Product has been added to the basket

The Chicxulub Discussion Part 1

This page contains discussion related to Keller, Adatte & Stinnesbeck's article, The Non-Smoking Gun. The more recent items in the discussion are to the top of the page. Go to Discussion page 2 for aubmissions after 22.12.03. If you wish to contribute to this moderated on-line discussion, please email the Editor.

Wider issues remain elusive

From Andrei Ol'khovatov, Moscow (posted 22.12.03)

Sir, As a non-expert, but an observer of the debates, it looks like we are gradually shifting from the original Chicxulub discussion into other items as the Rancho Nuevo discussion, etc. This is not surprising, as the debates are about a subject on which modern science has little knowledge (see my earlier contribution). There is (fortunately!) no real-life verified model of what happens in the case of a large asteroid (or comet) strike. Also, we have a little understanding what happens during such "super-eruptions" as the Deccan event. By the way, couldn't these "super-eruptions" be just a minor part (a "tip of an iceberg"?) of phenomena occuring with the Earth at such times? Mankind lives during a rather calm period of Earth history (and even it gives us a lot of puzzles), so it is not reliable to use an extrapolation of our modern experience to such extreme situations of the Earth history.

I think we can agree that although modern science is not ready to give a final answer, further research and these debates are bringing the answer a little bit closer.

Meanwhile, a Merry Christmas and Happy New Year to All!

Slumping evidence - Reply to Markus Harting

From Jan Smit (rec'd 17.12.03)

Sir, Markus Harting (15 Dec) may be fulminating against evidence for slumping, but the evidence remains unmistakable. I add one more picture (image1) showing slumps in the top layers of the Mendez here at Rancho Nuevo. Lower down the Mendez strata are here not deformed. The fluidization, as I said earlier, occurs for the most part in the top 1-2 meters of the Mendez. Still, some deformation must have taken place, as the turbidite beds are clearly about 20 degrees inclined (NOT horizontal as Harting claims), whereas the top sandstone layers here are almost horizontal. I also add a picture (image2) of one of the additional spherule layers at el Penon. Again, I do not argue against multiple "layers", but think they are doubled or tripled by syn-sedimentary slumping, fluidisation etc., which makes them all around the same age, not thousands of years separated in time.
Image.1 Image 1 Rancho Nuevo, image of a small turbidite fan, about 10m below the clastic beds. This image is taken of the same outcrop, but from a different angle as the image shown by Harting (included in image 1). The turbidite beds are inclined . However, the top layers of the clastic beds (In the background, not the turbidite fan shown in the foreground) are almost horizontal. The beds above the turbidite fan show evidence for slumping!
Figure.2 Image 2. Spherule layer at el Penon, some 5m below the base of the sandstone beds, where also spherules occur. On first sight there appear two layers. However, the lower one is truncated at both ends, without signs of channeling, and is clearly inclined versus the almost horizontal upper one at the base of the sandstone beds.

How to explain? There are various mechanisms thinkable, all related to synsedimentary deformation: package gliding, slumping, fluidization. Tectonic deformation is unlikely regarding the almost total absence of tectonic deformation in the area, except for slight tilting of beds and small scale normal faulting. Aggradational filling of a large channel is also unlikely here. Also unlikely is an explanation as a normal sedimentary layer, that would be correlatable over large distances as Keller claims. (click on image for a detailed view)

Maastrichtian greenhouses

From Lee Nordt (rec'd Monday 15 December 2003)

Sir, Recent terrestrial evidence using stable C and O isotopic compositions of paleosol carbonate point to two intense greenhouse events in the Maastrichtian between 69 to 68Ma ago and between 65.5 and 65Ma ago (Nordt et al., 2002; 2003; Atchley et al., 2004).

These two intervals, designated as the Mid-Maastrichtian Event (MME) and Late Maastrichtian Event (LME), were characterized by pCO2 levels of between 1000 and 1400 ppmV with associated regional temperatures (West Texas) of 21 to 23°C, both corresponding to marine transgressions. Climate models indicate that when atmospheric CO2 and temperatures are at these levels continental glaciers will not form. In contrast, when atmospheric CO2 levels drop below 1000 ppmV, with mean global temperatures of less than 20°C, continental glaciers will form. Intervals between the two Maastrichtian events as detected in west Texas, including the K-T boundary and into the Danian, meet the climate criteria for glaciation in association with marine regression.

Our work strongly indicates that climate shifts prior to the K-T mass extinction, as currently understood, were dramatic. Mounting evidence from both marine and terrestrial records indicate that Deccan trap volcanism was responsible for the LME, but the cause of the MME is unknown. Phase two of our research will involve studying paleosols specific to the two greenhouse Maastrichtian events and those formed during intervening cool greenhouse (icehouse?) intervals using other geochemical indices for ecosystem reconstruction to assess the magnitude of response of the terrestrial record to dramatically fluctuating pCO2 and temperatures.


Nordt, L., Atchley, S., and Dworkin, S. 2002. Paleosol stable carbon isotopes indicate extreme fluctuations in pCO2 across the K-T boundary. Geology, Vol. 30, p. 703-706.

Nordt, Atchley, S., and Dworkin, S. 2003. Terrestrial evidence for two intense greenhouse events in the latest Cretaceous. GSA Today, Vol. 13, p. 4-9.

Atchley, S., Nordt, L., and Dworkin, S. 2004. Eustatic control on alluvial sequence stratigraphy? The K/T transition in west Texas, USA. Journal of Sedimentary Research (May).

'Unmistakable Evidence' for slumping at Rancho Nuevo?

Reply to Jan Smit (see below) Received 15.12.03

From Markus Harting*

Sir, It is easy to find some sedimentological features that could be interpreted as slumps, in most stratigraphic sequences of any age or place. What is at stake here is Smit's sweeping interpretation that some isolated small slump on the side of the channelised deposits of the Rancho Nuevo outcrop represents 'unmistakable evidence' of large scale slumps or fluidized 'diapir-like' squeezing upward of strata due to shaking caused by the Chicxulub impact. Sound geologic interpretations of such regional scale are not based on isolated, small and local features, but must include the regional database.

In Keller et al's response (Debate round II) Fig. 1 shows the undisturbed horizontal stratification of Mendez marls at Rancho Nuevo. Large areas beneath and on the side of the channelised siliciclastic deposits are clearly not affected by fluidised or slump structures. I could not recognize these structures in the field, nor could they be observed by other investigators (López-Oliva 1996; Schulte 2003).

I also have grave reservations about Smit's negating the existence of multiple spherule layers, despite well-documented evidence to the contrary by many investigators, including myself. Smit's musings about fluidisation and slump structures to explain the multiple spherule layers are sheer speculations in the absence of detailed geochemistry, microscopy, mineralogy and laboratory studies.

In an earlier interpretation, Smit et al. (l992, l994) argued that the spherule layer of unit 1 'was deposited by the first tsunami waves, which reworked impact ejecta that had just arrived'. By Smit's hypothesis the overlying sandstone of unit 2 was deposited within about 2.5 hours, followed by the sand-silt-shale layers of unit 3 within another couple of days (Smit et al., l996). It has always puzzled me how reworking of sediments immediately after spherule deposition could have avoided the mixing in of the underlying spherules into the sandstone of units 2 and 3.

In his reply to my comment on multiple spherule layers at Rancho Nuevo Smit claims that "all spherule layers in northeastern Mexico are reworked - there is no primary layer left." This agrees with my observations of the unit 1 spherule layer throughout northeastern Mexico: unit 1 spherules are reworked from the spherule layer(s) within the Méndez marls. These spherule layers are older than the K/T boundary, which confirms Keller et al.´s observations of 'pre K/T-deposited ejecta', which is the likely source of spherules of unit 1.

There is no evidence that the existence of multiple spherule layers could be interpreted by Smit´s 'multiplets' or 'alternations of soft Méndez-clast-rich and spherule-rich aggradational layers'. Moreover, it is hard to imagine that such a random sedimentological process could be operating over an area of over 80 km (from Rancho Nuevo to El Penon) in a heterogeneous sedimentary basin, where the only evidence of disturbance are occasional small local slumps (e.g. Mesa Juan-Perez; Schulte, 2003). Smit's speculative interpretations fail to be convincing in view of the clear evidence to the contrary throughout northeastern Mexico.


López-Oliva, J.G. (1996): Stratigraphy of the Cretaceous/Tetiary (K/T) boundary transition in northeastern ans east-central Mexico. PHd-Princeton University.

Schulte, P. (2003): The Cretaceous-Paleogene transition and Chicxulub impact ejecta in the northwestern Gulf of Mexico: Paleoenvironments, sequence stratigraphic setting and target lithologies.  PHd-Karlsruhe University.

Smit, J., Montanari, A., Swinburne, N.H.M., Alvarez, W., Hildebrand, A.R., Margolis, S.V., Claeys, P., Lowrie, W., Asaro, F. (1992): Tektite-bearing, deep-water clastic unit at the Cretaceous-Tertiary boundary in northeastern Mexico. Geology, 20, 99-103.

Smit, J., Roep, Th.B., Alvarez, W., Claeys, Ph. & Montanari, A. (1994): Is there evidence for Cretaceous-Tertiary boundary-age deep-water deposits in the Caribbean and Gulf of Mexico : Comment and reply. Geology, 953-958.

Smit, J., Roep, T.B., Alvarez, W., Montanari, A., Claeys, P., Grajales-Nishimura, J.M. and Bermúdez, J., 1996. Coarse-grained, clastic sandstone complex at the K/T boundary around the Gulf of Mexico: Deposition by tsunami waves induced by the Chicxulub impact. Geological Society of America Special Paper 307,151-182.

*Institute of Mineralogy and Geochemistry, Karlsruhe University, Germany

Beata est veritas

From Keith James (15.12.03)

Sir, I support the idea that multiple impacts, related volcanism and environmental consequences caused a progressive extinction around the K/T boundary (1). I also support suggestions that impacts have contributed significantly to Earth history (2,3,4). However, the latter include ideas that large impacts initiated rifting and the formation of large igneous provinces. So I must ask why Chicxulub, apparently so large, did not have such dramatic effects?

I would like to contribute two observations for impact collectors. Satellite-derived bathymetry reveals a circular depression, some 180km in diameter, on the floor of the Venezuelan Basin, just SE of the Beata Ridge (intersection W70°-N15° falls on its eastern rim at 3 o'clock). I christened this the Beata Crater (5,6,7). I am tempted to relate it to a regional, abrupt and violent Middle Eocene event in which giant olistostromes became emplaced around the Caribbean and to the Eocene strewn field of the SE USA (8). This was also the beginning of easterly drift of the Caribbean Plate relative to both N and S America (9). However, the Beata Crater is unconfirmed by geophysical/geological investigation and its age is unknown. It could be an oceanic relative of Chicxulub. A further circular anomaly, which I tentatively name the Tierra del Fuego Crater, occurs at 64°30°W, 56°30°S, close to the tip of S America. It is clearly visible on the Global Topography & Bathymetry map published by GETECH (10).


Keller et al., Geoscientist, v. 13, no. 11, p. 8-11.

Marvin, U. B., 1990, Impact and its revolutionary implications for geology: In, Sharpton, V. L. and P. D. Ward, Global Catastrophies in Earth History, GSA Special Pub. 247 p. 147-154.

Rampino, M. R., 1999, Impact crises, mass extinctions, and galactic dynamics: The case for a unified theory: In, Dressler, B. O. and V. L. Sharpton, eds., Larger Meteor Impacts and Planetary Evolution II, GSA Special Paper 339,p. 241-248

Price, N. J., 2001, Major Impacts and Plate Tectonics: A Model for Phanerozoic Evolution of the Earth's Lithosphere: Routledge Publications, 368 p.

James, K. H., 1997, A possible impact crater in the Venezuelan Basin, Caribbean Sea: in Avé Lallement, H. G., Final Report: Evolution of the eastern Caribbean-South American plate boundary zone: Transpression, volcanic arc accretion, and orogenic float, Workshop, Rice University, Houston.

James, K. H, K. A. Soofi and A. C. Weinzapfel, 1998, A possible impact crater in the Venezuelan Basin, Caribbean Sea: poster at the 15th Caribbean Geological Conference, Jamaica, July 1998

James, K. H, K. A. Soofi and A. C. Weinzapfel, 1998, Ocean floor mapped from space: AAPG Explorer, October, 1998, p. 20-23.

James, K. H., 2002, Cretaceous-Eocene Flysch-Wildflysch Deposits of Northern South America: Implications for Tectonic History and Early Episodes of Hydrocarbon Generation, 16th Caribbean Geological Conference, Barbados, Abstracts, p. 39.

James, K. H., 2003, A Simple Synthesis of Caribbean Geology: AAPG International Conference, Barcelona, extended abstracts CD.

Geophysical Exploration Technology, C/O Department of Earth Sciences, University of Leeds, Leeds

Reply to Markus Harting (see below)

From Jan Smit

Sir, I have my reservations about how to recognize fluidization or and resulting slump structures in the Mendez marls by geochemistry, microscopy, mineralogy and laboratory studies. It is only in the field that one can recognize these and I invite Markus Harting to again have a closer look at the unmistakable evidence for slumping at Rancho Nuevo (image 1). On the side of the channels there is ample evidence for synsedimentary oversteepening of the channel walls. I include here an image of such a slump taken just below the criticized 'diapirs'

The picture (Fig. c of Smit) of the Rancho Nuevo channels was taken as much as possible parallel to the longest axis of the two channels (see inset image 2), in order to avoid such a "result of the unique view across, and as a cross section, of the channelized siliciclastic deposit". The diapiric movements of the top layers of the Mendez are real, not an artifact of the figure.

Also, the presence of "well preserved ejecta" (actually not well preserved, all tektite glass is altered no matter in what layer) in the supposed lower primary layer is also well visible at the Mimbral site. Well preserved (i.e. not flattened) spherules occur adjacent to sublayers with flattened green spherules. All spherule layers in eastern Mexico are reworked - there is no primary layer left. I have no objection against multiple spherule horizons. I have given my interpretation of the 'multiplets': they are alternations of soft Mendez-clast rich and spherule rich aggradational layers, not separate events separated by thousands of years.
Image.1 Image 1: Slumpfolds of one of the turbiditic layers in the Mendez shales just underlying the clastic sandstones (see inset and image 2 for location). Click on image for a larger version

Image.2 Image 2: Steepened sidewall, oversteepened due to the sinking of the channel into the soft Mendez marls, of the NE-most channel 2, filled with the clastic sandstones. Vertical striations and furrows on the sidewall indicate the upward movements of the Mendez marls against the channel. The horizontal lineations, perpendicular to the furrows, are synsedimentary faults that indicate similar movements within the sandstones itself. The top sandstone layer stretches across both channels and the top of the diapirized Mendez. Inset: fieldsketch of the orientation of the two channels, with the diapirized Mendez in between. Red arrow, image 1 slump. Blue arrow, image 2.

Chicxulub Debate Round II: Keller, Adatte, Stinnesbeck respond to Smit (1 of 4) 

Smit smited

"Fluidized, squeezed, diapir-like slumped and folded? Méndez-Formation at Rancho Nuevo?

From Markus Harting*

In Smit's recent reply to Keller et al. he claims that the Upper Cretaceous Méndez shales at the Rancho Nuevo section are fluidised and squeezed up, diapir-like. He further claims that no additional spherule layers are present in the obviously bedded and undisturbed Mendez marls below the spherule layer of unit 1 at the base of the siliciclastic deposit (Fig. c of Smit). No data are presented in support of these claims.

We have conducted a detailed investigation of the Rancho Nuevo outcrop based on field examinations and laboratory studies, including sedimentology, geochemistry, mineralogy and microscopy. Our studies revealed no evidence of fluidised or squeezed Méndez Marls, or squeezing at the edges of the sandstone unit. The outcrop feature interpreted by Smit as 'diapir-like' appears to be the result of the unique view across, and as a cross section, of the channelised siliciclastic deposit.

Smit's claim that no additional spherule layers are present below the unit 1 spherule deposit is also unconfirmed. Our investigation revealed at least one more spherule horizon approximately 60-70 cm below the base of the spherule-rich deposit (SRD) of Unit 1, which supports the existence of multiple spherule horizons (Keller et al., 2002, 2003).

Our geochemical and mineralogical analyses of the spherule unit 1 (Smit's primary ejecta layer) revealed altered ejecta material and secondary authigenic minerals due to hydrothermal influence, in contrast to well preserved ejecta within the Méndez Marls. These investigations, as well as the stratigraphic position, indicate a reworked origin for the uppermost spherule layer, which is commonly labelled unit 1 of the siliciclastic deposit.


Keller G. et al. 2002. 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. 2003a. Multiple impacts across the Cretaceous-Tertiary boundary. Earth-Science Reviews 62, 327-363.

*Institute of Mineralogy and Geochemistry, Karlsruhe University, Germany

Multiple impacts?

From Tommy Tyrberg* Received Monday 10 Nov

Sir, The problem with the data of Keller et al. is that they are all derived from the area in or around the Chicxulub crater where the sedimentation conditions were very disturbed and abnormal to put things mildly. More distant K-T boundary sites, of which there are now a considerable number, both marine and terrestrial invariably contain only a single impact layer.

In particular I would like to point to the sequence from the Denver basin described in Rocky Mountain Geology, Vol 38:1. This is a K-T boundary sequence that 'has everything', good magnetostratigraphy, good palynology, a fairly high and seemingly even sedimentation rate, several volcanic ash layers with good absolute dates and a boundary sequence with boundary clay, shocked quartz, iridium, spherules and a fern spike all in the usual order. The palynology indicates a large scale floral extinction at the boundary layer. There is even hadrosaur and ceratopsian fossils c. 4m below the boundary and and Protungulatum fossils about 12 metres above it showing that the turnover between Maastrichtian and Early Paleocene fauna must have happened quite close to the K-T boundary. There is no trace of multiple impacts.

Of course it is always possible that all impact layers except one have eroded completely away while the last one was left intact and that the sedimentation rate happened to vary in such a way as to hide this process but it does not seem very likely.

However there is one scenario that could result in multiple impact layers in the Caribbean and a single layer anywhere else. This is if there were multiple impacts quite closely spaced in time and space, say within a few hours and several hundred kilometres of each other. In that case the ejecta blankets emplaced by the ejecta curtains and/or base surges in the near zones would be superposed upon each other while further away where the ejecta arrived ballistically at the top of the atmosphere the impact signatures would merge. Such a pattern would be expected from a binary asteroid or a fragmented comet like Shoemaker-Levy.

*Norrkoping, Sweden.

Gerta Keller replies to Tyberg: (Rec'd 14.11.03)

Terrestrial record more complete?

Tyberg's claim that terrestrial sections are more complete than marine sequences comes as a complete surprise to us or any biostratigrapher, sedimentologist, palaeontologist or geochemist working with the temporal sedimentary record. It is well known that terrestrial environments are inherently incomplete and fragmented due to sporadic deposition in lakes, rivers or deltas where non-deposition and erosion predominate. This is particularly the case during the sea level regression of the Western Interior that characterizes deposition in the Denver Basin across the K/T transition. Although these sections are excellent terrestrial records, they are not temporally complete. The absence of impact evidence does not prove anything in areas of frequent erosion and non-deposition.

Tyberg states that 'there are several volcanic ash layers with good absolute dates and a boundary sequence with boundary clay, shocked quartz, iridium, spherules and a fern spike all in the usual order'. But what is the exact timing of these layers? Absolute dating of the ash layers varies from 65.03±0.26Ma to 65.74±0.43Ma (Raynolds and Johnson, 2001) with the error margin as large or larger than the age between the Chicxulub and K/T boundary impacts. To our knowledge there are no terrestrial sections where shocked quartz, iridium and spherules are found in precisely the same stratigraphic interval. In terrestrial sections it is quite difficult to evaluate the time of deposition of each of these layers, which are often amalgamated, reworked or simply eroded.

Mexico's sections disturbed?

Tyberg's claim that sites in and around Chicxulub are 'very disturbed and abnormal to put things mildly' is also curious. The northeastern Mexico sites are 2000km from Chicxulub and not disturbed. Hemipelagic marine sediments characterize these sequences and they are only interrupted by the sandstone-silt-shale unit below the K/T boundary and above the highest impact spherule layer. As noted in the position paper for this debate, this siliciclastic unit is also bioturbated at several discrete intervals, which indicates deposition of this unit occurred over an extended time interval during which marine burrowing organisms repeatedly colonized the ocean floor (Ekdale and Stinnesbeck, l998). The up to four Chicxulub impact spherule ejecta layers below this deposit therefore must predate the K/T boundary as also indicated by biostratigraphy.

K/T Ir anomaly & chicxulub ejecta not the same

Tyberg argues that since deep-sea sections contain only a single impact layer, this means that only one impact occurred. Most deep-sea sections contain a millimeter-thin red layer enriched in iridium at the K/T boundary and the mass extinction of tropical and subtropical planktic foraminifera characterizes and defines this interval. That same red layer and iridium enrichment is also present in the sections of northeastern Mexico, where it marks the K/T boundary and mass extinction - just like in the deep-sea sections. These records are therefore comparable and document the same event.

However, in the Mexican sections, the K/T boundary, red layer, Ir-enrichment and mass extinction of planktic foraminifera are stratigraphically and temporally well above the Chicxulub impact glass spherule ejecta layers, as well as above the siliciclastic unit. These spherule layers are interbedded in hemipelagic marls. As noted in our position paper, we consider the oldest and stratigraphically lowest spherule layer as the original ejecta depositional event and the other layers as reworked.

To date bona fide Chicxulub impact spherule ejecta has been documented from the Caribbean, Central America, Gulf of Mexico, southern USA to New Jersey, Blake Nose, and offshore Brazil. Why are they not present in the deep-sea record globally? There are a number of reasons.
  1. The Chicxulub impact with its crater size of about 120-150km is smaller than necessary for global ejecta dispersal.
  2. The search to date has concentrated on the K/T boundary and Ir anomaly based on the assumption that this is the Chicxulub impact signal.
  3. The older pre-K/T age of Chicxulub necessitates a search for impact ejecta in latest Maastrichitan sediments.
Tyberg's multiple impact scenario assumes that several impacts occurred over the same time and space. This is not the case. The stratigraphic evidence indicates three impact events distributed over about 500Ka:
  1. Chicxulub at 300Ka before the K/T boundary (evidence impact glass spherule ejecta).
  2. K/T boundary impact (global distribution of iridium enrichment). 3. early Danian P. eugubina zone impact (iridium enrichment).
We agree with Tom Van Flandern (this debate 5.11.03) that this pattern is not compatible with Jupiter's capture of Comet Shoemaker-Levy 9, but implies the explosion of a planet-sized parent body in the main asteroid belt.

A summary of the evidence for multiple impacts is presented in the article that kicked off the Chicxulub Debate. Detailed evidence and discussion is presented in Keller et al. (2003a,b) and many illustrations and brief summary are given on the website:


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

Keller, G., Stinnesbeck, W., Adatte, T. and Stueben, D., 2003. 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., C. de Leon and J. de la Cruz, 2003. Spherule deposits in Cretaceous-Tertiary boundary sediments in Belize and Guatemala. J. Geol. London, 160, 1-13.

Raynolds, R.G. and Johnson, K. R., 2001. Synopsis of the stratigraphy and paleontology of the uppermost Cretaceous and lower Tertiary strata in the Denver Basin, Colorado. Rocky Mountain Geology, 38(l), 171-181.

Wolfgang Stinnesbeck replies to Tommy Tyberg (rec'd 14.3.11)

What does Tyberg mean by "disturbed and abnormal"? Originally, these sections around the Gulf of Mexico have been chosen by Smit, Bourgeois, Izett and others because they were so expanded and show a thickness of several metres of layered and correlatable sediments (e.g., clastic deposits), whereas distal sections are generally condensed to a few cm or less, and often present considerable hiatuses (e.g., Gubbio). Also, the above authors agreed upon the fact, that the sections expose a sequence of events (which they interpreted as deposited within a few hours) at a position where distal sections only present a single layer.

The difference between their interpretation of Gulf of Mexico sections and ours is essentially the timing of events, not much more. While they lumped these expanded sequences and several events into a scenario of one catastrophic impact, we were able to differentiate events which are clearly separated in time. It is essentially this biostratigraphic difference in interpretation with events in the late Maastrichtian, at the K/T boundary, and in the Danian that made us conclude that there is nothing abnormal about these sections and that disturbances are minor, and where present, clearly recognizable.

We suggest that authors go back to distal sections and also start to look in more detail at stratigraphic signals. Biostratigraphic timing of events is excellent around the Gulf of Mexico as a result of refined planktic foraminiferal zonation, which allows separation of events of less than 100Ka apart. There is no comparable time resolution at present in terrestrial sections such as the sequence in the Denver basin, and in consequence it is difficult to evaluate the question whether a complete section is present there, or whether erosion led to an incomplete record and hiatus.

ENDS (Tyberg + replies)

One for the palaeomagicians...

From Matt Terry* - Received 6.11.03

Sir,Clube and Napier have suggested that passage of the Sun through denser than average regions of the Galaxy, such as molecular clouds (the remnants of the one we currently inhabit is called Gould's Belt), and as is also presently the case, when nearest the extremes of its orbital parameters (height above/below the plane and crossing the plane, proximity to spiral arm, and perigalacticon) would serve to inundate the inner solar system with cometary bodies, some of which are likely to be of a proportion unseen by scientists, (except those of the keenest perception). This process takes time, on the order of one to three million years after maximum stress on the Oort cloud, and/or gravitational acquisition of interstellar cometary material, to reach the inner solar system, and a subsequent similar scale for the inevitable disintegration of these super-massive bodies into a huge swarm of smaller comets. The Kreutz Group of sungrazers is one such presently observed swarm, which fortunately intersects only our intellect, not our planet.

I note that two million years is approximately the time between the beginning of the Deccan traps volcanism and the K/T impactor, and so I would propose to the geologists this question: Speculating that the Deccan event could have been antipodal to a very large impactor, has anyone investigated the location which lay opposite India in the late Permian? Does that site even exist today, or has it been subducted or otherwise erased? If still extant, it would seem a likely plausible candidate region for a crater search.

I think it should here be pointed out that comets fragment much more readily than do asteroids, and so consideration of multiple impact scenarios should reflect the much greater likelihood of such bodies being not asteroids but comets, which ice balls are, especially if cosmogenic, very different from rocky and metallic (Metallica Rocks!) asteroids.

(As with everything else in astronomy, there is a continuum of creation bridging comets and asteroids, but the number of such directly related objects is relatively small, although those in the Taurid meteor stream, and other Apollo asteroids, could still hit hard enough to finish all but the hardiest of us off.)

*Tampa, FL

13th impact crater associated with K/T boundary

From Tom Van Flandern* Received 5.11.03

Last year, a British team of scientists announced the discovery of a multi-ringed crater (named Silverpit) with a central peak beneath the floor of the North Sea, believed to have been caused by an asteroid impact between 60 million and 65 million years ago [1-4]. The crater has 10 concentric rings from 2-20 km in diameter. The rough dating suggests that the asteroid that caused Silverpit might have been a chunk that broke off the larger asteroid that hit the Yucatan Peninsula 65 million years ago. Together with the Chicxulub and other craters, this discovery gives new support to the idea that killer objects from outer space may have sometimes arrived in pairs or even swarms. "It's so clear," said Dr. Gerta Keller, a geologist and paleontologist at Princeton, who studies the links between cosmic bombardments and life upheavals. "A tremendous amount of new data has been accumulated over the past few years that points in the direction of multiple impacts."

Actually, that brings the list of impact craters sometimes associated with the K/T boundary (65Ma) to 13. These are: Beyenchime-Salaatin (Siberia), Eagle Butte (Oregon), Upheaval Dome (Utah), Manson (Iowa), Kara (Western Siberia), Kamensk (Siberia), Gusev (W. Russia near Ukraine), Unnamed (Pacific Ocean), Chicxulub (Yucatan), Belize (south of Yucatan), Haiti (Caribbean), Alvaro Obregón (N. Mexico), Silverpit (North Sea) [5-10]. However, the global distribution of these impacts argues strongly against chunks off a parent asteroid, which would be expected to have much less than a hemispheric distribution, assuming Earth's atmosphere is the agent responsible for the break up. Capture and tidal break up by Earth, similar to Jupiter's capture of Comet Shoemaker-Levy 9, is extremely unlikely (by a factor of over 10 million) because Earth has a much smaller mass and is closer to the Sun. The clear implication of this global terrestrial cratering cluster is the explosion of a planet-sized parent body in the main asteroid belt, a hypothesis for which considerable astronomical evidence already exists [11-22].

*Meta Research


(2002), Nature 418, 520-523.

(2002/11/06), CCNet.

(2002) &ei=5040&partner=MOREOVER


(1996), "Terrestrial impact crater list",

(1991), Lunar&PlanetarySci. XXII, abstracts, 961-962.

(1993), Nature 363, 670-671.

(1993), Nature 363, 615-617.

(1980), Nature 288, 651-656.

(1998/03/12), "More evidence points to impact as dinosaur killer", JPL Press Release 98-42, NASA HQ & Pasadena, CA.

T. Van Flandern, Dark Matter, Missing Planets and New Comets, North Atlantic Books, Berkeley, chapter 11, (1993; 2nd edition 1999) - synthesis of exploded planet hypothesis (EPH) evidence.
Icarus 36, 51-74 (1978) - technical justification for the EPH. 'Solar System' tab, 'EPH' sub-tab - recent updating and distilling of the most telling EPH evidence, and how its predictions have fared; to be published in 2002.

Mercury 11, 189-193 (1982) - the EPH as an alternative to the Oort cloud for the origin of comets.
Icarus 47, 480-486 (1981) - the EPH's 'satellite model' for comets as an alternative to the 'dirty snowball' model.

Science 203, 903-905 (1979) - asteroid satellite evidence, confirming an EPH prediction.

Science 211, 297-298 (1981) - technical comment on previous paper.

Asteroids, T. Gehrels, ed., U. of Ariz. Press, Tucson, 443-465 (1979) - theory and observations of asteroid satellites.

Dynamics of the Solar System, R.L. Duncombe, ed., Reidel, Dordrecht, 257-262 (1979) - short summary of selected EPH evidence.

Dynamics of Planets and Satellites and Theories of their Motion, V. Szebehely, ed., Reidel, Dordrecht, 89-99 (1978) -- short summary of selected EPH evidence.

Comets, Asteroids, Meteorites, A.H. Delsemme, ed., U. of Toledo, 475-481 (1977) - short summary of EPH evidence with technical critiques and author responses.

Science Digest 90, 78-82 + 94-95 (1982) - popular exposition of the EPH and its implications.

Multiple impacts not unusual

from Andrew Glikson* Received 5.11.03

Sir, The Archaean imapct records includes the following documented multiple impacts:
  1. At least 2 and possibly more 3.47Ga impacts in the Antractic Chert Member, Warrawoona Group, Pilbara (Glikson et al., in press) and probably a double 3.47Ga impact in the Hoogenoeg Group, Barberton, South Africa (Lowe et al., 2003).
  2. A double impact in the 3.24 Ga Mapepe Formation, Barberton, South Africa (S2 and S3) (Lowe et al., 2003).
  3. A double impact in the 2.56Ga Bee Gorge Member of the Wittenoom Formation, Hamersley Basin (SMB-1, SMB-2).
  4. A double impact in the 2.47-2.50Ga Dales Gorge Group, Brockman Iron Formation, Hamersley Basin, Western Australia (DG4 and DG7).
Major impact clusters are recorded in the late Devonian (Woodleigh, Siljan, Alamo Breccia, Charlevoix), late Jurassic (Gosses Bluff, Morokweng, Mjolnir), Late Eocene (Popigai, Chesapeake Bay). So far as the K-T boundary is concerned, already two impact structures are known (Chicxulub, Boltysh), as well as the fallout units recorded by Keller. The evidence suggests that double impacts and impact clusters are commonly recorded in the terrestrial impact record.

*Research School of Earth Science, Australian National University Canberra, ACT 0200

Smoking gun proponents "must defend themselves"

From Andrei Ol'khovatov, PhD Received 5.11.03

Sir, It is nice to read such interesting and open debates! This is the way Science should work! All participants can present their arguments (flawed arguments are quickly dismissed by opponents), and a scientific audience can compare the logic and weight of arguments of both sides.

My impressions, which are those of a person who is interested in Chicxulub but was never deeply involved in this research, are as follows. It seems that on the basis of "proved science", the arguments of G. Keller sound more logical, so her opponents have to defend themselves. It seems that their best argument is that our understanding of the processes discussed is not good enough for any solid conclusion to be drawn. This argument is hard to overcome, as indeed, here we are in terra incognita. For example, our understanding of all large impact phenomena is based on our knowledge of relatively small meteorite impacts extrapolated (with help of numerous calculations) to impacts many orders of magnitude larger.

Numerical calculations depend on a physical model of the event under consideration. In the case of smaller events, results could be verified just by comparing the outcome of one's calculations with reality. It is very much more complicated in the case of large impacts. So, all models of large impacts are "unverified" at the moment.

To sum up. It seems that nowadays a "complete victory" is not possible due to "being in the field of unknown". My impression is that more research should be conducted. And currently we should to recognise (at least to ourselves) that we really don't know why dinos disappeared!

*Moscow, Russia

The timing's right

From Michael Paine* Received 4.11.03

Sir, The debate over the timing of the Chicxulub impact is fascinating and indicates a healthy state of scientific research on this topic. Irrespective of the timimg of the event, however, the wide range of research that has been conducted since Chicxulub's discovery has shown that the average frequency of such impacts is around 50 to 100 million years and the multiple environmental consequences of impacts the size of Chicxulub can easily account for the observed mass extinctions. In other words, the Chicxulub impact is an entirely plausible explanation for the KT extinction event.

Maybe there were multiple impacts and traces of the others have been lost to subduction and erosion or maybe there are other causes. My money is still on Chicxulub (this has nothing to do with it being Melbourne Cup horse race day in Australia).

For more on environmental effects of impacts see

*The Planetary Society Australian Volunteers

More pluribus than unum

From Matt Terry*. Received Sunday 2 November

Sir, This vision of multiple large impacts fits better with the palaeontological protests that the dinos' decline was more stretched out than that postulated by a single global "Cosmic Winter" disaster. As Clube and Napier have made clear, the injection of one or more giant, likely cosmogenic, comets into the inner solar system necessarily implies its subsequent fragmentation into a swarm of hazards, as the Kreutz group gives current evidence for.

Three hundred thousand years is not so long a time, on deep-space timescales, and it is positable that the Chicxulub crater is one among many impactors from that time, again, in resonance with the slow decline of the Triassic hyper-fauna. Statistically, given the limits of our search and the oceanic plenitude, we should expect that Chicxulub is certainly not the only, or even the largest, impactor of its time.

*Tampa, FL, USA