Paper No. 2
Presentation Time: 8:20 AM
IMPORTANCE OF IMPACT-GENERATED MELT SYSTEMS FOR UNDERSTANDING THE FORMATION OF IMPACT STRUCTURES
Impact structures, particularly the largest ones known on Earth such as Vredefort (South Africa) and Sudbury (Canada), are characterized by melt systems that are highly variable in terms of occurrence, volume, chemical composition and, most importantly, process of generation. The most prominent melt systems range from impact melt sheets of several thousands of cubic kilometre volumes ponded in impact structures, associated dikes and melt breccias in target rocks, to allochthonous, regionally to globally distributed melt components in suevitic impact breccia, as well as tektites and spherules, respectively. All of these melt systems display in themselves strong compositional heterogeneity that depends on overall target rock composition (lithology, mineralogy, and textural parameters) and interaction with host rocks upon melting. The analysis of individual melt systems is important to determine projectile types and distribution, to determine the ages of impact events, to scale original sizes of partially eroded terrestrial impact structures, as well as the environmental effects of impact and mechanisms contributing to crater formation. As to the latter, there is considerable controversy regarding the formation of impact melt dikes and those of other melt breccia (so-called pseudotachylitic breccias) in target rocks underlying large impact structures, and the mechanical relationship(s) of dike breccias to crater-forming processes, and timing of breccia formation within the short-term impact cratering process. Shock melting, frictional sliding on structural discontinuities, and decompression melting are the dominant modes considered for the formation of impact melt or melt breccia dikes. These issues will be delineated by presenting new data from the Vredefort and Sudbury impact structures. Our studies show that there is a great demand in the systematic analysis of ground-truth and its transfer into numerical models.