Search

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.icarus.2009.06.003 Byline: Sarah J. Morrison, Peter C. Thomas, Matthew S. Tiscareno, Joseph A. Burns, Joseph Veverka Keywords: Saturn; Satellites; Surfaces; Satellites; Shapes; Satellites; Dynamics Abstract: High-resolution images from the Cassini Imaging Science Subsystem (ISS) show parallel sets of grooves on Epimetheus and Pandora. Grooves have previously been observed on other satellites and asteroids, including Phobos, Gaspra, Ida, Eros, and minor occurrences on Phoebe. Sets of parallel grooves are so far observed only on satellites known or likely to be subject to significant tidal stresses, such as forced librations. Grooves on asteroids and on satellites not subject to significant forced librations occur in more globally disorganized patterns that may reflect impacts, varying internal structures, or even thermal stresses. The patterns and individual morphologies of grooves on the tidally-affected satellites suggest fracturing in weak materials due to tidal stresses and forced librations. Author Affiliation: Cornell University, Center for Radiophysics and Space Research, Space Sciences Building, Ithaca, NY 14853, USA Article History: Received 23 April 2009; Revised 1 June 2009; Accepted 3 June 2009

Author(s): Carolyn C. Porco (corresponding author) [1]; Emily Baker [1]; John Barbara [2]; Kevin Beurle [3]; Andre Brahic [4]; Joseph A. Burns [5]; Sebastien Charnoz [4]; Nick Cooper [3]; Douglas [...]

Spacecraft have successfully landed on the Moon, Venus, and Mars, and have penetrated the atmosphere of Jupiter. On 2001 February 12, the Near-Earth Asteroid Rendezvous (NEAR) Shoemaker spacecraft landed on the surface of the asteroid (433) Eros after a year of observations in orbit about the asteroid. NEAR Shoemaker was not designed to land on an asteroid, complicating the design of operations needed to accomplish this feat. However, the NEAR Shoemaker team wanted to attempt a landing after the year of orbital operations that consumed most of the remaining spacecraft fuel, operations funding, and planned Deep Space Network tracking. This would be a fitting end to the mission, and it would be possible to obtain images at much greater resolution during the descent than could be obtained from orbit. The operations were more successful than the NEAR Shoemaker team had hoped, obtaining 70 high-resolution images during the descent and two weeks of gamma-ray spectrometer data from the surface after the successful soft landing. Key Words: asteroids; Eros; surface, asteroids; orbits.

NEAR multispectral imaging was obtained at seven wavelengths (450-1050 nm) to characterize hemispheric and regional color properties of Eros. The highest-resolution whole-disk data, 180 m/pixel, were obtained during the last pre-orbit insertion sequence on 12 February 2000. The same areas were imaged again in color at 10-20 m/pixel from high orbit in March-April 2000, and selected targets have been studied in color at resolutions as high as 4 m/pixel from low orbit. Whole-disk spectra are in close agreement with ground-based observations. These and the disk-resolved measurements show little variation in visible-wavelength color, but they do reveal spatial variation of several percent in the 950-nm/760-nm reflectance ratio, used here as a proxy for depth of the 1-[micro]m olivine-pyroxene absorption band. After photometric correction to i = 30 [degrees], e = 0 [degrees] using both a Hapke correction and a modified empirical phase function, the disk-resolved images show reproducible spatial variations in albedo and 950-nm/760-nm reflectance ratio. The northern hemisphere exhibits average reflectances at I = 30 [degrees], e = 0 [degrees] of 0.136 [+ or -] 0.007 at 760 nm and 0.115 [+ or -] 0.006 at 950 nm, equivalent to geometric albedos of 0.30 [+ or -] 0.02 and 0.26 + 0.02, respectively. There is more than factor of 2 spatial variation in reflectance, but only about 10% variation in the 950-nm/760-nm reflectance ratio. Reflectance and color ratio are highly correlated, with the highest reflectances in discrete, sharp-edged patches on slopes exceeding 20 [degrees], consistent with material being exposed by downslope movement. Eros is also conspicuously deficient in small, bright, spectrally distinctive craters which are found on the other two S asteroids, Gaspra and Ida, imaged by spacecraft at close range. Eros exhibits a larger range of albedos than other S asteroids, but its color variations are much more subtle: variation in the 950-nm/760-nm reflectance ratio with 760-nm reflectance is several times less than on those asteroids or in the lunar maria. Of the different mechanisms possibly responsible for reflectance and color differences on Eros, spatial differences in the extent of regolith aging by space weathering are most consistent with Eros's observed properties. However, the effects of this process are both qualitatively and quantitatively different than in the lunar maria. Key Words: asteroids; spectral properties; surfaces, Eros; NEAR.

Earth-based spectral measurements and NEAR Shoemaker magnetometer, X-ray, and near-infrared spectrometer data are all consistent with Eros having a bulk composition and mineralogy similar to ordinary chondrite meteorites (OC). By comparing the bulk density of 433 Eros (2.67 [+ or -] 0.03 g/[cm.sup.3]) with that of OCs (3.40 g/[cm.sup.3]), we estimate the total porosity of the asteroid to be 21-33%. Macro (or structural) porosity, best estimated to be ~20%, is constrained to be between 6 and 33%. We conclude that Eros is a heavily fractured body, but we find no evidence that it was ever catastrophically disrupted and reaccumulated into a rubble pile. Key Words: Eros; meteorites; surfaces, asteroids; interiors.

NEAR Shoemaker imaging of the asteroid Eros has revealed a variety of surface forms that probably reflect underlying structures. They range from global to local in scale. The most prominent positive topographic feature is a long ridge, Rahe Dorsum, which extends over 18 km and has the morphology of a thrust fault. Two large, ancient trough systems are also present. Grooves are ubiquitous, commonly occur in suborthogonal sets, and have linear or scalloped outlines. The grooves on Eros are similar in morphology and scale to those on other small bodies such as Gaspra, Ida, and Phobos and are likely to have a similar mode of origin, namely drainage of regolith into underlying fractures. Models that use groove dimensions as indicators of regolith thickness imply depths of up to several tens of meters. Many craters have been subjected to structural control, resulting in squared outlines. We find no evidence that specific groove sets are associated with individual impact events; however, evidence exists that preexisting structures have been reactivated by later impact activity. The large scale of Rahe Dorsum and many grooves on Eros's surface suggests that the asteroid is a largely coherent, but fractured body. Key Words: asteroids; Eros; impact processes; regoliths; asteroid surfaces; tectonics.

Images from the Mars Orbiter Camera on the Mars Global Surveyor spacecraft show crater populations extending the size distribution down to around 16 m. The results suggest a broad range of surface ages. Lava flows within the Arsia Mons caldera are estimated to be no more than 40-100 million years old, and it is suggested that on Mars, volcanism is a continuing process.

Images of Europa from the Galileo spacecraft show a surface with a complex history involving tectonic deformation, impact cratering, and possible emplacement of ice-rich materials and perhaps liquids on the surface. Differences in impact crater distributions suggest that some areas have been resurfaced more recently than others; Europa could experience current cryovolcanic and tectonic activity. Global-scale patterns of tectonic features suggest deformation resulting from non-synchronous rotation of Europa around Jupiter. Some regions of the lithosphere have been fractured, with icy plates separated and rotated into new positions. The dimensions of these plates suggest that the depth to liquid or mobile ice was only a few kilometers at the time of disruption. Some surfaces have also been upwarped, possibly by diapirs, cryomagmatic intrusions, or convective upwelling. In some places, this deformation has led to the development of chaotic terrain in which surface material has collapsed and/or been eroded. Key Words: Europa; Europa geology; Galileo; Galilean satellites; jovian satellites.

New, high-quality Galileo images of Io's dayside and Jupiterlit nightside, examined for diurnal changes in violet albedo patterns and diurnal changes in hemispheric color, do not provide any evidence for significant nighttime condensation of the satellite's thin S[O.sub.2] atmosphere. In particular, the Galileo images do not show the major change in Io's average hemispheric color between day and night that was reported by Buratti et al. (1995, Icarus 118, 418-422) in a study of Voyager data. The lack of detectable nighttime frost deposition in the Galileo images is not surprising, given the tenuous, 'patchy' nature of Io's S[O.sub.2] atmosphere and the associated difficulty in producing optically thick nighttime frost layers. Key Words: atmospheres, dynamics; Galileo spacecraft; Io; surfaces, satellite.

Galileo's Solid State Imaging camera recorded six images at wavelengths from 0.41 to 0.99 [[micro]meter] of Europa's trailing hemisphere ([approximately]1.6 km/pixel resolution) during the G1 orbit (1st orbit - target Ganymede) of the nominal mission. We have photometrically corrected these data and extracted spectra representing Europa's diverse geologic terrains. The goals of the analysis of these spectra are (1) to determine whether Europa's geologic units differ spectrally from one another, (2) to determine the number of color components necessary to explain Europa's dark material spectral behavior, and (3) to examine how europan dark materials may change in color with time. Our data indicate that europan dark spots, lineaments and triple band side materials represent a single 'dark' and reddish endmember component. We see no evidence for more than one dark endmember; however the dark endmember seen in these G1 data is probably not a 'pure' exposure of europan dark material. Mottled terrains, brighter lineaments, and aging triple bands can be modeled by a mixture of bright plains materials and the dark component. Exposures of europan dark materials may thus be brightening with time, eventually blending into the surrounding bright plains. Key Words: Europa; Galileo; multispectral imaging.

Clear-filter Viking images, and an accurate numerical model of the shape of Phobos, have been used to determine this satellite's photometric properties. A global-average Hapke function derived from disk-resolved data confirms previous indications that Phobos has a strong opposition surge. Photometrically corrected images were mosaicked into an albedo map; most of the resulting normal reflectances are in the range 0.06-0.10, and the brightest region on Phobos is the northeast rim of the crater Stickney, the portion of that rim with the highest concentration of grooves. Globally, there are three albedo classes, reasonably separated geographically: (1) Bright material is to the east and south of Stickney, corresponding approximately to the locations on Phobos having the highest, 'bluest' visible/near-IR ratio (Murchie et al. 1991). (2) The darkest material is to Stickney's west, correlating with material having an intermediate visible/NIR ratio (Murchie et al.'s 'bluish gray' unit). (3) Intermediate-albedo material dominates the anti-Stickney hemisphere, corresponding to material with a lower visible/NIR ratio (Murchie et al.'s 'reddish gray' unit). A search for variations in phase behavior across Phobos' surface shows few such effects overall, limited to isolated areas: (1) Stickney's floor darkens with increasing phase faster than average Phobos. This crater floor is both slightly more backscattering and significantly rougher than the global average; the latter effect may be related to the slumping hinted at in low-resolution images of Stickney. (2) We confirm that in many cases, the contrast between the bright rims of small craters and grooves and their surroundings drops noticeably with increasing phase (phase angles in use = 10 [degrees] - 50 [degrees]). However, these bright rims, overall, display a diversity of photometric behavior and are the most heterogeneous areas on Phobos in terms of regolith properties. (3) We confirm that dark deposits in the floors of smaller craters darken faster with increasing phase than their surroundings (cf., Goguen et al. 1978) and find that these deposits are more backscattering than average Phobos. Isolated regions with unusual phase behavior occur on Phobos but not on Deimos (Thomas et al. 1996). The variable photometric properties of Phobos' isolated craters and grooves, and the association of global albedo features with Stickney, reinforce interpretations that Phobos' regolith is emplaced and modified by discrete cratering events and is not mixed horizontally by extensive downslope creep as is apparently the case on Deimos.

We test the hypotheses that the lunar opposition effect is due to shadow-hiding, coherent backscatter, or some combination of the two phenomena. Hapke's photometric model is extended to include M. I. Mishchenko's (1993, Astrophys. J., 411,351-361) description of the coherent backscatter opposition effect. The model was fit simultaneously to V-filter ([Lambda] = 0.55/[[micro]meter]) whole-disk photometry of the Moon over many phase angles (2 [degrees] [less than or equal to] [Alpha] [less than or equal to] 143 [degrees]) and to disk-resolved observations of the lunar surface over a broad range of incidence, emission, and phase angles. The shape of the Moon's opposition surge is accurately represented by the combination of a narrow coherent-backscatter peak whose effect is most strongly defined at [Alpha] [less than] 2 [degrees] and a very broad shadow-hiding peak, best defined over [Alpha] [less than] 20 [degrees]. The relative angular widths of the shadow-hiding and coherent backscatter contributions can be used to estimate the grain sizes most important for scattering at small phase angles. Scatterers with sizes comparable to the wavelength of light are required for coherent backscattering and are present in the smallest size fraction of lunar soil, but are predicted only for a model in which coherent backscatter causes the narrow component and shadow-hiding the wide one. The amplitude of the shadow-hiding opposition effect measures grain transparency and our solution implies that the lunar regolith particles responsible for the shadow-hiding contribution behave like opaque grains, consistent with complex structure of lunar regolith particles at much larger scales than those which control the coherent backscatter contribution. Most other airless planetary surfaces exhibit opposition surges whose broad components also appear to be caused by shadow-hiding. We propose that the submicron sized grains that control coherent backscatter do not contribute to the shadow-hiding opposition surge because the minute shadows cast by these grains are illuminated by their own forward-scattered light. We suggest that coherent backscatter cannot be responsible for the broad component of the Moon's opposition surge because such forward scattered light would illuminate all particle shadows and thereby eliminate the shadow-hiding contribution.

Models for the shape and global photometric behavior of 243 Ida derived from Galileo images have been used to model the asteroid's telescopic lightcurves. We find excellent agreement between the calculated and observed times of lightcurve extrema, confirming the sidereal rotation period of 4.633632 [+ or -] 0.000007 hr determined by Binzel et al. (1993. Icarus 105, 310-325.). The shapes and amplitudes of the synthetic and observed lightcurves agree with each other to within [approximately equal to]0.05 mag, and the average offset in absolute photometry between model and observation is only 0.01 mag. This consistency confirms the accuracy of the shape model derived for Ida by Thomas et al. (1996. Icarus 120, 20-32.) and supports the model of the asteroid's photometric behavior developed by Helfenstein et al. (1996. Icarus 120, 48-65.).

Galileo flyby images of 951 Gaspra show a crater population dominated by fresh craters several hundred meters in diameter and smaller. They must represent a production population because their spatial density is low (few overlaps) and because degraded craters are underabundant; equilibrium may be attained at diameters near to or below the resolution limit of the best image. We have counted, measured, and classified craters from the highest resolution, 'high phase' image, which shows >600 craters in 90 [km.sup.2]. The differential population index (0.2-0.6 km) for the fresh, obvious craters is very 'steep' (-4.3 [+ or -] 0.3). It probably reflects the index of asteroidal projectiles; it is much steeper than the theoretical value of -3.5 for collisional equilibrium. Gaspra's crater population differs from that observed on Phobos but resembles those observed on the Moon and Mars at these sizes (consistent also with the near-Earth asteroid population). Gaspra's fresh craters are superposed on a landscape that appears 'smoothed' at a vertical scale of hundreds of meters. Some 'soft,' subdued crater-like features, commonly [approximately]>500 m across, are visible. Some of these are associated with the linear grooves on Gaspra and may be endogenic features. Many others are probably pre-existing impact craters deeply blanketed or otherwise much degraded. Gaspra's largest-scale shape is highly irregular, perhaps faceted. The biggest facet exceeds the largest crater (relative to body radius) ever observed on a satellite or expected from collisional fragmentation models. Facets cannot be successive crater-forming impacts; later scars would have destroyed earlier ones. Far-encounter images show a more lumpy than faceted visage of Gaspra; two craters are [approximately]3 km in diameter, but not even half the radius of Gaspra. We expect that Gaspra was created by collisional fragmentation of a larger parent body. Its gross configuration may reflect collisional spallation of the parent. Certainly, megaregolithic processes of reaccumulation and blanketing and/or shaking are evident, due to subsequent sub-catastrophic collisions. Gaspra's subdued craters peek through the effects of the last such collision. That smoothed surface has been cratered ever since by the steep production function, which, however must become shallower again below 10 m. Since the overall density of fresh craters is low, Gaspra must be relatively youthful. Scaled to a calculated 0.5 Gyr age for bodies of its size, based on asteroid collision models and assuming that Gaspra does not have metallic strength, its cratering lifetime is [approximately]0.2 Gyr, with large modeling uncertainties. The cumulative volume of all visible craters could create a regolith only

Dactyl was discovered in solid state imaging (SSI) data on February 17,1994, during the long playback of approach images from the Galileo spacecraft's encounter with the asteroid 243 Ida. Forty-seven images of the Ida-Dactyl pair were obtained. A detailed search for other satellites was made. No confirmed detections were made, all other candidate features being consistent with radiation hits. We deduce a manifold of osculating two-body orbits that approximate Dactyl's motion over the observed orbital arc depending on the assumed mass of Ida. At the time of Galileo's encounter, Dactyl was found to be 85 km from the center of Ida, moving at [approximately]6 m [center dot] [sec.sup.-1] in the same direction as Ida's retrograde spin. The inclination of its orbit is [approximately]172 [degrees] in Ida's equatorial system (IAU definition). It was not possible to obtain a definitive orbit or measure of Ida's mass from the observed motion even though supplemental techniques (search for Dactyl's shadow on Ida, changes in angular diameter and brightness, and attempts to determine the spin of Dactyl) were explored. The influence of Ida's irregular gravitational field and solar perturbations on two-body motion are evaluated and found to be undetectable in the observed orbital arc. These effects may, however, strongly influence the motion over orbital time scales. Limits to the value of Ida's gravitation parameter, GM, are derived. A robust lower limit, GM > 0.0023 [km.sup.3] [center dot] [sec.sup.-2], is obtained by requiring Dactyl's orbit to be bound. Hubble Space Telescope observations, which show no evidence of Dactyl on a hyperbolic orbit, excludes values of GM in the range 0.00216 < GM < 0.0023 [km.sup.3] [center dot] [sec.sup.-2]. An upper limit, GM < 0.0031 [km.sup.3] [center dot] [sec.sup.-2], deduced by requiring that the orbital motion has a long lifetime in a realistic approximation to Ida's gravitational field, is illustrated with numerical calculations. Ida's mass is therefore constrained to the range 4.2 [+ or -] 0.6 x [10.sup.19] g, which, together with a volume of 16,100 [+ or -] 1900 [km.sup.3] (Thomas P. C., M. J. S. Belton, B. Carcich, C. R. Chapman, M. E. Davies, R. Sullivan, and J. Veverka 1996. Icarus 120, 20-32.) yields a bulk density of 2.6 [+ or -] 0.5 g [center dot] [cm.sup.-3] (Belton, M. J. S., C. R. Chapman, P. C. Thomas, M. E. Davies, R. Greenberg, K. Klaasen, D. Byrnes, L. D'Amario, S. Synnott, T. V. Johnson, A. McEwen, W. Merline, D. R. Davis, J-M. Petit, A. Storrs, J. Veverka, and B. Zellner 1995. Nature 374, 785-788.).

We provide an overview of the execution, data, and results of the solid state imaging (SSI) experiment at the encounter of the Galileo spacecraft with the asteroid 243 Ida. Ninety-six images of the asteroid, representing 18 time samples during a rotation period (4.633 h), were transmitted to Earth as a result of the UT 1993 August 28.70284 encounter. This provided coverage of [approximately]95% of the surface and achieved ground resolutions as high as 25 m/pixel. Coverage of most of Ida's surface is available in four colors, with limited regions in five colors, at resolutions up to 105 m/pixel. A natural satellite of Ida, called Dactyl, was discovered in a prograde (with respect to Ida's spin), near-equatorial, orbit moving slowly ([approximately] 6 m/sec) with a separation of 85 km from Ida. Ida's shape is highly irregular; by comparison, Dactyl's global topography is quite smooth. The best fit ellipsoid to Ida's shape has principal dimensions 59.8 x 25.4 x 18.6 km, mean radius 15.7 km, and volume 16,100 [+ or -] 1900 [km.sup.3]. Dactyl's mean radius is only 0.7 km. Ida's spin axis (right ascension: 348.76 [degrees] [+ or -] 7.5 [degrees]; declination: 87.10 [degrees] [+ or -] 0.4 [degrees]; J2000) was found to align with the principal axis of inertia to within the error of measurement. This is consistent with a homogeneous density distribution. Dactyl's rotation rate is unknown, but its long axis was pointed in the direction of Ida at the time of observation, suggesting synchronism of its orbital motion and spin. Constraints on Dactyl's orbit yield 4.2 [+ or -] 0.6 x [10.sup.19] g for Ida's mass and 2.6 [+ or -] 0.5 g/[cm.sup.3] for its bulk density. Unless Ida's bulk porosity is exceptionally high, Ida has moderate to low NiFe content. Subtle color variations across the surface of Ida are associated with fresh craters, but, unlike the case for Asteroid 951 Gaspra, are not correlated with topographic features such as ridges. This difference may be a reflection of a deeper and/or more mobile regolith on Ida. Dactyl's spectral reflectance is similar to, but quantitatively distinct from the surface of Ida itself. This difference may reflect compositional differences between Dactyl and Ida, which in turn may have originated in an only partially differentiated Koronis parent body. Results on the origin, collisional history, and geology of Ida and Dactyl are the subject of many of the papers in this special issue. There is general agreement that these asteroids originated in the catastrophic breakup of the Koronis parent body and that the formation of asteroid-satellite systems may be relatively common in such events. The age and collisional history of the pair present a dilemma: using standard interpretations of the cratering record on Ida's surface, an age > 1 byr. is indicated. However, the lifetime of Dactyl against collisional disruption is many times less than this. Novel ideas are presented concerning the collisional history of these two small objects that may resolve this dilemma. These ideas result from analysis of the geological record on the surface of Ida, Dactyl, and, by comparison, Gaspra - all of which are examined in this special issue. The execution of the Galileo flybys of Gaspra, Ida, and Dactyl provide important lessons for future flybys of small bodies. We present our views on the limitations faced by the Galileo imaging experimenters and indicate how future missions can be made more quantitative and productive through the application of innovative electronic control systems and detector technology.

A control network containing 64 points identified on Galileo images obtained during the August 1993 flyby was used to determine the direction of Ida's north pole. We find the right ascension and declination (in J2000 coordinates) to be [[Alpha].sub.0] = 348.76 [degrees] [+ or -] 7.5 [degrees], [[Delta].sub.0] = 87.10 [degrees] [+ or -] 0.4 [degrees]. This result agrees reasonably well with earlier determinations from groundbased observations of Ida's lightcurve. We confirm that the rotation is retrograde, but we lack the time base to improve the rotation period of 0.1930680 days determined from lightcurve analysis (Binzel, R. P., S. M. Sivan, P. Magnusson, W. Z. Wisniewski, J. Drummond, K. Lumme, M. A. Barucci, E. Dotto, C. Angeli, D. Lazzaro, S. Mottola, M. Gonano-Beurer, T. Michalowski, G. De Angelis, D. J. Tholen, M. Di Martino, M. Hoffmann, E. H. Geyer, and F. Velichko 1993. Icarus 105, 310-325).

Seventeen positive relief features [approximately]45-150 m across are identified as probable blocks in Galileo high-resolution images of Ida. Their presence provides direct evidence for regolith retention on asteroids. The spatial distribution, maximum size, and integrated volume of the blocks are consistent with those of blocks associated with craters on the Earth, the Moon, Phobos, and Deimos. The concordance suggests that the features are impact ejecta fragments and that cratering mechanics on Ida, an object of average diameter [approximately]32 km, are similar to those applying on previously studied rocky bodies. The blocks that lie within or near the rims of craters Lascaux and Mammoth were likely mobilized in the low-velocity tail portion of the excavation flow that formed those craters. A few blocks located near smaller craters may have migrated some distance away from their source, possibly by impact-induced spallation, hopping, rolling, and/or sliding. Some blocks on Ida could be surviving fragments from the Koronis parent body, accreted after its breakup. The lifetime of [10.sup.2]-m sized boulders against collisional disruption is estimated to be in the [10.sup.8]-[10.sup.9] year range, consistent with ages considered for the largest and oldest craters on Ida. Extrapolation of successful ejecta scaling laws to other asteroids suggests that blocks [approximately]15 and 70 m across could be present on Dactyl and Gaspra, respectively (in both cases too small to be identified in available Galileo images). Blocks 100 m in size could be present on 433 Eros, and km-sized megablocks on 4 Vesta.

Craters on Ida have a size distribution consistent with equilibrium under conditions of crater saturation over the range from 100 m to 1 km in diameter. The density of craters is high and they follow a -3 differential slope. The craters show a full spectrum of degradational morphologies at both large and small sizes. Craters >2 km in diameter (and especially those >5 km) are more prevalent on one part of Ida; a sub-equilibrium, production population is seen on the other part, which rises steeply and reaches equilibrium for diameters