the leading and trailing hemisphere. The band depth varied by less than 10% in the regions measured. Part of this work was done at JPL/ CALTECH, under contract to NASA. *Visiting Scientist at the NASA Infrared Telescope Facility. Goguen J. D.* Veeder G. J. Matson D. L. Johnson T. V. Toomey D. A ~10 km Resolution Image of Loki at 3.8 Micrometers A major eruption at Loki Patera began in December 1990 and continued into March 1991. Simultaneous 3.8 and 10 micrometers radiometry at the IRTF of a series of occultations of Io by Europa during this period is used to develop a detailed quantitative picture of this eruption and its evolution. From the times of disappearance and reappearance of hot spot thermal emission behind Europa's limb, the latitude and longitude of the eruption site can be determined to better than 100 km. The site of the 1991 eruption was the SW corner of Loki Patera. In the Voyager images, the SW corner contains a ~20 x 50 km region that has the lowest albedo in Loki Patera, suggesting that this site may undergo recurrent activity. Specifically, there is no significant signature in the occultation data of thermal emission from the dark linear "fissure" 250 km NE of Loki Patera that was the site of 2 plumes detected in the Voyager images. The 3.8 micrometer occultation radiometry has sufficient signal-to-noise to reveal the one-dimensional structure of the thermal source region at a scale of ~10 km. This hottest component of the emission emanates from 2 narrow (10- 30 km wide) N-S oriented "fissures" 50-100 km in length and separated by 100 km. Pairs of one-dimensional scans with different orientations of Europa's limb corresponding to disappearance and reappearance were obtained on 5 different dates. We will attempt to reconstruct a two-dimensional image from these 10 scans, assuming constancy of the 3.8 micrometer source region. This work is part of research funded by NASA under contract to JPL/Caltech. Clarke J. T.* Ajello J. Ballester G. Luhmann J. McGrath M. Schneider N. HST UV Spectra of Io: Spatially Resolved Regions and New Eclipse Spectra New UV spectra of Io have been obtained with the FOS and GHRS instruments on the Hubble Space Telescope over February-August 1993, in addition to previously reported observations from January 1992. At the time of writing, we have obtained GHRS high-resolution spectra near times of Io eclipse entry that have resolved the emission multiplets of O and S emission, including the optically thick O triplet at 1302-1306 Angstroms and optically thin S doublet at 1900-1915 Angstroms. These spectra reveal that the 1304-Angstrom feature is dominated by oxygen emission, which decreased rapidly in intensity as Io entered eclipse. Three Io orbits later, the S emission decreased very little as Io entered eclipse. More recent observations, obtained but not yet reduced, are a series of 1-minute FOS spectra from 2270-3300 Angstroms as Io emerged from eclipse, and a N/S spatial scan across the Io disk near west elongation with a 0.3-arcsec-diameter aperture in the same wavelength range. New observations scheduled for July/August 1993, but not yet executed, include FOS spectral time series of the whole disk with Io emerging from eclipse over 1150-1600 Angstroms and 1600 2300 Angstroms, and 1600-3300 FOS spectra of selected regions on the Io disk near west elongation with the 0.3-arcsec aperture. The new spectra will be presented, along with an initial analysis of the data with relevance to Io's surface reflectivity, evidence for atmospheric absorption, and atmospheric emissions of atomic S and O. Burnett D. S.* Wiens R. C. Calaway W. F. Pellin M. J. Ion-Sputtering Products of Sodium Sulfate, and Implications for the Surface and Atmosphere of Io It was recently shown (Schneider et al.) that a significant fraction of the neutral Na encircling Jupiter is created downstream from Io, implicating Na- bearing molecules escaping from Io as the source. If Io's atmosphere proves to be thin, surface sputtering by co-rotating ions is likely to be the mechanism for ejecting Na and Na-bearing molecules from Io. Plausible surface constituents are Na2SO4 and Na-sulfides. The relative abundances of Na-oxide and Na-sulfide molecules in the Na cloud should be dependent on the relative importance of sulfates and sulfides on the surface, which in turn could provide a major constraint on Io geologic processes. We have therefore undertaken a study of the neutral sputtering products of Na2SO4 using the SARISA IV laser postionization secondary neutral mass spectrometer. Neutral sputter products were non-resonantly ionized with 6.4 eV photons and accelerated in a time-of-flight mass spectrometer. We have identified Na, Na2, NaO, NaS, Na20, NaSO, Na2S, Na202, as well as heavier species, with the most intense molecular signal at Na20. If the measured Na-oxide/atomic Na ratio is taken as a lower limit to the actual values, the fraction of Na atoms sputtered as constituents of oxide molecules is at least 20%. Velocity and energy distributions were measured for each major species. The Na20 velocity peak is ~1.4 km/s, which is a function of the binding energy, and independent of the primary ion energy above the sputtering threshold. The distribution has a significant tail (15% of the total) above the 2.6 km/s escape velocity of Io. The atomic Na distribution is broader than would be expected, and is possibly an experimental artifact resulting from photodissociation of some fraction of the Na-bearing molecules. Our results show that, if sodium sulfate is a major surface constituent, Na-oxides will be among those escaping Io. Meade P. E.* Jakosky B. M. Evolution of Io's Surface Due to Thermally Driven SO2 Diffusion The extremely high porosity of Io's surface has led to speculation that diffusion of atmospheric sulfur dioxide (SO2) into the surface could be an important process on the satellite. Matson and Nash (1983) proposed that atmospheric SO2 could be cold-trapped in the surface layer and that this process would result in layering of Io's surface. Surface layering is consistent with thermal observations of Io such as those of Morrison and Cruikshank (1973), Hansen (1973), and Sinton and Kaminski (1988). In Meade and Jakosky (1991) we developed a thermal diffusion model of Io's surface and concluded that the thermal gradients in the upper few centimeters of the surface would lead to downward diffusion of any SO2 present in the surface layer. We suggested that these thermally driven SO2 diffusion fluxes could cause the near-surface to become layered. Our present work extends our previous model to allow the thermophysical properties of the surface to evolve as the SO2 diffuses. We also relax the boundary condition at the surface to allow diffusion between the surface and atmosphere. We are especially interested in the changes with time of the porosity, thermal inertia, and mean particle size of the surface, and the effects of these changes on the diurnal temperature profile. Since the diffusion mechanism is insolation dependent, we anticipate important latitudinal variations. We also attempt to evaluate the efficiency at which the surface can act as a cold-trap for atmospheric SO2. Hapke B.* Io: Implications of the Observed Thin Atmosphere for the Nature of the Surface Recent measurements show that the SO2 atmosphere of Io is thin (surface pressure ~.01-40 nb). These pressures are much smaller than the vapor pressure of SO2 at the subsolar point, and imply that the atmosphere is probably temporary and local and that the abundance of volatiles may be more limited than in the generally accepted models. It is likely that the base of the exosphere is at the surface part of the time, especially at night. Attempts to explain the low SO2 pressure by invoking high-albedo regional cold traps or subsurface cold traps are in fundamental disagreement with the spectrophotometric properties of Io. The only model proposed to date that can adequately account for both the thin atmosphere and the spectral properties is the thin ephemeral layer model of Hapke (1993, Icarus, 79, 56), in which the spectrally active SO2 is contained in a thin (~1 micrometer) layer on the surface of the regolith. Whether or not the thin ephemeral layer model is correct is independent of whether the yellow material on Io is polysulfur oxide (PSo) (Hapke, op cit) or elemental sulfur. However, recent observations of Io with the Space Telescope show that minimal albedo changes have occurred since the Voyager flyby. Recent high temperatures observed for some hot spots imply silicate volcanism. Both observations are consistent with predictions of the PSO model. Ingersoll A. P.* LeBeau R. P. Jr. A Nightside Atmosphere of Io? Several models of the atmosphere of Io have indicated that there may exist a high velocity/supersonic flow of material away from the subsolar point towards the nightside (Ingersoll et al., 1985; Ingersoll, 1989; Moreno et al., 1990). Despite this large mass flux, the rapid condensation of sulfur dioxide to the surface in these cold regions probably would result in a negligible SO2 atmosphere on the darkside of Io. However, photodissociation of the sulfur dioxide into other chemical constituents (sulfur, oxygen, sulfur monoxide) on the dayside would result in a flux of non-condensing constituents to the nightside, which could yield a significant atmosphere. This study numerically models such a process in a sublimation-driven atmosphere, determining the chemical composition of the nightside atmosphere over a variety of subsolar and darkside temperatures. The results indicate that a non-S02 atmosphere on the nightside may reach pressures on the order of 10^-8 bar, significantly above that needed to explain the Pioneer 10 radio occultation data. Therefore, non-sulfur dioxide constituents of the atmosphere may provide an explanation for the excess pressure detected by Pioneer 10 at the Ionian terminator. Acknowledgments: This research was started at the Geophysical Fluid Dynamics Summer Study Program of the Woods Hole Oceanographic Institution and was further supported by the Planetary Atmospheres Program of NASA. POSTERS Howell R. R. Two-Dimensional Images of Loki from Mutual Event Occultations During the 1991 mutual events, numerous occultations of the thermal emission from the Loki hotspot were observed. Each occultation produces a disappearance and a reappearance trace that are one-dimensional projections of the hotspot's brightness distribution. The angle of each projection is determined by the angle of the occulting satellite's limb. In principal, if sufficient projections are observed then various tomographic techniques can be used to reconstruct an image. Such an image would be extremely valuable in deciphering the nature of the volcanism. However, numerous complications are present. For example, the hotspot emission can be treated as constant between two traces from a single occultation, but some changes are known to occur between occultations. The signal-to-noise ratio varies for different occultations. Furthermore, some were observed through thin clouds. Although the transparency is constant over the short time of the individual hotspot events, the overall satellite occultation lightcurve is distorted. The standard reconstruction techniques assume that only random noise is present, and we do not yet understand the results of systematic errors. I will present reconstructions obtained by a variety of methods to evaluate the reliability of such two-dimensional images. Nash D. On Io's 2.788 and 2.539 Micrometer Bands: Origin by SO2 Not H2O Io's recently reported band at 2.788 micrometers [1] is readily explained by solid S02 on Io's surface. Bregman et al. [l] attributed the band to H20 (trapped in S02 ice). But, as pointed out by Nash and Howell [2], lab S02 frost has an absorption band at 2.79 micrometers, the 2-nu(sub)1 + nu(sub)3 band of solid 32Sl6O2. Schmitt et al.[3] also argue that S02 explains the 2.788-micrometer band. The figure, which appears in the hard copy, shows reflectance spectra from my S02 frost experiments. Spectrum A is for cold substrate (81 K, sulfur) prior to depositing S02 frost on it: the broad absorption feature centered at 3.1 micrometers is from trace contaminant water; there is no narrow band feature in the vicinity of 2.79 micrometers from condensed water; the doublet near 3.4 micrometers is from trace hydrocarbon in the substrate. Spectrum B is for thin (~100 micrometers) S02 frost; it partly obscures the 3.4-micrometer substrate feature, but (due to too-little pathlength in S02 frost of this thickness) yields no absorption bands in the spectral region shown. Spectrum C is for thick (~2 mm) S02 frost; this has a distinct band at 2.789 (and one at 2.539 micrometers), with band depth of ~40? (at 4 cm^-1 resolution). This lab work shows that H20 frost does not have a band at 2.788 micrometers, but S02 frost does. Io's 2.788- micrometer band is consistent with condensed (solid) S02 on Io, not H20. Io observers should find another S02 band (2-nu(sub)3 + 2-nu(sub)1) at 2.539 micrometers. References: [1] Bregman, J., et al. (1993) B.A.A.S., 25, 851-852. [2] Nash, D. and R. Howell (1989) Science, 244, 454-457. [3] Schmitt, B. et al. (1993) Paper presented at Io--An Internat. Conf., June 22-25, 1993, San Juan Capistrano, California. Veeder G. J. Blaney D. L. Johnson T. V. Matson D. L. Goguen J. D. A Radiometric View of Io's Volcanology and Geophysics Infrared radiometry of Io as a function of longitude and time at 4.8, 8.7, and 20 micrometers characterizes the temperatures, the spatial distribution, and the temporal evolution of thermal anomalies. We present data collected with the NASA IRTF from 1983 to 1993. Roughly equal amounts of power are emitted from both the leading and trailing hemisphere in spite of the location on the trailing hemisphere of Loki (the single largest volcanic center on Io). Although thermal emission at 4.8 micrometers is highly variable, the total power output has been stable over the last decade. The average heat flow is 2.5 (s.d. = 0.29) W/m^2 or 1.05^14 (s.d. = 0.12) W with the maximum observed change between apparitions 20%. This is significantly higher than permitted for steady state by current models of the Jupiter-Io tidal interaction and Io's orbital evolution. The increase in heat flow over our previous estimates is due to the development of an improved background model for the surface of Io. This model allows for the recognition of lower temperature, cooler thermal anomalies than had been detected earlier. We have observed two outbursts. The first in 1986, suggested a temperature of 900 K. The second on January 9, 1990 is best modeled by a source at the location of Loki with a temperature of 1200 K and radius of 5.6 km at the beginning of the observations and a temperature of 700 K and radius of 13 km 3 hours later. These temperatures are too high to be explained by molten sulfur, but are quite reasonable for silicate eruptions and imply a high rate of resurfacing by lava flows. This work was done at JPL/ CALTECH, under contract to NASA. *Visiting Scientist at the NASA Infrared Telescope Facility. Sartoretti P. McGrath M. A. Disk-resolved Imaging of Io with the Hubble Space Telescope Ultraviolet and visible disk-resolved images of Io have been obtained in May, June, and July 1993 with the Faint Object Camera (FOC) of the Hubble Space Telescope at a resolution of ~230 km. They are analyzed together with early, near-UV FOC data obtained in March 1992 and with Voyager data. Images at 3500 Angstroms and at 4100 Angstroms show the reflecting surface and are compared with Voyager results for determining long term variations in the distribution of the surface materials, especially SO2 frost, due to the high volcanic resurfacing. Images at 2300 Angstroms and 2600 Angstroms provide a measure of the combined effects of intrinsic surface albedo and absorption by atmospheric SO2 gas allowing tighter constraints to be placed on both the surface composition and the atmospheric gas content. Ballester G. E. McGrath M. A. Strobel D. F. Zhu X. Feldman P. D. Moos H. W. The SO2 Atmosphere on Io as Observed with HST Ultraviolet (1975-2325 Angstrom) spectra of Io's trailing hemisphere, obtained with the Faint Object Spectrograph on the Hubble Space Telescope on March 1992, have provided a positive detection of SO2 in Io's atmosphere. The derived geometric albedo has been modeled with atmospheric SO2 absorption including optical thickness effects and gas temperature dependence, with the surface reflectance fitted empirically. Below 2150 Angstroms the albedo is dominated by SO2 gas absorption band signatures consistent with a global SO2 atmosphere of 48 x 10^15 cm^-2 column density for an average gas temperature of 110-500 K Similar fits are found for the global atmosphere and for spatially confined atmospheres down to a limiting case of -11% areal coverage and N = 7 x 10^l6 cm^-2. Although collisionally thick, these densities are 25- 50 times lower than the previous upper limit. These moderate values could reflect temporal variations, since a denser SO2 atmosphere has been detected with independent millimeter-wave observations, and coincidentally, Io's volcanic activity was very low at the time of the HST observations, as found with independent IR studies. A simple model of a volcanic atmosphere was found to be too spatially confined to generate the observed spectral contrast below 2150 Angstroms. But combining this volcanic component with a more extended, moderate-density atmospheric component (which could also be of volcanic origin) fits the data as well. Sublimation of SO2 frost as the atmospheric source implies a 107-115 K average surface temperature, low compared to previously postulated surface temperatures. Better agreement is obtained with atmospheric temperatures around 200-400 K, but better data are needed to confirm this result. Wednesday, October 19, 1993 URANUS AND NEPTUNE 8:30 - 10:00 AM North Ballroom Chair(s): P. N. Romani J. I. Moses Rages K. A.* Pollack J. B. Modeling Uranus' Temporal Variability Analysis of spatially resolved Voyager images of Uranus has shown that the subtle latitudinal banding seen on the planet is associated with substantial variations in the optical depth and single-scattering albedo of the methane cloud layer above 1.3 bar [1]. Due to Uranus extreme axial tilt, this latitudinal variation in the atmospheric scattering properties will account for some portion of the temporal variability seen by Lockwood et al. [2]. Models of Uranus scattering atmosphere are being derived for a number of additional latitudes, using the spatially resolved Voyager data. These models will then be combined in properly weighted sums over latitude to give Uranus geometric albedo and bolometric albedo as functions of orbital phase. The geometric albedos will be compared to the observations of Lockwood to determine if some additional temporal variation in the planetary scattering properties is required to reproduce the observed changes in the geometric albedo. [1] Rages et al. (1991) Icarus, 89, 359-367. [2] Lockwood et al. (1983) Astrophys. J., 266, 402-414. Holme R.* Ingersoll A. P. Baroclinic Instability in the Interiors of the Giant Planets: A Mechanism for Depleting the Internal Energy of Uranus We propose a quasi-geostrophic, non-hydrostatic, baroclinic model for heat transport within the interior of a stably stratified Jovian planet. Density varies from zero at the surface to a realistic value at the center. Motion is assumed to take place in thin cylindrical annuli. We find instability when the poles are hotter than the equator--the Uranus case--but not when the equator is hotter than the poles--the Neptune case. Since the interior is stably stratified, for the Uranus case the isentropes lie deeper at high latitudes than at low latitudes. If the slope of the isentropes is great enough, the convective heat transport is both upward and equatorward. The surface temperature is higher for a stably stratified interior--the Uranus case--than for one that is adiabatic--the Neptune case. The higher surface temperature leads to a higher rate of infrared emission and the planet cools faster. We propose that Uranus lost its internal heat sooner than Neptune because baroclinic motions were able to extract its internal heat while the surface was still warm. According to our model, such motions only occur for a planet with a large obliquity, for which the poles are hotter than the equator. Acknowledgments: This research was started at the Geophysical Fluid Dynamics Summer Study Program of the Woods Hole Oceanographic Institution and was further supported by the Planetary Atmospheres Program of NASA. Marten A.* Gautier D. Matthews H. E. Owen T. Sanders D. B. Ladd E. F. Deane J. R. New CO Observations of Neptune in the Millimeter and Sub-Millimeter Range Emission lines of carbon monoxide have been observed previously at 230 and 345 GHz with the JCMT and CSO radiotelescopes located in Hawaii (Marten et al, ApJ, 406, 285, 1993). The analysis suggests that CO originates from the interior of the planet, although its derived abundance (about 10^-6) is greater by several orders of magnitude than predicted by current thermochemical models. For comparison Uranus has also been observed in May 1992 and we estimated a CO upper limit 100 times lower than for Neptune. We report here on two new detections of CO lines on Neptune: the J=1-0 line at 115 GHz has been observed in absorption in the troposphere with the IRAM Interferometer at Plateau de Bure, France (Guilloteau et al., A&A, 1993, in press). The J=4-3 line at 461 GHz has been detected in emission in June 1993 with the JCMT and CSO. We present a summary of all observations and discuss uncertainties on the retrieval of the vertical distribution of CO. Our previous conclusion that carbon monoxide is uniformly mixed in the whole atmosphere of Neptune is reinforced by these recent detections. New constraints on the vertical temperature profile of the stratosphere are obtained from the comparison of the three CO emission lines we have now observed by single-dish heterodyne techniques. Courtin R.* Gautier D. Marten A. Strobel D. Edwards S. Detectability of CO, N2 and NH3 on Neptune from UV Spectroscopic Measurements The microwave detection of CO and HCN in Neptune's stratosphere (Marten et al., 1992) has stimulated studies of the disequilibrium chemistry in the atmosphere of that planet, particularly centered on the question of the origin of the nitrogen atoms. One model invokes the upwelling of N2 (and CO) from the deep troposphere, its dissociation at high altitudes by magnetospheric electron impact, and subsequent production of HCN via reactions between N atoms and hydrocarbon radicals. A potential test of this model is to search for N2 UV absorption in the reflection spectrum of Neptune. The Vegard-Kaplan band system, consisting of very weak narrow transitions shortward of 175 nm, offers such a possibility. We have modeled these transitions in Neptune's reflection spectrum, together with the Cameron 0-0 band of CO around 206 nm. The background spectrum is determined by Rayleigh + haze scattering and absorption (Raman scattering is neglected in a first approximation), and by C2H2 photoabsorption. Nitrogen chemistry on Neptune could also lead to substantial production of NH3 at high altitudes. The influence of NH3 on the reflection spectrum is therefore also investigated. Our model calculations are first compared with spectroscopic measurements from the IUE (Wagener et al., 1986; Caldwell et al., 1988). We will also attempt to compare them with soon-to-be-released FOS measurements by HST, in order to determine mole fractions or upper limits for CO, N2, and NH3. Hofstadter M. D.* Microwave Imaging of Neptune's Troposphere Neptune was observed with the Very Large Array in October of 1991, revealing horizontal and vertical variations in tropospheric properties. The planet was mapped at wavelengths of 3.6 and 6.2 cm (disk-averaged brightness temperatures are 173 K +- 10 and 191 K +- 15, respectively). With the assumptions that tropospheric temperature variations along surfaces of constant pressure are less than 10 K, and that the tropospheric lapse rate is approximately adiabatic, four conclusions can be drawn from these data: On average, the upper 20 bars of Neptune's troposphere is ~5 times more absorbing than the troposphere of Uranus. The south polar region of Neptune, and most likely northern high latitudes as well, are brighter than equatorial latitudes. The region from -65 degrees +- 15 to +25 degrees +- 25 appears dim. The absorber abundance varies with latitude by a factor of 2 or 3 at pressures less than ~15 bar, and varies by an order of magnitude or more at higher pressures. These variations are comparable to ones found for Uranus based on similar observations. Horizontal variations in atmospheric properties persist at least down to the 30-bar level. Analysis of these data is continuing, and at the time of the conference it is hoped that differences between Neptune's and Uranus' tropospheric lapse rate and/or cloud structure can be discussed. This work was done while the author held a National Research Council-Jet Propulsion Laboratory Research Associateship. Hammel H. B.* LeBeau R. P. Jr. Lawson S. L. The Relative Intensity of Neptune's 'Bright Companion' Feature in 1989 as a Function of Wavelength From 1986 to 1992, a single feature in Neptune's southern hemisphere dominated the planet's methane-band reflectivity. The 1989 Voyager 2 encounter provided high resolution imaging of Neptune; by comparing simultaneous spacecraft and groundbased imaging the dominant feature was identified as the "Bright Companion" south of the Great Dark Spot (Smith et al. (1989) Science, 246, 1422). However, spacecraft imaging was limited to wavelengths shorter than 60200 Angstroms; the PPS produced a single reconstructed image at 7270 Angstroms (Lane et al. (1989) Science, 246, 1450). We present a preliminary analysis of visible and near-IR images of Neptune obtained with the U. H. 2.24-m (Mauna Kea) during the week up to and including Voyager's closest approach on 24 August 1989. Our filters were centered on three methane bands (6190, 7270, and 8900 Angstroms) and three continuum regions (6340, 7490, and 8260 Angstroms). We detected the Bright Companion at all wavelengths in 1989, including continuum regions (see table); maximum contrast was at the methane band wavelengths. Photometry from images taken in 1986 with the same filters had not shown evidence of the dominant feature shortward of 8900 Angstroms within the precision of the photometry (Hammel et al. (1989) Icarus, 79, 14). The detections in 1989 are consistent with the report that Neptune experienced an atmospheric disturbance between 1986 and 1989, which significantly enhanced its rotational variability at shorter wavelengths (Hammel et al. (1992) Icarus 99, 363). These data will be used for constraining models of the vertical aerosol distribution in this feature, and for examining how the feature varies with time. Sromovsky L. A.* Fry P. M. Limaye S. S. Hubble Space Telescope Images of Cloud Features on Neptune During October and November 1991, the Wide Field/Planetary Camera (WF/PC) recorded 52 images of Neptune at visible and methane-band wavelengths. These images were made in the PC mode (0.043 arc seconds/pixel), for which Neptune's diameter spanned 53 pixels. Although the images have poor signal to noise ratios exacerbated by spherical aberration in the HST primary, they still provide evidence of banded cloud structure and isolated cloud features. From a sequence of five 8890 Angstrom images spanning 51 h we identified a feature near 49 degrees S with a rotation period (about Neptune's spin axis) of 16.41 +- 0.01 h. A sequence of five images taken 25 days later and spanning 6.43 h yielded a rotation period of 16.2 +- 0.3 h for a feature at nearly the same latitude. Within measurement errors these periods agree with each other and match the Voyager-based zonal wind profile of Limaye and Sromovsky (IGR 96, 18941-18960, 1991). However, the longitude change over the 25-day interval is inconsistent with a single feature having a constant rotational period. A plausible time-varying model that fits the observations contains oscillations that are very similar to those of Neptune's second Dark Spot (DS2) (Sromovsky, Science 254, 684-686, 1991) and suggests that the feature seen in HST images is the DS2. The HST images also contain evidence for the continued presence of South Polar Features (SPF) at latitudes of 60-70 degrees S, with longitudes 180 degrees out of phase with the proposed DS2 feature, a configuration also identified in the Voyager 2 data (Sromovsky et al., Icarus, in press, 1993). We could not establish rotation periods for features seen near 30-45 degrees N, and near 40 degrees S. The latter might be the companion of the Great Dark Spot, but, if so, is far less prominent than would be expected from simultaneous Voyager 2 and groundbased observations (Smith et al., Science 246,1422-1454, 1989). This research was partially supported by a grant from the Neptune Data Analysis Program. Based on observations with the NASA/ESA Hubble Space Telescope, obtained at the Space Telescope Science Institute, operated by AURA under NASA contract NAS5-26555. Moses J. I.* Rages K. Pollack J. B. Voyager Imaging Analysis of Neptune's Stratospheric Haze We have analyzed Voyager imaging data to determine the properties of the stratospheric hazes on Neptune. The high spatial resolution (~1 km/pixel or ~5 km/pixel for the narrow-angle and wide-angle camera images respectively) of the high-phase-angle images enables us to invert profiles of specific intensity across Neptune's limb to retrieve information about the atmospheric extinction as a function of altitude and the scattering phase function at a reference altitude. We then use a spherical radiative transfer algorithm combined with a detailed cloud microphysical model to simulate Neptune's stratosphere. The theoretical models help us to separate the contributions from molecular Rayleigh and aerosol scattering, and help determine the variation of the aerosol size, concentration, and scattering properties with altitude. Further comparisons between models and data allow us to constrain the location and composition of the hazes, the concentration and downward flux of certain condensable hydrocarbon gases, the eddy diffusion coefficient in the lower stratosphere, and the thermal profile in the Rayleigh-scattering- dominated portion of Neptune's stratosphere. We will compare our results with the results from the Voyager PPS experiment and discuss differences between Neptune and Uranus. Wang Y.* Yelle R. V. Modeling the Thermal Structure of Neptune's Stratosphere We have constructed the thermal structure of Neptune's stratosphere and lower thermosphere with pressures between 10^-3 microbar and 100 mbar using a radiative-conductive model that includes solar UV and EUV heating, non-LTE cooling by hydrocarbon fundamental bands, cooling by H2 collisional-induced opacities, and heating by the CH4 near and far infrared bands. The model uses newly developed algorithms for calculating the solar IR heating through CH4 bands. Using the methane mole fraction of 1.3 x 10^-3 determined by Yelle et al. (Icarus, in press, 1993) we find that the calculations match the observations well at pressures greater than 20 mbar. However, at lower pressures the calculated temperature profile is 30 K or more colder than the observed thermal structure, when only gaseous opacities are considered. We have attempted to reconcile the calculations and the data by adding two additional heat sources, parameterized with Chapman profiles. The two heat sources are required to have integrated heat fluxes of 9.0 ergs/cm2/sec and 4.2 ergs/cm2/sec with maxima heating located at 8 mbar and 20 microbar respectively. We believe that heating by stratospheric aerosols may provide a reasonable explanation for the deeper of these two heat sources. The total optical depth for the aerosols layer at 8 mbar is only 0.01. This appears to be consistent with current observational constraints. In addition, 8 mbar is approximately the level at which hydrocarbons produced by photochemistry at low pressures will condense and form aerosols. POSTERS Wild W. Carter B. Kibblewhite E. Wittman D. Lloyd-Hart M. Angel R. Baines K. Observation and Interpretation of a Near Infrared Spot Feature on Uranus Images of Uranus acquired 30 May 1993 utilizing J, H, and short K band filters with a NICMOS array on a single 1.8m MMT telescope showed a small dark spot near 35 degrees south latitude. The observations were conducted using the University of Arizona adaptive optics (ACME) system which, in the configuration used for a single mirror, stabilized image motion during the exposure. Additional features in the images include signifilcant limb brightening in the polar region and what appears to be an irregular dark band near the pole. Analysis of the various aerosol structures observed in these images will be presented, based on perturbations of the Baines and Bergstralh (1986 Icarus 65, 406-411) global models. In particular, analysis of center-to- limb profiles of K band imagery will deduce the stratospheric aerosol content, and place constraints on its spatial variability. Woodney L. M. McGrath M. A. Long-term Behavior of Ly-Alpha Emission from Uranus Ten years of data with the International Ultraviolet Explorer satellite have been used to study the disk-averaged HI lambda 1216 (Ly-alpha) emission from Uranus. The primary production mechanism for this emission is still uncertain. Study of Ly-alpha emission provides a means to help understand several key atmospheric processes including the relative strength of dayglow vs. auroral production processes, the production rate of H atoms from H2 and CH4 dissociation, hydrocarbon photochemistry, and the vertical mixing in the thermal structure of the atmosphere. Due to the unusual orientations of the spin and magnetic axes of Uranus both dayglow and auroral processes might be expected to make important contributions to the disk-averaged Ly-alpha emission. Our work shows that over the past 10 years this emission has averaged ~1.5 kR, and it exhibits no significant correlation with long-term solar activity, in sharp contrast to the dayglow Ly-alpha emission from Jupiter and Saturn, both of which are well correlated with the solar cycle. Variations in Ly-alpha brightness as large as a factor of 2 on timescales shorter than the rotation period of the planet (17.24 hr) have been observed on three occasions: April 24, 1984, February 26, 1985, and March 4, 1985. These results imply that either the dayglow Ly-alpha process on Uranus is very different than that on Jupiter ad Saturn, or that charged particle precipitation is an important source of the emission, exceeding what can be explained by resonant scattering. Herbert F. Hall D. T. Spatial Structure of the Uranian H Corona Deduced from Voyager Ultraviolet Spectrometer Observations A 35-hour subset of the Voyager 2 Ultraviolet Spectrometer (UVS) observations of the extended Uranian H corona has been analyzed in order to determine the spatial distribution of H. This least squares analysis uses the singular value decomposition (SVD) approach, so that fitting accuracy might be assessed. The fit functions for coronal H density n are of the form Y(sub)l^m/r^2, with {y(sub)l^m} the spherical harmonics. Varying the coefficients of the basis functions for n, values of the computed H Lyman alpha intensity, with g values from Strobel et al. (1991, Uranus, Bergstralh, p. 65), plus several forms of background, were fit to ~33,000 individual UVS measurements of H Lyman alpha intensity. This work represents an exploratory foray into the full dataset, so the results are still preliminary. Moreover, because of low signal-to-noise, only low-order fits could be determined. The initial determinations of density are around 1900(R(sub)U/r)^2 cm^-3 at the sun-side spin-axis pole, and about half that value at the opposite pole. A spherically symmetric distribution (80 cm^- 3 at 3.8 R(sub)U) deduced by Strobel et al. (1991) lies intermediate between these polar extremes. We find the geographic maximum at ~40 degrees from the sunlit pole at ~30 degrees W, not far from the sunside magnetic pole. The excess of density toward the sunward side agrees with predictions of the electroglow model (Broadfoot et al., 1986, Science, 233, 74), but is contrary to the first results of a magnetospheric model (Herbert, 1993, JGR, 98, 3999). However, the location of the coronal geographic maximum near the magnetic pole with the brighter aurora (Nerbert and Sandel, submitted) suggests an auroral source. Baines K. H. Mickelson M. E. Larson L. E. Ferguson D. W. The Abundances of CH4 and Ortho/Para H2 in Neptune and Uranus: Implications of New H2 4-0 Linestrength Measurements The ortho/para hydrogen ratio and tropospheric methane molar fraction (f(sub)CH4) are presented for Uranus and Neptune based on new linestrength determinations for key hydrogen features recently reported by Ferguson et al. (1993, J. Mol. Spec 160, 315-325). In comparison to analyses adopting theoretical linestrengths, the relatively weak laboratory linestrengths (approximately 30% and 15% smaller than the theoretical 4-0 S(0) and S(1) linestrengths, respectively) results in a ~30% decrease in the tropospheric methane ratio and a comparable increase in the pressure level of the optically thick cloudtop marking the bottom of the visible atmosphere (P(sub)cld). The increase in the ratio of S(1)/S(0) linestrengths from 4.4 (theoretical) to ~5.9 (measured) results in a decrease in the range of viable ortho/para ratios; an equilibrium hydrogen distribution is now the best fit for both planets. For Uranus, we find f(sub)CH4 = 0.019 +- 0.005, P(sub)cld = 3 4 +- 0.5 bars, and 0.88 < f(sub)eH2 < 1.00 (where f(sub)eH and 1-f(sub)eH denote the fraction of H(sub)2 in the equilibrium and normal state, respectively) compared to the Baines and Bergstralh (1986, Icarus 65, 406-441) values of f(sub)cH4 ~0 030 +- 0.013, P(sub)cld = 2.7 +- 0.5 bars, and 0.63 < f(sub)eH2 < 0.95. For Neptune, we find f(sub)cH4 = 0.023 +- 0.005, P(sub)cld = 3.8 +- 0.5 bars, and 0.89 < f(sub)eH2 < 1.0 compared to the Baines and Smith (1990 Icarus 85, 65-108) values f(sub)CH4 = 0.03 +- 0.007, P(sub)cld ~3.3 +- 0.3 bars, and 0.85 < f(sub)eH2 < 1.0. The methane mixing ratios derived here are in agreement with values derived from Voyager radio occultation measurements, i.e., 0.023 for both Uranus (Lindal et al., 1987, JGR 92, 14987-15001) and Neptune (Lindal 1992, AstronJ 193, 967-982). LeBeau R. P. Jr. Fischer A. S. Backes K. A. Charrette E. E. Dowling T. E. Comparisons of the Atmospheric Dynamics of Uranus and Neptune Two-layer, shallow-water computational models have proved useful in simulating vortices in the atmosphere of jovian planets, notably the Great Red Spot of Jupiter (Dowling and Ingersoll, 1989, J. Atmos. Sci.). Similar simulations of the atmospheric dynamics of Uranus and Neptune have indicated that differences in the zonal wind structure of these two planets have a critical role in the maintenance of these vortices, potentially explaining the lack of vortices on Uranus as opposed to the presence of the Great Dark Spot and other vortex features on Neptune (LeBeau and Dowling, 1992, Conference on Variable Phenomena in Jovian Systems). The impact of the zonal wind structure on the atmospheric dynamics of these two planets has now been studied with simulations using the Explicit Planetary IsentropicCoordinate model (Dowling, this conference). The primary objective of these modeling efforts is to further test the effect of zonal wind structure on the formation and sustenance of vortices. Other potentially dynamics-related features of these atmospheres are investigated as well. Maguire W. An Upper Limit to the Abundance of HCN on Neptune from Voyager 2 Infrared Observations The Voyager 2 encounter with Neptune occurred in August 1989. Several thousand infrared spectra, ranging from 30 degrees N to 80 degrees S, measured between 6-2.5 1 days before and 2.5-6 days after the encounter were averaged to increase the S/N in a search for minor constituents. Previously, C2H2, C2H4 and C2H6 have been identified from spectral features and their stratospheric abundances determined. On the other hand, some atmospheric constituents expected to be present on the basis of photochemical models, such as HCN, have not been detected. Using the nu(sub)2 HCN fundamental at 712 cm^-1 and assuming a uniform distribution for this gas above a saturation region, an upper limit of 2 x 10^-9 mole fraction has been determined. Romani P. N. Bishop J. Bezard B. Atreya S. Methane Photochemistry on Neptune: Ethane and Acetylene Mixing Ratios and Haze Production We have used a one-dimensional methane photochemical model to analyze Voyager observations of hydrocarbons and hazes in the stratosphere of Neptune. The model predicted hydrocarbons are very sensitive to the height profile of the eddy diffusion coefficient (K). For both IRIS and UVS data sets K varying inversely with the atmospheric number density to a power, produced poor results. Good agreement with the data requires that K must be weak in the lower stratosphere (K ~ 2 x 10^3 cm^2 sec^-1 for p >~ 2 mbar) but fairly vigorous in the upper stratosphere (K >= 5 x 10^7 cm^2 sec^-1 for p <~ 0.5 mbar). We are able to reproduce the IRIS C2H2 and C2H6 emission features well, less so the UVS occultation lightcurves. For solar maximum fluxes (Voyager encounter conditions) the model production rate is 1 x 10^-14 g cm^2 sec^-1. C2H6 is the dominant haze component (75%), with the remainder coming from C2H2 and C3 and C4 compounds. Balancing our haze production rate by the sedimentation rate for 0.25 micrometer radius particles (upper size limit from PPS observations) yields a total haze column burden 24% above the PPS upper limit. However lifetime analysis indicates that the model haze production rate should be averaged over solar minimum and maximum conditions. Under these conditions the model haze density is consistent with the PPS data. The predicted C4H2 and C2H6 haze densities are consistent with the lack of ice signatures in the IRIS spectra. Lodders K. Fegley B. The Origin of Carbon Monoxide in Neptune's Atmosphere Marten et al. (1991, 1993) and Rosenqvist et al. (1992) detected about 1 ppm of CO in Neptune's atmosphere. Here we show that CO on Neptune is plausibly explained by rapid vertical mixing from its deep atmosphere. Our results require that Neptune has large enrichments of heavy elements (Z >= 3) over solar composition. CO is produced from CH4 via the net thermochemical reaction CH4 + H2O = CO + 3H2. The CO mixing ratio (X(sub)CO) is given by K(sub)eq(X(sub)CH4X(sub)H2O/X^3(sub)H2) phi (1/P^2(sub)T) where K(sub)eq is the equilibrium constant for the reaction, X(sub)i is the mixing ratio of i, phi is the fugacity coefficient quotient, and PT is the total pressure (Fegley et al., 1992, in Uranus, Bergstralh et al., eds.). Using observed mixing ratios and upper limits for Uranus and Neptune, the ratio of the CO mixing ratios on the two planets is X(sub)N/X(sub)U = (X(sub)H2O)(sub)N/ (X(sub)H2O)(sub)U to a first approximation since the quench temperatures (and hence the equilibrium constants), the CH4 mixing ratios, the H2 mixing ratios, the fugacity coefficient quotient, and the total pressure at the quench level are almost the same on Uranus and Neptune. The data of Rosenqvist et al. show that the ratio X(sub)N/X(sub)U >= 16 because there is only an upper limit for CO on Uranus. Assuming a ratio of 16 and that water and methane are equally enriched on Uranus, the observed methane enrichment factors of 17-42 times solar on Uranus (Fegley et al., 1992) imply a water enrichment factor on Neptune of about 272-672 times the solar value (H2O/H2 ~ 1.56 x 10^-3). Thus, the observed CO on Neptune can be supplied by rapid vertical mixing (K(sub)eddy ~ 10^7-10^9 cm^2s^-1) from the deep atmosphere of Neptune if it is water rich. The amount of water is reduced if more CH4 is present on Neptune than on Uranus because X(sub)CO scales as E^2 where E is the heavy element enrichment over solar. The amount of water on Neptune is also reduced if water is already enriched more than methane on Uranus. This is plausible if most of the carbon on Uranus was originally accreted as volatile ices, because water ice is more refractory than either CH4 or CO ices (or clathrates). Our model is testable by observations of the water abundance in the deep atmosphere of Neptune, the CH4 abundance in the deep atmosphere, and by the vertical profile of CO in Neptune's atmosphere. Acknowledgments. Supported by the NASA Planetary Atmospheres Program. Dobrijevic M. Morel T. Parisot J. P. Brillet J. Gautier D. Study of Transmitted Light Through Neptune's Atmosphere The first step in photochemical modeling is the study of the transmission of solar flux through the atmosphere. Instead of using a classical radiative transfer program, we developed a Monte-Carlo code, which is more applicable to the Mie scattering modeling. The physical effects we have taken into account are Rayleigh and Mie scattering and absorption by atmospheric gases and aerosols. The composition and the profile temperature are derived from Lindal (1992) and the distribution of aerosols from Baines and Smith (1990). The results are presented in terms of solar flux variations in the atmosphere as a function of altitude and zenithal angle of incident photons. Finally, in order to compare our model with observations, the albedo of Neptune is presented. DeBoer D. R. Steffes P. G. Effects of the Centimeter Wavelength Opacity of H2S on Propagation and Emission in the Atmospheres of the Outer Planets Recently, measurements of the microwave properties of H2S have been taken under simulated conditions for the outer planets (De Boer and Steffes (1993) Laboratory Research for Planetary Atmospheres Conference). This is especially significant for Uranus and Neptune since they appear to be depleted in NH3 and thus H2S may significantly affect the emission from those planets (de Pater et al., (1991)Icarus, 91:220). These measurements show values that are significantly greater than values predicted by Van Vleck-Weisskopf models, even using the new value for the H2S line broadening parameter developed by Joiner et al. (1992, IEEE Trans. MTT, 40:1101). In addition to the strong opacity (relative to the Van Vleck-Weisskopf line shape function) the hyper- refractivity of the H2S molecule (8.85 x 10^17 N-units/molecule/cm^3-8 times greater than that of nitrogen) has consequences for the interpretation of observational data. Radiative transfer models, which utilize these results are being developed for Uranus and Neptune. The most recent radiative transfer models will be presented and the conclusions to date will be summarized. 1This work was supported by the NASA Planetary Atmosphere Program under grant NAGW-533. TITLE-ONLY PRESENTATIONS Lam H. A. Miller S. Joseph R. D. Tennyson J. H3+ Distribution on Uranus L window studies of Uranus show that although the planet is essentially invisible when broadband imaging is attempted, c.v.f images taken at wavelengths sensitive to the H^3+ molecular ion are able to discriminate against the sky's thermal background. The images were obtained on April 22 and April 23, 1993 (U.T.) using the ProtoCAM infrared camera on the NASA Infrared Telescope Facility on Mauna Kea, Hawaii. The new images show that H^3+ is widely distributed across the Uranian disk. The total power output is comparable with that obtained in 1992 by Trafton et al., and shows little variation between the two nights. The images are smeared due to seeing and tracking (about 0.75") and the rotation of the planet (between 20 and 30 degrees). However, they show up some structure. When Trafton et al. obtained the first H^3+ spectrum from Uranus last year, their spatial resolution was unsufficient to say if the radiation they observed was due to auroral effects or a more diffuse ionospheric emission. Our images indicate that the latter is the case, although one of the images could be interpreted by allowing auroral effects to account for an enhancement of up to 50% above the general background. Wednesday, October 20, 1993 IO II: ATMOSPHERE TO TORUS 10:30 - 12:00 NOON South Ballroom Chair(s): F. Bagenal N. Schneider McGrath M. A.* Sartoretti P. Neutral Sulfur Emission from the Io Atmosphere Ultraviolet emissions from neutral sulfur have been studied in detail to provide information on the spatial extent and abundance of sulfur relative to other constituents (such as Na, O, and SO2) in the Io atmosphere. Spatial scans in both the N-S and E-W directions across the satellite, as well as disk-averaged spectroscopy have been performed with Io near western elongation using both the Faint Object and High Resolution Spectrographs of the Hubble Space Telescope in March and May 1992. Results from these observations will be compared with HST measurements made when Io was entering Jupiter eclipse (Clarke et al. 1992), and with previous results obtained with the IUE satellite (Ballester 1989). Preliminary work indicates that the neutral sulfur emissions are narrowly confined spatially to within ~0.35" (~1200 km) of the Io surface, a factor of ~3 better spatial resolution than was obtainable with IUE. In addition, using assumptions similar to those used to derive the Na abundance (Schneider et al. 1991), sulfur appears to be ~10^3-10^4 times more abundant than Na in the near-Io atmosphere. Wilson J. K.* Schneider N. M. Trauger J. T. High Molecular Abundances in Io's Atmosphere The discovery of sodium-bearing molecular ions in the Io torus, and more recent findings from sodium emission imaging near Jupiter, have important implications for the production, dynamics, and destruction of Io's atmosphere. Our data indicate that Io's atmosphere is collisionally thick over at least part of Io's surface. The amount of Na exposed to plasma bombardment is about an order of magnitude greater than a collisionally thin exosphere can account for. The source of mass to Io's atmosphere must be primarily molecular (SO2, NaX), and the vertical transport timescale to the "interaction region," the place where plasma in the torus mixes with the atmosphere, can be no longer than a few hours. Our observations require a significant abundance of Na-bearing molecules in Io's atmosphere. Atomic S and O react with and destroy candidate NaX molecules very quickly, ruling out an atomic atmosphere. An SO2 atmosphere exposed to solar UV radiation for more than a few hours produces enough atomic S and O from photodissociation to destroy nearly all Na-bearing molecules. The high molecular abundance in the atmosphere means molecular processes play an important role in the destruction of Io's atmosphere, and in the mass budget of the plasma torus. Most SO2 molecules in the "interaction region" are dissociated by electron impact, leading to a primarily atomic interaction region. However, the molecules force the mass supply to the interaction region to have an O/S ratio of close to 2:1. SO2+ and SO+ may account for nearly one- third of the net ion production in the torus, or one-half of the net mass supply to the torus. These ions are quickly dissociated in the torus, producing neutral S and O atoms which escape the jovian system. This work has been supported by NASA grants NAGW-2484, NAGW-2492, NGT-51091, and NSF-9157751. Scherb F.* Smyth W. H. Variability of [OI]6300 Angstrom Emission Near Io We obtained observations of [OI]6300 Angstroms emission near Io in 1990,1991,and 1992 using the solar-stellar spectrograph on the McMath-Pierce telescope at Kitt Peak. High-resolution spectra with a resolving power of about 1.2 x 105 were obtained with an integration time of 10-15 minutes each. The viewing aperture for the observations was 5.2 arcsec x 5.2 arcsec centered on Io with spatial resolution limited within this area by seeing conditions. These observations have been reduced to obtain average brightness values over the aperture, which range from ~200 to 1000 Rayleighs for a number of different Io phase angles and Io System III longitudes. The brightness of the [OI]6300 Angstrom emission near Io is substantial and also quite variable indicating both an east-west and a System III longitude intensity asymmetry(1). For the [OI]6300 Angstrom emission, the O(^1D) state may be excited by electron impact of atomic oxygen and by electron impact dissociation of SO. The molecule SO may be present at the exobase or may be produced above the exobase as the dominent product of S02 dissociation by electron impact. Preliminary assessment indicates that production of O(^1D) by molecular dissociation may be more important. The [OI]6300 Angstrom emission may therefore provide a remote signature for monitoring (1) the upward transport rates of molecular species in Io's atmosphere, (2) the relative abundance and time-variable dissociation of SO2 and/or SO at the exobase and in the corona of Io, and (3) the spatial distribution of these escaping molecular and atomic species and their ion production rates in the planetary magnetosphere. New observations, obtained in May, 1993, are being analyzed and will be presented with previous results. References: (1) Scherb F. and Smyth W. H. (1993) JGR, Planets, in press. Smyth W. H.* Marconi M. L. The Neutral Clouds of Io In the summer of 1972, the discovery of sodium gas at Io marked the first of four (Na, K, O, S) extended atmospheres and neutral clouds that have been detected to date for the satellite. A fifth cloud of neutral S02 is expected, because of the detection by Voyager PLS of a small ion density of mass to charge 64 (likely S02^+) in the inner plasma torus and because of groundbased observations very near Io of a significant brightness of 6300 Angstrom emission from atomic oxygen likely excited by molecular dissociation. The S02 cloud has not been detected directly however. Of all the neutral clouds, the sodium cloud has been almost exclusively observed because of its much larger brightness and hence groundbased accessibility. Although sodium is by a wide margin the brightest of the neutral clouds because of its extremely efficient resonance scattering of sunlight at the peak of the intensity of the solar spectrum, the neutral gas clouds for oxygen and sulfur (and perhaps S02) have much larger densities and hence will provide the dominant contributions to the heavy ion plasma of the magnetosphere. The spatial structure and density of the neutral clouds in the circumplanetary space depend upon the absolute gas source rate at the satellite, the initial velocity distribution for this source (which determines the ensuing orbital dynamics of the atoms and molecules), and the neutral sink, which is dominated in the vicinity of Io's orbit by space-time dependent electron-impact ionization of atoms (or dissociation of molecules) and charge-exchange processes in the plasma torus. Neutral cloud models that calculate the three-dimensional spatial distribution of Na, K, O, S, and S02 in the satellite coronae, extended atmospheres, and neutral clouds have been developed and are being applied to observational data to determine the neutral source conditions at the satellite exobase and to assess the impact of these neutrals on the structure and properties of the plasma torus. Modeling studies that have been performed for select sodium observations and some limited observational data for the other neutral clouds will be summarized. Estimates of the gas source rates from Io for the neutral clouds will be given. Calculations for the space-time dependent nature and impact of the dominant neutral species on the ion loading, mass loading, pick- up electrical conductivity, and energy budget of the Io plasma torus will be presented. Connerney J. E. P.* Baron R. Satoh T. Owen T. Images of Excited H3+ at the Foot of the Io Flux Tube in Jupiter's Atmosphere The electrodynamic interaction between Jupiter's satellite Io and the jovian magnetosphere is fundamental to the understanding of jovian magnetospheric phenomena. This interaction drives a system of currents to and from the planet's ionosphere, where H3+ emission is excited at the foot of the Io flux tube. The first direct images of this phenomenon were obtained with the ProtoCAM infrared camera at NASA's 3-m Infrared Telescope Facility. These emissions are localized to the instantaneous foot of the Io flux tube and appear ~8 degrees equatorward of the more intense auroral H3+ emission that is therefore associated with higher magnetic latitudes. The foot of the Io flux tube (IFT) can be accurately tracked on the surface of the planet whenever viewing conditions are favorable. The IFT foot leads that of (undisturbed) model magnetic field lines passing through Io by 15 to 20 degrees in longitude, and is less visible in the northern hemisphere at longitudes where the surface magnetic field strength is greatest. These observations provide the first direct evidence of the coupling between Io and the jovian ionosphere. These data favor theoretical models of the Io interaction in which the current induced by Io's motion through the magnetosphere closes in the jovian ionosphere, and provide new insight into the source location and generation of jovian decameter radio emission. The measured path of the Io foot in both polar regions is close to that predicted by a current (GSFC O6) magnetic field model. This footprint provides a much-needed fiducial trace of the foot of the IFT on the surface of the planet with which magnetic field models may be improved and with which other infrared and ultraviolet polar emissions may be organized relative to source regions in Jupiter's magnetosphere. Hall D. T.* Moos W. Feldman P. Strobel D. McGrath M. Clarke J. Waite H. Gladstone R. Bagenal F. Schneider N. Shemansky D. EUVE Observations of the Io Plasma Torus The first observation of the Jovian system using the Extreme Ultraviolet Explorer satellite shows a rich emission spectrum from the Io plasma torus. At least 12 features between 350 and 750 Angstroms are apparent in the 62,000 second exposure obtained between March 30 and April 1, 1993. Several features lie below the short wavelength cutoff of the Voyager Spacecract UV instruments and have not been observed previously. Preliminary analysis identifies most of the features as multiplets of oxygen and sulfur ions. The emissions are spatially resolved along the long dimension of Io's projected orbital ellipse and all of the ansa emissions show a distinct dawn-dusk asymmetry. Several O+ and S++ ion emission multiplets are detected and provide an estimate of the relative ion mixing ratios. Analyses to identify other emitting ions (such as Na+), to determine the temperature of electrons exciting the emissions and to estimate the radiated power are continuing. Brown M. E.* The Io Plasma Torus: A View from Above A major problem confronting remote observations of the Io plasma torus is that these observations can only view the torus integrated along a line of sight. Because of the rapidly varying geometry and large numbers of possible configurations, this problem causes considerable uncertainty in values of torus parameters at specific locations. Deprojection of observations of the torus is important for understanding which torus properties are simply due to geometric effects and which are due to intrinsic variations in the torus. We have used a large number of high-resolution (lambda/delta lambda ~ 40,000) very long slit (slit length ~6 arc minutes) spectra of the 6717- and 6731- Angstrom [SII] lines of the torus to deproject the average behavior of the torus over the eight-month period from November 1991 until June 1992. The spectra were obtained using the 0.6-m coude auxiliary telescope feeding the Hamilton echelle spectrograph at Lick Observatory. We will discuss the deprojection method and show the results of the deprojection of the torus intensities, ion temperature, rotation velocity, and electron density. Causes for observed variations in torus properties will be discussed. Thomas N.* The Spatial Distribution of Major Ion Species in the Io Plasma Torus In March 1993, a series of high resolution spectra of the four major ion species in the Io plasma torus (IPI) were obtained using the Cassegrain Echelle Spectrograph (CASPEC) of the European Southern Observatory mounted on the 3.6 m telescope on La Silla, Chile. In addition to [SII] at lambda labmda 6716,6731 and [SIII] at lambda 6312, observations of both [OII] lambda lambda 3726,3729, and [OIII] lambda 5007 were obtained. The observation of [OIII] lambda 5007 was the first detection of this emission from the IPI. All data were taken within a 72 hour period. The long slit spectra provide excellent spatial resolution of the distribution of the ions with changing jovicentric distance. Comparisons with modeled spectra based on the results from the Voyager PLS and WS experiments show reasonably good agreement. The spectral resolution was high enough to derive ion temperatures. For [SII], the core ion temperature was found to drop rapidly inside 5.8 R(sub)J. The other ion species did not show this behavior. Measurements of the electron temperature were also obtained using the lambda lambda 4069,4076 [SII] lines. A summary of the data acquired will be presented. Banaszkiewicz M.* McGrath M. Thomas N. Determination of the Io Plasma Torus Parameters from the Visual and UV Observations Using recently obtained high-resolution optical and UV spectra of the Io torus, an attempt is made to determine the plasma parameters for a few radial distances between 5 and 7 R(sub)J. The mass and energy balance equations for the five major ion constituents of the torus (OII, OIII, SII, SIII, and SIV) are solved for as a function of five external parameters: the neutral oxygen and sulfur densities, the densities of bulk and suprathermal electrons, and the density of hot ions. The intensities of the collisionally excited lines of the sulfur and oxygen ions are then computed taking into account the vertical structure of the torus and the slit orientation with respect to the torus. The residuals between the measured and calculated intensities of the UV and optical emissions in [OII] lambda lambda 3726 and 3729, [OIII] lambda 5007, [SII] lambda lambda 1256, 6716, and 6731, [SIII] lambda lambda 1729 and 6312 and [SIV] lambda 1406 are minimized with respect to the external parameters. In addition to calculated intensities, outputs from the optimization code include the ion densities and mean temperatures as well as the electron temperature and the plasma diffusion time. POSTERS Matheson P. L. Shemansky D. E. Transport and Composition as a Function of Radius in the Io Plasma Torus As Io torus plasma is transported away from Io, it is modified in several ways. Mass loading increases the ponderance of a given volume of plasma subjected to centrifugal forces and further ionization increases the charge state of the plasma, The average magnetic moment per ion increases, showing either the non-adiabatic nature of the transport, or the presence of a remarkably broad source region. Critically, at the torus edge near 7 R(sub)J the ionization rates for the torus species increase by two orders of magnitude due to rapidly rising electron temperature. One dimensional radial transport of the quantity, plasma per unit magnetic flux NL^2, fails to account for the changing composition of the plasma. The large ionization rates near 7 R(sub)J and transport time scales longer than a few days could readily produce observable amounts of OIV and SV, particularly if this region, containing the plasma ramp, represents an impounded plasma (Siscoe et al., 1981). We intend to present a more complete set of constraints on transport in the Io torus based on the composition implied by the Voyager UVS observations. In order to avoid the build up of excessive amounts of high charge state ions the drop in plasma content in the plasma ramp must be intrepreted as a plasma sluice gate in which torus plasma is rapidly drawn off the torus boundary rather than as a region of decreased diffusive transport. Other diagnostic quantities such as the observable ratio of [SII]/[SIII] serve to constrain the rate at which plasma must move through the torus. We also consider the effect of ion temperatures on the transport problem and the ramifications of this study to the source rate of material from Io. Bird M. K. Funke O. Patzold M. Asmar S. W. Edenhofer P. Io Plasma Torus: Density and Temperature Inferred from Ulysses Radio Occultation Observations Dual-frequency Doppler measurements of the ULYSSES down-links were recorded during the spacecraft's passage through the Io Plasma Torus (IPT) shortly after closest approach to Jupiter. The electron column density of the IPT was derived from the differential phase shift between the S-band (Lambda = 13.1 cm) and X-band (Lambda = 3.6 cm) signals. Corrections for non-Jovian contributions to the change in electron content along the interplanetary propagation path and in the Earth's ionosphere are important and must be applied. In achieving its high inclination post-encounter orbit, ULYSSES moved behind the IPT from north to south during the Jupiter encounter. This geometry was favorable for probing the maximum density regions of the IPT with the radio links to Earth, particularly for investigating the IPT latitudinal and longitudinal structure. These radio occultation observations imply that the IPT in early 1992 had about the same same peak electron density, but a lower plasma temperature than during the VOYAGER flybys in 1979. The signal ray path traversed distinctly bounded sectors of the IPT on the dayside and nightside of Jupiter, respectively. Azimuthally symmetric IPT models provide generally fair agreement with the observations. The "best-fit" model parameters derived for these two restricted IPT sectors, however, were found to be quite different. This implication of an azimuthal asymmetry is supported by the appearance of localized enhancements in the electron content profile that could be associated with "blobs" in the IPT. Taylor M. H. Bagenal F. Schneider N. M. Sandel B. R. Matheson P. L. Comparison of Io Torus Temporal and Longitudinal Variations with the Bagenal Plasma Model Bagenal has recently published an updated version of her Io torus plasma model (JGR, submitted) that combines Voyager 1 in situ plasma measurements with ion composition derived from the ultraviolet spectrometer remote observations. The Bagenal model is longitudinally symmetric and is, therefore, intended to be a starting point for developing more complex models of the torus that include longitudinal and local-time variations. We have taken an initial step in this development by using the Bagenal model and a computer model called CITEP (Colorado Io Torus Emissions Package) to simulate torus emissions observed by the ultraviolet spectrometer (UVS) aboard Voyager 1. CITEP is designed to simulate torus emissions using accurate viewing geometry and up-to-date atomic data, and is available for distribution. A surprising result of the analysis is the large difference between the modeled and observed average brightness of the torus at the S++ 685-Angstrom feature of the UVS spectra. The measured average brightness of the torus is a factor of ~3 brighter than the modeled brightness. We present an analysis of the longitudinal and temporal variations of the torus 685-Angstrom brightness during the Voyager 1 encounter and attempt to determine if such variations can account for the difference between the modeled and observed brightness. Sandel B. R. Dessler A. J. Cooling Time of the Io Plasma Torus Published theories for the cooling time of plasma within the Io torus agree on values near 20 hours. The cooling time is loosely defined as the time for the plasma temperature to drop by a factor of (1-1/e) from its initial temperature after its source of heat has been turned off. Using data from the Voyager UVS experiment, we present empirical evidence showing that the actual cooling time is about one order of magnitude shorter than the theoretical cooling time. The data show: (1) rapid time variations in torus brightness (t > 2 hours), (2) an Io phase effect that shows characteristic times of both heating and cooling that are considerably shorter than 20 hours, and (3) the discovery that most of the EUV emission comes from a narrow (presumably ribbonlike) feature within the torus. The significance of this third finding is that if one takes the electron kinetic energy within the volume of a ribbonlike feature of the torus and divides by the power radiated from this feature, a cooling time of about 2 hours is obtained. The short cooling time inferred from these observations impacts present theories of heating of the Io plasma torus. Many of these problems would evaporate if power could be transferred from ions to electrons more quickly than allowed by present theoretical models. Ryabov B. P. The Io Flux Tube Model According to Jovian S-Emission Data On the basis of recent Jupiter DAM observations performed by the world's largest decameter telescope, UTR-2, we propose a physical model and geometrical configuration for the Io Flux Tube (IFT). An attempt has been made to take into consideration the majority of the frequency-time dependencies, the occurrence probability and the spatial directivity characteristics of S- radiation. The surface layer of the IFT is comprised of bunched plasma current filaments having small cross-sections and oriented along the local helical magnetic field around the IFT. The strong asymmetry of the angular thicknesses of the visible S-emission cone walls and arc formation processes in the dynamic spectra can be explained by curvature of te Io tube axis along the DAM radiation zone. The curvature radius is minimized at an S-source longitude near 220 degrees. At the same time, periodic diameter modulation of the Io tube along its axis shows the periodic frequency dependence of the S-burst occurrence probability revealed by our observations as well as by measured lifetimes of individual S-bursts. It follows that S-emission sources are located mainly at widening zones in the Io tube. The intrinsic size of active IFT intervals, in the framework of the model given, increases with distance from the planet. Wednesday, October 20, 1993 VENUS SURFACE 10:30 - 12:00 NOON North Ballroom Chair(s): G. Schubert M. Bullock Strom R. G.* Schaber G. G. Dawson D. D. The Global Resurfacing of Venus The impact cratering record on Venus is unique among the terrestrial planets. Fully 63% of the craters are in pristine condition and only 7% are heavily fractured. Remarkably only <3% of the craters and crater-related features are embayed by lava, although intense volcanism and tectonism has effected the entire planet. Furthermore, the spatial distribution of the craters is consistent with a completely random one, including stochastic variations. Monte Carlo simulations indicate that the equilibrium resurfacing model is not consistent with the small number of embayed craters or the random distribution of craters for resurfacing areas between 0.03% and 100% of the planet. It is also not consistent with the number and non-random distribution of volcanoes, or the non-random distribution of embayed and heavily fractured craters. The cratering record strongly indicates Venus experienced a global resurfacing event about 300 million years ago followed by a significant reduction in volcanism and tectonism and the accumulation of the present craters. Several thermal history models suggest that earlier episodic global events probably occurred. Monte Carlo simulations indicate that about 4-6 percent of the planet has been volcanically resurfaced since the global event, and that the lava production rate was between 0.01 to 0.15 km^3/yr during this time. This rate is significantly less than the current rate of intraplate volcanism on Earth (0.33-0.5 km3/yr). Most of the recent volcanism occurs in the Beta-Atla- Themis region, and most of the recent tectonism is associated with the major global-scale tectonic disruption zones that lie within and connect the equatorial highlands. Episodic resurfacing events that had global effects also occurred on Earth, e.g., the mid-Cretaceous superplume, and possibly on Mars. On Mars they may have been the trigger for the catastrophic release of water that formed the outflow channels. Schaber G. G.* Strom R. G. Geologic Activity on Venus: The Past 300 m.y. The density of impact craters with elevation on Venus is in remarkable agreement with the elevation density distribution (or hypsometric function) of the planet. The impact craters are thus consistent with a statistically random distribution in 3-dimensions--both spatially (as reported earlier) and with elevation. This added dimension to the randomness of the craters is interpreted as strong evidence in support of the global resurfacing of the planet about 300 my ago followed by a significant and geologically rapid decline in tectonism and especially volcanism. The most widespread geologic activity that postdated global resurfacing was along the major fracture and rift belts between latitude 30 degrees N and 30 degrees S and longitude 300 degrees and 60 degrees E (33.3% of planet). Here, heavily fractured craters are concentrated at a level that is about 2 times higher than that of the average planet. Lava-embayed craters are concentrated around Atla and Beta Regio. Twenty-one percent of all heavily fractured craters are concentrated in the 15% of the surface that includes the Ishtar Terra highlands and is bounded by latitudes 45 degrees and 90 degrees N. Fifty-five percent of all slightly fractured craters and 83% of all heavily fractured craters lie either on tessera terrain or on local- to global-scale fracture belts, suggesting some low level of ongoing extension planetwide. However, Monte Carlo computer simulations of global resurfacing using the observed cratering record suggest that only 4%-6% of the planet has been resurfaced and 4%-6% of the craters (35-55) obliterated in the last 300 my. About 42% (9 of 33) of the extremely rare (4%) lava-embayed craters on Venus lie at the mean planetary radius (6051.84 km) to 500 m above, suggesting that recent volcanic activity is concentrated within this elevation interval. Craters <=16 km in diameter on tesserae (about 11% of the surface) have exactly the same density as those on the average surface of the planet (about 80% plains), which suggests that the earlier cratering records on both the tesserae and the plains were completely reset (perhaps by different geologic processes) by the global resurfacing event. About 50% of the craters <=16 km in diameter formed on tesserae may have been preferentially obliterated since the resurfacing event by a scale- dependent geologic process operating in that terrain. Pettengill G. H.* Ford P. G. Venus: More Thoughts on the Anomalous Surface Emissivity and Radar Backscattering Regions Magellan spacecraft observations have confirmed earlier findings that a few regions on Venus, primarily located at high elevations, possess unexpectedly low values of radiothermal emissivity, occasionally reaching as low as 0.3. Associated values of radar backscatter cross section show enhancements of as much as a factor of 10 to 100 (depending on angle of incidence) over corresponding values for the typical Venus surface. Many of the general characteristics of the observed anomalous scattering behavior can be explained either as (1) emission or reflection from a highly reflective, but roughened, single interface between the atmosphere and a surface material having a bulk dielectric permittivity of order 80; or as (2) emission or reflection from the surface of a lowloss medium having a permittivity of order 5, but which contains many voids enabling efficient internal multiple scattering. Neither of these mechanisms, however, can explain the relatively high backscatter enhancements seen at all angles of incidence (coherent backscatter or COBE can explain at most only a factor of 2). To generate these enhancements requires a mechanism similar to that seen in optically reflective coatings used as road markers (involving small beads of refractive material), or the equivalent in radar corner reflectors, although neither of these mechanisms by itself displays the appropriate polarization behavior. It seems likely that the actual phenomena responsible represent a combination of processes (1) and (2), above, invoking multiple scattering (including COBE) from a deep, complex surface possibly having high dielectric perrnittivity. What are needed are observations in a bistatic geometry, in which both orthogonal senses of scattered polarization are preserved. Lovell A. J.* Schloerb F. P. McGill G. E. A Method for Determining Impact Crater Depths on Venus Magellan's altimeter is providing some of the highest resolution topography of Venus achieved to date. Nevertheless, it is often important to determine the topography of features that are comparable to or smaller than the altimeter footprint. In this paper, we demonstrate a possible method for determining topographic relief by fitting Magellan altimetry echo profiles to simple topographic models. Model surfaces are constructed on a Cartesian grid, and the heights of the elements in the grid are adjusted to simulate a particular surface feature. A model echo is constructed by summing the contributions of scattering from each surface element, assuming that each individual element scatters according to Hagfors' law. Topographic relief below the typical value within the footprint creates additional echo features on the trailing edge of the echo profile, while echo features due to elevated relief appear on the leading edge. The time difference between the echo features is directly related to the topography, and by measuring this difference, the relative elevations of features within an altimeter footprint can be determined. In general, we find that echo profiles from simple topographic features often reproduce the complex Magellan profiles that are observed, and we are now testing our method further by applying it to the determination of impact crater depths. Impact features are especially good candidates for this procedure since their basic geometry is well understood, and knowledge of their depth-diameter relationship contains important information about the nature of the crust of Venus. Results of a preliminary investigation of approximately 100 craters spanning the diameter range 10-100 kilometers will be presented. Rosenblatt P.* Pinet P. C. Thouvenot E. Hypsometric Analysis of Earth and Venus Topographic Distributions: Inferences on the Thermal Lithospheric Thickness The comparison between Earth and Venus global hypsometric distributions is reinvestigated on the basis of the high-resolution data provided, on one hand, by the recent Magellan global topographic coverage of Venus' surface [1], and on the other hand, by the detailed NCAR 5' x 5' grid topographic database currently available for the Earth's surface. For both datasets the spatial resolution is about 10 x 10 km2 and the vertical accuracy is around 50 to 100 m. For the purpose of comparison the terrestrial bathymetry is unloaded from the water column weight. The study of the cumulative hypsometric curve for both planets reveals, as noted earlier, that there exists a domain of elevation for which the relationship between the elevation and the square root of cumulative area is linear. This linear domain covers about 80% of the mapped venusian surface and nearly all the terrestrial oceanic surface, i.e., 55% of the Earth's surface. For the Earth, the elevation of the upper and lower thresholds representing the borders of the linear domain is respectively -1950 m +/- 100 m and -3950 m +/- 100 m. For Venus, the upper threshold is at the planetary radius of 6052.8 km +/- 100 m and the lower at 6051 km +/- 100 m. Calculation of the mean depth of the whole system of midoceanic ridges gives -195O m, with a 500-m standard deviation and corresponds to the elevation of the upper threshold in the terrestrial case. The geophysical meaning of the topographic level of the midoceanic ridges depth, in terms of thermal lithospheric structure, is that it represents the level at which the thermal lithospheric thickness tends towards zero. It is a natural proposition to think that, whatever the resurfacing history of a planetary surface is, as long as it is related to the thermal internal activity, such a geophysical level should exist. On this ground, both upper thresholds of terrestrial oceanic and venusian linear domain are taken as a reference level for both Earth and Venus differential hypsometric curves. One then notes that the terrestrial oceanic peak and the venusian peak do coincide very closely. In addition, both terrestrial and venusian hypsometric distributions associated to the areas of which elevation is lower than the reference level are quite similar, both in their amplitude and modal distribution, and display a pronounced symmetry. Although the identification of the venusian peak with the terrestrial oceanic peak is a striking feature, it does not necessarily mean the occurrence of plate tectonics on Venus. Indeed, as discussed in the companion abstract (Pinet et al., this volume), the pattern of the associated spatial distributions is quite different and is strongly in favor of a generalized large-scale hot-spot tectonics occurring at Venus [2,3]. The present observation only calls for the existence of a similar thermal isostasy process acting at the planetary scale and places an upper bound at 90 km for the modal and nearly maximal thickness of the thermal lithosphere of Venus, if the Venus mantle thennal conditions are earthlike, this estimate being reduced if hotter conditions occur. References: [1] Ford P. and Pettengill (1992) G. H., JGR, 97, 13, 103. [2] Morgan and Philipps (1983) JGR, 88, B10, 8305. [3] Janle P. et al. (1992) Earth, Moon, Planets, 58, 1. Pinet P. C.* Rosenblatt P. Thouvenot E. Venus Global Geodynamics from Magellan Data Hypsometry On the basis of the comparative hypsometric analysis presented in the companion abstract (Rosenblatt et al., this volume) and identifying the peak of the unimodal topographic distribution of Venus with the oceanic peak of the terrestrial distribution, it is of great interest to examine the spatial distribution of the areas whose elevation is respectively above and below the upper and lower thresholds, corresponding to the limits of the domain of elevation for which the relationship between elevation and the square root of cumulative area is linear. On Earth, the mapping of the upper threshold elevation range corresponds to the midoceanic ridges that form a linear network around the planet. It is worth noting that immediately above this level, the spatial connectivity of the hypsometric distribution is interrupted and that the higher elevation range corresponds to the continents, the continental shelves, and the oceanic islands. The areas below the lower threshold are, for one part, associated with subduction zones but are also found surrounding several oceanic hot spots features such as the Emperor-Hawaiian chain and the Bermuda and Cape Verde islands. These low areas represent about 3 to 6% of the oceanic surface, i.e., 2 to 4% of the Earth's surface and appear spatially disconnected at global scale. On Venus, there is no interruption in the regional spatial connectivity of the topographic distribution, at the level of the upper threshold that encircles every venusian highland. The high areas represent about 13% of the mapped venusian surface. However, there is no planetwide linear network, which is the case for the low-lying areas (below the lower threshold). These low areas represent about 8 to 10% of the mapped venusian surface; the associated curvilinear interconnected network is around 100,000 km long and about 750 km to 2000 km wide. It defines a limited number of surface units each containing one or more venusian highland. The resulting observed patchwork covers the entire venusian surface and each surface unit so defined reproduces at larger scale the situation observed in the case of the terrestrial oceanic hot spots. Besides, there exists a high correlation between the venusian surface topography and gravity features [e.g., 1] and the majority of the high areas identified have high geoid anomaly and high geoid-to-topography ratio, suggesting that their topography is partially to totally dynamically supported by the mantle [2,3]. This strongly suggests that the broad-scale venusian topography reflects the convective circulation of the underlying mantle, with the high and low elevation areas being respectively a surface manifestation of the mantle upwellings and downwellings. The observed pattern is in general agreement with recent results obtained in spherical models of mantle convection, taking into account the effect of an endothermic phase transition at the 670-km disconhnuity [4]. [1] Mc Namee J. B. et al. (1993) JGR, 98, 9113. [2] Smrekar S. E. and Phillips R. J. (1991) EPSL, 107, 582. [3] Kiefer W. S. and Hager B. H. (1992) Geophys. J. Int., 108, 198. [4] Tackley P. J. et al. (1993) Nature, 361, 699. Wong P. B.* Tyler G. L. Simpson R. A. Finite Difference-Time Domain Model for Surface Scattering with Application to Magellan at Venus A major difficulty in physical interpretation of Magellan microwave backscatter observations is absence of data on the signature of typical geologic surface features, e.g., rocks. Numerical solution of Maxwell's Equations offers procedures for determination of the signature of realistic surface features up to several wavelengths in size. The FiniteDifference Time-Domain (FD-TD) numerical method has been widely studied and has proved to be convenient for solving scattering problems that involve conducting obstacles and/or inhomogeneous dielectric and magnetic components placed in a homogeneous medium. We introduce a new FD-TD methodology for calculating the scattered field due to inhomogeneities on or inside the lossless/lossy dielectric plane, such as a rock on a planetary surface (see figure, which appears in the hard copy). For application to Magellan, the method involves establishing total field/scattered field interfaces that "control" the electromagnetic environment of the scatterer. Preliminary results will be presented. Maurer M. J.* Tyler G. L. Simpson R. A. Comparison of Magellan Venus Differential Altimetry with Observed Doppler Anomalies Measurements of the Venus surface made by the Magellan altimeter may be processed in several ways. The most straightforward is to obtain estimates of the terrain elevation directly below the spacecraft [1]. Refinements of this method can also determine other surface properties such as reflectivity and roughness. Our own analysis of the data has consistently shown unexpected shifts in the frequency of the peak of the echo spectrum, indicating that the strongest signal was not coming from directly below the spacecraft but from somewhat ahead of or behind it. We have proposed several mechanisms for this effect including largescale tilt of the surface, inhomogeneities in reflectivity over the echo footprint, and anisotropy of scattering properties due, perhaps, to patterned ground. The large-scale tilt effect can be estimated using the observed altimetry data along the track of the spacecraft. We have computed a predicted Doppler offset based on the known large-scale topography, and subtracted this effect from the observed frequency offsets. The resulting Doppler anomalies still show organized patterns and correlation with some geological features, indicating that topography is not their only source. Future work will attempt to eliminate reflectivity inhomogeneities and isolate the effects of anisotropic scattering. [1] Ford, P. G. and Pettingill, G. H. (1992) J. Geophys. Res., 97-E8, 13103-13114. Yoder C. F.* Venusian Obliquity as a Tidally Evolved End State Dynamical models of Venus' spin pole orientation indicate a ~1.5 degree remnant free obliquity sigma which is enigmatic based on conventional models of tidal and core mantle friction. The slow retrograde spin of Venus is believed to be the result of a balance between solid friction tides, which tend to drive Venus toward synchronous rotation and atmospheric thermal tides, which drive the spin away from synchronous rotation. The spin stabilizes at a specific retrograde (or prograde) rate because of the expected inverse frequency dependence of the semi-diurnal, S2 thermal tidal constituent. The obliquity of Venus is affected by an additional frictional force, core mantle friction (CMF) at a core mantle boundary (CMB). The Reynolds number Re=omega R(sub)c^2sin^2Delta/epsilon/nu, where Delta epsilon is the differential, core mantle obliquity and nu is the fluid core kinematic viscosity is so large that the CMB fluid stress is turbulent rather than laminar, and this stress is proportional to sin^2Delta epsilon. Turbulent CMF rapidly damps free obliquity on a time scale of order 10^7 to 10^8 yr. Both solid and thermal tides (dominated by surface heating) can cause the obliquity to increase at a rate proportional to sin Delta epsilon. Usual estimates of this CMF torque show that it is of order 10^3to 10^4 stronger than the other secular torques. I have found that CMF can be reduced by a factor of (e(sub)o/e(sub)c)^3(n/omega)^6 where e(sub)c and e(sub)o are core and mantle dynamic ellipticities (e=(C-1/2(B+A))/C), respectively. The differing dependencies of these torques on obliquity implies there is the opportunity for balance of the three secular torques as they effect obliquity at a specific (and nonzero) value of the obliquity determined largely by the parameter e(sub)o/e(sub)c. I find this balance occurs for e(sub)c~20e(sub)o=2.7 x 10^-4. For comparison, Earth's nonhydrostatic core and mantle ellipticities are 1.2 x 10^-4 and 3.5 x 10^-5, respectively. The strong dependence of CMF on spin rate indicates that CMF is effective only for slow spins. Additionally, I find that Venus can begin with a prograde spin and evolve to a retrograde state. I shall describe in more detail the potential geophysical meaning of this large venusian core ellipticity, which is apparently supported by ~400km thick thermal boundary layer just above the CMB, and additional constraints on tidal evolution history. POSTERS Lenardic A. Kaula W. M. Mantle Plume Morphology on Venus and Earth: Potential Differences and Speculations as to Cause Observations of broad topographic swells believed to be surface manifestations of thermal mantle plumes on Earth and Venus suggest that plumes in the two planets differ morphologically. Comparison of the largest swells combined with theoretical studies modeling the features as plume-induced uplift elucidate key differences that may exist between Earth and Venus plumes. Mantle convection models suggest that these differences can be attributed to the absence of subducting tectonic plates on Venus. Models incorporating platelike behavior lead to a larger temperature drop across a lower thermal boundary layer from which plumes form than do models lacking platelike behavior. The different plume types expected for systems with large and small boundary layer temperature drops are consistent with inferred differences in Earth and Venus plume morphology. We argue that the majority of variation between plume- related swells on Earth and Venus can be attributed to the respective presence/absence of active plate tectonics on these planets. Thompson T. W. Magellan Flight Team Magellan Mission Progress Report Magellan was launched from Cape Kennedy on May 4, 1989, and was inserted into orbit around Venus on August 10, 1990. Radar imaging, altimetric and radiometric mapping of the venusian surface commenced on September 15, 1990, and continued until September 15, 1992 (mission cycles 1,2,3). Some 98% of the surface has been mapped with resolutions on the order of 120 m. Then, high- resolution Doppler tracking of the spacecraft from September 16, 1992, until May 24, 1993 (mission cycle 4), provided gravity observations of venusian near-equatorial regions. Some 950 gravity orbits spanning 360 degrees of longitude were obtained. These radar and gravity observations satisfy the Magellan science objectives of (l) improving the knowledge of the geologic history of Venus by analysis of surface morphology and (2) improving the knowledge of the geophysics of Venus, principally its density distribution and dynamics. A new mission phase started on May 25, 1993, when Magellan's periapsis was lowered and aerobraking began. This will continue until mid August. The appoapsis altitude, which was about 8500 km when aerobraking started, will be lowered to 500 to 700 km. Doppler tracking in this near-circular orbit will provide high-resolution gravity data for the polar and near-polar areas of Venus. Magellan radar products are available as analog photographs and digital compact disks (CD-ROMS) at the National Space Science Data Center (NSSDC), Goddard Space Flight Center, Greenbelt, Maryland. Over 1200 mosaics are available. In addition, the altimetric and radiometric data products from MIT and the cartographic maps from U.S.G.S. are available at the NSSDC. Magellan gravity products, line-of-sight acceleration profiles and spherical harmonic fields are at NSSDC also. Wednesday, October 20, 1993 1993 UREY PRIZE LECTURE 1:30 - 2:20 PM Grand Ballroom Chair(s): R. Beebe Yelle R. V.* Pluto's Atmosphere: Where It Comes From and Where It's Going No abstract available. Wednesday, October 20, 1993 INVITED TALK 2:20 - 3:00 PM Grand Ballroom Chair(s): J. Lunine Phillips R. J.* Magellan Gravity Experiment and Geodynamical Investigations Magellan completed its successful imaging mission in the summer of 1992, and commenced gravity mapping in September of that year. Data were acquired by tracking Magellan in its elliptical orbit; data quality is superior to Pioneer Venus gravity because of lower orbital eccentricity and superior signal-to- noise ratio (X-band vs. S-band tracking). In May 1993, 360 degrees of longitudinal tracking was completed, and the spacecraft performed a maneuver to lower periapsis altitude and then used atmospheric drag to circularize the orbit by August of 1993 with a planned altitude of ~250-300 km. This circularized orbit will provide a high-resolution gravity field over the entire planet and allow a series of investigations heretofore not possible. We know from past gravity investigations that the relationship between long- wavelength gravity and topography on Venus is remarkably different from on Earth, and that this probably results from the lack of an asthenosphere or low-viscosity zone in the upper mantle of Venus. This allows direct coupling of mantle convective flow into the lithosphere and provides dynamic support for some of the larger highland features such as volcanic rises. Gravity analyses have led to the concept that much of the tectonic deformation observed in the images can be traced directly to mantle convection. However, for features such as crustal plateaus, there is no agreement on the magnitude or even sign (upgoing or downgoing plume?) of mechanical support. High- resolution Magellan gravity data will provide new constraints for this problem. Additionally, the boundaries of some large coronae have been proposed as the sites of lithospheric subduction. If this idea is correct, then Venus will join the Earth as one of two planets where this process is known to occur. Gravity data from the circular orbit offers the strongest means of testing this hypothesis, as well as the notion that there is subduction on the boundaries of Ishtar Terra. Wednesday, October 20, 1993 MOON AND MERCURY 3:30 - 5:30 PM South Ballroom Chair(s): J. K. Harmon D. Domingue Hawke B. R.* Blewett D. T. Campbell B. A. Spectral and Radar Studies of the Schiller-Schickard Region: Implications for the Local Mixing Hypothesis In the post-Apollo era, Oberbeck and coworkers proposed that local mixing was an important process in the formation and emplacement of the distal deposits of large, lunar impact structures. According to the local mixing hypothesis, secondary-forming projectiles are ejected from the pre-impact target site and travel in ballistic trajectories away from the parent basins or craters. When these projectiles strike the lunar surface they excavate local material and incorporate this non-primary debris into the resulting deposit. At relatively large distances from the original impact site, the local mixing hypothesis predicts that local, non-primary material will dominate the ejecta deposit. The Schiller-Schickard (SS) region is ideally suited to test the local mixing hypothesis. This region is located ~1300 km from the center of the Orientale impact basin, and secondary-forming projectiles excavated and incorporated preexisting mare basalt into the distal deposits of the Orientale basin. Near- IR (0.6-2.5 micrometers) reflectance spectra were obtained for a variety of geologic units in the SS region and analyzed in order to extract compositional information. Both band analysis and mixing model techniques were utilized. The results for spectra obtained for small areas on light plains in the SS region indicate that these deposits contain 40-60% mare basalt. To gain greater spatial coverage, mixing model studies were also conducted using multispectral images obtained in 12 filters (0.4-0.99 micrometers). The average amount of mare basalt in light plains pixels is 45%. The 3.8 cm, 70 cm, and 7.5 m radar datasets were also analyzed. No anomalously low 3.8 cm returns were found for areas previously mapped as dark mantle. Hence, the existence of an extensive, thick dark-mantle deposit of pyroclastic origin is unlikely. Portions of the SS region exhibit relatively low values in the 70 cm depolarized images. These anomalous areas correlate in part with cryptomare deposits. Low 70 cm returns are associated with some Orientale secondary crater clusters. Much of the SS region also exhibits low values in the 7.5 m dataset. Chevrel S. D.* Pinet P. C. Martin P. Copernicus: A Regional Probe of the Lunar Interior The spatial distribution of crater materials within Copernicus is presented at the subkilometer scale on the basis of their spectral ultraviolet (UV)-visible (VIS)-near infrared (NIR) characteristics. Data are high spatial (0.7 km) and spectral (R = 100) resolution CCD images obtained from the 2-m telescope of the Pic du Midi Observatory in France. Preliminary work has shown a general consistency (within 1 to 2% mean deviation) between the spectra retrieved from our CCD data and earlier reflectance spectra available for particular spots within Copernicus, both in terms of overall continuum slope and absorption features. On the basis of this agreement, a detailed spectromineralogical map of the crater interior and its surroundings has been produced by applying a spectral mixing analysis based on the endmember technique. It has been shown that a mixing of three components can account for nearly all the total spectral variance of the image. From previous spectroscopic studies [1,2], one end member representing an olivine-bearing lithology has been selected among the central peaks. Our analysis then reveals that olivine is detected as the primary mafic component not only within the three central peaks but also along a significant portion of the rim and wall terraces of the crater. The two other selected end members may respectively represent marelike and impact- meltlike materials. Indeed, the spectral characteristics of the third end member are consistent with optical alteration effects observed in laboratory spectra obtained from materials having undergone melting and vitrification processes [3]. The associated fraction image maps out vast areas spectrally homogeneous, distributed across the crater floor and on the external rim and represents a first-order mapping of the impact melt component at Copernicus. Most of the interior and surroundings of the crater can be described as a linear combination of either ejectalike material (with a variable amount of impact-melt-) or marelike material and olivine-bearing rocks, in variable proportion. Our results suggest that the petrology of the target was uniform throughout the stratigraphic horizons from which the rim and the peak materials originated. Consequently, the preimpact target had an olivine- bearing lithology at shallower depth than previously thought [2]. In addition, the linear mixing results suggest that from the near-surface to the depth of the proposed troctolitic/dunitic horizon, a progressive change in the pyroxene/olivine ratio may occur in the target lithology. Although it is difficult to discriminate among the various hypotheses concerning the structure of the crust at the Copernicus site, these results support the idea that the lower crust and possibly the lunar mantle are regionally at shallow depth in this region of the Moon. [1] Pieters C. M. (1982) Science, 215, 59. [2] Pieters C. M. and Wilhelms D. E. (1985) JGR, 90, C415. [3] Clark B. E. et al. (1992) Icarus, 97, 288. Henderson B. G.* Jakosky B. M. Near-Surface Thermal Gradients and Mid-IR Emission Spectra of Planetary Surfaces In evacuated particulate materials, the thermal conductivity is extremely low, and radiation is an important heat-transfer mechanism. Under high-temperature conditions, the top two or three e-folding absorption skin depths can cool via radiation to space, and significant thermal gradients will form in the top few hundred micrometers of the surface where the thermal emission is generated. The effects will show up in an emission spectrum due to the variation in opacity with wavelength; at different wavelengths, the thermal energy comes from material at different depths with different temperatures. The resulting spectrum will show effects not related to the true surface emissivity, complicating compositional interpretation. We are modeling the conductive and radiative heat transfer in the uppermost layer of a planetary surface to determine in which remote planetary environments near-surface thermal gradients exist and what effects they have on midinfrared emission spectra. Calculations thus far have been performed for basal-heated and solar-heated materials in which the medium is purely absorbing and emitting. Future models will include scattering. Results to date indicate that near-surface thermal gradients are important in the surfaces of the Moon and Mercury (40-80 K/100 micrometers), small but not negligible in the martian surface (<= ~5 K/100 micrometers), and negligible for Earth (<1 K/100 micrometers). Morgan T. H.* Potter A. E. Variation of the Sodium Lunar Exosphere with Lunar Phase The lunar sodium exosphere above the equatorial lunar limb was measured over a sequence of days on Dec 25-27, 1988, Dec 2-7, 1990, and Sept 23-29, 1991. Sodium emission was measured as a function of altitude up to about 500 km in most cases. These datasets have a common feature in that the measurement period included observations made near full Moon. In each of these cases the emission intensity near full Moon was extremely small. Near full Moon, sodium emission was so small as to be detectable only at low lunar altitudes, just above the lunar surface. The emission intensity increased as the lunar phase angle increased. The scale height temperature of the sodium also increased. These results apparently contradict our previously published observation (Potter and Morgan, 1991) on February 22, 1989, where the sodium emission near full Moon was not significantly different than at other phases. However, consideration of particle fluxes in the magnetotail can provide an explanation. When the Moon passes through a lobe of the Earth's magnetotail, the energy and flux of particles striking the lunar surface is small (Hardy, et al., 1976). However, sometimes the Moon will pass through the plasma sheet at the center of the magnetotail, where particle energy is comparable to the solar wind. If particle sputtering is the dominant source process for exospheric sodium, then, when the Moon passes through a lobe region, relatively little sodium should be produced. When the Moon passes through the plasma sheet, sodium production should be comparable to that outside of the magnetotail. The position and dimensions of the magnetotail are variable, so that both types of behavior should be observed. References: [1] Potter A. E. and Morgan T. H. (1991) GRL, 18, 2089-2092. [2] Hardy D. A. et al. (1976) Magnetospheric Particles and Fields (B. M. McConnac, ed.), 89-98. Wiens R. C.* Burnett D. S. Calaway W. F. Pellin M. J. Experimental Studies of the Role of Photodesorption in the Formation of Planetary Na Atmospheres Sodium atmospheres have been discovered surrounding Mercury, the Moon, and Io. Photodesorption has been suggested as the dominant mechanism for Na liberation from the Moon and Mercury. Observations of the lunar Na atmosphere (e.g., Potter and Morgan, 1988) have suggested a two-component distribution, one with a scale height indicative of surface-equilibrated temperatures, and an energetic component corresponding roughly to 2000 degrees K. Recently, Sprague et al. (1992) has suggested that the surface-temperature component, which is most abundant at the subsolar point, is due to thermal desorption, while the energetic component is due to photodesorption. Limited experimental data are available on photodesorption, particularly in the low-intensity regime where thermal effects are negligible. We therefore initiated an experiment studying Na removed from Na2SO4 by a low-intensity 308 nm XeCl excimer laser beam. The experiment was carried out in Argonne's SARISA IV laser postionization mass spectrometer using a 30 ns pulsed photodesorption beam, followed by a pulsed 6.4 eV laser ionizing beam parallel to the surface, and time-of-flight detection. Over the range from 0.7 to (3 x 10^-6) mJ/cm^2/pulse the response was linear (at 1 x 10^-7 Na/photon), ruling out thermal desorption. The photodesorption cross section of 4 x 10^-23 cm^2 is within the range of earlier estimates, confirming that photodesorption is a significant source for Na atmospheres, assuming the results also apply to more realistic regolith materials. The weakest intensity, 280 mW/cm^2 averaged over the pulse length, was only two orders of magnitude greater than the integrated <= 308 nm solar UV intensity at Earth. The peak of the velocity distribution corresponds to thermal (<800 degrees K) temperatures, indicating that photodesorption may be indistinguishable from thermal desorption in its peak velocity. Since this experiment was done in the near UV, it is unlikely that the whole solar spectrum would give a much higher peak velocity for a given surface temperature. Wildey R. L.* Kieffer H. H. A New Facility for Absolute Photometric Imaging of the Moon A new automated lunar photometry observatory has been constructed. An objective of this work is to provide new radiance information needed to use the Moon as a well-defined radiometric source for calibration of Earth- orbiting instruments. Instrumentation consists of a fully digital mount, a 20-cm aperture Ritchey- Cretien telescope of 0.88m focal length, a photometer that can hold up to 34 filters, a 512 x 512 Silicon CCD, and a factor-of-100 geometric optical stop. Inside the dome is a full-aperture calibration plaque and a NIST-traceable 1000 watt spectral-irradiance standard lamp. All of this equipment is computer controlled. The lunar diameter subtends about 490 pixels. Optical filters are chosen variously to approximate the standard stellar magnitude systems, accommodate wavelenqths of lunar spectral contrast, and reproduce passbands of planned Earth-orbiting instruments (ASTER, LANDSAT-6, MISR, MODIS, SeaWIFS). The same equipment is used for lunar and stellar observation, with the optional use of the optical stop. A typical night run involves observation of about 100 photometric stars and the Moon 5 times; 10 of the photometric stars are observed repeatedly to provide determination of extinction. Observations will be made on every photometric night during the bright half of the month for at least 4.5 years, to adequately cover phase and libration variation. Routine observations are scheduled to begin in Sept. 1993. Each lunar image will be reduced to absolute exo-atmospheric radiance, and reprojected to a fixed selenographic grid system. The collection of these images at various librations and phase angles will be reduced to photometric models for each of the ~120,000 points in the lunar grid, for each filter. Radiance models of the Moon can then be produced for the precise geometry of an orbiting instrument observation. Mitchell D. L.* de Pater I. The Very Low Microwave Opacity of Mercury's Regolith: Implications on Surface Mineralogy We have imaged Mercury's thermal emission at wavelengths from 0.3 to 20.5 cm. These observations reveal that Mercury's surface is remarkably transparent to microwaves: the regolith is at least 3 times more transparent than the lunar maria, and at least 40% more transparent than the lunar highlands. The microwave opacities of returned lunar samples have been shown to increase with the abundance of ilmenite (Gold et al. 1976), which is the most common titanium-bearing mineral on the Moon (ideal formula: FeTiO3). Ilmenite, which is also opaque at optical wavelengths, is concentrated in the lunar maria and is largely responsible for the albedo differences between the maria and highlands (Pieters 1978). The fact that Mercury's regolith is very transparent to microwaves, together with the Mariner 10 observation that the albedos of Mercury's surface geologic units are sytematically higher than corresponding units of the lunar surface (Hapke et al. 1975), strongly suggests that Mercury's regolith has a very low abundance of ilmenite. None of the models of Mercury's origin that have been proposed to account for the planet's large mean density (volatilization of the surface, selective accretion, giant impact) currently predict low abundances of Fe and Ti at the surface (Lewis 1988). Mercury's surface mineralogy may result from differentiation of Fe and Ti to deeper levels than on the Moon as a result of more extensive lithospheric melting and a larger surface gravity (Strom 1984). The extreme differentiation of Mercury's crust, even when compared with the lunar highlands, may have occurred following a giant impact, which should have caused global melting of surface. Harmon J. K.* Slade M. A. Velez R. Crespo A. Dryer M. New Radar Images of the Poles of Mercury New radar reflectivity images of the poles of Mercury have been made from a reanalysis of 1991-92 Arecibo data. These images provide improved estimates of the locations and radar properties of the north and south polar anomalies, recently discovered radar-bright features that may be associated with enhanced backscatter from polar ice deposits (Slade et al., 1992; Harmon and Slade, 1992). The north (left) and south (right) polar images are shown below, with darker shading corresponding to higher reflectivity. Each of these images was constructed by summing polar images from sixteen observing runs spread over eight different days. Much of the north polar feature is concentrated in several distinct bright spots less than 50 km in size. Parts of these spots show the inverted polarization characteristic of enhanced backscatter from ice. The south polar anomaly is dominated by a 100-km-diameter circular feature in the floor of crater Chao Meng-Fu, although some smaller features can also be seen. The implications of these images for the icy-crater hypothesis (Paige et al, 1992; Ingersoll et al., 1992; Butler et al., 1993) will be discussed. The mapped northern bright spots smear out rapidly as the assumption of zero polar obliquity is relaxed; this has been used to set an upper limit of 1 degree for the obliquity. Paige D. A.* Wood S. E. Vasavada A. R. Radiosity Modeling of High-Latitude Impact Craters on Mercury and the Moon: Implications for Water Ice Recent radar observations of Mercury have revealed the presence of anomalous radar reflectivity and polarization features near the north and south poles, which are most easily explained by the presence of relatively pure ice deposits (1-5). We have constructed a detailed thermal model that calculates the temperatures of surfaces within bowl-shaped and flat-floored impact craters on Mercury and the Moon that includes the effects of multiple scattering at solar wavelengths and re-radiation at infrared wavelengths using the radiosity method (6,7). The results confirm the findings of previous studies that show that water ice should be stable to evaporation within the permanently shadowed regions of larger high-latitude impact craters on both bodies. We have also found that because of their orbital and axial orientations, portions of these permanently shadowed regions on both bodies can be viewed directly from Earth. These results make an even stronger case for the presence of ice on Mercury and the Moon. References: [1] Harmon J. K. and Slade M. A. (1992) Science, 258, 640. [2] Slade M. A. et al. (1992) Science, 258, 635. [3] Paige D. A. et al. (1992) Science, 258, 643. [4] Ingersoll A. P. (1992) Icarus, 100, 40. [5] Butler B. J. (1992) JGR. [6] Hodges R. R. (1980) Proc. LPSC, 11, 2463. [7] Goral C. M. et al. (1984) Computer Graphics, 18, 213. Slade M. A.* Jurgens R. F. Anderson J. D. Lau E. L. Harmon J. K. Campbell D. B. Chandler J. F. Shapiro I. I. The Equatorial Topography and the Orbit of Mercury from Goldstone and Arecibo Radar Ranging Several proposed Discovery-class missions to Mercury hold out the promise of completing the reconnaissance of the terrestrial planets. Navigation of these spacecraft to their various encounters with Mercury requires the maintenance of its ephemeris. Fortunately, since 1988, a program of cooperative radar ranging to Mercury has been in operation. This program endeavors to produce pairs of ranges at two different places in the Mercury orbit when the sub- Earth points are identical for each set of ranges. The common topography cancels in the differenced range, leaving extremely precise information about the orbit. While the aims of this program have been primarily in the realm of gravitational physics (such as setting limits on a time variation of G; directly measuring the solar gravitational oblateness, etc.), a useful byproduct of this radar ranging program is the determination of topography of an equatorial zone between roughly +-15 degrees Latitude. Topography results from the many instances in which the so-called "closure pair" does not get "closed". In addition, observations are performed to calibrate the two ranging system of the two telescopes. Also some observations are performed for radar- imaging of the Mercury disk (e.g., Harmon and Slade, 1992). The topography thus determined is from a global solution of all appropriately weighted observations of Mercury, so that the heights are relative to a greatly improved datum for Mercury. Mariner 10 ranging and Doppler measurements, and radar ranging previous to 1988, both from Arecibo and Goldstone, have been included in the solutions (Anderson et al., 1992a; Anderson et al., 1992b, Harmon et al., 1986). The global topography has been computed in two ways: (1) the residuals resulting from solving for just a common Mercurian radius are interpolated to a regular grid and least-squares-fitted cubic splines are passed through the set of points (2) the "global" solution includes a parameterization of the topography T: T(phi,lambda) = R [ 1 + Sigma Sigma (A(sub)mcos m lambda + B(sub)lm sin m lambda) P(sub)m (sin phi)] The topography will be compared with radar-determined albedo variations and with knowledge of the gravitational potential of Mercury. Domingue D.* Sprague A. Kubala A. Lee S. Error in Sodium Abundance Estimations in Mercury's Atmosphere Using Hapke's Surface Reflectance Model Sodium abundances in Mercury's atmosphere are estimated using an absolute calibration that relies on Hapke's bidirectional reflectance function to convert data numbers (dn) from CCD measurements into photon counts. Using global Hapke parameters from Veverka et al. (1988), reflectances are calculated for areas of the surface imaged by CCD. This calculated reflectance is converted to numbers of photons. A calibration factor converting CCD dns to photon counts is derived by comparing photon counts estimated from Hapke's model to the CCD dns at wavelengths within the continuum. The dns measured within the emission wavelengths include light reflected from the surface plus photons emitted from atmospheric soduim atoms. Estimates of the number of photons due to surface reflections are subtracted from the dn counts within the emission wavelength region. The remaining dn counts are converted into sodium abundances using the calibration factor derived from the continuum as described above. The accuracy of the sodium abundance is dependent on the calibration factors derived using global Hapke parameters for a sub-area of Mercury's surface. The global Hapke parameters do not necessarily describe each sub-area on the surface accurately. Preliminary comparisons of Veverka et al.'s (1988) solutions 1 and 2 for the disk-averaged brightness of Mercury (where BO was set to 1 for both solutions) show that differences in calculated reflectance of up to 10% can be seen at some viewing geometries. Chamberlain J. W.* Bishop J. Planetary Exospheres: Closed Solutions for the Evolution of Orbital Elements for Perturbations by Radiation Pressure We present the exact solutions in three dimensions for the orbital elements of a satellite particle continuously exposed to weak solar radiation. The solutions are obtained from a set of auxiliary variables, which are separable and therefore integrable. We have illustrated various examples of the behavior of the elements as they evolve cyclically. In addition, there are three items worthy of note: (1) The planar case, wherein the orbital plane contains the Sun, has been solved previously, but it is here shown to be unstable, quickly evolving out of its original plane. This behavior is a consequence of radiation pressure being a repulsive perturbation. (2) In the general, three- dimensional case the orbital elements not only have components that vary with time, as expected, in a uniform sinusoidal manner. They possess, as well, non- uniform sinusoidal variations that have the same period as the uniform variations. (3) The semi-major axis, a, which is related to the total energy, may be regarded as constant in the first order, on the grounds that a particle returns essentially to its starting point after one orbit. Here we derive the actual time variation of a from the extended Laplace constant. That constant was obtained in Paper I from the separation of variables that is allowed when the perturbed, Keplerian problem is formulated in parabolic coordinates, and it remains a valid constant even when the total Keplerian energy is not. The relative amplitude, Delta a(t)/a(O) varies as (a/r(sub)x)^4, where r(sub)x is the distance from the planet where the gravitational and radiative acceleration are equal (i.e., the radius of the exosphere). Hence, the variation in a(t) is very small, except for orbits comparable in size to the exosphere itself. POSTERS Marconi M. L. Smyth W. H. The Sodium and Potassium Atmospheres of the Moon Numerical models for the lunar sodium and potassium exospheric atmospheres have been developed to study the importance of various atmospheric and surface processes and to analyze groundbased observations of these lunar gases in their solar resonance scattered D-line emissions (sodium: 5890Angstroms, 5896Angstroms; potassium: 7665Angstroms, 7699Angstroms). The models calculate the three-dimensional sodium and potassium densities and also their projected two-dimensional column densities or solar resonance scattered emission images (and high-resolution line profiles) on the sky viewing plane for an initial surface source flux velocity-distribution that may be both spatially (i.e., longitude and latitude) and time dependent. The subsolar temperature (~400 K) and the photoionization lifetimes for sodium (~15 hrs) and potassium (~10 hrs) are nearly constant since the Moon-Sun distance does not vary significantly. The density contributions of thermal atoms are locally determined since the photoionization lifetime is smaller than the lateral migration time. Observations of sodium at large distances from the surface, however, indicate that there is a significant abundance of nonthermal source atoms in the atmosphere and an escaping anti-sunward comet-like coma. Solar radiation acceleration is minimum for no atom doppler shift (i.e., a zero atom-Sun radial speed) out of the bottom of the solar Fraunhofer features and is 2.71 cm/s^2 (i.e., 1.7% of surface gravity) for sodium and 5.33 cm/s^2 (i.e., 3.3% of surface gravity) for potassium. The solar radiation acceleration is, however, time variable and may be several times its minimum value because of the doppler shift produced by the vector sum of the nonthermal atom-Moon motions (~1-2.5 km/s), the lunar orbital motion about the Earth (~1 km/s), and the radial Earth-Sun and other three-body motions (~0.5 km/s). This variable acceleration produces an expanded atmosphere/coma at first quarter in the spring and a contracted atmosphere/coma at third quarter in the fall, which, for the sodium antisunward coma, has been recently verified by groundbased observations. For the sodium sunward atmosphere, an excellent model fit to the observed vertical profiles of Potter and Morgan (1988, GRL 15, 1515, altitudes up to 1200 km) and Mendillo et al. (1991, GRL_, 2097, altitudes between ~3170 and 9000 km) is obtained for a 1000 K Maxwell-Boltzmann hemispherical surface source and for 100% gas sticking at the surface. Robinson M. S. Hawke B. R. Lucey P. G. Edwards K. Mariner 10 Multispectral Images of the Moon and Mercury The Mariner 10 spacecraft acquired multispectral images of both the Moon and Mercury. We have developed a calibration scheme to make these images useful for quantitative color studies. This radiometric calibration includes dark current removal and a nonuniformity/nonlinearity correction. We have also devised a strategy to improve the accuracy of the spacecraft pointing. This is necessary to allow for accurate calculation of image geometry: the emission, incidence and phase angles for each image. From a color ratio image (blue/orange) we have extracted values for individual mare units on the eastern limb and farside of the Moon. This color ratio, for mare units, is predominately controlled by TiO2 abundance. All maria examined have a ratio less than that seen in northeastern Fecunditatis, which has an Earth-based derived TiO2 level of about 5%. Thus, mare basalts examined in this region of the Moon all exhibit assumed TiO2 values <5%. Galileo image data for the western farside also indicate that the majority of mare units examined in that portion of the Moon have TiO2 values <5%. From the new Mariner 10 and Galileo results we infer that no high TiO2 mare basalts were extruded on the farside and interpret this to be due to an inability of higher density TiO2 rich melts to make excursions to the surface of the thicker farside crust. Additionally, we have constructed a preliminary normalized albedo map of the northeast nearside (NEN) region from 4 frames (FDS 2660,61,67,68) at resolution of 1 km. These calibrated images were transformed to relative albedo using a Hapke function. Due to temperature anomalies at closest encounter we found problems with the calibration in certain regions of both cameras. However, we found that selective masking of these areas results in a qualitatively acceptable match. Our analysis of this preliminary map indicates (1) Some portions of the "light plains" deposits east and north of Mare Frigoris exhibit unusually low albedo, compatible with mare units. Mare volcanism in this region may have been more extensive than has previously been thought; (2) New albedo data confirm the existence of dark mantle deposits in the NEN region; (3) Localized pyroclastic deposits occur in the interior of Gauss crater; ongoing analysis may reveal the presence of additional dark mantling units. In addition to our lunar work we are currently processing both low (~20 km/pixel) and medium resolution (5 km/pixel) multispectral images of Mercury (uv, blue, orange, uv polarized wavelengths). These calibrated data will be used to examine the distribution of color units for ~50% of the mercurian surface. Budney C. J. Lucey P. G. Near-Infrared Imaging Survey of the Moon Recent advances in IR detector technology have made relatively large IR imaging arrays available at major observatones. In December 1990, during four nights of photometric quality, we collected over 2000 images of the Moon at the University of Hawaii 2.24 m telescope on Mauna Kea in four infrared wavelengths. The filters used have centers of 1.28, 1.50, 1.65, and 1.99 micrometers, with bandwidths of about 20 nanometers. Spatial coverage extends from approximately 30 degrees N to 30 degrees S and has a resolution of about 2 km/pixel. We have performed flat field, sky subtraction, bad pixel mask, and photometric collections on the data. Flat fields were constructed using both principal component analysis of a set of images and the method of Kuhn et al. (1991) [1]. The photometric corrections are relative to the Apollo 16 standard site, which was imaged about every twenty minutes. We then performed a least squares fit to align the images to within a fraction of a pixel and constructed 7 mosaic swaths of the 1.50-micrometer images. Images in the other wavelengths were fitted to this mosaic to obtain image cubes in four wavelengths. The images at 1.99 micrometers required magnification correction before the image cubes could be constructed. Ratio images and principal component analysis show structures in the mare that may be individual flows or compositional units. We will present a color mosaic of the Moon showing different spectral units. References: [1] Kuhn J. R. et al. (1991) Astro. Soc. Pacific, 103, 1097-1108. Niiler T. A. Mendillo M. Modeling the Scattering of Moonlight in the Earth's Atmosphere Accounting for scattered moonlight is a crucial factor in imaging studies of the extended sodium atmosphere of the Moon. Observationally, on- and off-band data have been used to make this correction. We also have developed a new phase-dependent model of the atmospheric scattering of lunar light. The model is based on all earlier model by Krisciunas and Schaefer (1991), which accounted for both aerosol and molecular scattering, but also assumed the Moon to be a point source at distances that are large compared to the lunar radius (RM). Using the earlier model via the principle of linear super-position and the observed shape of the emitting fraction of the lunar disk, we are able to include in the model the non-axisymmetric nature of the scattering at points close to the Moon. In addition, our model allows predictions of scattered light at wavelengths outside the standard visual band. Initial comparisons with image datasets show good agreement at near-Moon distances (r <~4 R(sub)M), with less accuracy beyond. Vorder Bruegge R. W. Shoemaker E. M. The Clementine Mission to the Moon and NEA 1620 Geographos The Clementine Mission is being built and flown by the Naval Research Laboratory under the sponsorship of the Ballistic Missile Defense Organization of the United States Department of Defense in cooperation with NASA, and will explore the Moon and the near-Earth asteroid (NEA) 1620 Geographos with lightweight sensors developed by the Lawrence Livermore National Laboratory. A NASA Science Team for this mission was selected by way of a NRA in April 1993. The instrument suite includes imaging cameras that cover a spectral range from the near-ultraviolet to the mid-infrared, a laser ranger, and a charged particle telescope. To be launched in early 1994, Clementine will be in lunar orbit from February through May 1994, at which time it will depart the Moon for a flyby of 1620 Geographos in August 1994. This mission represents an outstanding opportunity for scientists interested in the Moon and asteroids. It is anticipated that the data returned from this mission will permit an assessment of (1) global lunar crustal heterogeneity at a resolution of less than 1 km, (2) lithologic heterogeneity of Geographos at a scale of 100 m or better, and (3) of surface processes on Geographos on the order of 10 m. This poster describes the basic mission of Clementine and some of the key scientific questions that will be addressed. Davies M. E. Colvin T. R. Meyer D. L. Nelson S. The 1993 Unified Lunar Control Network The objective of the unified lunar control network is to combine a series of control networks into one compatible network with its origin at the center-of- mass of the Moon and its coordinates referred to the mean Earth/polar axis system. The derivation of the initial unified coordinate system was described in Davies et al., 1987. This network gives new coordinates of points from Meyers, 1980 telescopic control network. The 1993 unified lunar control network modifies and extends the 1987 network. Coordinates of points north of the Apollo region have been recomputed based on Galileo images from the second Earth-Moon flyby. Mariner 10 points have been remeasured and integrated into the control. The Catalog of Lunar Positions, 1975 contains illustrations and coordinates of control points covering both the near side and far side. The coordinates were derived from Lunar Orbiter images. Lunar Position points have been identified on Galileo and Apollo pictures and added to the control network. Thus, there are far side points from the Apollo and Mariner 10 regions. In tables of coordinates, the 1159 (1987) network points have no prefix, the 90 Mariner 10 points have the prefix M, and the 172 Lunar Position points have the prefix P. References: [1] Davies M. E. et al. (1987) JGR, 92, B13, 14,177-14,184. [2] Meyer D. L. (1980) DMA, TR 80-001, Aerospace Center, St. Louis. [3] Catalog of Lunar Positions Based on The Lunar Positional Reference System (1975) Defense Mapping Agency, Aerospace Center, St. Louis. Sprague A. Deutsch L. Hoffman B. Hora J. Kozlowski R. Fazio G. Shivanandan K. Mercury: Mid-Infrared Imaging with MIRAC We have imaged Mercury with the Mid-Infrared Array Camera (MIRAC) and the Steward Observatory 2.3 meter telescope during the period of April 19-22, 1992. We obtained images in 23 wavelengths falling between 7.5 and 13.5 micrometers. A CVF with 1.8% resolution was used. For each wavelength 50 images were obtained utilizing a 53 msec integration time. This allowed us to minimize the effects of atmospheric turbulence and pointing instabilities and to maximize our spatial resolution. The array used was 20 by 32 pixels with a pixel scale of 0.67 arc seconds per pixel. Mercury was ~8.5 arc seconds in diameter and ~0.41 illuminated during this observing period. Because the wavelength region of this study is that of the strong Reststrahlen bands in silicates, we hope to obtain spatially resolved compositional differences on the surface. Under way are preliminary reductions at four wavelengths, which bracket and characterize a known emission minimum of olivine. Horn L. Nelson R. Weiss J. Smythe W. Evans M. Gatz E. Kuo S. Lane A. Linick S. Lopes-Gautier R. Manatt K. Martin W. Morris R. Ocampo A. Spradlin G. Wallis B. Yen C. Danielson G. Garvin J. Guest J. Hapke B. McClintock W. Simmons K. Russell C. Cruz M. Freitag J. Mitchler L. Hermes Global Orbiter: Mission to Mercury The planet Mercury, at the innermost limit of our Sun's planetary system, provides a technically challenging target for spacecraft exploration. Mariner 10, the only spacecraft to fly by Mercury, mapped about 40% of the planet's surface two decades ago. Over half of Mercury remains to be explored. Today, NASA is funding a study called the Hermes Global Orbiter (HGO) to place a spacecraft in polar orbit around Mercury. The Hermes Global Orbiter is a proposed Discovery-class mission. Hermes will be launched aboard a Delta II rocket in 1999 and will be placed in an elliptical polar orbit about Mercury. The spacecraft is a TRW Eagle Class C Lightsat that will be modified for Mercury's environment. Remote sensing measurements of the planet's surface, atmosphere, and magnetosphere will be performed. The preliminary Hermes payload consists of an imaging system, a laser altimeter/photopolarimeter system, an ultraviolet spectrometer and a magnetometer. Key mission goals include mapping the entire surface at 1 km resolution, characterizing the surface composition, texture, and topography, searching for water ice at the poles, characterizing the atmosphere, and constraining the interior structure. The Hermes mission will address important scientific questions regarding Mercury. These problems include possible volcanic origin for smooth plains, search for iron in the crust, identification and mapping of hypothesized icy polar caps, role of impact cratering in surface evolution, distribution of atmospheric constituents and their production and loss rates, nature of magnetic and gravitational fields, and their relation to the interior. This work was supported by the NASA Discovery Program. TITLE-ONLY PRESENTATIONS Lucey P. G. Bruno B. C. Domergue-Schmidt N. Mid-Infrared Spectroscopy of Lunar Surface Features: Some Upper Limits on Spectral Contrast Two observing runs were conducted to obtain mid-IR (7-12 micrometers) spectra of selected lunar regions. The instrument used was a scanning grating spectrograph equipped with a LN2 cooled HgCdTe detector. The spectral resolution was 50 nm yielding an average lambda/Delta lambda of about 200. The first of the two runs was aimed at attempting to quantify noncompositional influences on mid-IR spectra of the Moon by observing sites with similar compositions based on near-IR spectra or lunar geologic inferences, but at differing emission angles and temperatures. The sites chosen were a variety of mature mare surfaces. The photometric conditions during the run were quite poor but a small number of very high quality (better than 1% relative precision) ratio spectra were obtained. Of these ratios no spectral features were identified that could not be attributed to expected temperature differences between sites. There are undoubtedly minor compositional differences between the mare basalts observed, but no differences could be inferred from these rather high quality spectra. The second of the two runs was aimed at detecting spectral differences between areas of known or inferred extreme compositions. Unlike the former run, the maximum compositional contrast was sought. We observed (1) a pyroclastic deposit on the Aristarchus Plateau, which is very dark in the visible and exhibits a well-developed glass band in the near-IR; (2) two locations known or suspected to be enriched in olivine, the central peaks of Copernicus and a massif on the Aristarchus Plateau called Herodotus X; and (3) a variety of so- called lunar "red spots," which are UV anomalies and have been suggested, based on morphological grounds, to be possibly composed of silicic rocks types perhaps containing alkali feldspars or quartz. Unfortunately, while the conditions during this run were photometric (though humid) the instrument suffered from low sensitivity and data precision of only 10% was achieved. In these data no unequivocal spectral features were observed. These results, while negative, are consistent with earlier measurements, which showed that high precision is required to detect lunar mid-IR spectral features. This project has extended the range of compositional variation, precision, and spatial resolution for which mid-IR spectra have been obtained. Sprague A. L. Kozlowski R. W. H. Witteborn F. C. Graybill C. A. Mercury: Mid-Infrared Spectroscopy and Surface Composition Mid-Infrared (7.2-13.5 micrometers), groundbased observations of Mercury's surface with the High Resolution Faint Object Grating Spectrometer (HIFOGS) have resulted in high signal to noise ratio spectra, which can be interpreted in terms of Mercury's surface composition. Data were obtained during December 7-10, 1990 and July 9-12, 1992 observation periods at the Infrared Telescope Facility on Mauna Kea. Exceptionally good observing conditions prevailed on both the December 8 and 10, 1990 and July 11, 1992 observing runs. Data from these dates have been reduced and analyzed. Low humidity, relatively constant sky conditions, and good equipment performance combined to give very satisfactory datasets. All data were corrected for atmospheric absorptions in several ways for comparison. Spectral maxima at 8.0, 9.2, 10.8, and 12.6 micrometers and minima at 11.3 micrometers persisted in the Dec 8 and 10, 1990 data despite reduction with several different calibration stars and telluric correction techniques. Identification of features awaits further modeling and laboratory measurements. July, 1992 data show a persistent emission maximum at 8.0 micrometers and less overall spectral contrast than the Dec. 1990 data when different Mercurian longitudes were observed. We have also corrected for Si-O absorptions common in late stage infrared standard stars used for the telluric correction procedure. A thermal model of Mercury was used to determine a reasonable temperature for removal of the thermal continuum. Wednesday, October 20, 1993 VENUS ATMOSPHERE 3:30 - 5:50 PM North Ballroom Chair(s): R. Young D. Grinspoon Crisp D.* Meadows V. S. Allen D. A. Near-Infrared Observations of the Venus Surface and Lower Atmosphere Near-infrared (NIR) thermal emission from the night side of Venus at wavelengths near 1.0, 1.10, and 1.18-micrometers originates primarily from the surface and lowest scale height of the atmosphere. The Infrared Imaging Spectrometer at the Anglo-Australian Telescope was used to acquire moderate- resolution lambda/Delta lambda=400), NIR (0.9 to 2.5 micrometers) spectral image cubes of the night side during the 1991 and 1993 inferior conjunctions. Images extracted at wavelengths near 1.0, 1.1, and 1.18-micrometers reveal intensity variations produced by cloud opacity differences and elevated topographic features including Beta Regio, Phoebe Regio, and Aphrodite Terra^1. High-elevation regions are 20 to 50% darker than the surrounding plains because they are up to 40 K cooler. Large reductions in the surface emissivity at high-elevations (>50%), like those seen at microwave wavelengths, could also modulate the observed emission. These low-emissivity regions have been attributed to either conductive surface weathering products, or volume scatterers embedded within the surface. The first of these could have a distinct NIR signature, while the second should not. Scattering by the H2SO4 clouds limits the spatial resolution to 100 km, but atmospheric seeing further limits our spatial resolution to 250 km. Scattering by the clouds and continuum absorption by gases also attenuate the observed emission and reduce the contrast, but these data still resolve regions with elevation differences of 1 km. We used a radiative transfer model to simulate the observed intensities and contrasts for a variety of topographic elevations (0 to 6 km), surface emissivities (0.06 to 0.5), geoid temperatures (725 to 740 K), vertical temperature gradients (7 to 9 K/km) and H2O amounts. We find that the atmospheric temperature lapse rates are marginally stable (8K/km) in all regions occupied by large-scale topography. The H2O mixing ratios appear to decrease slightly with altitude, from about 45 ppmv at the surface to about 30 ppmv at the top of the first scale height (16 km), but this conclusion is very dependent on the assumed CO2 far-wing and continuum opacity near the surface. We find no evidence for large NIR surface emissivity variations (>10%) that are spatially correlated with known microwave low-emissivity regions. Jenkins J. M.* Steffes P. G. Twicken J. Hinson D. H. Tyler G. L. Atmospheric Profiles and Sulfuric Acid Vapor (H2SO4) Profiles from the October 1991 Magellan Orbiter Radio Occultation Experiments at Venus On October 5 and 6, 1991, dual-frequency radio occultation measurements of the Venus atmosphere were conducted on three successive orbits using the telecommunications system aboard the Magellan spacecraft and the 70-m DSN antenna at Tidbinbilla, Australia. The experiments probed between 56 degrees and 68 degrees N latitude and a solar zenith angle of approximately 112 degrees. The high radiated power (EIRP) from the spacecraft, plus the accurate pointing of the spacecraft antenna, made it possible to develop highly accurate profiles of atmospheric refractivity and absorptivity down to the 36 km level at 3.6 cm, and down to the 34 km level at 13 cm (above a mean radius of 6052 km). The refractivity profiles have yielded vertical profiles of temperature and pressure in the neutral atmosphere, while the combination of refractivity and absorptivity profiles have yielded profiles of H2S04(g). The analysis of the datasets includes error bars derived using the Standard Propagation of Errors. New techniques for processing radio occultation data were used to derive the profiles. These include a recently developed iterative method for solving the 2-way radio occultation problem, and the use of optimal digital filters for deriving absorptivity and refractivity profiles. The temperature and pressure profiles are compared to the Pioneer Venus probe measurements and the VIRA model. The three sets of profiles are compared to investigate atmospheric variability over the spatial scales spanned between occultations (the atmosphere rotated between 6 and 12 degrees between successive orbits.) Dalton J. B.* Pollack J. B. Grinspoon D. H. Bezard B. de Bergh C. Chlorine Abundances in the Deep Venus Atmosphere as Constrained by Near-Infrared Spectroscopic Observations Near-infrared spectra of Venus' nightside thermal emission have proven to be useful for obtaining unique and important information on gas abundances below the main clouds and the radiative properties of the clouds themselves. The emission originates within the hot, lower atmosphere of Venus. This emission is subsequently attenuated by scattering and absorption within the main cloud deck and is concentrated within spectral "window" regions where the lower atmosphere is most transparent. We have modeled new high-resolution spectra taken by Bezard and de Bergh in the 1.7-micrometer window using a radiative transfer program that allows for emission, absorption, and scattering by atmospheric gases and particles. The new spectrum represents a major improvement over previously obtained high- resolution spectra of the 1.7-micrometer window in that calibration procedures have enabled removal of telluric features, removal of scattered light from the dayside, and the establishment of accurate spectral intensities. This window is particularly sensitive to the abundances and distributions of water vapor and HCl in the lower atmosphere between 14 and 23 km altitude. Our current study provides improved estimates of the HCl abundances in this region of the deep Venus atmosphere. Widemann T.* Bertaux J. L. Moroz V. I. Ekonomov P. Vega-1 and 2 Decent Modules: In-situ Measurements of Ultraviolet Absorption and Relationship of SO2 with Presently Active Volcanism on Venus In 1985, the Vega-1 and Vega-2 spacecrafts launched two probing missions through the atmosphere of Venus. Onboard, the ISAV ultraviolet spectroscopy experiment, consisting in a UV light source absorbed by atmospheric constituents circulating freely into a tube, generated a wealth of spectra in the 220-400 nm range with an unprecedented vertical resolution (165 to 70 m). Due to a renewed interest towards SO2 variability and Venusian volcanism since the Magellan surface mapping, as well as the still unresolved question of UV albedo contrasts in the 300-400 nm region and the nature of the second absorber, those data were reanalyzed in light of a recent suggestion that croconic acid (C(sub)5O(sub)tH(sub)2), a polymer of carbon monoxide and a possible contaminant of H2SO4 cloud droplets, might absorb light and explain the spherical albedo of Venus in the 300-450 nm range. We present these results. The thermochemical equilibrium between SO2 in the lower atmosphere of Venus and the rocky minerals at 735 K implies a maximum value of about 20 ppmv for SO2. After the measurements of Pioneer Venus and Venera-12 in 1978 yielding about 150 ppmv, it has been argued that this high value, in contradiction with geochemical equilibrium, called for a recent source of SO2 in the atmosphere: active volcanoes on Venus. Present ISAV measurements in 1985 are quite compatible with surface equilibrium: at 22 km, we confirm a mixing ratio n(sub)so2 (z) = (37 +- 4) ppm and n(sub)so2 (z) = (34 +- 5) ppm, and therefore this controversial argument for presently active volcanoes on Venus is no longer valid. Grinspoon D. H.* Bullock M. A. Head J. W. III Resurfacing History of Venus and Implications for Atmospheric Evolution We will report on the results of a program of Monte Carlo modeling of impact cratering and volcanic resurfacing of Venus. Paying particular attention to the number and spatial distribution of partially embayed craters has allowed us to place first-order constraints on the history and rates of volcanic resurfacing. Combining this with informed, albeit rough, estimates of the volatile content of venusian lava flows allows estimates of the endogenous source flux of volcanogenic trace gasses. Implications for atmospheric and climate history will be discussed. Bullock M. A.* Grinspoon D. H. Pollack J. B. Perturbations to the Venus Greenhouse Effect Due to Mineral Equilibrium Buffering The high temperatures and pressures at the surface of Venus have led to the suggestion that surface/atmosphere interactions may play an important role in buffering CO2 and other volatiles [1,2]. If this is the case, perturbations to the atmospheric inventory of CO2 and other volatiles, caused by volcanic eruptions, may have a significant impact on the climate of Venus, and upon the stability of the greenhouse effect. Fegley and Treiman [3] have shown that the surface temperature and pressure on Venus coincide approximately with the P-T equilibrium of the calcite-wollastonite mineral reaction. If this reaction is indeed buffering atmospheric CO2 at the surface of Venus, and if reaction rates are small on geologic timescales, it is of interest to assess the impact it may have on the greenhouse effect. For example, it appears that a small increase in atmospheric CO2 would increase the magnitude of the greenhouse effect, and shift the mineral equilibrium to one of both a higher temperature and pressure. In this way, the buffering effect of the surface mineral equilibrium can produce an important feedback on the greenhouse effect. A simple nongray radiative-convective greenhouse model of Venus atmosphere has been developed that couples the radiative properties of the atmosphere to the volatile abundances that are achieved through mineral equilibrium reactions. Perturbations to the model are considered by increasing the abundances of trace gases, and by calculating the new radiation field that results from constraining the pressure and temperature at the surface to those predicted by several candidate volatile buffering reactions. Results that have implications for possible alternate Venus climate regimes will be discussed. References: [1] Urey, H. C. (1952) The Planets, Yale Univ., New Haven. [2] Adamcik, J. A. and Draper A. L. (1963) Planet. Space Sci. 11, 1303-1307. [3] Fegley, B. and Treiman A. H. (1991) Proceedings of the Chapman Conference of Comparative Study of Venus and Mars. Fegley B.* Lodders K. Klingelhofer G. Kinetics and Mechanism of Pyrite Decomposition on the Surface of Venus Ten years ago pyrite (py) decomposition was proposed as a source of reduced sulfur gases in the atmosphere of Venus. Here we present experimental results on py decomposition under Venus-like conditions. Cubes of natural py were cut and polished into slices of known weight and surface area. The slices were isothermally heated at atmospheric pressure in CO2 and CO2 gas mixtures for known time periods. Temperatures ranged from 390 degrees C to 531 degrees C, which spans the <0 to 9 km altitude range on the surface of Venus. The reaction rate was determined by measuring the weight loss of the reacted samples after removal from the furnace. The samples were characterized by X-ray diffraction, scanning electron microscopy, electron microprobe analysis, and Mossbauer spectroscopy. A cross section of a py sasnple partially reacted to pyrrhotite (po) in pure CO2 is shown in the figure, which appears in the hard copy. Py is at the top, po next, and epoxy at the bottom. Thermal decomposition of py to po follows contracting volume kinetics (shown in the figure for the 416 degree C runs). The slope of the line is the rate constant for py decomposition to po at 416 degrees C. The measured rate constants from 390 degrees C - 531 degrees C yield pyrite lifetimes of 292 days (9 km) to 17 days (0 km) on the surface of Venus. The activation energy derived from an Arrhenius plot of the rate constant data is 153 +/- 19 kJ/mole. Py decomposes to hematite (hem) + po in CO2 (96%)-CO (2%)-SO2 (2%) gas mixtures. This reaction apparently also follows contracting volume kinetics. Experiments done in CO2 binary gas mixtures with CO and SO2 also show pyrite decomposition to po +/- hem. The experimental results confirm the proposed pyrite decomposition in the Venus sulfur cycle. Acknowledgments. Supported by the NASA Planetary Atmospheres and VDAP Programs. Rosenqvist J.* Guibert B. Lellouch E. Encrenaz T. Absolute Wind Measurements and CO Distribution in the Mesosphere of Venus in 1993 Venus was observed on May 6 and June 1-2, 1993, in the J = 1-0 and J = 2-1 12CO and J = 2-1 13CO rotational lines at 115, 230, and 220 GHz respectively, using the IRAM 30-m radiotelescope at Pico Veleta, Spain. The telescope was equipped with a SIS receiver characterized by a system temperature between 400 and 800 K depending on the sky conditions and elevation of Venus. A dozen points were recorded as the Venus size was ~40" in May and ~25" in June. For each line, we used simultaneously two filter banks: a 512-channel 1-MHz resolution bank, and a 256-channel 100-kHz resolution bank or an autocorrelator 10-kHz resolution bank. At 230 GHz, the 10-kHz resolution corresponds to a line-of-sight velocity resolution of 13 m/s. However, the signal-to-noise (S/N) ratio will allow us to get a somewhat higher velocity resolution. These data will be presented and compared with previous observations of Venus performed in 1991 (Lellouch et al., DPS, 1992). As opposed to 1991 observations, a first-step study of the 13CO J = 2-1 line contrast shows an asymmetry between the western and eastern limbs. This contrast will be interpreted as CO variations along the disk and discussed in terms of Venus circulation near the 100-km altitude level. Roos M. C.* Drossart P. Lellouch E. Encrenaz Th. Carlson R. W. Baines K. H. Taylor F. W. Pollack J. The Thermal Structure of the Venusian Atmosphere Above the Cloud Tops from the Galileo NIMS Spectra The 4.3-micrometer and 4.8-micrometer CO2 absorption bands, recorded by the NIMS experiment during the GALILEO Venus flyby in February 1990, are being analyzed. The bands contain information about the thermal structure of the atmosphere above the cloud tops, i.e., from 70 to 90 kilometers. Taking the Venus International Reference Atmosphere (VIRA) thermal profiles as a starting point, the 4.3-micrometer CO2 band is fitted. Radiation from the cloud tops, that forms the continuum, has been analyzed in a previous paper (Roos et al., Planetary and Space Science, in press), and the resulting cloud temperatures are used in this analysis to fit to the continuum. An iterative method is used in order to obtain a least square fit to the observations. In the equatorial region, the derived thermal profile shows a tendency to be several degrees cooler than VIRA between 70 and 80 kilometers, and to be about 15 degrees hotter above. At high northern latitudes, corresponding to what is called the Northern Collar, the thermal profile appears isotherm above the cloud tops up to 75 kilometers, and then follows approximately the structure as found in the equatorial region. Maillard J. P.* Bezard B. Domisse L. Crisp D. Simons D. Spectro-Imaging of the Dark Side of Venus in the 1.27 Micrometer O2 Emission with an Imaging FTS The nightside of Venus has been imaged at 1.27 micrometers through a filter covering the 1delta-g band of oxygen airglow and an atmospheric thermal window, with an imaging Fourier Transform Spectrometer. The facility FTS installed at the f/35 focus of the CFH Telescope has been coupled to the observatory infrared camera (Simons 1993) built around a Hg:Cd:Te NICMOS3 256 x 256 infrared array, making it possible to record within a 24-arcsec aperture the spectra of all the points in the field of view with a sampling of the images of 0.33 arcsec/pixel. For the Venus data a spectral resolution of 1800 was achieved. A first test of the method was made on October 18, 1991, two weeks before the greatest western elongation of the planet (disk size 29 arcsec), while the last observation was conducted in March 8, 1993, when the planet was stationary with a 48 arcsec diameter. The two observations confirm the nonuniform and changing aspect of the O2 airglow. The October 1991 observation, which covered most of the darkside, shows an intense aiglow phenomenon with three bright spots (maximum area of ~2 arcsec diameter), with two at the longitude of the local midnight, 21 degrees N and 80 degrees S respectively, and with an intensity of 9.5 +/- 1 MRayleigh (after correction of the limb brightening). The region of minimum airglow is located at the equator and has a brightness of 3.5 +/- 1 MR. In March 8, 1993, only half of the darkside was viewed, with the aperture centered on the northern hemisphere. Comparatively, the activity was low, almost uniform over the accessible portion of the planet, with no significant bright spot, the maximum being however in the equator, consistent with Crisp's observations (1991) made at the AAT. A rotational temperature can be deduced from the spectra. A uniform temperature of 190 K +/- 10 is indicative of a constant altitude for the airglow. Observations over several consecutive nights are required to build a comprehensive model of the excitation mechanism to reconcile these opposite pictures. This method of spectro-imaging offers a powerful tool to address this question, providing there is enough telescope time. As a by-product, a mapping of the thermal emission in the 1.27-m window is also possible, with a spectrum of this emission on each point. Stewart A. I. F.* Alexander M. J. Pryor W. R. The Morphology of Far-UV Emissions from Carbon Monoxide in the Venus Thermosphere The Pioneer Venus Ultraviolet Spectrometer obtained extensive series of images of Venus in the 4th Positive system of carbon monoxide. The (0,1) band at 160 nm is excited by ionospheric processes and by resonance fluorescence of sunlight, while the (14,4) and (14,3) bands at 139 nm and 136 nm are produced by an accidental resonance between the (14,0) band and solar Lyman-alpha (the 136-nm feature is also contaminated by the photoelectron-excited 3P-5S multiplet of atomic oxygen). The 160-nm feature is emitted near 150 km, whereas the 139-nm and 136-nm features are emitted two or three scale heights lower. Thus the morphology of the images depends on the horizontal distribution of carbon monoxide in the lower and middle thermosphere. An earlier study (Alexander et al., 1993) of images in the 3P-3S resonance triplet of atomic oxygen at 130 nm showed that in the upper thermosphere, oxygen at high latitudes is more abundant in the afternoon than in the morning, but that on the equator this effect is less pronounced. Their interpretation invoked the dissipation of gravity waves generated in the middle atmosphere, a mechanism proposed by Alexander (1992) to explain the thermospheric superrotation. A preliminary inspection of the carbon monoxide images at all three wavelengths shows similar trends. Since the variations of the densities of CO and O across the sunlit disk are dynamically controlled, we will examine the images for evidence of changes in the dynamical regime with position and altitude. Referemces: Alexander M. J. (1992) Geophys. Res. Lett., 19, 2207. Alexander M. J. et al. (1993) JGR, 98, 10849. Schubert G.* Limonadi D. Newman W. I. Greeley R. Bender K. Wind Streaks and Atmospheric Circulation on Venus Magellan radar images reveal a variety of surface features on Venus attributed to aeolian processes. More than 5830 wind streaks have been identified and mapped for ~98% of the surface imaged by Magellan. The locations and orientations of the wind streaks enable the first global mapping of near- surface wind patterns on Venus and can be used to assess models of circulation in the lower atmosphere. Wind streaks on Venus can be grouped into two major classes. One class consists of streaks associated with parabolic ejecta blankets, while the second consists of all other streaks. Most of the streaks associated with parabolic ejecta blankets have lengths in excess of 100 km, exhibit predominantly westward azimuths aligned parallel to the symmetry axes of the parabolas, and are generally found within the borders of the ejecta blankets. These streaks have almost exclusively dark or mixed radar backscatter signatures and occur as striped patterns with cross-streak wavelengths of 3-10 km. The characteristics of these streaks suggest that they formed by the deposition of impact ejecta transported downstream (westward) by the retrograde superrotation of the atmosphere and influenced by near surface roll type convection cells generated by impact heating of the surface and oriented east-west by the zonal winds. Wind streaks not associated with parabolic ejecta blankets exhibit a broad predominantly equatorward azimuth orientation in the northern hemisphere. In the southern hemisphere these features exhibit a predominantly northeastward azimuth orientation. These observations support the presence of a Hadley cell circulation regime that extends to middle and perhaps high latitudes in both hemispheres. Westward streaks not related to parabolic ejecta blankets constitute a smaller fraction of the total streaks in both hemispheres; they indicate some influence of the westward zonal wind near the surface. Young R. E.* Walterscheid R. L. Schubert G. Pfister L. Houben H. Bindschadler D. L. Large Amplitude Stationary Gravity Waves Generated Near the Surface of Venus This study focuses on finite amplitude effects associated with large amplitude stationary gravity waves generated near the surface of Venus. Computed waves are forced near the Venus surface by periodic forcing. The height-dependent profiles of static stability and mean wind in the Venus atmosphere play a very important role in the evolution of the nonlinear behavior of the waves, just as they do in small amplitude (linear) wave solutions. Certain wave properties are qualitatively consistent with linear wave theory, such as wave trapping, resonance, and wave evanescence for short horizontal wavelengths. However, the finite amplitude solutions also exhibit many other interesting features. In particular, for forcing amplitudes representative of those that could be expected in mountainous regions such as Aphrodite Terra, waves generated near the surface can reach large amplitudes at and above cloud levels, with clear signatures in the circulation pattern. At still higher levels, the waves can reach large enough amplitude to break, unless damping rates above the clouds are sufficient to limit wave amplitude growth. Well below cloud levels the waves develop complex flow patterns as the result of finite amplitude wave- wave interactions, and waves are generated having considerably shorter horizontal wavelengths than that associated with the forcing near the surface. Nonlinear interactions can excite waves that are resonant with the background wind and static stability fields even when the primary surface forcing does not, and these waves can dominate the wave spectrum near cloud levels. A global map of Venus topographic slopes derived from Magellan altimetry data shows that slopes of magnitude comparable to or exceeding that used to force the model are ubiquitous over the surface. Hinson D. P.* Jenkins J. M. Magellan Radio Occultation Measurements of Atmospheric Waves on Venus Dual-frequency radio occultation experiments were performed on three successive orbits of the Magellan spacecraft in October, 1991, yielding high- quality data on the thermal structure, composition, and dynamics of Venus' atmosphere. We have recently begun an investigation of atmospheric waves based on analysis and interpretation of these data. Initial results include accurate vertical profiles of number density, pressure, temperature, and buoyancy (or Brunt-Vaisala) frequency at altitudes 35-95 km near 65 degrees N latitude. The temperature profiles exhibit wave-like oscillations throughout this altitude interval. The vertical wavelength is about 3-5 km. Wave amplitudes are smallest (a few tenths of a K) in the vicinity of the middle cloud layer (50- 55 km) where N approaches zero and the static stability is low. Larger amplitudes (1-4 K) are observed in regions of high stability (near 45 km and above 60 km). The atmosphere rotates roughly 10 degrees in longitude during one spacecraft orbit so that data from successive occultations provide constraints on the zonal wavenumber and important clues concerning wave identity and origin. Initial comparisons show a high degree of correlation among the temperature oscillations observed in the three profiles. POSTERS Lasica S. J. Colwell W. B. Pryor W. R. Stewart A. I. F. Ajello J. M. Solar Cycle Effect on the Venus Lyman-alpha Corona The Venus hydrogen Lyman-alpha corona was first observed by Mariner V (Barth, et al., Science, 158, 1675, 1967). Anderson (JGR, 81, 1213, 1976) used a multiple scattering code to fit the data with a two-component exospheric hydrogen density model. The lifetime of the Pioneer Venus Orbiter (PVO) has allowed a long-term examination of Venus emissions. We present preliminary analysis of PVO ultraviolet spectrometer (PVOUVS) Lyman-alpha (1216 Angstoms) observations from 1978 to 1990, which indicates that the coronal signal varies with solar activity. Our study spans a full solar cycle and reveals both a 27-day and a solar-cycle variation in the coronal signal. The coronal signal is noticeable above the interplanetary background beyond six Venus radii at solar maximum, but not at solar minimum. We compare the coronal signal with VF10.7, a solar Lyman-alpha and EUV index (Brace et al., JGR, 93, 7282,1988), and deduce the extent of solar-linked fluctuations in the exospheric hydrogen density itself. This research has been supported by the Pioneer Venus Guest Investigator Program. Na C. Y. Barker E. S. Stern S. A. Observations of Venus SO2 in 1993 Recent IUE observations of Venus made during greatest elongations in January and June 1993 are presented. We obtained a total of 10 exposures ranging from 0.5 sec to 20 min using the LWP camera during 9-12 January 1993. During 8-11 June 1993, eight additional exposures were obtained using both the large and small apertures of the IUE telescope. A preliminary analysis of the January 1993 observations indicates that the amount of sulfur dioxide at the cloud tops of Venus is around ~50 ppb, and the scale height at the same altitude is ~3 km. These values are similar to the results from previous IUE observations made in 1987 and 1988 [Na et al. 1990], thus indicating that there have been no recent changes in SO2 abundance above the clouds of Venus. On the same days as the IUE observations in January and June 1993, we obtained groundbased, spatially resolved spectra and images of Venus in the 3000- to 3650-Angstrom region. These spectra will also be used to estimate the SO2 abundance, and a comparison between the groundbased measurements and the IUE observations will be presented. Pryor W. R. McClintock W. E. Stewart A. I. F. Hord C. W. Barth C. A. Taylor M. H. Models of CO 4th Positive Band Emissions on Venus A sounding rocket flown on October 7, 1983 obtained a 3-Angstrom resolution disc-integrated spectrum of Venus covering 1300-1800 Angstroms, dominated by the CO 4th Positive Band System. The Galileo Ultraviolet Spectrometer Experiment obtained 6-Angstrom resolution spectra from 1150-1900 Angstroms on February 10,1990. We model these data with a synthetic spectrum similar to work by Durrance et al. (Geophys. Res. Lett., 7, 222, 1980) and Swanson (Ph.D. Thesis, U. of Colorado, 1980). The model includes solar fluorescence, self- absorption in CO, and absorption by CO2. These spectra are of sufficiently high spectral resolution to allow us to constrain the vertical profile of CO. The Galileo spectra also contain spatial information on the CO distribution. Kliore A. J. Pioneer Radio Occultation Observations of the Temporal Diurnal and Spatial Variability of the Venus Middle Atmosphere The Pioneer Venus Orbiter Spacecraft was used to obtain approximately 250 vertical profiles of temperature in the middle atmosphere of Venus (about 40 km to 90 km). These measurements were taken during periods of high solar activity (1978-83), low solar activity (1984-86), and again high solar activity (1989-91). Comparisons have been made between data taken during these different observation periods as well as between nightside and dayside observations during a certain observation period. Nightside observations yield higher temperatures at altitudes above 80 km for all three observation periods, but it must be pointed out that dayside and nightside measurements are also separated in time and latitude due to the nature of the radio occultation measurements. Encrenaz Th. Lellouch E. Cernicharo J. Paubert G. Gulkis S. First Detection of the 183 GHz Water Line in the Atmosphere of Venus On January 13, 1991, the water vapor transition at 183 GHz was detected on Venus, with the 30m millimeter antenna of IRAM at Pico Veleta, under extremely dry weather conditions. The Doppler shift of Venus was 3 MHz. The contribution due to the terrestrial mesospheric water was removed by using a spectrum of Saturn recorded the same day. The observed Venus absorption line has a depth of 1.6% and a FWHM of about 15 MHz. We used both the depth and the width of the line to constrain the water vapor abundance above the clouds and the temperature at the level z=95 km. A preliminary reduction indicates that the observed profile is compatible with a H2O mixing ratio of 1.5 ppm above the clouds, and a minimum temperature of 140 K at z=95 km. If this result is confirmed, the water vapor abundance would then be smaller than the value retrieved 4 months earlier (August 30, 1990) from the 226 GHz HDO line (Encrenaz et al., 1991), by a factor of about 2. Hansell S. A. Wells W. K. Hunten D. M. A Search for Venus Lightning A search for lightning in the clouds of Venus has been made. Coronagraphic image reduction optics were used to image the dark side of the planet during 25 February-14 March, 1993. The University of Arizona Obserratory's 1.5 meter telescope (Mt. Bigelow Station) was used for this program. A frame transfer mode CDD camera was used to record the images at rates as high as 18.8 frames per second. The integrated live time for the observations was approximately 5 hours over 5 separate observing nights. A total of over two hundred thousand images was obtained during this period. The images were taken through a narrow band filter centered at 777.4 nm. This corresponds to the light emission maximum for simulated lightning in a model Venus atmosphere (Borucki W. J. et. al) 1983, Geophys. Res. Lett., 10, 961}. Calculations show that an Earth- typical lightning flash in the clouds of Venus would be roughly equivalent to magnitude 11 for the duration of one of our single observations at 777.4 nm (Williams et al (1982} Icarus, 52, 166. The duration of a single flash is on the order of 1 ms. Multiple strokes are commonly separated in time by approximately 200 ms. Thus, any one lightning strike will likely be in only one image. Reslts of our ongoing search for lightning in the atmosphere of Venus will be presented. Strangeway R. J. Russell C. T. Ho C. M. Detection of Lightning Generated ELF Waves at Very Low Altitude in the Venus Nightside Ionosphere The Orbiter Electric Field Detector (OEFD) detected many ELF/VLF wave bursts at low altitudes within the nightside ionosphere during the early phase of the Pioneer Venus mission. These wave bursts were attributed to atmospheric discharges (i.e., lightning). ELF waves, at 100 Hz, appear to be whistler-mode waves escaping from the atmosphere in regions of enhanced ambient magnetic field and reduced plasma density, known as "holes." These waves are thought to have propagated some distance from the source in the surface-ionosphere waveguide. At higher frequencies VLF bursts are also detected. These bursts are more localized in both local time and altitude, and further occur at frequencies within the propagation stop-band for cold plasma wave modes. These signals are probably a "prompt" response to lightning occurring below the spacecraft. In the late summer and fall of 1992 the Pioneer Venus Orbiter once again acquired data at low altitudes (<200 km) within the Venus nightside ionosphere. The last two orbits of the spacecraft provided data at altitudes never previously sampled by the orbiter, 130 km. Intense 100-Hz wave bursts were detected near periapsis on these orbits. Although spacecraft interaction effects cannot be completely discounted, many features of the signals indicate that the bursts are naturally occurring. The scale height for decay of the signals is consistent with that expected for propagation in a collisional plasma, and these wave bursts could be the direct measurement of electromagnetic radiation from lightning discharges within the atmosphere of Venus. TITLE-ONLY PRESENTATIONS Krasnopolsky V. A. Pollack J. B. Modeling of OCS, CO, and H2SO4 Profiles in Venus' Troposphere Recent measurements of profiles of seven species (Pollack et al., 1993) and H2SO4 (Jenkins et al., 1993) stimulate chemical kinetics modeling of Venus' troposphere. Results of the first phase of this work, which includes (1) the problem of diffusion and condensation of H2SO4 and H2O in the cloud layer and (2) modeling of OCS and CO profiles, are given here. A method to solve a coupled diffusion/condensation problem for the H2SO4/H2O system has been developed. The model shows that the water vapor mixing ratio decreases from 18 ppm at 47 km (the lower cloud boundary) to 3 ppm at 60 km. Sulfuric acid vapor mixing ratio has a maximum of 9 ppm at 47 km and becomes smaller than 1 ppm above 51 km. The H2O/H2SO4 ratio in the cloud particles varies from 0.076 at 47 km to 0.87 at 60 km. The column rate of photochemical formation of H2SO4 is equal to 5 x 10^12 cm^-2s^-1. Considering problem (2), a model has been developed that reproduces fairly well the measured OCS and CO mixing ratios and their gradients for rate coefficients of the reactions SO3 + CO --> CO2 + SO2, k = 10^-11 exp(-9500/T) cm^3/s SO3 + OCS --> CO2 + (SO)2, k = 10^-11 exp(-12000/T) cm^3/s for the downward flux of CO of 4.5 x 10^12 cm^-2s^-l at 47 km, and for eddy diffusion of (5 +/- 2) x 10^3 cm^2/s below 35 km. The model predicts the OCS mixing ratio increasing from 4.5 ppm at 33 km to 28 +/- 2 ppm near the surface, while that of CO decreases from 35 ppm at 47 km to 25 ppm at 36 km and to 9 ppm at 20 km. Thursday, October 21, 1993 OUTER PLANET SATELLITES I 8:30 - 10:00 AM South Ballroom Chair(s): A. Coustenis T. S. Zwier Muhleman D. O.* Grossman A. W. Slade M. A. Butler B. J. Titan's Radar Reflectivity and Rotation Extensive new observations of Titan will be made in August, 1993 in which we will attempt to measure the disk averaged radar reflectivity over the entire orbit except for 3 days at inferior and superior conjunctions where confusion from emission of Saturn complicates the measurements. The distribution of reflectivity in three longitudinal strips parallel to the rotation axis will be measured each day and an accurate orientation of the projection of Titan's rotational axis will be determined, including the direction of rotation. These measurements should provide a better representation of the surface variations than the individual measurements for the oppositions of 1989, 90, 91, and 92. The global mean reflectivity for the the hemisphere facing the Earth around eastern elongations has been accurately measured. These results appear in the table below as the average of all data, including individual days where the measured reflectivities were greater than 50 percent. (The reflectivities shown are in the "expected sense", i.e., right circular polarization transmitted, left circular received). The Titan longitudes in the table are the averages for each year assuming that Titan rotates synchronously with its orbit about Saturn. We feel that the second value of 0.217 +- 0.094 is our best estimation of Titan's reflectivity at a wavelength of 3.5 cm. This value is, of course, completely inconsistent with the presence of a significant body of liquid hydrocarbons in this hemisphere of Titan. A global hydrocarbon ocean deeper than a few 100 meters would have a reflectivity less than 0.02 and the reflection would be specular, not uniform over the satellite's disk as shown by our measurements. Statistically low reflectivities were measured on 2 days. It is possible that these were caused by miss-pointing the transmitting antenna at Goldstone, for which we have no definitive check. The reflectivities on days when they were high are statistically significant (over 2 sigmas from mean) but those of the low days are not significantly different from the mean. A reflectivity of 0.2 is similar to that of Callisto, but smaller than that of Europa and Ganymede. This suggests that Titan's surface is "dirty" ice but the nature of the dirt is unknown. Sears W. D.* Lunine J. I. Greenberg R. Nonsynchronous Rotation of Titan A satellite on an eccentric orbit experiences a zero net tidal torque when the rotation is slightly faster than synchronous, if there is no significant permanent asymmetry (Greenberg & Weidenschilling 1984, Icarus, 58, 186). We examine various models for tidal dissipation including Darwin's model and linear and quadratic extensions to MacDonald's model. Using these models we calculate that for Titan the equilibrium nonsynchronous rotation period should be in the range 15.82 to 15.93 days. This range includes the 15.911 day period recently determined by radar observations (Muhleman et al. 1992 BAAS, 24, 954). We also examine the allowed maximum permanent asymmetry and find that it is possible to have a topographic feature which is large enough to stand above the calculated 100-200 m layer of photochemical debris so as to show up as a radar bright spot and still be small enough so as to not cause tidal locking. A single circular ice feature which was 100 m high providing the maximum allowed asymmetry would have a radius of approximately 180 km. Sagan C.* McDonald G. D. Thompson W. R. The Titan Haze Revisited--Origin and Production Rates Recent coupled radiative transfer-coagulation/sedimentation models (see McKay and Owen, 1992, ESA SP-388, 185) have borne out the consistency of the optical properties of Titan tholin with the scattering properties of Titan's haze [first noted by Sagan et al. 1983, BMS, 15, 843 and Thompson 1984 (see Sagan, Thompson and Khare 1992, Acc. Chem. Res., 2,5 286)]. However, assumed production efficiencies and a confusion over the very different roles of plasma and high-energy electrons have led to the impression that Titan tholin provides insufficient mass flux for these models. Yet photochemical polymers (specifically polyacetylene and poly-HCN) that could provide sufficient particle mass flux have optical properties divergent from those inferred for Titan's haze (McKay et al. 1993, BAAS, 24, 949). By quantifying the yield of Titan tholin produced in plasma discharge (as we have previously done for gas-phase species) and by reassessing the relative roles of plasma and high-energy electrons, we offer a partial solution to this quandary. Titan tholin yields are large--comparable to the mass yields per unit energy for all gas-phase species combined. Combining this result with the flux of high-energy electrons (which are delivered to Titan's upper stratosphere with much higher efficiency than plasma electrons) indicates that sufficient mass flux to satisfy scattering/sedimentation models can easily be provided by Titan tholin alone. In reality, however, both radiation-chemical tholin and photochemical polymers or heteropolymers must contribute to Titan's haze population, and the challenge for further models lies in finding solutions consistent with the production rates and optical properties of both kinds of materials. Hutzell W. T.* McKay C. P. Toon O. B. Two Dimensional Modeling of Titan's Stratospheric Haze We will present preliminary results £rom a two-dimensional model of Titan's stratospheric haze, which we have constructed. Our haze model is adapted from an earlier one dimensional haze model (Toon et al. 1992 and Hutsell et al. 1993). The microphysical processes assume spherical haze particles, and include sedimentation and coagulation whose efficiency is determined by aerosol electrical charging. For haze transport, we have constructed an arbitrary but consistent pole to pole circulation cell that is based on estimations of the stratospheric circulation of Titan (Flasar et al. 1981, and Flasar and Conrath 1990). The haze model calculates the meridional cross sections of the aerosol site distribution and their optical properties. The resultant geometric albedo and the latitudinal varariation of brightness are also determined. These modeling results will be used to investigate the effects of constant and time varying wind fields on Titan's stratospheIic haze. Erom the time varying calculations, we hope to determine possible explanations for the northern-southern hemispheric brightness contrast (Sromovsky et al. 1981 and Caldwell et al. 1992) and geometric albedo variations (Lockwood et al. 1986). Lemmon M. T.* Tomasko M. West R. A. Titan's Haze as Fractal Aggregates: Observational Constraints Recent work regarding Titan's atmosphere has suggested that the presence of fractal aggregate aerosols in the main haze layer may explain the observations of strong forward scattering and high polarization. We model the aerosols as cluster-cluster aggregates with a fractal dimension near 2 and monomer size near 0.06 micrometers. The single scattering properties of these polymer