Product has been added to the basket

The Great Chicxulub Debate concludes - where next?

The Chicxulub Debate - what next?

KellerGerta_small.jpgGerta Keller

The debate, Chicxulub - the Non-Smoking Gun, was sponsored by the Geological Society of London and kicked off with a provocative article here and in Geoscientist (November 2003), with debate comments published on this Web site.

Dr Ted Nield was the instigator, and the heart and soul of this debate. He and his colleagues at and Geoscientist deserve the highest praise for devising a forum to encourage open discussion of such controversial issues, both with this debate and the earlier Mantle Plumes debate.

The Chicxulub controversy has resulted in a lively on-line debate, with the result that no readers will ever again view the K/T impact theory as a 'done deal'. For the very first time the debate allowed contrary evidence to the popular impact theory to be publicised, and it has raised the level of awareness among Earth scientists and the general public as a whole. I have experienced even six year-olds who now question the popular impact theory.

This is a singular success, regardless of which side of the debate you are on. It allows us now to move forward with the search for the real reasons for the K/T mass extinction, whether single or multiple impacts, major volcanism and climate change, or a coincidence of impacts and volcanism.

The Chicxulub Debate was conceived in response to new results from NE Mexico and the Chicxulub crater core Yaxcopoil-1 by Keller, Adatte and Stinnesbeck, showing evidence that indicates this impact predates the Cretaceous/Tertiary boundary and was not the cause for the mass extinction. The debate centred on three questions:

  • What is the age of the Chicxulub crater?
  • Can the clastic deposits in NE Mexico be interpreted as impact-tsunami deposits?
  • Are there multiple impacts?
These queries directly question the commonly accepted Alvarez KT impact theory and Smit et al.'s impact-tsunami theory.

We presented data from impact spherule deposits in NE Mexico and from the crater core Yax-1 that indicate that the Chicxulub impact predates the K/T boundary by about 300Ka and therefore was not related to the mass extinction. We presented evidence of bioturbation throughout the clastic deposits that effectively rule out deposition via impact tsunami over a period of just hours or days. And we presented evidence of multiple impacts, including the pre-K/T Chicxulub and the K/T boundary impacts.

Readers' comments were generally supportive. The only serious challenge came from Jan Smit in three lively and lengthy contributions defending the impact-tsunami theory and a K/T age for the Chicxulub impact. Smit's defence rested on denying the evidence of bioturbation, claiming widespread slumping caused impact ejecta to be folded into late Maastrichtian sediments, and claiming that late Maastrichtian foraminifera identified from the laminated limestone above the impact breccia in the Chicxulub core Yax-1 are just dolomite rhomb crystals.

What next?

Clearly, this debate did not change Smit's mind. Our findings are revolutionary; if we are right they require rethinking the way we view mass extinctions and particularly the biotic effects of impacts. To move the science of mass extinctions and the biotic effects of impacts and volcanism forward, we need to shed previous prejudices and assumptions.

Smit's proposal to conduct a blind test on the Yax-1 core samples will not solve the controversy. Blind tests in the past have been total failures with at best inconclusive results (e.g. Ir anomaly at Gubbio, El Kef mass extinction). They are poor substitutes for real in-depth scientific investigations. The controversy also does not rest on the presence or absence of microfossils in the Yax-1 core, but also on the magnetostratigraphy, sedimentology and mineralogy that provide no evidence for chaotic high-energy backwash and crater infill. Moreover, the controversy does not rest upon the Yax-1 core. What is even more important are the many outcrops in NE Mexico with multiple spherule layers interbedded in late Maastrichtian marls and the bioturbation within the so-called impact tsunami deposits of Smit and others. These sections provide conclusive evidence that Chicxulub predates the K/T boundary impact.

Ultimately, it is the reproducibility of these mineralogic, paleomagnetic, biostratigraphic and geochemical results that will determine who is right. Science is not done by consensus; what is relevant are reproducible results. To this end the Yax-1 samples and the K/T sections of NE Mexico should be made available to investigators willing to conduct the research.

Field Trip to NE Mexico

We propose to lead a field trip to the NE Mexico sections to show all interested parties the evidence we have presented in this debate and to let them draw their own conclusions. This field trip would also facilitate collection of samples for further studies by interested parties. Ideally, such a field trip could be an extension of this debate. Could it perhaps be sponsored by the Geological Society of London?


I have two regrets I about this debate. First, it got narrowly stuck with evidence from NE Mexico and the Chicxulub crater and never broadened to the wider question of spherules in the Danian of southern Mexico, Guatemala, Belize and Haiti.

Second, the debate never touched the question of the mass extinction, or the biotic effects of impacts and volcanism. However, this is a topic worthy of a Debate of its own, with wide interest and a guaranteed multitude of participants across all fields of palaeontology.

Princeton, NJ January 2004

Back to Debate front page 

The Debate has ended - for now!

Smitsmall.jpgJan Smit

To the uninitiated reader of the Chicxulub Debate, it may have appeared futile - the participants did not agree in the end! But I think that is not true. The debate has served an important purpose. It has brought out the main controversies, and what geological data are all about. Geological data are messy, by necessity. There is always a devil in paradise. There are nowhere long, detailed, uninterrupted, undisturbed, well preserved sedimentary sequences with high sedimentation rates (=resolution), which remain in the same facies throughout their lateral extent. Even the most continuous cores from the Ocean Drilling Program are interrupted, not complete and disturbed. It is simply not possible to collect a series of samples and compare them on a one-to-one basis at face value. Each sample a geologist takes is mingled with biases; each has a long history before one can read the original record correctly.

Time to review the evidence

Consequently, what we call evidence has to be taken with a grain of salt, because such evidence remains a personal interpretation. Take, for instance, our debate on the 'burrows' in the clastic deposits at the K/T boundary. For Keller it refutes the tsunami hypothesis. But before you can chalk this up as a strong point in case, a tremendous number of questions have to be resolved first. Is it a burrow or an artefact? Where does it come from? Where does it occur? How many are there? How quickly was it made? What animal lived in it?

The same goes for the foraminifera above the clastic beds in and outside Chicxulub crater. Are they in situ? Are they reworked? Are they identified correctly? Is the type of rock they occur in compatible? Are they made of dolomite or calcite?

Time to review the peers

How does peer review deal with such evidence? With the enormous amount of data and papers nowadays reviewers simply lack the time and expertise to screen all the evidence adequately. The peer review system is ready for revision, and internet debates such as these may play an important role in the change. The European Geosciences Union EGU is already doing this. Papers appear on the web for a few months before publication, for all to scrutinise, besides classical peer review. A debate such as this may call for another type of referee. One who independently checks the validity of data.

Mike Simmons (Posted 7.01.04) has taken up the challenge, and I gladly accept his offer and would like to turn my polished thin sections over to him (under supervision of the Geological Society?) to scrutinise. It is unfortunate that Keller evades such test so far, and offers instead to look at other sections (with new problems) or to questionable data (Ward et al., 1995) that are irreproducible, because all samples from that core are lost.


Luis Alvarez once said that what made the impact theory for him as physicist so attractive, is its testability, and the possibility of rejection. At the start of the impact extinction theory in 1978, a number of predictions were made, that could be tested. Theories are strengthened when their predictions come true, and what came true is:
  • The global occurrence of only one ejecta layer, synchronous in time
  • The finding of impact shock markers such as shocked quartz in the ejecta layer
  • The occurrence Ni-rich spinels, microkrystites and impact glass in the ejecta layer
  • The same platinum metal abundance ratios in the ejecta layer as in meteorites
  • The finding of tiny diamonds
  • The finding of chromium isotope ratios in the ejecta layer the same as known from carbonaceous chondrite meteorites, but unknown from any source on Earth (not mentioned in the last contribution by Tim Reston)
  • The finding of a single ejecta layer on land, just above the last dinosaur footprints
An all encompassing, elegant theory is preferred by most scientists - Occam's razor's principle if you will. If one theory explains all your data, is this theory not preferable above separate theories each explaining part of the problem? Why is plate tectonic theory so successful? Because it explains geological phenomena in one coherent, encompassing theory that were explained by many mechanisms before. Why are physicists looking for the Grand Unifying Theory? Why is string theory so appealing? Because physicists believe it may bridge the gap, and even unify two separate theories.

Impact - adhesive?

Why is the Impact extinction theory appealing? Because up to now it explains all the phenomena at the K/T boundary in one coherent, elegant theory, while Keller prefers to explain the different phenomena each by different mechanisms.

If the tests for the impact theory fail to support it, we should abandon it. But so far I believe we can explain all the known data within the single impact theory, and the purported data accumulated by Keller to show otherwise, will not hold when critically evaluated. But that is not mine to decide, others will make that decision for her.

Amsterdam, Netherlands January 2004

Back to Debate front page 

Ward, W. C., Keller, G., Stinnesbeck, W., and Adatte, T., 1995, Yucatan subsurface stratigraphy: Implications and constraints for the Chicxulub impact: Geology, v. 23, p. 873-877.