Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “bolides”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 127 records · Page 7

Impact cratering calculations

In the course of carrying out finite difference calculations, it was discovered that for large craters, a previously unrecognized type of crater (diameter) growth occurred which was called lip wave propagation. This type of growth is illustrated for an impact of a 1000 km (2a) silicate bolide at 12 km/sec (U) onto a silicate half-space at earth gravity (1 g). The von Misses crustal strength is 2.4 kbar. The motion at the crater lip associated with this wave type phenomena is up, outward, and then down, similar to the particle motion of a surface wave. It is shown that the crater diameter has grown d/a of approximately 25 to d/a of approximately 4 via lip propagation from Ut/a = 5.56 to 17.0 during the time when rebound occurs. A new code is being used to study partitioning of energy and momentum and cratering efficiency with self gravity for finite-sized objects rather than the previously discussed planetary half-space problems. These are important and fundamental subjects which can be addressed with smoothed particle hydrodynamic (SPH) codes. The SPH method was used to model various problems in astrophysics and planetary physics. The initial work demonstrates that the energy budget for normal and oblique impacts are distinctly different than earlier calculations for silicate projectile impact on a silicate half space. Motivated by the first striking radar images of Venus obtained by Magellan, the effect of the atmosphere on impact cratering was studied. In order the further quantify the processes of meteor break-up and trajectory scattering upon break-up, the reentry physics of meteors striking Venus' atmosphere versus that of the Earth were studied.

Ahrens, Thomas J.↗

Structural review of the Vredefort dome

The structure of the older-than-3.2-Ga Archean basement and Archean-to-Precambrian sedimentary/volcanic rocks (3.07 to ca. 2.2 Ga) in the center of the Witwatersrand Basin to the southwest of Johannesburg (South Africa) is dominated by the ca. 2.0-Ga megascopic Vredefort 'Dome' structure. The effect of the 'Vredefort event' is demonstrably large and is evident within a northerly arc of about 100 km radius around the granitic core of the structure. Northerly asymmetric overturning of the strata is observed within the first 17 km (strata is horizontal in the south), followed by a 40-km-wide rim synclinorium. Fold and fault structures (normal, reverse, and strike-slip) are locally as well as regionally concentrically arranged with respect to the northern and western sides of the structure. The unusual category of brittle deformation, the so-called 'shock deformation', observed in the collar strata has attracted worldwide attention over the past two decades. These deformation phenomena include the presence of coesite and stishovite, mylonites, and pseudotachylites, cataclasis at a microscopic scale, and the ubiquitous development of multiply striated joint surfaces (which include shatter cones, orthogonal, curviplanar, and conjugate fractures). The macroscopic to microscopic deformation features have led to the formulation of various hypotheses to account for the origin of the Vredefort structure: (1) tectonic hypotheses--deep crustal shear model, doming and N-directed thrust fault model, fold interference model, and diapir model; (2) the exogenous bolide impact hypothesis; and (3) the endogenous cryptoexplosion model.

Colliston, W. P.↗

Simulated meteorite impacts and volcanic explosions: Ejecta analyses and planetary implications

Past cratering studies have focused primarily on crater morphology. However, important questions remain about the nature of crater deposits. Phenomena that need to be studied include the distribution of shock effects in crater deposits and crater walls; the origin of mono- and polymict breccia; differences between local and distal ejecta; deformation induced by explosive volcanism; and the production of unshocked, high-speed ejecta that could form the lunar and martian meteorites found on the Earth. To study these phenomena, one must characterize ejecta and crater wall materials from impacts produced under controlled conditions. New efforts at LLNL simulate impacts and volcanism and study resultant deformation. All experiments use the two-stage light-gas gun facility at LLNL to accelerate projectiles to velocities of 0.2 to 4.3 km/s, including shock pressures of 0.9 to 50 GPa. We use granite targets and novel experimental geometries to unravel cratering processes in crystalline rocks. We have thus far conducted three types of simulations: soft recovery of ejecta, 'frozen crater' experiments, and an 'artificial volcano. Our ejecta recovery experiments produced a useful separation of impactites. Material originally below the projectile remained trapped there, embedded in the soft metal of the flyer plate. In contrast, material directly adjacent to the projectile was jetted away from the impact, producing an ejecta cone that was trapped in the foam recovery fixture. We find that a significant component of crater ejecta shows no signs of strong shock; this material comes from the near-surface 'interference zone' surrounding the impact site. This phenomenon explains the existence of unshocked meteorites on the Earth of lunar and martian origin. Impact of a large bolide on neighboring planets will produce high-speed, weakly shocked ejecta, which may be trapped by the Earth's gravitational field. 'Frozen crater' experiments show that the interference zone is highly localized; indeed, disaggregation does not extend beyond approx. 1.5 crater radii. A cone-shaped region extending downward from the impact site is completely disaggregated, including powdered rock that escaped into the projectile tube. Petrographic analysis of crater ejecta and wall material will be presented. Finally, study of ejecta from 0.9- and 1.3-GPa simulations of volcanic explosions reveal a complete lack of shock metamorphism. The ejecta shows no evidence of PDF's, amorphization, high-pressure phases, or mosaicism. Instead, all deformation was brittle, with fractures irregular (not planar) and most intergranular. The extent of fracturing was remarkable, with the entire sample reduced to fragments of gravel size and smaller.

Gratz, A. J.↗

Ar-40 to Ar-39 ages of the large impact structures Kara and Manicouagan and their relevance to the Cretaceous-Tertiary and the Triassic-Jurassic boundary

Since the discovery of the Ir enrichment in Cretaceous-Tertiary boundary clays in 1980, the effects of a 10-km asteroid impacting on the Earth 65 Ma ago have been discussed as the possible reason for the mass extinction--including the extinction of the dinosaurs--at the end of the Cretaceous. But up to now no crater of this age that is large enough (ca. 200 km in diameter) has been found. One candidate is the Kara Crater in northern Siberia. Kolesnikov et al. determined a K-Ar isochron of 65.6 +/- 0.5 Ma, indistinguishable from the age of the K-T boundary and interpreted this as confirmation of earlier proposals that the Kara bolide would have been at least one of the K-T impactors. Koeberl et al. determined Ar-40 to Ar-39 ages ranging from 70 to 82 Ma and suggested an association to the Campanian-Maastrichtian boundary, another important extinction horizon 73 Ma ago. We dated four impact melts, KA2-306, KA2-305, SA1-302, and AN9-182. Results from the investigation are discussed.

Trieloff, M.↗

The Tapanui region of New Zealand: Site of a Tunguska around 800 years ago?

Evidence is discussed that the wide-spread fires ca. 800 years ago which denuded the southern provinces of the South Island of New Zealand of the extensive forests present at that time were due to the entry of a large bolide into the atmosphere, the conflagration being ignited by the intense heat generated as this extraterrestrial projectile ablated/detonated in a similar manner to that of the Tunguska object of 1908. These fires led to the extinction of the giant terrestrial bird known as the Moa, and the end of the archaic epoch of Maori history known as the Moa Hunter period. This interpretation is well attested to in Maori myth and legend.

Steel, Duncan↗

Ar-40/Ar-39 age spectra of Apollo 15 impact melt rocks by laser step-heating and their bearing on the history of lunar basin formation

Results are reported on 26 high-resolution (16-51 steps) Ar-40/Ar-39 age spectra obtained on 12 Apollo-15 melt rocks of different composition using a continuous laser system on submg fragments of recrystallized melt and single-crystal plagioclase clasts from impact melt rocks collected at the Apennine Front where the Imbrium and Serenitatis basins intersect. A table is presented with the summary of the Ar-40/Ar-39 spectrum data, which represent 891 individual temperature step analyses. Also presented are 20 of the 26 age spectra along with their respective K/Ca plots. Melt rock fragments and plagioclase clasts from seven of the 12 samples analyzed yielded reproducible, intermediate-T Ar-40/Ar-39 age spectrum plateaus, which were interpreted as crystallization ages that represent the times of impact of bolides onto the lunar surface.

Dalrymple, G. B.↗

Stratigraphy and sedimentology of the K/T boundary deposit in Haiti

The K/T boundary sequence is exposed in uplifted carbonate sediments of the southwest peninsula of Haiti. It is found at 15 localities within the Beloc formation, a sequence of limestone and marls interpreted as a monoclinal nappe structure thrust to the north. This tectonic deformation has affected the K/T boundary deposit to varying degrees. In some cases the less competent K/T deposit has acted as a slip plane leading to extensive shearing of the boundary layer, as well as duplication of the section. The presence of glassy tektites, shocked quartz, and an Ir anomaly directly link the deposit to a bolide impact. Stratigraphic and sedimentological features of the tripartite sequence indicate that it was formed by deposition from ballistic fallout of coarse tektites, emplacement of particle gravity flows and fine grained fallout of widely dispersed impact ejecta.

Carey, S.↗

Shock induced reaction in Chicxulub target materials (CaSO4 and SiO2) and their relation to extinctions

The global platinum element rich layer, the presence of shocked quartz grains (in some cases with stishovite), and the observation of a tektite-rich layer, precisely at the K-T boundary, are the three major arguments for the extinction bolide impact hypothesis of Alvarez et al. Tektites (spherules) from Beloc in Haiti and Mimbral in Mexico received particular interest because of their geological proximity to the Chicxulub impact structure, which is a leading candidate for at least one of the K-T impact craters. Although in most localities the original glass has weathered to clay minerals, some shock-induced glass is found in outcrops and drill cores which is used for Ar-38/Ar-39 dating. The glassy tektites were found to be chemically similar and coeval at 65.0 Ma with Chicxulub melt rock. Two kinds of K-T spherules were discovered: (1) a silic black glass; and (2) a yellow glass, enriched in Ca, Mg, and S. The high sulfur content of the glass and the abundance of anhydrite (CaSO4) in the carbonate-evaporite sequence observed in Drill Holes Y-1 and Y-2 at Chicxulub prompted studies of calcium sulfate devolatization. Further discussion of our experiments is presented.

Chen, Guangqing↗

Dynamics of large scale impacts on Venus and Earth

Large scale impacts are a key aspect of the accretion and growth of the planets, the evolution of their atmospheres, and the viability of their life forms. We have performed an extensive series of numerical calculations that examined the mechanics of impacts over a broad range of conditions and are now extending these to account for the effects of the planetary atmosphere. We have examined the effects of large scale impacts in which the trapping and compression of an atmosphere during impact is a significant factor in the transfer of energy to the atmosphere. The various energy transfer regimes and where conventional drag and trapping and subsequent compression of atmosphere between the bolide and planetary surface are significant are shown.

Okeefe, John D.↗

Delimitation of terrestrial impact craters by way of pseudotachylytic rock distribution

The determination of the shape and size of terrestrial impact craters is problematic, yet is critical to understanding cratering mechanics and for scaling bolide mass, volume, and impact velocity with crater size and target response. The problem is particularly difficult in older geological terrains (e.g. Precambrian) which are more likely to have suffered post-impact deformation and hence distortion of the original structure and/or where weathering may have partly removed or obscured its original shape. Traditionally, a number of features are used to assist us in determining the shape and size of an impact structure. These include the following: (1) the occurrence of faults, especially those disposed concentrically relative to the crater--the outermost ring faults being interpreted as indicating a viable minimum diameter; and (2) the development of so-called breccias, some of which are also associated with faults (e.g. the Sudbury Breccia developed within the target rocks of the Sudbury Structure of Onta rio, Canada). 'Breccia' is not a satisfactory term because a number of breccia-types exist at impact sites (e.g. fall-back breccias and in-situ brecciated target material). Of relevance to crater diameter determination is the recognition of discrete zones and fault- and shock-related pseudotachylyte. Pseudotachylyte is a rock type comprising a fine-grained, usually dark matrix containing clasts of minerals and/or rock derived from the country rock target material. It origin is normally attributed to high-speed slip (including vibration) along a slip surface (i.e. fault) or to the passage of a shock wave through the host material. The clasts can occur as angular fragments (i.e. like a breccia), but are more commonly developed as rounded to sub-rounded fragments. Significantly, the scale of these pseudotachylytes can range from sub-millimeter thick veinlets to dyke-like bodies up to 1 km or more thick. It is the latter, larger occurrence which has been referred to as 'breccia.' The smaller-sized occurrence is generally not recognized in the field, nor is it traditionally associated with its larger counterpart.

Spray, John G.↗

Impact-induced devolatilization of CaSO4 anhydrite and implications for K-T extinctions: Preliminary results

The recent suggestions that the target area for the K-T bolide may have been a sulfate-rich evaporite and that the resulting sulfuric acid-rich aerosol was responsible for the subsequent cooling of the Earth and the resulting biological extinctions has prompted us to experimentally examine the impact-induced devolatization of the sulfate minerals anhydrite (CaSO4) and gypsum (CaSO4(2H2O)). Preliminary results for anhydrite are reported. Up to 42 GPa peak shock pressure, little or no devolatilization occurs, consistent with chemical thermodynamic calculations. Calculation of the influence of the partial pressure of the gas species on impact-induced devolatilization suggests that an even greater amount of sulfur than that proposed by Brett could have been released to the atmosphere by an impact into a sulfate-rich layer. Solid recovery, impact-induced devolatilization experiments were performed on the Caltech 20mm gun using vented, stainless steel sample assemblies.

Tyburczy, James A.↗

Impacts, extinctions, volcanism, glaciations, and tectonics: Matches and mismatches

The debate concerning possible reactions between impacts, extinction events, and volcanism has recently taken a new turn. Diamictites and associated sedimentary deposits long regarded by geologists as glaciogenic, have been reinterpreted as impact-related. Going further, the Permo-Carboniferous diamictites that are widespread in the southern continents and India are not put forward as evidence that fragmentation of the Gondwana supercontinent in the Mesozoic was a direct result of meteorite impact. In an abstract at a meeting of the American Geophysical Union, and in an article in the popular press, one member of the earth science community has made a specific claim to identify the site of the supercontinent-destroying bolide on the Falkland/Malvinas Plateau. It is claimed by this scientist that the Cape Fold belt in Africa represents a 'breaking wave' of deformation resulting from this impact, and that fractures in the clasts of the Dwyka diamictite in southern Africa represent impact-induced cataclasis of the target rock. These hypotheses fly in the face of the well-established tectonic history of the Gondwana supercontinent in several respects.

Dalziel, I. W. D.↗

Radiative signals from impact of Shoemaker-Levy on Jupiter

The temperature and internal energy fields calculated by Takata et al. in the plume are used to calculate the greybody thermal radiation emitted versus wavelength to predict what might be observed by several spectral sensors operating from different platforms when fragments of Comet Shoemaker-Levy 9 (SL-9) impact Jupiter in July 1994. A SPH code was used by Takata et al. to calculate the full three dimensional flow and thermodynamic fields in the comet fragment and the atmosphere of Jupiter. We determined the fragment penetration depth, energy partitioning between the atmosphere and the impactor, and energy density deposited per unit length over the trajectory. Once the impactor had disintegrated and stopped, and the strong atmospheric shock decayed, the flow is driven by buoyancy effects. We then used our SPH code to calculate the flow and thermodynamic fields: pressure, article velocity, temperature, and internal energy distributions in the plume. The calculations for 2 and 10 km cometary fragments yield maximum deposition depths of approximately 175 and 525 km, respectively (1 bar = 0 km depth). We also calculated that 0.7 and 0.6 of the initial kinetic energy of the 10 and 2 km bolides, respectively, are deposited as internal energy in Jupiter's atmosphere.

Ahrens, Thomas J.↗

The collision of Jupiter and Comet Shoemaker-Levy 9

A simple description of the disruption and deceleration of 100-m- to 5-km-diameter comets striking Jupiter is combined with numerical simulations of the subsequent explosions to predict the fate of Comet Shoemaker-Levy 9. Detailed numerical simulations of the first 3 min of the explosion were performed. Our numerical simulations begin either with hot cylinders with dimensions suggested by the disruption and deceleration model or with an initial wake constructed from a moving line charge. In all cases, extensive plumes of hot gas are expelled from the atmosphere. The models with wakes evolve about twice as fast as the initially confined models. Models of both types generate similar pressure waves into the planet. Temperatures and negative hydrogen ion opacities were computed. For atmospheric entry, light curves were computed. On entry the largest bolides could be very bright, possibly as bright as Jupiter for observers placed to see them, although for kilometer-size impactors the luminosity peak is obscured by clouds. For the fireball, light curves were computed from the numerical simulations. Metals from the vaporized comet provide electrons that dramatically increase the opacity of Jovian air at low temperature. The fireball rises through and above the atmosphere, brightening at first as its surface area increases, but later fading to invisibility as its temperature drops and its opacity plummets.

Zahnle, Kevin↗

Geology and landscape evolution of the Hellas region of Mars

Hellas basin on Mars has been the site of volcanism, tectonism, and modification by fluvial, mass-wasting, and eolian processes over its more than 4-b.y. existence. Our detailed geologic mapping and related studies have resulted in the following new interpretations. The asymmetric distribution of highland massifs and other structures that define the uplifted basin rim suggest a formation of the basin by the impact of a low-angle bolide having a trajectory heading S60E. During the Late Noachian, the basin was infilled, perhaps by lava flows, that were sufficiently thick (greater than 1 km) to produce wrinkle ridges on the fill material and extensional faulting along the west rim of the basin. At about the same time, deposits buried northern Malea Planum, which are interpreted to be pyroclastic flows from Amphitrites and Peneus Paterae on the basis of their degraded morphology, topology, and the application of a previous model for pyroclastic volcanism on Mars. Peneus forms a distinctive caldera structure that indicates eruption of massive volumes of magma, whereas Amphitrites is a less distinct circular feature surrounded by a broad, low, dissected shield that suggests generally smaller volume eruptions. During the Early Hesperian, an approximately 1-to 2km-thick sequence of primarily fined-grained, eolian material was deposited on the floor of Hellas basin. Subsequently, the deposit was deeply eroded, except where armored by crater ejecta, and it retreated as much as 200-300 km along its western margin, leaving behind pedestal craters and knobby outliers of the deposit. Local debris flows within the deposit attest to concentrations of groundwater, perhaps in part brought in by outflow floods along the east rim of the basin. These floods may have deposited approximately 100-200m of sediment, subduing wrinkle ridges in the eastern part of the basin floor. During the Late Hesperian and Amazonian, eolian mantles were emplaced on the basin rim and floor and surrounding highlands. Their subsequent erosion resulted in pitted and etched plains and crater fill, irregular mesas, and pedestal craters. Local evidence occurs for the possible former presence of ground ice or ice sheets approximately 100 km across; however, we disagree with a hypothesis that suggest that the entire south rim and much of the floor of Hellas have been glaciated. Orientations of dune fields and yardangs in lower parts of Hellas basin follow directions of the strongest winds predicted by a recently published general circulation model (GCM). Transient frost and dust splotches in the region are, by contrast, related to the GCM prediction for the season in which the images they appear in were taken.

Tanaka, Kenneth L.↗

Palomar observations of the collision of comet Shoemaker-Levy 9 with Jupiter

We present two papers of the near-infrared observations from Palomar observatory of the impact of fragment R of comet Shoemaker-Levy 9 with Jupiter on July 21, 1994. The first paper is concerned with the lightcurves found from these observations: Two instruments were used to image the event at 3.2 and 4.5 microns simultaneously. The lightcurves from these image sequences both show two faint precursor flashes, a bright main peak, and several oscillations over the following hour. We identify the precursor flashes with the entry of the bolide into Jupiter's upper atmosphere, and with the post-impact ejecta plume rising above the planet's limb. The main peak is due to the re-entry of the collapsing plume in Jupiter's atmosphere and the resultant shock heating. The second paper is concerned with the low-resolution spectra from the lightcurves: Low-resolution 8-13 micron spectra taken near the peak of the lightcurve show a broad emission feature that resembles the delicate silicate feature commonly seen in comets and the interstellar medium. We use this feature to estimate the dust content of the impact plume. The overall infrared spectral energy distribution at the time of peak brightness is consistent with emission from an optically-thin layer of small particles at approximately 600 K. Integrating over the spectrum and the lightcurve, we obtain a total radiated energy from the R impact of greater than or equal to 10(exp 25) ergs and a plume mass of greater than or equal to 3 x 10(exp 13) g.

Neugebauer, G.↗

Impact Cratering Calculations

Understanding the physical processes of impact cratering on planetary surfaces and atmospheres as well as collisions of finite-size self-gravitating objects is vitally important to planetary science. The observation has often been made that craters are the most ubiquitous landform on the solid planets and the satellites. The density of craters is used to date surfaces on planets and satellites. For large ringed basin craters (e.g. Chicxulub), the issue of identification of exactly what 'diameter' transient crater is associated with this structure is exemplified by the arguments of Sharpton et al. (1993) versus those of Hildebrand et al. (1995). The size of a transient crater, such as the K/T extinction crater at Yucatan, Mexico, which is thought to be the source of SO,-induced sulfuric acid aerosol that globally acidified surface waters as the result of massive vaporization of CASO, in the target rock, is addressed by our present project. The impact process excavates samples of planetary interiors. The degree to which this occurs (e.g. how deeply does excavation occur for a given crater diameter) has been of interest, both with regard to exposing mantle rocks in crater floors, as well as launching samples into space which become part of the terrestrial meteorite collection (e.g. lunar meteorites, SNC's from Mars). Only in the case of the Earth can we test calculations in the laboratory and field. Previous calculations predict, independent of diameter, that the depth of excavation, normalized by crater diameter, is d(sub ex)/D = 0.085 (O'Keefe and Ahrens, 1993). For Comet Shoemaker-Levy 9 (SL9) fragments impacting Jupiter, predicted excavation depths of different gas-rich layers in the atmosphere, were much larger. The trajectory and fate of highly shocked material from a large impact on the Earth, such as the K/T bolide is of interest. Melosh et al. (1990) proposed that the condensed material from the impact upon reentering the Earth's atmosphere induced. radiative heating, and producing global firestorms. The observed reentry splash of the SL-9 impact-induced plumes that reimpact Jupiter (Boslough et al., 1994) supported Melosh's K/T model. The fate of early primitive planetary atmospheres during the latter stages of planetary accretion, resulting from impactors in the 100 to 103 km diameter require modeling, e.g. Newman et al. (1997). Ahrens (1990; 1993) and Chen and Ahrens (1997) found that upon delivery of most of the impact energy to the solid planet, very large ground motions arise, which couple sufficient kinetic energy to the atmosphere to cause substantial atmospheric escape. The trade-off of this model with that of Cameron (1997) who suggests that atmospheric blow-off occurs as a result of the massive impact-induced heating of the atmosphere and Pepin (1997) who uses this heating event to model differential hydrodynamic loss of lighter atmospheric gases, requires further research.

Ahrens, Thomas J.↗

Geology and landscape evolution of the Hellas region of Mars

Hellas basin on Mars has been the site of volcanism, tectonism, and modification by fluvial, mass-wasting, and eolian processes over its more than 4-b.y. existence. Our detailed geologic mapping and related studies have resulted in the following new interpretations. The asymmetric distribution of highland massifs and other structures that define the uplifted basin rim suggest a formation of the basin by the impact of a low-angle bolide having a trajectory heading S 60 deg E. During the Late Noachian, the basin was infilled, perhaps by lava flows, that were sufficiently thick (>1 km) to produce wrinkle ridges on the fill material and extensional faulting along the west rim of the basin. At about the same time, deposits buried northern Malea Planum, which are interpreted to be pyroclastic flows from Amphitrites and Peneus Paterae on the basis of their degraded morphology, topography, and the application of a previous model for pyroclastic volcanism on Mars. Peneus forms a distinctive caldera structure that indicates eruption of massive volumes of magma, whereas Amphitrites is a less distinct circular feature surrounded by a broad, low, dissected shield that suggests generally smaller volume eruptions. During the Early Hesperian, an approx. 1- to 2-km-thick sequence of primarily fined-grained, eolian material was deposited on the floor of Hellas basin. Subsequently, the deposit was deeply eroded, except where armored by crater ejecta, and it retreated as much as 200-300 km along its western margin, leaving behind pedestal craters and knobby outliers of the deposit. Local debris flows within the deposit attest to concentrations of groundwater, perhaps in part brought in by outflow floods along the east rim of the basin. These floods may have deposited approx. 100-200 m of sediment, subduing wrinkle ridges in the eastern part of the basin floor. During the Late Hesperian and Amazonian, eolian mantles were emplaced on the basin rim and floor and surrounding highlands. Their subsequent erosion resulted in pitted and etched plains and crater fill, irregular mesas, and pedestal craters. Local evidence occurs for the possible former presence of ground ice or ice sheets approx. 100 km across; however, we disagree with a hypothesis that suggests that the entire south rim and much of the floor of Hellas have been glaciated. Orientations of dune fields and yardangs in lower parts of Hellas basin follow directions of the strongest winds predicted by a recently published general circulation model (GCM). Transient frost and dust splotches in the region are, by contrast, related to the GCM prediction for the season in which the images they appear in were taken.

Tanaka, Kenneth L.↗