[Show abstract][Hide abstract] ABSTRACT: Among the many Cassini ISS (Imaging Science Subsystem) images of
Enceladus are a few severely-underexposed, motion-blurred images that
were acquired on “boresight-drag” events on the closest
flybys. During boresight-drags, ISS is statically aimed at a point that
intercepts the predicted path of Enceladus’ across the sky. The
ISS Narrow angle (NAC) and Wide Angle (WAC) cameras are repeatedly
triggered together in hope of serendipitously capturing a close-up
“BOTSIM” image-pair of the body as it passes. Because the
events are so fast, the surface footprints and lighting geometry cannot
be predicted in advance - a cascade of images are just quickly shuttered
at the minimum 5 msec exposure. On each of four boresight-drags, surface
images were captured. However, the two most recent (image-pair
W/N1669812043 in November 2010 and W/N1713106405 in April 2012,
respectively) were poorly illuminated -- three of four images only in
Saturnshine. Despite their poor signal quality, they are rare images of
Enceladus’ surface obtained with spatial resolutions better than a
few meters/pixel. Careful use of Fourier filtering and spatial
reconstruction techniques was needed to eliminate image noise and
residual electronic banding that was not removed during routine
radiometric calibration of the images. Fourier motion debluring
techniques were then applied to correct for significant motion smear.
Images W/N1669812043 (55.1°N, 20.2°W) are in old cratered
terrain, inside a prominent 23 km sized impact crater along the rise of
its updomed floor. They show a system of parallel ~250m wide mesas
trending around the dome’s circumference. Smooth detritus
inundates mesas and valleys near the dome summit and the mesa surfaces
are otherwise mantled with regolith that is finely cratered down to the
~2 m/pixel NAC resolution limit. W/N1713106405 (66.9°S, 29.5°W)
show the chaotically fractured margin of the active South Polar Terrain
- an area divided by parallel ridges and troughs with relatively smooth
flanks and valley floors. Quasi-linear arrangements of ice-blocks, each
block tens of meters or smaller, are found mostly near ridge-tops.
[Show abstract][Hide abstract] ABSTRACT: More than a dozen small (<150 km mean radius) satellites occupy
distinct dynamical positions extending from within Saturn's classical
rings to the orbit of Dione. The Cassini mission has gradually
accumulated image and spectral coverage of these objects to the point
where some generalizations on surface morphology may be made. Objects in
different dynamical niches have different surface morphologies.
Satellites within the main rings display equatorial ridges. The F-ring
shepherding satellites show structural forms and heavily cratered
surfaces. The co-orbitals Janus and Epimetheus are the most lunar-like
of the small satellites. Satellites occupying libration zones (Trojan
satellites) have deep covering of debris subject to downslope transport.
Small satellites embedded in ring arcs are distinctively smooth
ellipsoids that are unique among small, well-observed Solar System
bodies and are probably relaxed, effectively fluid equilibrium shapes
indicative of mean densities of about 300 kg m‑3.
[Show abstract][Hide abstract] ABSTRACT: a b s t r a c t The nucleus of comet Tempel 1 has been investigated at close range during two spacecraft missions sep-arated by one comet orbit of the Sun, 5½ years. The combined imaging covers $70% of the surface of this object which has a mean radius of 2.83 ± 0.1 km. The surface can be divided into two terrain types: rough, pitted terrain and smoother regions of varying local topography. The rough surface has round depressions from resolution limits ($10 m/pixel) up to $1 km across, spanning forms from crisp steep-walled pits, to subtle albedo rings, to topographic rings, with all ranges of morphologic gradation. Three gravitationally low regions of the comet have smoother terrain, parts of which appear to be deposits from minimally modified flows, with other parts likely to be heavily eroded portions of multiple layer piles. Changes observed between the two missions are primarily due to backwasting of scarps bounding one of these probable flow deposits. This style of erosion is also suggested by remnant mesa forms in other areas of smoother terrain. The two distinct terrains suggest either an evolutionary change in processes, topo-graphically-controlled processes, or a continuing interaction of erosion and deposition.
[Show abstract][Hide abstract] ABSTRACT: a b s t r a c t On February 14, 2011 Stardust-NExT (SN) flew by Comet Tempel 1, the target of the Deep Impact (DI) mission in 2005, obtaining dust measurements and high-resolution images of areas surrounding the 2005 impact site, and extending image coverage to almost two thirds of the nucleus surface. The nucleus has an average radius of 2.83 ± 0.1 km and a uniform geometric albedo of about 6% at visible wavelengths. Local elevation differences on the nucleus reach up to 830 m. At the time of encounter the spin rate was 213° per day (period = 40.6 h) and the comet was producing some 130 kg of dust per second. Some 30% of the nucleus is covered by smooth flow-like deposits and related materials, restricted to gravitational lows. This distribution is consistent with the view that the smooth areas represent material erupted from the subsurface and date from a time after the nucleus achieved its current shape. It is possible that some of these eruptions occurred after 1609 when the comet's perihelion distance decreased from 3.5 AU to the current 1.5 AU. Much of the surface displays evidence of layering: some related to the smooth flows and some possibly dating back to the accretion of the nucleus. Pitted terrain covers approximately half the nucleus surface. The pits range up to 850 m in diameter. Due to their large number, they are unlikely to be impact scars: rather they probably result from volatile outbursts and sublimational erosion. The DI impact site shows a subdued depression some 50 m in diameter implying surface properties similar to those of dry, loose snow. It is possible that the 50-m depression is all that remains of an initially larger crater. In the region of overlapping DI and SN coverage most of the surface remained unchanged between 2005 and 2011 in albedo, photometric properties and morphology. Significant changes took place only along the edges of a prominent smooth flow estimated to be 10–15 m thick, the margins of which receded in places by up to 50 m. Coma and jet activity were lower in 2011 than in 2005. Most of the jets observed during the SN flyby can be traced back to an apparently eroding terraced scarp. The dust instruments detected bursts of impacts consistent with a process by which larger aggregates of material emitted from the nucleus subsequently fragment into smaller particles within the coma.
[Show abstract][Hide abstract] ABSTRACT: a b s t r a c t We present results from our study of the Stardust-NExT NAVCAM images of Comet 9P/Tempel 1, in which we analyze the dust coma and derive the locations and directions of 11 isolated jets detected around clos-est approach. Seven of the jets form a cluster that is associated with terraced terrain near the equator. Other jets arise from the nightside of the nucleus, having been in darkness for as long as 4 h, indicating that thermal lags continue to drive activity long after sunset. We compare the coma features observed here to those seen during the Deep Impact encounter on the previous apparition, and argue that much of the isolated jet activity is associated with steep slopes and the edges of smooth areas. We estimate that the cluster of jets produces 7–20% of the total dust in the coma, indicating that isolated sources play a significant role in the comet's activity. We measured an average dust production rate A(a)fq = 42 ± 6 cm at an approach phase angle of 79°, corresponding to a dust mass loss of approximately 264 kg s À1 . Our analysis also indicates that the Stardust-NExT spacecraft did not pass through any dust jets during the flyby.
[Show abstract][Hide abstract] ABSTRACT: a b s t r a c t Observations from the second encounter of Comet 9P/Tempel 1 by the Stardust-NExT spacecraft provide an improved shape model and rotational pole for the nucleus (Thomas, P.C. et al. . Icarus 222, 453– 466) that allows us to greatly improve our knowledge of its rotational evolution beyond that outlined earlier in Belton et al. (Belton, M.J.S. et al. . Icarus 213, 345–368). Model light curves are shown to fit observations at both perihelia with a single pole direction indicating that polar precession during a single perihelion passage is small. We show that the rotational phasing associated with observations taken far from perihelion in the previous work was incorrectly assessed by approximately half a cycle leading us to a significant reassessment of the evolution of the non-gravitational torques acting on the nucleus. We present an updated spin rate profile (torque model) for the 2005 perihelion passage and show that retardation of the spin rate well before perihelion is no longer a required feature. With the exception of the spin rate before the 2000 perihelion passage, the evolution of rotational rates through the three most recent perihelion passages is largely unaffected as is the prediction of the rotational phase of the comet's nucleus at the Stardust-NExT near-perihelion encounter. We find a spin rate of 209.4 ± 0.01°/d likely applies in the quiescent period before the 2000 perihelion, a 0.2% change, and that the rotational period shortened by 12.3 ± 0.2 min during the 2000 perihelion passage. We present an analysis of Stardust-NExT time-series photometry that yields a spin rate near 213.3 ± 0.8°/d at the time of encounter. An application of the 2005 torque model suggests that, while roughly similar, the torques were probably weaker during the 2011 perihelion passage.
[Show abstract][Hide abstract] ABSTRACT: The Stardust-NExT mission imaged the impact crater formed by the DI
impact probe. The size of the crater is estimated to be 150 m to 200 m
in diameter based on the observed crater and disruption of ejecta as
seen in images from the DI flyby in 2005.
[Show abstract][Hide abstract] ABSTRACT: On 14 February 2011, the Stardust spacecraft performed its second
cometary flyby, following the successful completion of primary mission
to return coma samples from comet 81P/Wild 2. Stardust NExT passed
within 200 km of the nucleus of comet 9P/Tempel 1, first visited by the
Deep Impact spacecraft in 2005, providing an opportunity to make in-situ
measurements of a second cometary coma with the same dust detection
instruments. We present new results from the Dust Flux Monitor
Instrument at the Tempel 1 flyby which detected bursts of impacts
consistent with measurements at Wild 2, interpreted as the fragmentation
of larger aggregates of material emitted from the nucleus into smaller
particles within the coma.
[Show abstract][Hide abstract] ABSTRACT: On February 14, 2011 Stardust-NExT flew by Comet Tempel 1, the target of
the Deep Impact mission in 2005, obtaining dust measurements and
highresolution images of areas surrounding the 2005 impact site, and
extending coverage to almost two thirds of the nucleus surface. No
large-scale morphologic or photometric changes have occurred in the
region of common coverage during the past six years. The most
significant changes took place along the edges of a prominent smooth,
flow-like feature, the margins of which receded in places by up to 50
meters. The amount of material removed is sufficient to account for
about a percent of the estimated mass loss of the comet per perihelion
passage. The DI impact site shows a subdued impact scar some 50 meters
in diameter implying surface properties similar to those of dry, loose
snow. Smooth areas cover about half of the newly imaged face of the
comet. On the nucleus smooth regions are restricted to gravitational
lows, consistent with the view that they originated from eruptions onto
the surface and date from a time after the nucleus had achieved its
present shape. Much of the surface of the comet displays evidence of
pervasive layering. Two types of layers may be involved: some related to
the smooth flows and others possibly dating back to the accretion of the
nucleus. Coma and jet activity were lower in 2011 than in 2005. As was
the case at the Wild 2 flyby in 2004, the dust instruments detected
bursts of impacts consistent with a process by which larger aggregates
of material emitted from the nucleus fragment into smaller particles
within the coma.
[Show abstract][Hide abstract] ABSTRACT: New data obtained during the Stardust-NExT mission encounter has led to
a substantial increase in the accuracy of the spin pole determination
and shape model of the nucleus of comet 9P/Tempel 1. This has allowed
improved determination of the locations of the mini-outbursts  that
were discovered, but not imaged, around the Deep Impact mission
encounter. The expanded mapping of the surface accomplished by
Stardust-NExT includes high-resolution (~20 m/pixel) coverage of one of
these areas under good lighting conditions thus providing views of
candidate features caused by outburst eruptions and, possibly, a surface
feature associated with the pre-eruption state. Plausible candidates for
post-eruption features, located within 10 of the nominal location, are
several irregularly shaped, fresh looking, depressions with central
peaks. Also seen in the same region are two relatively smooth positive
elevation features that we postulate may be associated with the
pre-eruptive state. We compare these features to the full range of
depression morphologies seen elsewhere on the imaged surface and discuss
the possibility that most of the depressions seen on the surface were
caused by outburst activity and subsequently eroded by sublimation. The
mechanism responsible for outburst activity proposed earlier by Belton
and Melosh  is briefly discussed in light of these new findings.
[Show abstract][Hide abstract] ABSTRACT: Tectonic shear along tiger stripes and other fractures at the South Pole of Enceladus is a key mechanism in current theories about active cryovolcanism and the tectonic evolution of the South Polar Terrain (SPT) region. Frictional heating of tiger stripe faults has been proposed to provide energy to drive cryovolcanic venting . Tidal flexing of Enceladus's lithosphere, which has been proposed to control periodic opening and closing of volcanic fissures , results in oscillating stresses which, when resolved along tiger stripes, have both normal and shear components. The efficacy of frictional heating and strike-slip motion varies with the compressive or extensional intensity of the resolved normal stress component. Transpression is a state of stress that occurs when the surface experiences simultaneous compression and shear, while transtension occurs when it experiences simultaneous tension and shear. Significant lateral displacements along strike are expected to accompany shear failure in the tiger stripes , and large lateral offsets can accumulate over time due to the process of "tidal walking"  in which alternate periods of transtension and transpression on each tidal half-cycle provides the means by which one side of the rift can progressively ratchet laterally from the other. The identification and measurement of shear indicators along South Polar fractures may provide critical constraints on the thickness of Enceladus' lithosphere  and its possible librational history . Transpression and transtension may also have been important in tectonism related to non-synchronous rotation [7,8]. In this study, we extend our earlier work [9,10] to identify and analyze geomorphologic indicators of transtension and transpression within tectonic features of the active SPT region on Enceladus. In addition to numerous strike-slip offsets visible along fractures, our Rosetta stone for interpreting these geomorphologic indicators is a prominent tectonic stepover structure at the distal ends of the tiger stripes on the Saturn-facing hemisphere (Fig. 1). Near Damascus Sulcus, the stepover features a (transtensional) releasing bend that is identified with a prominent negative flower structure. The opposite end of the stepover, near Cairo Sulcus is identified by a parallel system of narrow, sub-kilometer high ridges that curve ~90 that appear to form a (transpressional) restraining bend. Suggested axial discontinuities along the three visible tiger stripes in Fig. 1 each exhibit an apparent CCW rotational twist suggesting that they have all been deformed in response a left-lateral system of shear parallel to overall trend of the fractures and often mirrored by quasi-parallel, curvilinear patterns in the ropy plains between the tiger stripes. The presence of a distinct, 750m-high pair of narrow ridges centered on the Damascus discontinuity suggests that narrow dorsa within the tiger stripes (i.e. "shark fins") may be transpressional features, like those we've interpreted to form within restraining bends. This is also suggested by the morphological resemblance of stratigraphically old, degraded medial dorsa within sections of Baghdad Sulcus to positive flower structures [9,10]. References:  Nimmo, F. et al. (2007). Nature 447, 289-291;  Hurford, T. et al.  Helfenstein, P. et al. (2011). Tectonism and Terrain Evolution on Enceladus: I. Tectonic features and patterns (submitted);  Helfenstein, P. et al. (2011). Tectonism and Terrain Evolution on Enceladus: II. Interpretation and Hypotheses (submitted). FIGURE 1 (next page): Tectonic unit map (left) and corresponding 28m/pixel photomosaic (right) show tectonic patterns of the South Polar Terrain region. Proposed axial discontinuities along tiger stripes are shown by curved arrows labeled "A" (dashed arrows indicate least certain interpretation). From top to bottom are Damascus, Baghdad, and Cairo Sulci. Stepover extends from a releasing bend near Damascus Sulcus to a restraining bend at the corresponding end of Baghdad.
2011 Enceladus Focus Group Meeting, SETI Institute, Mountain View, CA; 05/2011
[Show abstract][Hide abstract] ABSTRACT: The Amor mission will rendezvous and land at the triple Near-Earth Asteroid system (153591) 2001 SN263 and execute detailed, in-situ science investigations. The spacecraft reaches 2001 SN263 by using a two-year DeltaVEGA (DeltaV-Earth Gravity Assist) trajectory with a relatively low launch C3 of 33.5 km2/s2. Rendezvous will enable reconnaissance activities including global and regional imaging, shape modeling, system dynamics, and compositional mapping. After landing, Amor will conduct in-situ imaging (panoramic to microscopic scale) and compositional measurements to include elemental abundance. The main objectives are to 1) establish in-situ the long-hypothesized link between C-type asteroids and the primitive carbonaceous chondrite (CC) meteorites, 2) investigate the nature, origin and evolution of C-type asteroids, and 3) investigate the origin and evolution of a multiple asteroid system. The mission also addresses the distribution of volatiles and organic materials, impact hazards, and resources for future exploration. Amor is managed by NASA Ames Research Center in partnership with Orbital Sciences, KinetX, MDA, and Draper with heritage instruments provided by Ball Aerospace, JHU/APL, and Firestar Engineering. The science team brings experience from NEAR, Hayabusa, Deep Impact, Dawn, LCROSS, Kepler, and Mars missions. In this paper, we describe the science, mission design, and main operational challenges of performing in-situ science at this triple asteroid system. Challenges include landing on the asteroid components, thermal environment, short day-night cycles, and the operation of deployed instruments in a low gravity (10^-5 g) environment.
[Show abstract][Hide abstract] ABSTRACT: High-resolution mosaics of the leading-hemisphere of Enceladus were acquired for the first time by Cassini's ISS Narrow Angle Camera (NAC) during two close flybys, one on November 21, 2009 and another on May 18, 2010, respectively. Additional imaging of the leading-hemisphere is planned for a close (2800 km) flyby on August 13, 2010 (orbit 136). Low resolution imaging obtained earlier in the Cassini mission showed that the leading hemisphere is heavily modified by tectonism, but the extent to which the tectonic styles on the leading hemisphere might be similar to those elsewhere on Enceladus was unclear. The new mosaics show that the leading side is subdivided into distinct geological provinces that exhibit different cratering histories and diverse tectonic styles. The highly tectonised terrains are bounded by a prominent broad annulus of grooved and striated terrains that ranges from about 60 km to over 140 km in width. It surrounds a complex arrangement of tectonic structures, including a conspicuous province near 30°N, 90°W of curvilinear ridges and approximately orthogonal-trending ridged-troughs. Among the most significant new findings is a region near 10°S, 60°W of terrain that is covered by a distinct pattern of rounded, rope-like sub-parallel ridges. These ridges, in appearance and scale, are remarkably similar to ropy (funiscular) plains materials that previously have been found only in the South Polar Terrain region adjacent to active tiger stripes. We suggest that the pattern of ropy ridges on the leading hemisphere arose from a similar style of tectonic deformation that produced the South Polar funiscular plains - a terrain that is closely related to possible folding and tectonic spreading associated with the tiger stripes. These features may thus record an ancient episode of South Polar style tectonism and volcanism near the equator.
[Show abstract][Hide abstract] ABSTRACT: We investigate color variations of Prometheus and Calypso using recent high resolution calibrated Cassini Imaging Subsystem (ISS) images with the Narrow Angle Camera (NAC) broadband filters UV3 (338 nm) , GRN (568 nm), and IR3 (930 nm) in comparison to those of Pandora, Janus, Epimetheus, and Telesto examined previously (Morrison et al., 2009, DPS 45.04). The average UV3/IR3 ratio values are as follows: Prometheus- 0.47, Pandora- 0.53, Janus- 0.57, Epimetheus- 0.58, Telesto- 0.95, and Calypso- 0.77 with formal uncertainties < 0.01 due to noise. While both Prometheus and Calypso continue the trend of increasing blueness toward the E-ring, Calypso differs in color from Telesto with a lower UV3/IR3 ratio and a mean UV3/GRN ratio value of 0.7 compared to Telesto's mean UV3/GRN value of 0.9 despite both being Trojans of Tethys. Calypso's surface also exhibits wider color and albedo variations than on Telesto and distinct UV3-bright units undergoing downslope transport with UV3/IR3 ratio values of 0.05 greater than the mean. Calypso's mean global CLR (611 nm) albedo at normal incidence and 35° phase is 13% higher than Telesto's, but its albedo range is broader and overlaps Telesto's. The color differences between these two moons may reflect varying degrees of surface burial and contamination from ring particles.
[Show abstract][Hide abstract] ABSTRACT: The scientific goals of the Stardust-NExT mission to comet 9P/Tempel 1 include imaging 25% of the area imaged by the Deep Impact mission at a resolution of at least 80 m/pixel looking for changes to the surface, imaging, at the highest possible resolution, of the artificial crater created during the Deep Impact encounter (349W, -26), and imaging coverage of a portion of the terrain not yet seen. Fulfilling these goals required detailed knowledge of the spin state of 9P's nucleus approximately one year ahead of the nominal time of encounter, Feb 14, 2011. Since the encounter occurs 33 days after perihelion passage it was necessary to develop a model of the acceleration of the comet's spin state through perihelion passage in order to make a prediction. Determination of the spin state to the required accuracy was achieved in January, 2010, and was accomplished using observations taken from the Hubble Space Telescope and from ground-based observatories. This information was then used as the basis of trajectory correction maneuver in Feb, 2010. We describe the independent development of two spin rate acceleration models based on 13 years of observations of the rate and phase of spin of the comet's nucleus through two perihelion passages obtained with the Hubble and Spitzer Space Telescopes, many ground-based telescopes within the organization of a International World-Wide campaign (Meech et al., 2005), and the Deep Impact mission. We present a visualization of the encounter that shows the anticipated coverage of the surface of 9P/Tempel 1 that should be attained and emphasizes the proportion of the surface previously seen with the Deep Impact cameras and the new terrain that will be covered. Meech, K.J, and 210 coauthors. 2005. Deep Impact: Observations from a World-Wide Earth-based Campaign. Science 310, 265-269.
[Show abstract][Hide abstract] ABSTRACT: The Deep Impact mission created an artificial crater on Comet 9P/Tempel 1, but did not succeed in obtaining imagery of the crater. The Stardust-NExT mission in now on course to visit Tempel 1 on Feb. 14, 2011 and, among other objectives, hopefully capture images of the crater. This goal requires that the spacecraft arrive at a time that ensures that the comet is oriented such the crater is well lit and visible from the Stardust-NExT spacecraft. We report on two independent efforts to model the rotation state of the comet in order to predict the comet orientation and estimate the optimal time of arrival for the spacecraft. We have used a massive data set of over 2500 ground-based photometric measurements spread over 11 years, plus Hubble Space Telescope measurements in 1997, 2004 and 2009, as well as Spitzer Space Telescope observations in 2004. This large and diverse data set of photometric light curve data reveal that the spin rate of Comet 9P/Tempel 1 has increased by statistically significant amounts at each of the past three perihelion passages. We believe that this spin rate increase is due to torques associated with cometary out-gassing activity, and we show that the torque history required to match the observations requires two or more discrete torque-producing source regions acting on the nucleus. We have developed an approximate torque profile for the 2011 encounter that we use to estimate the optimal spacecraft arrival time.
[Show abstract][Hide abstract] ABSTRACT: Introduction: Before lunar samples were returned to Earth, it was shown that linear polarization curves and albedo measurements of the moon and asteroids could be used to predict the regolith surface texture of these bodies (1). The prediction that lunar mare rego- lith is texturally similar to finely pulverized lava basalts was confirmed by returned lunar samples (2). Our ultimate objective is to use Cassini Imaging Science Subsystem (ISS) Narrow Angle Camera (NAC) images of icy Saturnian satellites - acquired through broadband color and polarization filters - to measure the linear polarizations and corresponding al- bedos of icy terrains and to characterize implied sur- face textural properties. In this preliminary study, we demonstrate the sensitivity of ISS NAC polarization images for the analysis of icy satellite surfaces by test- ing if these images can detect Umov's law, a well- known inverse relationship between the linear-polariz- ation and albedo of regolith-covered surfaces (3). Ia- petus is our test satellite because its terrain provides a broad range of albedos over which we expect linear polarization to systematically vary. In particular, we will use Cassini Regio's stark contrast with Iapetus' remaining brighter surface to demonstrate Umov's law. Our hypothesis is that the linear polarization of all materials on Iapetus will vary inversely and mono- tonically with their corresponding albedo. Method: Each ISS NAC polarization observation is a three-image sequence obtained for 0, 60, and 90 degree orientations of the electric field vector, respect- ively. The polarization filters are used in combination with broadband color filters from a tandem filter wheel, so that changes in polarization with wavelength can be studied. Each image sequence is obtained at approximately constant phase angle. First, the images are radio-metrically calibrated so that pixel DN value represents radiance factors. Camera point- ing information is updated using control points and limb locations on the satellite's surfaces. The separate images are spatially co-registered and combined using an algorithm that produces a linear polarization image and a corresponding total intensity (i.e. non-polarized) image (4). To produce albedo images, the total intensity im- age must be corrected for the gradient in shading across the visible disk due to gradual systematic grad- ations in the incidence and emission angles. We mod- el the photometric gradient across the disk with fits of the Minnaert photometric function: ) ( cos
[Show abstract][Hide abstract] ABSTRACT: Hall is a proposed NASA-led New Frontiers-class international robotic lander and sample return mission to explore and return samples from the two moons of Mars, Phobos and Deimos.