Geoscientist Online

This is the response of Keller et al., to Jan Smit's original riposte of November 12. This was received 18 November 2003.

Keller (picture), Adatte & Stinnesbeck write: We welcome Jan Smit's contribution to the Chicxulub Debate. Smit, Alvarez and others in l992 originally proposed the Chicxulub K/T impact-tsunami theory to explain the sediments around the Gulf of Mexico. We propose the multiple impact theory to explain the new evidence uncovered over the past 12 years.

In his defense of the old K/T impact-tsunami theory, Smit selectively chooses outlier data while ignoring the main body of evidence, or simply denies that this body of (published) evidence even exists. This denial of evidence includes:

• The well documented multiple impact spherule layers in zone CF1 which spans the last 300 ky of the Maastrichtian (see review in Keller et al., 2003).
• The presence of several horizons of bioturbation in his so-called tsunami deposits (Ekdale and Stinnesbeck, l998; Keller et al., 2002).
• The presence of late Maastrichtian planktic foraminifera of zone CF1 in micritic limestones overlying the impact breccia in the new Yaxcopoil-1 core of the Chicxulub crater, including paleomagnetic data indicating chron 29R and late Maastrichtian stable isotopes.

1. Multiple impact spherule layers in NE Mexico

Evidence of multiple spherule ejecta layers from the Chicxulub impact was discovered by our team in the late l990's, including five masters and three doctoral students from the Universities of Neuchatel, Karlsruhe and Princeton (Stinnesbeck et al., 2000; Keller et al., 2002, 2003). These students mapped and analyzed the K/T boundary, the siliciclastic units, the spherule ejecta layers directly beneath it, and the underlying late Maastrichtian Mendez Formation marls over an area spanning about 60km2 and encompassing more than 40 outcrops (Figure 1).

Figure 1. Locations of K/T boundary sections with multiple impact spherule layers in marls of the late Maastrichtian Mendez Formation. To date the multiple spherule layers have been documented in over 40 outcrops from El Penon to Rancho Nuevo, a distance of about 100km, but only small-scale slumps in one small area (Mesa Juan Perez outcrops) of La Sierrita.

This first detailed investigation of these Mendez Formation marls revealed the presence of up to three additional spherule layers below the one directly underlying the siliciclastic unit. To date, the multiple spherule layers have been documented over 100 km from El Penon to Rancho Nuevo (Fig. 1). Within this region, small-scale internal slumps, spanning a few meters, were observed at the Mesa Juan Perez outcrops of the La Sierrita area (Schulte et al., 2003).

In the best outcrop exposures the spherule layers are interbedded in 10-12m of pelagic marls of planktic foraminiferal zone CF1, which spans the last 300,000 years of the Maastrichtian (see Figure 2 of position paper The Non-smoking Gun, this debate). Deposition of the lowermost spherule layer, and original deposition of the impact ejecta, occurred near the base of zone CF1, which indicates that Chicxulub predates the K/T boundary by about 300,000 years. We interpret all stratigraphically younger layers as reworked because they contain abundant reworked marl and spherule-rich clasts. In contrast, the lowermost spherule layer consists almost entirely of spherules and glass fragments, frequent convex contacts and agglutination between spherules, which indicates that they were still hot and ductile during transport and primary deposition.

Are the spherule layers correlatable?

Smit claims that none of the spherule layers can be correlated beyond 10m distance. In support of this claim he cites the small scale slumps of the Mesa Juan Perez outcrops, but ignores the over 40 other sequences that are undisturbed.

Not all sections contain four spherule layers. We would only expect that the original spherule deposit is correlatable, but not the reworked or slumped layers. Moreover, outcrop exposures are limited to the top few meters of the Mendez marls in most sections (Fig. 2). However, where full outcrop exposure is available, such as at El Penon and at Loma Cerca of the La Sierrita area, the lowermost spherule layer occurs 12-13m below the siliclastic deposit, the second and third layers are closely spaced about 2m above, and 4-6m of marls separate these two spherule layers from the topmost layer below the siliciclastic units (Fig. 2). Thus, we can show that even the reworked spherule layers can be correlated over an area of at least 35km from El Penon to Loma Cerca.

Figure 2. Stratigraphic and lithologic correlation of late Maastrichtian sections spaning an area of about 35km from El Penon to La Sierrita. Note that limited outcrop exposures have permitted examination of the top 15m of the Mendez marl Formation only in the El Penon and La Sierrita (Loma Cerca outcrop b) areas to date. However, in these two outcrops the spherule layers are correlatable and within the same zone CF1 with the lowermost layer predating the K/T boundary by about 300Ka.

Are multiple spherule layers due to mass flows?

In his critique, Smit seized on the presence of the small-scale slumps at Mesa Juan Perez to assert that the multiple spherule layers over the entire region were caused by impact triggered mass wasting. Where is the evidence for his interpretation?

Minor slumps in one small outcrop area can't be generalized to the more than 40 sections already analyzed over a 100km in northeastern Mexico. Even immediately adjacent to the Mesa Juan Perez slump section, despite limited outcrop exposures, the spherule layers can be traced over more 20m interlayered in Mendez marls, undisturbed and without signs of slumps. The absence of major slumps or tectonic disturbance in any of the northeastern Mexico sections argues against widespread mass failure of the Mendez basin margin.

Is there hemipelagic sedimentation between spherule layers?

Smit flatly denies that the marls between the multiple spherule layers represent hemipelagic sedimentation and contents that these marls are part of the mass flows. Where is the evidence for his interpretation. Smit presents none.

We published mineralogical data of these marl layers that reveal the characteristic nature of marls of the Mendez Formation (Adatte et al., l996; Stinnesbeck et al., 2000). The marls are lithologically uniform and show no structural features of either slumps or repeated faulting. There is no difference between marls below, between, or above the various spherule layers. The mottled character of the marls indicates bioturbation, and individual trace fossils (e.g., Chondrites, Planolites) are occasionally recognizable. All of this indicates normal pelagic sedimentation, rather than slump deposits.

We also published planktic foraminiferal data that show faunal assemblages and species abundances that are consistent with late Maastrichtian zone CF1 with little evidence of reworking, except for the spherule layers above the stratigraphically oldest one near the base of zone CF1 (Keller et al., 2002). The microfossil assemblages show species trends that are characteristic of zone CF1 and do not support a series of rapid and repeated mass flows.

How likely is repeated stacking by mass flows?

By Smit's interpretation the original spherule layer was deposited at the K/T boundary and, prior to the arrival of tsunami waves depositing the siliciclastic units, a series of rapid mass flows resulted in repeated stacking of slumped spherule and marl layers and thus artificially creating the illusion of normal Cretaceous hemipelagic sedimentation.Where's the evidence for this interpretation. None is provided.

This interpretation seems physically impossible. How could the unconsolidated spherule and marl layers remain separate entities during mass flows? And remain so repeatedly with one couplet deposited above another - presumably over a period of only hours to at most two days after the impact event? How could the planktic foraminiferal assemblages remain in tact and show long-term trends that are consistent with zone CF1? How could consolidated spherule-rich marl clasts form within hours to days? The answer - it is not possible.

There is strong evidence that the reworking of spherule layers occurred after the original spherule deposit lithified, which would have taken a long time period. This is indicated by the presence of consolidated spherule-rich marl clasts above the basal/oldest spherule layer and in some sections interspersed in the Mendez marls. This reveals that erosion, transport and redeposition of the original spherule layer occurred a long time after the initial deposition.

2. Is there a Tsunami Deposit?

Smit et al. (l992, l996) proposed the old impact-tsunami hypothesis to explain the siliciclastic deposits that separate the spherule layers from the overlying K/T boundary and Ir anomaly (Fig. 3a). In this scenario the glass spherules settled out first, followed by a megatsunami depositing the siliciclastic units and finally settling of fines depositing the Ir anomaly.

Figure 3a.The K/T transition at El Mimbral, NE Mexico, showing the siliciclastic deposit with the spherule unit 1 at the base, the thick sandstone unit 2, the alternating sand-silt-shale layers of unit 3 and K/T boundary above it. Erosive contacts separate these units and are also found within the units.

Critical evidence challenging the impact-tsunami interpretation surfaced almost from the very beginning - namely, bioturbation (churning caused by burrowing organisms) and erosional disconformities within the siliciclastic deposit.

The first challenge came during the l994 LPI-sponsored field trip (led by Keller, Stinnesbeck and Adatte) by the world reknown trace fossil expert, Toni Ekdale, who was invited to examine these deposits. At El Mimbral, he discovered trace fossils and bioturbation within the siliciclastic deposit. This indicated that invertebrates colonized the ocean floor and therefore deposition could not have been rapid or due to a tsunami. Although he was effectively booed for this observation by non-experts, who could not conceive that the impact spherules may not be of K/T age, he later returned to Mexico to study many of the classic K-T localities and document several horizons of bioturbation (Ekdale and Stinnesbeck, l998, Fig. 3b).

Trace fossils identified by Ekdale include Chondrites, Ophiomorpha, Planolites and Zoophycos. Ophiomorpha and other burrows are truncated by overlying sand beds in several intervals through the siliciclastic deposit, which indicates pulses of rapid deposition alternating with normal pelagic sedimentation. Several of these burrowed layers are present in the alternating sand-shale-silt (unit 3) in the upper part (Fig. 3b).

Figure 3b. Trace fossils from various horizons of units 2 and 3 of the siliciclastic deposits. Bioturbation is common particularly Chondrites in the fine interlayers of unit 3 (A, C). In general, Thalassinoides, Zoophycos and Ophiomorpha are common in the upper layers of unit 3, and J-shaped burrows infilled with spherules are found in the sandstone unit 2 and the sandy limestone layer of unit 1.

In the underlying sandstone (unit 2) there are truncated 10cm high J-shaped and spherule-infilled burrows probably made by crabs (Fig. 4).

Figure 4. J-shaped burrow infilled with spherules and truncated by erosion from the lower part of the sandstone unit 2 at El Penon.

Recently we discovered the same type of spherule-infilled burrows in the sandy limestone layer that separates the spherule layer at the base of the siliciclastic deposit at El Penon (Figs. 5a, 5b) and El Mimbral (Figs. 5, 6).

Smit criticised the use of the term 'sandy limestone layer', saying that it is essentially a sandstone cemented by sparry calcite at a later time. But terminology aside, he ignored the presence of truncated J-shaped spherule-filled burrows, which indicate deposition was interrupted and long-term.

Figure 5a. Unit 1 spherule layer separated by a sandy limestone layer with J-shaped burrows indicates that deposition of the spherules occurred in two phases separated by normal sedimentation.

Figure 5b. (below, left) J-shaped burrow infilled with spherules and truncated by erosion in the sandy limestone layer of the spherule unit 1 at El Penon.

Figure 6a. (below, right) Sandy limestone layer within spherule deposit at Mimbral.

These bioturbated horizons within the siliciclastic deposits effectively invalidate the old impact-tsunami hypothesis and also show that the underlying impact spherules cannot be coeval with the K/T boundary above the siliciclastic deposit.

Moreover, the burrowed sandy limestone layer separating the underlying spherules into two layers indicates deposition occurred in two phases separated by an extended time period during which limestone formed and organisms burrowed the ocean floor. This spherule ejecta could therefore not be of the same origin and age as the Ir anomaly above the siliciclastic unit.

Figure 6b. Close-up of sandy limestone layer within spherule deposit at Mimbral.

Smit denies evidence of bioturbation

Smit's response to the evidence of bioturbation is denial of their existence, except for the top 15cm of the siliciclastic unit, which is an extremely heavily burrowed hardground. As for the bioturbation documented and illustrated by the trace fossil expert Toni Ekdale, Smit states he couldn't find any. He helpfully suggests that the trace fossil expert misinterpreted 'pseudo-burrow types' that could have been mistaken for recent root traces, mud wasp nests, or synsedimentary fault and flame structures. He concludes: 'We believe for these reasons that true burrows at those lower levels are non-existent'.

It is inconceivable to us that non-experts, like Smit and collaborators, even suggest that a well-known trace fossil expert doesn't know the difference between fossil burrows and root traces, wasp nets or small faults. And they make these assertions despite the well-illustrated evidence and without any evidence to the contrary.

Smit cites current direction as strongest support for tsunami.

Smit states that the siliciclastic layers bear some similarities to turbidites were it not for their variable current directions, which he interprets as a series of different tsunami waves (Smit et al.,1996).

We have measured paleocurrent directions on ripple-marks and flute-casts in the sandstones of units 2 and 3 of the siliciclastic deposits (Fig. 7). All data range between N30-N200, giving a mean direction of N117. These data are not consistent with Smit's observations, who interpreted opposite current directions as a repetition of tsunami mega-waves along the coasts of the Gulf of Mexico (up-surge/back-surge).

Moreover, only three opposite directions have been observed in 94 measurements. These opposite direction are due to refraction against the edges of the channel. The spatial distribution of the clastic deposits indicates that the siliciclastic units are channelized and trend in a SSE direction, with differentiated path lines for each lithological unit. Thus, the clastic influx cannot be related to the direction of the postulated crater.

Figure 7. Palaeocurrent measurements do not support Smit's contention of impact-tsunami waves.

Our data on mineralogic, sedimentologic, trace fossils and current directions are not consistent with a tsunami depositional model. We suggest that the siliciclastic deposits represent non-catastrophic multiple gravity flow deposits, spanning a considerable amount of time. These deposits are not directly related to the K/T boundary above it, or the spherule ejecta layers below.

3. Chicxulub predates K/T

Age of Sediments above Impact Breccia

The new Chicxulub core Yaxcopoil-1 was supposed to settle the controversial issue of the age of the impact breccia and hence the age of the Chicxulub impact.

Within this 1511m core a 100m thick suevite breccia (894-794m) identifies the Chicxulub impact event. Above the suevite breccia are dolomitic, micritic limestone and limestones. The K/T boundary is 50cm above the suevite breccia (Fig. 8). In order to have a common origin for the suevite breccia and the K/T boundary, this 50cm layer must be interpreted as part of the impact event, such as backwash and crater infill, as argued by Smit.

But this critical 50cm interval between the breccia and the K/T boundary is late Maastrichtian zone CF1 in age, as indicated by (a) planktic foraminiferal assemblages of zone CF1, (b) magnetostratigraphy showing chron 29R below the K/T boundary, and (c) typical late Maastrichtian carbon isotopes values followed by the characteristic negative excursion at the K/T boundary. All three age proxies indicate that deposition of this interval predates the K/Tboundary.

Figure 8. Lithology of the critical 50cm interval between the impact breccia and the K/T bondary in well Yaxcopoil-1. If the impact breccia is K/T age, then this 50cm interval of sediments must be interpreted as backwash and crater infill deposited over a geologically instantaneous time span. Sediment and microfossil analyses indicate deposition occurred during the last 300Ka of the Maastrichtian in variable but low energy environments.

The planktic foraminiferal assemblages are present in the laminated micritic limestones, where they are common to abundant and characteristic of zone CF1 assemblages, which span the last 300 kyr of the Maastrichtian. These assemblages include Globotruncana stuarti, G. insignis, G. arca, G. falsocalcarata, Abathomphalus mayaroensis, Rosita contusa, R. petaloidea, R. walfishensis, Rugoglobigerina rugosa, R. macrocephala, Plummerita hantkeninoides, Globotruncanella petaloidea, Heterohelix, Hedbergella sp. and Globigerinelloides aspera (Fig. 9).

Figure 9. Planktic foraminifera from the micritic limestones above the suevite breccia and the early Danian in Yax-1, Chicxulub. Early Danian zone Pla: Woodringina hornerstownensis, 2. P. eugubina, 3. Parasubbotina pseudobulloides, 4. Morozovella inconstans. Late Maastrichtian zone CF1: 5. Plummerita hantkeninoides(sample 20), 6. Rugoglobigerina macrocephala (sample 9), 7. R. rugosa (sample 12), 8. R. hexacamerata (sample 10), 9 & 10 Globotruncana insignis (sample 20), 11. Rosita contusa (sample 9).

These foraminiferal assemblages indicate deposition occurred after the Chicxulub impact event and prior to the K/T boundary mass extinction, sometime during the last 300 kyr of the Maastrichtian, similar to the age of spherule deposition in NE Mexico.

Foraminfera or dolomite crystals?

Smit et al contest the presence of planktic foraminifera, but not the paleomagnetic data or carbon isotopes. Smit states that neither he nor Antonio Arz (Assistant Professor at the University of Zaragoza) were able to find any determinable foraminiferal remains in any of the samples. But that they found dolomite overgrowth that has the same thickness as a foraminifer shell wall. They conclude that these dolomite rhombs resemble foraminifera in size and thickness and were mistaken as such by Keller - just as Ekdale mistook rootlets, wasp nests and small faults as trace fossils.

But foraminifera do not look like dolomite rhombs! This is obvious from the illustrations in Figures 9 and 10. Sediments that are partially or completely dolomitized (as illustrated by Smit) do not preserve foraminiferal shells. Moreover, dolomite rhombs never resemble test walls of foraminifers.

Identifying foraminifera in thin sections is difficult even in well-preserved limestones and it takes experience to identify them in recrystallized micritic limestones. It should therefore not be surprising that Smit and Arz encountered difficulties since neither has much experience in this respect. This is evident from their search in dolomitized intervals, where no foraminifera are preserved. If they concentrate their search in the micritic limestone intervals, they will find plenty of foraminifera.

Reworked microfossils?

Smit suggests that if late Maastrichtian planktic foraminifera turn out to be present, they must be reworked. He states that Arz found Albian planktic foraminifera reworked from the older Cretaceous below in the core, but provides no illustrations. Our analysis of the entire core has revealed no Albian planktic foraminifera. The only interval with planktics are from the late Cenomanian Rotalipora cushmani zone. He also states he found some benthics within breccia clasts, but does not illustrate them. We found no breccia clasts, and the only benthic foraminifera observed are small buliminellids.

This argument seems more designed to safeguard his old K/T impact-tsunami hypothesis than search for evidence of the real depositional history of these rocks. The planktic foraminifera are highly unlikely to be reworked since there is no source of older sediments with these assemblages in the rocks below the breccia. The older Cretaceous rocks on the Yucatan platform were deposited in shallow lagoonal to subtidal environments that did not support planktic foraminifera. Moreover, the assemblages are too narrowly constraints within zone CF1 and too diverse. Any reworking would have included species from different older rocks.

Sandstone or micritic limestone?

Smit argues that the critical 50cm interval is not micritic limestone, but rather coarse and fine grained sandstone and cross-bedded sandstones - and therefore by definition reworked. The only evidence he offers is a backscatter electron micropgraph of dolomite overgrowth of sand grains. Though no indication is given from which sample this image was taken, it is from the basal 5cm directly overlying the impact breccia with its cross-bedded sand layers in the top meters. There is no reason why there shouldn't be any sand grains in this interval (Fig. 11a). The micritic limestone begins above it.

What Smit calls coarse and fine-grained sand layers in this interval are actually alternating micritic limestone and the coarser dolomitized layers giving the appearance of size grading (Figs 11b, c, d). How do we know that these are not sand grains. Sand grains would be major constituents of insoluble residues. Our analysis of insoluble residues revealed none.

Figure 10. Limestone lithologies from Yax-1, Chicxulub. Numbers refer to samples indicated in Fig. 8. 12. Laminated microlayers in micritic limestone. 16 & 18.. Laminated microlayers in micritic limestones enhanced by partial dolomitization of some layers giving the appearance of grain size changes. 21. Dolostone, advanced diagenetic alteration. This is only present in the fivecm above the suevite breccia and may contain some quartz grains.

Cross bedding - backwash and crater infill?

• Near the base of the 50cm interval are three 1cm thick layers of oblique bedding within an 8cm thick interval between 794.45 and 794.53m (Fig. 8), which Smit interprets as cross-bedded dolomitic sandstones. But he presents no evidence (insoluble residues) that these layers are sandstone and not dolomite rhombs as we demonstrate from the rest of the section. We have no samples from this interval and so could not perform the test.
• However, even if there is sand present, three1cm thick layers would indicate slightly agitated bottom currents, but not the high-energy current deposition necessary to interpret the 50cm interval as backwash and crater infill.
• Backwash and crater infill is expected to erode, transport and redeposit material from the surrounding crater walls via slumps or gravity flows. To validate this interpretation the sediments must contain evidence of:
• High energy currents (e.g. cross-bedding, grain size grading), which we show is not present or at best very minor in three 1cm intervals near the base.
• Highly variable lithoclasts from the diverse underlying strata (e.g. shallow water limestones, gypsum, dolomite, anhydrite, breccia), which is rare to non-existent.
• A significant amount of melt rock either as glass fragments or glass altered to Cheto smectite. Glass fragments are very rare and no Cheto smectite is present.
• Reworked microfossil and macrofossil faunas from various ages of the underlying lagoonal to subtidal sediments. None are present - or extremely rare.

Glauconite - Conclusive evidence of long-term depsosition

There are five green clay and glauconite layers at the K/T boundary and below it (Fig. 8), which have been interpreted by Smit and others as evidence of abundant altered impact glass.

Impact glass alters to a characteristic, almost pure Cheto smectite (Keller et al., 2003b). XRD and ESEM analyses indicate that Cheto smectite is present in the impact breccia. However, the green clay at the K/T boundary, or in the other four intervals, is of glauconitic origin and represents mature glauconite (Fig. 11A)(Odin, l988). The insoluble residue grains from the green clay layer in sample 8, as well as samples 13 and 17 show similar glauconitic compositions. Thus there is no altered glass in these clay layers, rather they are mature glauconite layers.

Figure 11. Yaxcopoil-1. A) thin section micrograph of the green K/T clay layer (sample 8) with insert marking location of analysis. The XRD diffractogram of this green clay indicates the presence of mature glauconite. In contrast, the XRD of a sample from the breccia shows the presence of well-crystallized Cheto smectite, which is a typical altered glass product. B) ESEM micrograph and EDX analysis of the green clay show similar glauconitic compositions for the insoluble residue grains from the green layers of samles 13 and 17. The glauconite reference standard from the SEM petrology Atlas is shown for comparison.

Glauconite forms at the sediment-water interface in environments with very slow detrital accumulation as a result of sediment winnowing by gentle bottom currents and clast generation. The five green layers therefore indicate long pauses in the overall quiet depositional environment with slightly more active, though still minor current activity and small-scale transport.

Glauconite layers are characteristically burrowed and the five thin layers in the 50cm thick interval above the impact breccia are no exception. Although burrowing (Thalassinoides, Spreiten) is strongest in the 8cm below the K/T boundary glauconite layer, each of the four lower glauconite layers is also burrowed.

The five layers of glauconite formation thus effectively rule out a scenario of rapid deposition, high-energy currents, backwash and crater infill.

US Terrestrial sections more complete?

Until now the sections in NE Mexico and the Chicxulub crater have been considered as containing the most continuous sedimentation records and most convincing evidence for the K/T impact-tsunami hypothesis. With the contrary evidence accumulating, Smit now argues that the best evidence comes from the terrestrial sections of the US western interior. But terrestrial records are always highly condensed, often incomplete, and cannot be reliably correlated with the marine record (see our reply to Tyberg, this website).

There is no age control on the deposition of the closely spaced two clay layers of spherules and iridium anomaly Smit now considers as his best supporting evidence for a single impact. Smit states that they present one and the same event because 'not even a single layer of one fall season of leaves or plant material occurs between the two layers.' This is fantasy. There is no way of knowing what time interval may be missing between these clay layers. If the spherules are of Chicxulub origin, then the best age correlation is with the spherule layers in NE Mexico, which would place them in the late Maastrichtian.

Conclusions

The evidence in the Chicxulub core Yax-1 overwhelmingly points to Chicxulub as predating the K/T boundary impact event. All evidence in northeastern Mexico sections, including the multiple spherule layers interbedded in late Maastichtian zone CF1 marls, the several horizons of bioturbation in the siliciclastic deposit, the K/T boundary and Ir anomaly above it, point to multiple impacts: the Chicxulub impact in the late Maastrichtian about 300Ka before the K/T boundary, and the K/T boundary impact represented by the global Ir anomaly.

Figure 12. Multiple impact scenario based on impact glass spherulelayers and Ir anomalies surrounding the Gulf of Mexico, Caribbean and Central America. Note that the Ir anomaly at the K/T boundary is found worldwide. The Pla(l) zone Ir anomaly has so far been documented from Guatemala, Haiti, Mexico and possibly ODP Site 1049.

The old impact-tsunami hypothesis of Smit et al. requires ever more complex scenarios and denial of evidence in order to tie the Chicxulub impact to the multiple spherule layers in late Maastrichtian sediments and the siliciclastic deposit to the K/T boundary and irdium anomaly above it. The new Chixculub core Yax-1 requires still further complexity and denial of evidence to sustain this old hypothesis.

The K/T impact-tsunami scenario now requires the rapid redeposition of the original impact spherule layer by repeated mass flows depositing spherules in concentrated layers and separated by meters of undisturbed hemipelagic marls in the region surrounding the Gulf of Mexico. This regionwide repeated collapse of the basin margin would occur prior to impact tsunami depositing the siliciclastic sediments with bioturbation (ignored in this scenario), followed by the settling of the iridium. All this would have happened within hours to two days. In the Chicxulub crater, infill and backwash are supposed to explain laminated micritic limestones and five thin bioturbated glauconite clay layers.

This scenario is within the realm of fantasy. There is a complete lack of evidence supporting either repeated mass wasting, large-scale slumping, tsunami deposition or backwash and crater infill. For the most part these interpretations are based on beliefs and unsupported assertions. The basic purpose of this scenario appears to be the keeping alive of the old hypothesis - that Chicxulub is the K/T impact event.

 

References

Adatte, T., Stinnesbeck, W. and Keller, G., 1996. Lithostratigraphic and mineralogical correlations of near-K/T boundary clastic sediments in NE Mexico: Implications for origin and nature of deposition. Geol. Society Amer., Special Paper 307, p. 211-226.

Ekdale A.A. and Stinnesbeck W. l998. Ichnology of Cretaceous/Tertiary boundary beds in northeastern Mexico. Palaios 13: 593-602.

Keller G. et al. 2002a. Multiple spherule layers in the late Maastrichtian of northeastern

Mexico. Geological Society of America Special Paper 356: 145-161.

Keller G., Stinnesbeck W., Adatte T. and Stueben D. 2003b. Multiple impacts across the Cretaceous-Tertiary boundary. Earth-Science Reviews 62, 327-363.

Keller G., Stinnesbeck, W., Adatte, T. Holland, B., Stueben, D., Harting, M., De Leon, C. De la Cruz, J., 2003. Spherule deposits in Cretaceous-Tertiary boundary sediments in Belize and Guatemala. Jounal of the Geological Society, 160, 1-13.

Odin, G.S., l989. Les glauconies, Ph.D. Thesis, Pierre et Marie Curie University, Paris, France, 251p.

Smit J. et al. l992. Tektite-bearing deep-water clastic unit at the Cretaceous-Tertiary boundary in northeastern Mexico. Geology 20: 99-104.

Smit J. et al. l996. Coarse-grained clastic sandstone complex at the K/T boundary around the Gulf of Mexico: Deposition by tsunami waves induced by the Chicxulub impact. In The Cretaceous-Tertiary Event and other Catastrophes in Earth History, edited by Ryder G., Fastovsky D. and Gartner S. Geological Society of America Special Paper 307: 151-182.

Stinnesbeck W. et al. 2001. Late Maastrichtian age of spherule deposits in northeastern Mexico: Implication for Chicxulub scenario. Canadian Journal of Earth Sciences 38: 229-238.

• The Great Chicxulub Debate 2004
• View the general discussion page
• Read Jan Smit's response to Keller et al.'s first article
• Read what Jan Smit said about this piece