Abstract Economic deposits associated with terrestrial impact structures range from world-class to relatively localized occurrences. The more significant deposits are introduced under the classification: progenetic, syngenetic or epigenetic, with respect to the impact event. However, there is increasing evidence that post-impact hydrothermal systems at large impact structures have remobilized some progenetic deposits, such as some of the Witwatersrand gold deposits at the Vredefort impact structure. Impact-related hydrothermal activity may also have had a significant role in the formation of ores at such syngenetic ‘magmatic’ deposits as the Cu-Ni-platinum-group elements ores associated with the Sudbury impact structure. Although Vredefort and Sudbury contain world-class mineral deposits, in economic terms hydrocarbon production dominates natural resource deposits found at impact structures. The total value of impact-related resources in North America is estimated at US$18 billion per year. Many impact structures remain to be discovered and, as targets for resource exploration, their relatively invariant, but scale-dependent properties, may provide an aid to exploration strategies.

Three of the principal variables in scaling impact-crater dimensions are the impact velocity, the projectile size, and the gravitational acceleration of the target body. The amount of impact melt generated by an impact, however, is independent of gravity, but will grow in direct proportion to the projectile dimensions and as an increasing function of the impact velocity. Thus, if the impact velocity and gravitational acceleration were held constant and projectiles of increasing size were considered, the amount of melt generated relative to the dimensions of the final crater would grow at a steady rate. Using the Earth and the Moon for comparison, this paper examines the effects of differential scaling on the depth of origin of central-peak material, on the amount of stratigraphic uplift associated with the formation of those peaks, and on the clast contents of impact melts. When craters of similar size are compared, central peaks should be derived from greater depths on Earth because of relatively deeper melting. The amount of stratigraphic uplift, however, should be greater on the Moon. A lunar crater will be larger than its terrestrial counterpart formed by an identical projectile, but the terrestrial crater will be accompanied by substantially more impact melt. As a large fraction of the melt would have lined the transient cavity during the excavation stage of the impact event, a greater fraction of the lunar melt will have been in contact with clastic materials on the cavity wall. Thus, the clast contents of lunar impact melts should be higher than in those in terrestrial craters of similar size.

Approximately one hundred terrestrial hypervelocity impact structures, ranging in diameter from a few tens of meters to ∼ 140 km, are currently known. All are on land, with major concentrations occurring on the North American and European cratons. With the exception of three, all recognized structures are Phanerozoic in age, with ∼35% being <100 my old. There is no known structure commensurate in age with an impact event of the magnitude proposed in the hypothesis of Alvarez and others (1980) for an impact-induced biological crisis at the Cretaceous-Tertiary boundary. A size-frequency distribution of N∝D −2 for large terrestrial impact structures and a calculated cratering rate of 0.35 × 10 −14 km −2 y −1 for structures with D >20 km, indicates, however, that an impact event capable of producing a ≲ 200 km-sized structure may be expected somewhere on the earth’s land surface every 65 m.y. There are three or four known structures with D >25 km and ages of 65 ± 5 m.y. This, however, may not represent a statistically significant increase in the cratering rate at the end of the Cretaceous. Consideration of cratering mechanics indicates that if the proposed Cretaceous-Tertiary impact event occurred in the ocean it had the potential to locally excavate the oceanic crust and bring upper mantle material to the surface, thus producing an as yet to be detected geophysical anomaly. Although cratering efficiency calculations suggest that ~10 3 projectile masses of material may be excavated in a major impact event, the bulk of this ejecta is locally confined. If, as has been suggested, the siderophile-enrichments in the Cretaceous-Tertiary boundary layer indicate the presence of projectile-contaminated ejecta from a major impact, then the source of this material is most likely early-time ejecta accelerated upwards as the projectile is penetrating the target rocks. This is supported by the relative abundance and distribution of meteoritic siderophile elements in impact melt rocks, which indicate that the bulk (~95%) of the projectile mass may be lost from the immediate area of the crater. The difficulties in defining projectile types from siderophile anomalies in the relatively well-known environment of impact melt rocks suggest that more detailed geochemistry and mineralogy will be necessary before the siderophile enrichments at the Cretaceous-Tertiary boundary can be linked to a specific meteoritic compositional class. In general, the record of terrestrial cratering is not inconsistent with the proposal that a major impact event occurred at the Cretaceous-Tertiary boundary. It does not, however, supply any direct confirmation that such an event produced a biological crisis.