Modeling the formation of the K–Pg boundary layer

In this paper we investigate the formation of the Cretaceous–Paleogene (K–Pg) boundary layer through numerical modeling. The K–Pg layer is widely agreed to be composed of meteoritic material and target rock from the Chicxulub impact site, that has been ejected around the globe and mixed with local material during final deposition. The observed composition and thickness of the K–Pg boundary layer changes with azimuth and distance from the impact site. We have run a suite of numerical simulations to investigate whether we can replicate the observational data, with a focus on the distal K–Pg layer and the impact glasses at proximal sites such as Beloc, Haiti. Previous models of the K–Pg ejecta have assumed an initial velocity distribution and tracked the ejecta to its final destination. Here, we attempt to model the entire process, from impact to the arrival of the ejecta around the globe. Our models replicate the observed ejecta thickness at proximal sites, and the modeled ejecta is composed of sediments and silicate basement rocks, in agreement with observational data. Models that use a 45° impact angle are able to replicate the total ejecta and iridium volume at distal sites, and the majority of the ejecta is composed of meteorite and target sediments. Sub-vertical impacts generate too little iridium, and oblique impacts of ⩽30 degrees generate too much. However, in contrast to observations, models that involve ballistic transport of ejecta lead to ejecta thickness decreasing with increasing distance, and are unable to transport shocked minerals (quartz and zircon) from the Chicxulub basement rocks around the globe. We suggest that much of the K–Pg ejecta is transported non-ballistically, and that the most plausible mechanism is through re-distribution from a hot, expanding atmosphere. The results are important for future investigations of the environmental effects of the Chicxulub impact.