CHICXULUB IMPACT - NO MASS EXTINCTION
The Chicxulub impact, which left a crater of about 180 km-in-diameter, is commonly believed to have caused the K-T mass extinction. In previous studies we have shown that this impact predates the K-T boundary by about 300,000 years. Here we evaluate the biotic effects of the Chicxulub impact in NE Mexico, about ~600 km from the impact crater on Yucatan, and in Texas along the Brazos River about 1000 km from the impact crater.
Figure 1. Locations of localities studied with K-T sequences containing Chicxulub impact ejecta.
In each of these localities we evaluated the planktic foraminiferal assemblages above and below the impact ejecta layer (impact glass spherules) in terms of species diversity and abundance changes in each species population. Samples were analyzed at 10-20 cm intervals based on quantitative analyses of large (>150µ) and small (63-150µ) size fractions. Bulk and clay mineralogy, stable isotopes and platinum group elements were also analyzed. Stable isotope data is not useful because the original signals are obliterated in these diagenetically altered sediments. Platinum group elements (Ir, Pd, Pt) show no changes across the spherule layer.
In NE Mexico, sediments at the El Peñon section were deposited at >500 m depth in an upper slope environment. In this region, a sedimentary influx from continental erosion was high due to the rising mountains of the Sierra Madre Oriental and the sediments were funneled across the continental shelf and down the slope via submarine canyons. At this locality the Chicxulub impact is represented by a nearly 2 m thick spherule layer interbedded in undisturbed marls 4 m below the 8 m thick sandstone complex that infills a submarine canyon below the K-T boundary. Reworked impact spherules are present at the base of this submarine canyon fill.
In earlier studies the sandstone complex that infills the submarine canyons was interpreted as the result of a mega-tsunami generated by the Chicxulub impact. In this scenario sediment deposition occurred within hours to days of the impact. However, burrows are present through much of the canyon deposit and a limestone layer with burrows in-filled with spherules separates two spherule layers (Fig. 2). Limestone takes thousands of years to accumulate and invertebrates established colonies on the ocean floor repeatedly during deposition of these sediments. This means deposition of the submarine canyon sandstone complex occurred over a very long time and could not have been due an impact-tsunami.
Figure 2. Litholog of the El Penon section showing the sandstone complex with two reworked spherule layers at the base, and the original Chicxulub impact deposit in late Maastrichtian sediments over 4 m below.
Within the nearly 2 m thick spherule deposit more than 4 m below the submarine canyon sandstone deposit, there are four upward fining spherule layers that suggest wave action and suspension settling. At the base of each unit spherules are densely packed and compressed, or partly welded in a calcite matrix (Fig. 3). No detritus or foraminifera are present. These features suggest rapid settling after the Chicxulub impact. The absence of detritus indicates that these sediments were not reworked and transported from shallow waters, similar to the spherule layers at the base of the sandstone complex. This nearly 2 m thick spherule unit therefore may well represent the time of the Chicxulub impact and the immediate rapid settling and deposition of the ejacta fallout.
Figure 3. A-C: reworked Chicxulub impact spherules from the base of the sandstone complex (see Fig. 2). These spherules are in a matrix of detrital grains, reworked shallow water debris and foraminifera. D-P: Chicxulub impact spherules from the 1.8 m thick spherules unit of the late Maastrichtian more than 4 m below the sandstone complex. These spherules are in a matrix of calcite cement and show no signs of reworked shallow water debris. Abundant rounded (D-F), elongate and compressed spherules (G-K) with concave-convex contacts (L, M) and vesicular glass (N-P) are characteristic of the Chicxulub spherule ejecta layer. The cement matrix and absence of clastic grains indicate that no reworked component is present. The compressed and welded glass indicates that deposition occurred rapidly while the glass was still hot. Spherules range in size from 2-5 mm.
The latest Maastrichtian is identified by the presence of Plummerita hantkeninoides, a species that evolved in magnetochron C29r about 300,000 years before the K-T boundary and became a casualty of the mass extinction (Pardo et al.,1996). The presence or absence of this species in middle and low latitudes is a very reliable indicator for evaluating the continuity and completeness of the sedimentation record. At El Penon P. hantkeninoides first appears 8.25 m below the unconformity at the base of the submarine canyon sandstone complex, and 1.25 m below the base of the nearly 2 m thick Chicxulub spherule deposit (Fig. 2). This constrains the age of the Chicxulub impact to the late Maastrichtian and predates the mass extinction by about 300,000 years (Keller et al., 2003). No other species evolved during this interval and while a number of environmentally sensitive species disappeared or became very rare, none can be reliably shown to be extinct globally. The same age was determined for the Chicxulub impact layer at Brazos, Texas (Keller et al., 2007) and in the crater core Yaxcopoil-1 on Yucatan (Keller et al., 2004a,b).
Figure 4. High resolution biostratigraphic scheme for the KT transition.
Species richness, a census of the number of species present at any given time, and the relative abundance of individual species populations are two commonly used proxies to assess environmental changes. Both of these proxies were analyzed at El Peñon. Relative species abundances were analyzed in two size fractions in order to evaluate the response of the small (63-150µ) and large (>150µ) species. Large species comprise a very diverse group of generally complex, ornamented and highly specialized K-strategists that thrived in tropical and subtropical environments, but were intolerant of environmental changes and hence prone to extinction (Abramovich et al., 2003; Keller and Abramovich, in press). The biotic effects of the Chicxulub impact should thus be most apparent in the K-strategists. Small species are less diverse, ecologic generalists, or r-strategists, and generally tolerant of environmental perturbations, including variations in temperature, salinity, oxygen and nutrients (Keller and Abramovich, in press. Some of these species respond to environmental catastrophes by opportunistic blooms, such as observed for Heterohelix and Guembelitria species.
A total of 52 species are present in the >150µ size fraction at El Peñon during the late Maastrichtian. Of these 75% (39 species) are K-strategists and 25% (13 species) are r-strategists (Fig. 5). Across the Chicxulub impact spherule layer species richness remains unchanged - the same species present below the spherule layer are also present above it. Not a single species went extinct.
About 2 m above the spherule layer species richness decreases to 42-44 species, rising only at the base of the unconformity at the base of the sandstone complex probably due to reworking. The variability in species richness is due to the rare and sporadic occurrences of 9 (K-strategy) species, or 17% of the total assemblage. Their increasingly sporadic occurrences may be the result of environmental changes and/or preservation.
The bulk of the species (83%) are continuously present. These data indicate that the decrease in species richness cannot be assigned to the biotic effects of the Chicxulub impact because (1) it occurs much later, (2) the species that are very rare and sporadically present are already endangered species below the spherule layer, and (3) all of these species are known to have survived to the K-T boundary elsewhere.
Figure 5. Species richness and relative abundances of specialized large species show no significant changes across the Chicxulub impact spherule layer. This means that no species went extinct as a result of the Chicxulub impact and no significant environmental changes are evident on the geological time scale.
Species richness in the smaller (63-150µ) size fraction totals 39 species, of which 64% (25 species) are K-strategists and 36% (14 species) are r-strategists (Fig. 6). Species richness remains unchanged across the impact spherule layer and throughout the section, with a low variability of 34-36 species, in contrast to the slight decrease in the larger size fraction (Fig. 5). The maximum number present (38 species) is observed at the unconformity at the base of the sandstone complex with the reworked spherule layer (similar to the >150µ size fraction) and is likely the result of reworking. Variability is due to five K-strategy species, which are rare and sporadically present.
Figure 6. Species richness and relative species abundances in the smaller non-specialized species show no significant variations and no extinctions. The Chicxulub impact appears to have had no catastrophic effect on the geological time scale.
Relative abundance changes in individual species populations are more sensitive indicators of environmental changes than the presence or absence of species. During the late Maastrichtian, K-strategy species in the >150µ size fraction show normal diversity and abundances. Nearly half of the K-strategists are common with the assemblages dominated (10-20%) by Pseudoguembelina costulata, Rugoglobigerina rugosa and R. scotti (Fig. 5). Also common are pseudotextularids, other rugoglobigerinidsand globotruncanids (e.g., arca, aegyptiaca, rosetta, orientalis, stuarti). Among r-strategists, the larger morphotypes of Heterohelix globulosa are common in this assemblage. Relative species abundance variations above and below the spherule layer are within normal fluctuations of the section with no significant changes. The only significant abundance change occurs in the upper 2 m of the section where H. globulosa decreases and Pseudotextularia deformis and Globotruna stuarti increase. No specific biotic effects in K-strategists can be attributed to the Chicxulub impact.
Species abundances in the small size fractionare dominated by the small biserial r-strategist Heterohelix navarroensis, which varies between 40-50% across the spherule layer and decreases in the upper part to an average of 40% (Fig. 6). Other r-strategists vary between 5 and 15% and consist of small heterohelicids, globigerinellids, and hedbergellids. The disaster opportunist Guembelitria is a minor component (<5%). K-strategists are dominated by Pseudoguembelina costulata and costellifera. All other K-species are rare (<1%). The relative species abundance changes show no significant variations across the impact spherule layer, except for two species. Pseudoguembelina costellifera, a surface dweller, decreases 8% above the spherule layer, concurrent with a decrease in H. navarroensis, a low oxygen tolerant species. This abundance variation suggests a change in the watermass stratification, though whether this relatively minor biotic change was related to the Chicxulub impact is unclear.
If Chicxulub caused the K-T mass extinction, then the spherule ejecta should be found at the same stratigraphic layer as the mass extinction. This is not the case. The K-T boundary in NE Mexico is well represented and always above the sandstone complex, and thus up to 15 m above the spherule layer in the late Maastrichtian.
The best and most continuous K-T transitions can be found by laterally tracing the sandstone complex 50-150 m beyond the submarine canyons where only the topmost thin (10-25 cm) sandstone is present, such as at La Parida, La Sierrita, and El Mimbral (Keller et al., 1997). This is shown for La Sierrita and El Mimbral (Fig. 7A, B) where a thin clay and K-T characteristic red layer are present with iridium concentrations of 0.3 and 0.8 pbb, respectively. This clay and red layer mark the basal Danian planktic foraminiferal zone P0 (Keller et al., 1994). Elevated Ir concentrations between 0.2-0.8 ppb at the K-T boundary were also reported from eight sections in NE Mexico (Stueben et al., 2005).
Figures 7 and 8. The K-T boundary red layer at El Mimbral (Fig. 7) and La Sierrita (Fig. 8) is enriched in Iridium and marks the mass extinction in planktic foraminifera. The K-T boundary and red layer can only be found in areas away from the submarine canyon deposits, as for example by tracing the top of the deposit laterally away from the canyons where normal sedimentation occurred.
Figure 9. The mass extinction at La Sierrita coincides with the Ir anomaly and the negative excursion in carbon isotopes.
At La Sierrita, the section was collected where a 5 cm thick calcareous sandy layer is the only representative of the submarine sandstone complex. Above it is a thin clay and mm thin red layer, which contains an Ir anomaly. The K-T defining negative shift in carbon isotopes and the mass extinction of all tropical and subtropical foraminifera coincide with this clay layer. These characteristic mark the K-T boundary worldwide.
Figure 10. La Parid K-T boundary transition shows a thin calcareous sandstone remnant of the submarine canyon sandstone complex and 10cm thick marl with late Maastrichtian assemblages above it, but below the mass extinction horizon. This 10 cm thick marl layer indicates that the sandstone complex of the submarine canyon was deposited prior to the KT boundary.
At La Parida, the thin K-T clay layer is missing (Fig. 10). This section is interesting, however, for its 10 cm thick layer of Late Maastrichtian marls with Late Maastrichtian zone CF1 assemblages that overlies the remnant calcareous sand of the sandstone complex, but is below the K-T extinction horizon. A thin Late Maastrichtian marl layer overlying the sandstone complex was also observed in several other localities, including La Lajilla and El Mulatto (Lopez-Oliva and Keller, 1996). This suggests that the sandstone complex predates the K-T boundary.
Danian grey shale conformably overly this marl layer. Planktic foraminifera in the basal grey shale contain the early Danian Parvularugoglobigerina eugubina zone (P1a(1) assemblages (Fig. 6). The K-T boundary is thus marked by a short hiatus.
The mass extinction in planktic foraminifera has been documented in various sequences in Mexico (e.g., Keller et al., 1994, 1997; Lopez-Oliva and Keller, 1996; Stinnesbeck et al. 2002) and all show extinction and evolution patterns similar to La Parida (Fig. 10). From a maximum of about 52 species during the late Maastrichtian at the time of the Chicxulub impact at least 86% (45 species) survived to the end of the Maastrichtian in Mexico. The 7 species missing at La Parida may be result of local disappearances or failure to record them due to their rare and sporadic occurrences. Another 7 species are rare and sporadically present. In the 1 m below the K-T boundary at La Parida, rare species account for 22% (10 species) of the assemblages.
At the K-T catastrophe 69% (31 species) went extinct, all of them specialized tropical and subtropical large, complex K-strategists. Ten of the species (22%) present are known to have survived the catastrophe for at least some time, all of them r-strategists, tolerant of environmental fluctuations (heterohelicids, hedbergellids, globigerinellids). One species, the disaster opportunist Guembelitria cretacea, thrived in the immediate aftermath of the catastrophe globally. The evolution of new species began almost immediately after the mass extinction; all new species were small, unornamented and with simple biserial, triserial or trochospiral chamber arrangements. This mass extinction pattern is characteristic in planktic foraminiferal assemblages throughout the Tethys, though species abundances may vary depending on regional conditions.
Planktic foraminifera, which suffered the most dramatic mass extinction at the K-T boundary with 2/3 of the species extinct, experienced no significant biotic effects as a result of the Chicxulub impact. No species went extinct and no species population decreased or increased significantly as a result of this large impact (Figs. 5, 6). This observation comes as a surprise mainly because we have assumed that the Chixculub impact caused the K-T mass extinction by associating this impact with the K-T boundary. A survey of the impact crater and mass extinction records over the past 500 m.y. reveals that no impact crater can be associated with any mass extinction (review in Keller, 2005).
The Chicxulub crater with a diameter between 150-180 km is the largest known impact. Other well studied impacts that show no significant species extinctions or other biotic effects include the 90-100 km in diameter late Eocene Chesapeake Bay and Popigai craters dated at 35.7±0.2 and 35.6±0.2 Ma (Keller et al., 1983; Montanari and Koeberl, 2000; Pusz et al., 2006), the late Triassic Manicouagan crater dated at 214±1 Ma, the 100-120 km in diameter late Devonian Alamo (382.8-385.3 Ma) and Woodleigh (359±4 Ma) impacts (review in Keller, 2005). When none of these large impacts (90-120 km diameter craters) caused significant biotic and environmental effects, it should not be surprising that the same is true for the Chicxulub impact, which was not much larger with a crater of at most 180 km in diameter.
The Chicxulub impact and K-T mass extinction are thus two separate and unrelated events. What are likely alternative causes for the K-T mass extinction? The global Ir anomaly at the K-T boundary suggests another large impact, if the iridium is of extraterrestrial origin. But volcanism is another source for enhanced iridium. Recent studies suggest that the main phase (80%) of Deccan eruptions may have been very rapid and ended at the K-T mass extinction. These intriguing data call for a re-evaluation of the current K-T impact mass extinction theory.
1. Abramovich, S., Keller, G., Stueben, D., Berner, Z., 2003. Characterization of late Campanian and Maastrichtian planktonic foraminiferal depth habitats and vital activities based on stable isotopes. Palaeogeography, Palaeoclimatology, Palaeoecology 202, 1-29.
2. Keller, G., 2005. Impacts, volcanism and mass extinctions: random coincidence or cause
3. Keller, G., D’Hondt, S., Vallier, T.L., l983. Multiple microtektite horizons in upper Eocene marine sediments: No evidence for mass extinctions. Science, 221, 150-152.
4. Keller, G., Stinnesbeck, W., and Lopez-Oliva, J.G., l994a, Age, deposition and bioticeffects of the Cretaceous/Tertiary boundary event at Mimbral, NE Mexico: Palaios,v. 9, p. 144-157.
5. Keller, G., Li, L., and MacLeod, N. 1995, The Cretaceous/Tertiary boundary stratotype section at El Kef, Tunisia: how catastrophic was the mass extinction?Palaeogeography, Palaeoclimatology, Palaeoecology, v. 119, p. 221-254.
6. Keller, G., Lopez-Oliva, J.G., Stinnebeck, W. and Adatte, T., l997. Age, stratigraphy and deposition of near-K/T siliciclastic deposits in Mexico: relation to bolide impact? Geological Society of America Bulletin 109, 410-428.
7. Keller, G., Adatte, T., Stinnesbeck, W., Affolter, M., Schilli, L., and Lopez-Oliva, J.G., 2002. Multiple spherule layers in the late Maastrichtian of northeastern Mexico. Geological Society of America Special Paper 356, 145-161.
8. Keller, G. and Abramovich, S., 2008. Lilliput Effect in late Maastrichtian planktic Foraminifera: Response to Environmental Stress. Paleogeogr., Paleoclimatol., Paleoecol., in press.
9. Keller G., Stinnesbeck W., Adatte T. and Stueben D. 2003. Multiple impacts across the Cretaceous-Tertiary boundary. Earth-Science Reviews 1283: 1-37.
10. Keller, G., Adatte, T., Stinnesbeck, W., Rebolledo-Vieyra M., Urrutia Fuccugauchi, J., Kramar, G., and Stueben, D., 2004a. Chicxulub predates the K/T boundary mass extinction. Proceedings of the National Academy of Sciences 101, 3753-37-58.
11. Keller, G., Adatte, T., Stinnesbeck, W., Stüben, D., Berner, Z., Harting, M., 2004b, More evidence that the Chicxulub impact predates the K/T mass extinction: Meteoritics & Planetary Science, v. 39(7), p, 1127-1144.
12. Keller, G., Adatte, T. , Berner, Z., Harting, M., Baum, G., Prauss, M., Tantawy, A.A. and Stueben, D.,2007. Chicxulub impact predates K-T boundary: New evidence from Brazos, Texas, Earth Planet. Sci. Lett. 255, 339-356.
13. Lopez-Oliva, J. G., and Keller, G., 1996. Age and stratigraphy of near-K/T boundary clastic deposits in NE Mexico. Geol. Soc. Amer., Special Paper 307. p. 227-242.
14. Montanari, A., Koeberl, C., 2000, Impact stratigraphy. Lecture Notes in Earth Sciences, 93, Springer, Heidelberg, Germany, 364 pp.
15. Pardo, A., Ortiz, N. and Keller, G., 1996. Latest Maastrichtian and K/T boundary foraminiferal turnover and environmental changes at Agost, Spain. In, MacLeod, N. and Keller, G., (eds.), The Cretaceous-Tertiary Mass Extinction: Biotic and Environmental Effects, Norton Press, New York, p. 157-191.
16. Pusz, A.E., Miller, K.G., Kent, D.V., Wright, J.D., Wade, B.S., 2007. Global Effects of Late Eocene Impacts, AGU Joint Assembly, Acapulco, Mexico (2007) p.
17. Stinnesbeck, W., Keller, G., Schulte, P., Stüben, D., Berner, Z., Kramar, U., Lopez-Oliva, J.G., 2002. The Cretaceous-Tertiary (K/T) Boundary transition at Coxquihui, state of Veracruz, Mexico: evidence for an early Danian impact event? Am. J. S. Amer. Res. 15, 497-509.
18. Stüben, D., Kramar, U., Harting, M., Stinnesbeck, W., Keller, G., 2005, High-resolution geochemical record of Cretaceous-Tertiary boundary sections in Mexico: New constraints on the K/T and Chicxulub events: Geochemica et Cosmochimica Acta. v. 69 (10), p. 2559-2579.