ABSTRACT Structural analysis of overturned metasedimentary strata of the lower Witwatersrand Supergroup in the inner collar of the Vredefort Dome reveals the presence of tangential folds and faults associated with the 2.02 Ga impact. The folds are distinct from previously identified subradially oriented, vertical to plunging-inclined, gentle folds that are interpreted as the products of convergent flow (constriction) during the initial stages of central uplift formation. The tangential folds comprise disharmonic, open, asymmetric, horizontal to plunging-inclined anticline-syncline pairs with centripetally dipping axial planes and right-way-up intermediate limbs. They display centripetal-down vergence (anticline radially outward of the syncline) that is consistent with steep inward-directed shear of the overturned strata. We attribute this kinematic pattern to subvertical collapse of the Vredefort central uplift during the latter stages of crater modification. The folds are cut by pseudotachylite-bearing steep to vertical tangential faults that display center-down slip of <10 m up to ~150 m. Both the tangential folds and the faults suggest that the large-scale overturning of strata related to outward collapse of the Vredefort central uplift was accompanied by a component of inward-directed collapse via layer-parallel shearing and folding, followed by faulting. Subradially oriented faults with conjugate strike separations of 1–2 km in the NNE collar of the dome suggest penecontemporaneous tangential extension of the inner collar rocks. This evidence indicates that second-order structures in the metasedimentary collar of the Vredefort Dome preserve a complex, multistage record of evolving strain associated with both initial convergent and upward flow (constriction) related to central uplift rise and later divergent and downward flow (flattening) linked to its collapse, and that centripetally directed collapse features may be important components of the structural inventory of very large central uplifts.

Partial melting of metapelitic rocks beneath the mafic–ultramafic Rustenburg Layered Suite of the Bushveld Complex in the vicinity of the periclinal Schwerin Fold resulted in a structurally controlled distribution of granitic leucosomes in the upper metamorphic aureole. In the core of the pericline, subvertical structures facilitated the rise of buoyant leucosome through the aureole towards the contact with the Bushveld Complex, with leucosomes accumulating in en-echelon tension gashes. In a subhorizontal syn-metamorphic shear zone to the southeast of the pericline, leucosomes accumulated in subhorizontal dilational structural sites. The kinematics of this shear zone are consistent with slumping of material off the southeastern limb of the rising Schwerin pericline. The syndeformational timing of leucosome emplacement supports a syn-intrusive, density-driven origin for the Schwerin Fold. Modelling of the cooling of the Rustenburg Layered Suite and heating of the floor rocks using a multiple intrusion model indicates that temperatures above the solidus were maintained for >600,000 years up to 300 m from the contact, in agreement with rheological modelling of floor-rock diapirs that indicate growth rates on the order of 8 mm/year for the Schwerin Fold.

The four largest well-preserved impact basins in the solar system, Borealis, Hellas, and Utopia on Mars, and South Pole–Aitken on the Moon, are all significantly elongated, with aspect ratios >1.2. This population stands in contrast to experimental studies of impact cratering that predict <1% of craters should be elliptical, and the observation that ~5% of the small crater population on the terrestrial planets is elliptical. Here, we develop a simple geometric model to represent elliptical crater formation and apply it to understanding the observed population of elliptical craters and basins. A projectile impacting the surface at an oblique angle leaves an elongated impact footprint. We assume that the crater expands equally in all directions from the scaled footprint until it reaches the mean diameter predicted by scaling relationships, allowing an estimate of the aspect ratio of the final crater. For projectiles that are large relative to the size of the target planet, the curvature of the planetary surface increases the elongation of the projectile footprint for even moderate impact angles, thus increasing the likelihood of elliptical basin formation. The results suggest that Hellas, Utopia, and South Pole–Aitken were formed by impacts inclined at angles less than ~45° from horizontal, with a probability of occurrence of ~0.5. For the Borealis Basin on Mars, the projectile would likely have been decapitated, with the topmost portion of the projectile on a trajectory that did not intersect with the surface of the planet.

Central pit craters are common on ice-rich bodies, such as Mars, Ganymede, and Callisto. Mars and Ganymede represent the two end members regarding target characteristics (mixed ice and soil for Mars vs. almost pure ice for Ganymede). Comparisons of central pit craters on these two bodies can provide insights into the environmental conditions under which these craters form and provide constraints on the proposed formation models. This analysis includes 1604 central pit craters on Mars and 471 central pit craters on Ganymede. Martian central pit craters are divided into floor pits and summit pits, whereas all central pit craters on Ganymede are floor pits. Central pit craters form in similar-diameter ranges on both Mars and Ganymede when gravity differences are considered, and both bodies show no regional variations in pit crater distribution within the ±60° latitude zone. Martian floor pits are larger relative to their parent crater than summit pits, but the Ganymede pit/crater diameter ratio is larger than for either central pit type on Mars. Central pits have formed over the entire history of both bodies, and there is no indication that excavation depths have varied over time. Lack of crater floor updoming in Martian floor pit craters indicates that low concentrations of ice (estimated at ~20%) still allow production of central pits. The results of this study argue against central peak collapse as the formation mechanism for central pit craters. Excavation into a subsurface liquid layer cannot be ruled out but is difficult to support based on the distributions and consistencies in excavation depth on both bodies. These results support the model of vaporization and gas escape for central pit formation on both Mars and Ganymede.

This paper is a current status report on a project focused on understanding the formation of large impact basins on terrestrial planetary bodies. A set of preliminary two-dimensional axisymmetric numerical models of collisions of asteroids with diameters from 150 to 800 km with the Moon, Mars, and Mercury illustrates the main mechanical effects of planetary-scale impacts. The target body is modeled on a regular grid with a spatial resolution of 5–10 km. Self-gravity is included in the hydrocode. The main consequence of such an impact is a deep melt pool at the center of the basin. Model results are tentatively compared with known impact basins such as South Pole–Aitken on the Moon and Hellas on Mars.

Impact crater formation is sometimes affected by preexisting target inhomogeneities like faults or joint sets in ways that cause the plan view of the crater to deviate from the idealized circular shape. The resulting polygonal impact craters have been known to exist on the Moon for over a century, and they have been subsequently identified on all types of solid surface bodies in the solar system, including all the terrestrial planets. Newly identified polygonal impact craters in the central southern near-side highlands of the Moon display a size distribution clearly different from the nonpolygonal craters, confirming earlier results: lunar polygonal impact craters “favor” the size range from ~20 km to 45 km. Similar results have been obtained also from Martian and Venusian impact craters. When polygonal impact crater diameter data from all three bodies are combined, it becomes apparent that something in the formation mechanics of impact craters apparently drives small complex craters to become polygonal more easily than simple craters or larger complex craters.

We compared the target types and the morphologies and morphometries of various features within fresh complex craters on Mars to assess target dependence. The wide scatter in depth-diameter data from Martian craters is more pronounced than for lunar or Mercurian craters. This was previously assumed to be predominantly due to significant degrees of denudation and secondary infilling of the Martian craters. However, our data for fresh craters still exhibit a wide variation, which we interpret to be the result of comparatively higher target heterogeneity on Mars. Complex central peaks exhibit some crater diameter dependence, preferentially occurring in craters >50 km. Neither peak complexity nor geometry shows any statistical correlation with target type. Although central peak heights and aspect ratios do not exhibit any clear target dependence, they do appear to be correlated—higher peaks possess narrower aspect ratios. Floor and summit pits appear to be more common on lava targets than sedimentary targets, contrary to earlier studies with smaller sample sizes. This observation imposes additional constraints on models proposed for the origin of pits, especially those models that require the presence of volatiles in the target. The ability to correlate target type with crater morphologies/morphometries is highly contingent upon both the surface geology and the actual geology at depth. Some weak correlations may reflect our current limited understanding of the sub-surface geology of Mars. Information on the deeper lithologies acquired through future missions may help resolve the true effect of subsurface competence on intracrater structure.

Water resurge into newly excavated impact craters causes both erosion and conspicuous graded deposits in those cases where the water is deep enough to overrun the elevated crater rim. We compare published information on resurge deposits from mainly the Lockne, Tvären, and Chesapeake Bay structures with new results from low-velocity impact experiments and numerical simulations. Notwithstanding the limitations of each of the analytical methods (observation, experiment, and simulation), we can visualize the resurge process for various initial impact-target configurations, for which one single method would have been insufficient. The focus is on the ways in which variations in impact angle and target water depth affect water-cavity collapse, the initiation and continuation of the resurge, its transformation into a central water plume, and subsequent antiresurge, as well as tsunami generation. We show that (1) the resurge at oblique impacts, as well as impacts into a target with a varied water depth, becomes strongly asymmetrical, which greatly influences the development of the central water plume and sediment deposition; (2) the resurge may cause a central peak–like debris cumulate at the location of the collapsing central water plume; (3) at relatively deep target waters, the resurge proper is eventually prevented from reaching the crater center by the force of the antiresurge; (4) the antiresurge is separated into an upper and a lower component; (5) the resurge from the deep-water side at an impact into water of varied depth may overcome the resurge from the shallow-water side and push it back out of the crater; and (6) contrary to rim-wave tsunamis, a collapse-wave tsunami requires deeper relative water depth than that of Lockne, the crater-forming impact event with the currently deepest known target water depth.

Chicxulub is the only known impact structure on Earth with a fully preserved peak ring, and it forms an important natural laboratory for the study of large impact structures and understanding of large-scale cratering on Earth and other planets. Seismic data collected in 1996 and 2005 reveal detailed images of the uppermost crater in the central basin at Chicxulub. Seismic reflection profiles show a reflective layer ~1 km beneath the apparent crater floor, topped by upwardly concave reflectors interpreted as saucer-shaped sills. The upper part of this reflective layer is coincident with a thin high-velocity layer identified by analyzing refractions on the 6 km seismic streamer data. The high-velocity layer is almost horizontal and appears to be contained within the peak ring structure. We argue that this reflective layer is the predicted coherent melt sheet formed during impact, and it may be comparable with the unit known as the Sudbury Igneous Complex at the Sudbury impact structure. The Sudbury Igneous Complex, interpreted as a differentiated impact melt sheet, appears to have a similar scale and geometry, and an uppermost lithological sequence consisting of a high velocity layer at the top and a velocity inversion beneath. This comparison suggests that the Chicxulub impact structure also contains a coherent differentiated melt sheet.

Impact modeling and post-impact cooling studies predict a unique fracture and post-impact temperature distribution within the crater floor of large meteorite impact structures. The integration of numerical modeling results and their application to the observed geophysical and current topographic data provides new insights into the early evolution of the deeply eroded Sudbury Structure. The modeling shows a maximum depth of melting of 30–40 km (depending on impact angle and impact velocity). However, melt from upper target layers (< 10 km) is mainly ejected during the excavation stage of crater formation, and the remaining melt is strongly biased to melt derived from lower crustal material. Two-dimensional thermal evolution modeling with various granophyre/norite thickness ratios shows that irrespective of the granophyre/norite thickness ratio, the hottest part of the Sudbury Igneous Complex (SIC) was near the crater center at the melt-pool bottom and within the crater floor, which supports precipitation of sulfides toward the crater floor. The 2D cooling models give compelling evidence for longevity of melt at the bottom of the SIC and partial remelting of the crater floor. The numerical model results are compared with observed topographic, seismic and magnetic data and provide important constraints on their interpretation. A unique slow cooling history is manifested in the broad magnetic signature of the SIC and the adjacent crater floor, and its pronounced remanent magnetization. The vast damage zone and the complex fracture pattern predicted for the crater floor is well preserved in the new high-resolution topographic data for the Sudbury Structure. These regional topographic data allow the distinction between inside-basin fabric (radial topographic lineaments) and crater-floor topographic fabric (radial and contact parallel lineaments), which corroborates the numerical modeling results of radial and concentric faults propagating up to tens of kilometers from the crater center.

A ground-penetrating radar (GPR) survey was carried out in order to characterize reflection patterns at Paluküla Hill, at the NE portion of the rim of the 4-km-diameter, Upper Ordovician Kärdla impact crater. The results allow us to distinguish between the Quaternary overburden and layered postimpact marine sedimentary rocks that cover the crystalline rim. The bedrock is indicated by reflections parallel to bedding that are tilted inward and outward with respect to the crater along the inner and outer rim slope, respectively. It is obvious that the uppermost part of the Paluküla crystalline rim was eroded and leveled in the Late Ordovician prior to having been covered by subhorizontally layered sediments. The unexpected position and low height of the crystalline rim at the northernmost edge of Paluküla Hill indicate that the rim of the Kärdla structure has collapsed to different degrees. This is consistent with the different heights of the crystalline rim. Our results favor the interpretation that the observed geology is not due to erosion by sea resurging into the crater but is rather a result of collapse of the crater rim.

A detailed total intensity magnetic survey of a local negative magnetic anomaly located in the southern sector of the inner ring in the Ries impact structure was carried out in 2006–2007. As the suevite of the Ries crater is known to have an often strong reverse remanent magnetization causing negative magnetic anomalies, a suevite body lying below shallow lake sediments upon the crystalline basement rocks of the inner ring was suspected to be the cause of the anomaly. A drilling program conducted by the Geological Service of Bavaria offered the opportunity to drill a 100-m-deep core hole into this anomaly in 2006. The core stratigraphy involves from 0 to 4.5 m fluviatile Quaternary lake sediments, from 4.5 to 21 m Neogene clays of the Ries crater lake, and from 21 to 100 m suevite and impact melt rock. The suevite and the impact melt rock have a strong reverse remanent magnetization and very high Koenigsberger ratios. Thermomagnetic and coercivity analyses indicate that magnetite is the dominant carrier of the magnetization. The borehole unfortunately did not penetrate the crystalline basement rocks of the inner ring, but modeling of the magnetic source body indicates that the bottom of the hole could not be far from the contact. A macroscopic survey shows suevite from 21 to 87 m, highly diverse in terms of suevite types, and a gradational transition to massive impact melt rock constituting the lowermost 13 m of the drill core. A detailed macroscopic description and first results of microscopic observations reveal that suevite groundmass is substantially altered to secondary phyllosilicates (mostly smectite, minor chlorite) and locally extensive development of calcite. Crystalline basement–derived lithic clasts and minerals dominate the clast population, and only traces of clastic material derived from the upper sediment parts of the target could be recorded. Macroscopically and microscopically, melt fragments have mostly irregular shapes, which lead to the tentative conclusion that only part of the melt—and by implication suevite—mass is derived from fallout of the ejecta curtain. On the other hand, most melt fragments and larger lithic clasts are seemingly oriented subperpendicular to the core axis. This could be interpreted as being due alternatively to settling through air or lateral movement within the actual crater. The gradational zone between proper suevite and massive impact melt rock is characterized by increasing enrichment of melt component and concomitant reduction of suevitic groundmass, until in the uppermost impact melt rock, only millimeter-wide stringers of groundmass remain between densely packed centimeter- to decimeter-size melt fragments.

The Archean granites of the Vredefort impact structure show a high intensity of natural remanent magnetization (NRM) and a random dispersion of directions of high-coercivity components on the centimeter scale. It has been suggested that this anomalous remanence is carried by rod-shaped single-domain (SD) magnetites along planar deformation features (PDFs) in shocked quartz produced as a consequence of the impact event. To determine the carriers of this NRM, we conducted surface magnetic field observations using scanning magneto-impedance (MI) magnetic microscopy during stepwise alternating field (AF) demagnetization over a 1-mm-thick slice of Vredefort granite. We found that the stable component after demagnetization gives rise to just three strong magnetic anomalies. Progressive thinning of the scanned section and micro-Raman spectroscopy revealed that the source of these magnetic anomalies, the highly coercive remanence-carrying mineral, is an assemblage of relatively coarse-grained (1–200 μm) magnetite in biotite, not single-domain magnetite embedded along PDF lamella.

Cerro do Jarau is a prominent, ~13.5-km-wide, circular landform rising >200 m above the plains of the “pampas” in southern Brazil. The name (meaning Jarau hills) comes from the prominent crests of silicified sandstones, which form a semiring of elevated hills in the northern part of the structure. The origin of this structure has been debated for decades, and varied suggestions of its formation include either endogenous tectonic processes or large meteorite impact. However, no conclusive evidence to support either hypothesis has been presented to date. This structure was formed in Mesozoic volcano-sedimentary rocks of the Paraná Basin and consists of the Jurassic-Cretaceous Guará (sandstones), Botucatu (sandstones), and Serra Geral (basalts) formations. The Botucatu Formation sandstones are intensely silicified and deformed, and were subject to radial and annular faulting. Our investigations at Cerro do Jarau identified the occurrence of parautochthonous monomict lithic breccia and polymict breccias resembling suevite and striated joint surfaces resembling crude shatter cones in sandstones and basalts. In addition, our first mineral deformation studies show the presence of rare planar features in quartz clasts in polymict breccias. The identification of these features at Cerro do Jarau, for the first time, is suggestive of an impact origin for the structure. If confirmed by further investigation of possible shock features, Cerro do Jarau would become the sixth known impact structure in Brazil, as well as the fifth basalt-hosted impact structure on Earth.

Aeromagnetic data reveal a possible 15–20 km wide impact structure in the Okavango Delta, Botswana. The identified structure is located at 19°07′40.0″S, 23°18′12.7″E, and it is marked by a circular region of subdued magnetic intensity ~18 km across with a central magnetic peak. It does not appear to be visible in either Google Earth imagery or Shuttle Radar Topography Mission (SRTM) topographic data, and gravity data are too sparse in this region to image the structure. The regional magnetic fabric continues through the anomaly. The central magnetic peak has an amplitude of >700 nT. Modeling of the magnetic data indicates that the structure might be buried beneath ~200–500 m of sediments, and it has a magnetically modeled central uplift that is ~5 km wide.

The Phanerozoic Parnaíba sedimentary basin in the north-northeastern region of Brazil covers an area of ~400,000 km 2 and contains a number of circular structures, four of which are of possible impact origin: Serra da Cangalha, Santa Marta, Riachão, and São Miguel do Tapuio. All four exhibit a central morphological feature resembling a central uplift, characteristic of complex impact structures. A recently acquired regional aerogeophysical survey provided magnetic and gravity data for the entire basin. The magnetic and gravity characteristics of the four possible impact structures of the Parnaíba Basin were analyzed in comparison with impact structures elsewhere in the world. The analysis shows that, except for the São Miguel do Tapuio structure, three of the structures exhibit geophysical characteristics similar to the signatures found in some known impact structures of comparable sizes that have formed in clastic sedimentary rocks. Serra da Cangalha structure exhibits magnetic highs and gravimetric lows, Santa Marta structure exhibits magnetic lows and gravimetric highs, Riachão structure depicts subdued gravimetric and magnetic highs, and the São Miguel do Tapuio structure shows a magnetic high and a complex gravimetric signature. Based on the observed geophysical signatures presented here, Serra da Cangalha, Santa Marta, and Riachão could be regarded as potential impact structures, whereas São Miguel do Tapuio is the least likely of these four structures to have been formed by an impact event.

The Tookoonooka impact structure is a subsurface structure of the Eromanga Basin in Australia. Impact ejecta have recently been discovered in the stratigraphy proximal to the structure. The ejecta includes accretionary and armored impacto-clasts. They are observed at multiple locations in drill core across central Australia, spanning 375,000 km 2 within possible impact tsunami deposits. Typical characteristics of the accretionary impactoclasts include a distinctive brownish-gray color, flattened shapes, concentric zonation, and a variety of morphologies with and without obvious nuclei. Some complex accretionary impactoclasts include melt components. Apparent diameters of these impactoclasts in drill core are commonly less than 2 cm, but may be up to 9 cm. They occur in a variety of depositional contexts, including clast-supported breccia-conglomerate layers and “floating” within massive and planar-bedded sandstones. Microscopic and geochemical investigations reveal that they are pervasively altered. Many resemble the types of accretionary lapilli recognized from hydroclastic volcanic environments, which implies the presence of significant water at the time of impact. Tookoonooka is interpreted to have been a marine (likely paralic to shallow) impact event. It is proposed that hydroclastic types of accretionary impactoclasts at impact sites may be an indicator of wet or marine targets. Complex forms of accretionary impactoclasts may also lead to new understanding of impact vapor plume processes. The impactoclasts studied at Tookoonooka are consistent with an impact origin of the candidate ejecta. The consistent first occurrence of the impactoclasts at the base of the Wyandra Sandstone Member stratigraphically constrains the Tookoonooka impact age to 125 ± 1 Ma in the Lower Cretaceous.

Eight outcrops of chaotic debrisite containing ejecta from the 1850 Ma Sudbury impact event have been identified in and near the city of Thunder Bay, Ontario, 650 km west of the center of the Sudbury crater. Ejecta features include devitrified vesicular impact glass, spherules, accretionary lapilli, microtektites and tektites, and shocked quartz grains containing relict planar features including planar deformation features (PDFs) and planar fractures. The original volume of ejecta has been significantly reduced by carbonate replacement and recrystallization, so that today it only makes up ~20% of the debrisite volume. Volumetrically, the primary component of the debrisite is ripped-up clasts of the local bedrock, including carbonate grainstones, stromatolites, and chert of the 1878 Ma Gunflint Formation. These boulder- to coarse-sand–sized clasts commonly fine upward, in marked contrast to the chaotic nature of the remainder of the debrisite. Seven of the eight sites have had the upper portion of the impact layer removed by glaciation. The eighth site shows a complete stratigraphic section from the Gunflint Formation, through the ejecta bearing layer, and into the overlying 1832 Ma Rove Formation. Zircons extracted from one of the ejecta sites, and probably derived from Gunflint tuffaceous material entrained by the turbulent flow, reveal U-Pb ages clustered around 1856 Ma. The sequence of events deduced from these outcrops begins with mafic volcanic ash being deposited and reworked in a carbonate-dominated, nearshore environment that supported microbial mat growth and stromatolites. During subsequent subaerial exposure, blocky, meteoric calcite cements formed in this material. Upon impact, earthquakes fractured this bedrock as well as underlying Gunflint Formation cherts and carbonates. Impact-generated base surges, or less likely, tsunamis, stripped the area of loose sediment and ground up the Gunflint carbonates into fine-sand– to fine-gravel–sized clasts. This debris was mixed with the ejecta, along with ripped-up, fractured Gunflint chert-carbonate clasts, before being deposited as a chaotic, laterally variable layer. In an ensuing period of subaerial exposure lasting <18 m. y., weathering, erosive reworking, and cementation modified the deposits. The Animikie Ocean then transgressed the area, depositing the overlying Rove Formation carbonaceous shale.

The Campanian Manson impact structure of Iowa represents the best-preserved, large-diameter complex crater within the continental United States. The related bolide struck from the southeast at a low angle, potentially distributing ejecta downrange to the northwest across the Western Interior Cretaceous Seaway. Here, we (1) examine possible correlation of Manson impact horizons across the Cretaceous seaway to terrestrial formations of Montana, and (2) test a large hadrosaur bone bed from the Two Medicine Formation for evidence indicative of the Manson impact. The study includes geochronology; palynomorph, soot, and geochemical analyses; and physical searches for impact ejecta. The impact ejecta–bearing Crow Creek Member of the marine Pierre Shale can be correlated to the SB2 discontinuity in the Judith River and Two Medicine Formations of Montana based on radiometric dates, ammonite zonation, and an association with the onset of the Bearpaw transgression. A 40 Ar/ 39 Ar analysis of an associated bentonite bed dates the hadrosaur bone bed (TM-003) to 75.92 ± 0.32 Ma referenced to MMhb-1 at 523.1 Ma. This bentonite and associated lacustrine units suggest a potential correlation with the SB2 and the Crow Creek Member. However, our examination of the bone bed produced no definitive impact evidence. The combined analyses did reveal three unusual aspects: (1) an abundance of Ulmoideipites sp., (2) a high soot content, and (3) elemental and mineralogical changes suggestive of distinct geochemical units. A major wildfire followed by a postcatastrophe bloom dominated by Ulmoideipites sp. likely preceded the eventual debris flow that generated the bone bed. The SB2 discontinuity and the 33n.3r magnetic subzone represent traceable stratigraphic markers that could serve as guides in future exploration for Manson impact evidence in terrestrial formations west of the seaway.