against a background of exceptionally well defined 41,000- and 100,000-yr Milankovitch climate cyclicity and increasing oxygen restriction in the world's oceans (OAE II) with expansion and possibly benthic touchdown of the oceanic oxygen minimum zone(s). Yet the question remains: What initiated these dynamic changes on such a short time scale?
Whereas Orth et al. (1987, 1990, 1993) have cautiously suggested that the enriched late Cenomanian trace element suite, including Ir, was probably derived from oceanic sources (e.g., mantle outgassing during rapid Cenomanian-Turonian plate rearrangements, and/or from the last phases of outgassing of the Pacific superplume; Larson, 1991a,b), extraterrestrial sources cannot be ruled out. Whereas impacting is a stochastic process, and predictably spatially random (Grieve, 1982), it is not always temporally random. For example, Grieve (1982) has noted four large middle Cretaceous terrestrial impact craters with age error bars overlapping the Cenomanian-Turonian boundary: these are the Deep Bay Crater (100 ± 50 Ma), the Logoisk Crater (100 ± 20 Ma), the Steen River Crater (95 ± 7 Ma), and the West Hawk Crater (100 ± 50 Ma). Two additional craters have late Albian to Cenomanian ages of 100 and 100 ± 5 Ma. Alvarez and Muller (1984) and Strothers and Rampino (1990) have both noted that these craters comprise a statistical impact cluster, one of several with a proposed periodicity of 30 to 34 Ma. The late Albian-early Turonian (but predominantly Cenomanian) cluster of crater ages may therefore represent an impact storm, or shower (sensu Hut et al., 1987). Kauffman (1988b) pointed out that not only was the terrestrial impact record conservative for any temporal cluster of craters because of loss of record by subduction; weathering and erosion; the extent of younger sedimentary, vegetative, and ice cover on land; as well as the fact that many areas are still poorly explored (Grieve, 1982), but also that most extraterrestrial objects hitting Earth would fall into the sea. During greenhouse intervals (Fischer and Arthur, 1977) with elevated global sea-level, as in the Cenomanian-Turonian boundary interval, up to 80% or more of the Earth was covered by water. Thus, for any single terrestrial impact recorded at these times, an additional four would be predicted to have landed in the sea, and the timespan of the impact shower would be expanded over that of the terrestrial impacting record by at least a million years. These predictions are conservative because of the loss of the shallow water and terrestrial impact record through subduction, weathering, erosion, and younger sedimentary cover. However, if they are correct, the predominantly late Albian-Cenomanian impact shower would have extended across the C-T boundary. A series of meteorites or comets impacting in the sea over a relatively short period would probably result in massive evacuation of water (as steam) and debris along the impact path, major shock waves with their resultant marine tectonic effects, compression and heating of the surrounding oceanic water mass, giant tsunamis, and ultimately rapid mixing of oceanic water masses (Melosh, 1982). Oceanic feedback from such a geologically instantaneous perturbation might produce overturn of oceanic water masses, chemical advection events, and dramatic changes in the thermal and chemical character of these water masses. These changes would be preserved in the sedimentary record as dramatic excursions in the stable isotope, trace element, and Corg record beyond background levels (Kauffman, 1988b). Such oceanic perturbations would be expected to result in a prolonged and complex series of feedback processes (loops) as the ocean system sought equilibrium after each impact event. Could the dramatic chemical and thermal changes in the oceans across the Cenomanian-Turonian extinction interval have been initiated by oceanic impact, as would be predicted by the known impacting record? Emerging new data support this hypothesis. Latest Cenomanian bulk sediment samples collected by Michael Rampino and the author north of Boulder, Colorado, have yielded a few shocked quartz grains, some with multiple shock lamellae (Rampino et al., 1993; M. Rampino, personal communication, August, 1993), within an interval also characterized elsewhere by trace element (including iridium) enrichment (Figure 3.3). Further, Tomas Villamil and Claudia Arango (University of Colorado) have collected numerous microtektite-shaped microspheres, possibly glass, from at least two latest Cenomanian horizons in the Villeta Group of central Colombia, one associated with the Rotalipora foraminifer extinction, the other with a major iridium spike and molluscan extinction event. The composition of these grains is currently being analyzed; the results will bear heavily on the hypothesis of impacting as a catalyst for dramatic C-T ocean-climate perturbations and resultant steps of mass extinction.
The high-resolution biological, geochemical, and sedimentological data base developed internationally for the C-T mass extinction interval is one of the most comprehensive data sets available for the study of an ancient biodiversity crisis. These global data provide a wealth of information that can be correlated precisely, at a high level of resolution, among different continental margins, epicontinental seas, and ocean basins through lineage and assemblage biostratigraphy, event and cycle chronostratigraphy, and on a broader scale, geochronology and sea-level history. We are now in a position to study and model the rates, patterns, causal mechanisms, and ecological and evolutionary aftermath of an ancient (C-T) mass extinction in a greenhouse world, at a very fine level of temporal resolution. This will, in turn, help us to understand the possible consequences of the modern biodiversity crisis on Earth.