Wednesday, March 31, 2010

K-Pg Mass Extinction: Asteroid Impact Opponents Are Not Giving Up

Just weeks after a mega-paper in Science concluded that the Chicxulub asteroid impact was responsible for the late Cretaceous mass extinction, opponents of that theory are at it again.

Dr. Michael Prauss has submitting some work on the Brazos section in Texas which according to him shows that the Chixculub impact was not the sole cause of the extinction.

The basic argument of the opponents has been:

In stratigraphic sections spanning the late Cretaceous to Paleogene north of the impact site in parts of Mexico, Texas and New Mexico, there is sediment between the impact layer and the geochemically and micro-paleontologically defined Cretaceous-Paleogene boundary. This sediment accumulated gradually according to them. They estimate that the impact occurred a good 300,000 years before the K-Pg boundary crisis and thus the asteroid impact could not have been the sole cause of the mass extinction.

Opposing this view, the proponents of the impact theory say that this sediment is an event deposit which formed within hours to days of the impact and therefore the impact layer is the K-Pg boundary.

In this recent study published in Palaeogeography, Palaeoclimatology, Palaeoecology (I haven't read the paper -behind pay-wall) and described in Science Daily Dr.Prauss  has shown that in the sediment between the impact layer (he calls it ED for event deposit) and the K-Pg boundary there is a gradual increase to peak abundance in trilete spores. This he interprets as indicative of recolonization of damaged landscapes possibly after the asteroid impact by pioneering fern species but well before the Paleogene.

And below the ED in the entire studied section spanning upper Maastrichtian to lower Danian there seems to be signatures of long term sea-level fluctuations and also fluctuations in the oxygen and carbon isotopes matched by fluctuations in the types of dinocyst - a type of marine protist- assemblages.

He concludes that the data shows there to be a significant time span between the impact layer and the K-Pg boundary layer (as judged by the gradual increase in trilete spores) and that the K-Pg boundary was preceded by a long period of high frequency environmental changes and ecosystem degradation going back to a time even before the asteroid impact, a pattern inconsistent with a single catastrophic cause for the mass extinction.

Its really remarkable how two teams can come to conclusions which are such polar opposites. Supporters of the meteorite impact theory have interpreted the said sediment as an event deposit. Their opponents have read long term sea-level changes in the same section and have included the sediment in a sequence stratigraphy framework! There seemingly cannot be a reconciliation of these views. One reading of the data is completely wrong. Both sides will vociferously state that they are going strictly by the evidence but you just wonder how much of an influence does allegiance to a pet theory have on your reading of the data.

Earlier the opponents of the impact theory (Gerta Keller and her team) had based a lot of their conclusions on the presence of sedimentological features suggesting gradual deposition in a quiet environment and on the presence of late Cretaceous foraminifera in the sediment above the impact layer.

This line of reasoning was rejected by the supporters of the impact theory on the grounds that the foraminifera were not really foraminifera but recrystallized dolomite which looked like foraminifera! Alternatively some suggested that the presence of late Cretaceous foraminifera was due to the mixing of materials of different ages during the violent reworking and redeposition of sediment just after the asteroid impact.

All this is very confusing. The evidence collected in favor of the asteroid impact theory is massive. On the other hand it will be interesting to see how supporters of the impact theory explain the apparent preservation of correlated sedimentological, geochemical and palynological trends signifying long term change in environmental conditions.

And that pattern of gradual increase in spore content should not be preserved in a reworked event deposit.


  1. What puzzles me about this whole dispute is that I can't see any attempt in this paper to look for the iridium anomaly. Surely that would help to estimate the timescale of deposition, by seeing if it confined to the very bottom or spread through the whole sequence?

  2. I've found the work supporting the event deposit (vs. the Keller group interpretations) to be plausible and even convincing when combined with all of the other evidence. But this work (the Prauss paper) complicates things. The response should be interesting. Also, I agree with CJR that we should really know where the Ir anomaly sits in this section.

  3. Based on previous work on Brazos , Keller et al claim that no pronounced Ir anomaly is associated with the spherule layer or the overlying event deposits. They take it to indicate that the Chicxulub asteroid was not Ir enriched! Their work does show a more Ir anomaly coincident with the K-T boundary as defined by C13 shift and first Danian formaminifera. That they have interpreted as having caused by another impact!

  4. Here is an excerpt from Prauss' abstract:

    "These data [climate/oceanographic proxies] suggest that prominent, high frequency palaeoenvironmental changes precede the K/Pg boundary, which is inconsistent with a single "catastrophic" impact as the cause for the K/Pg boundary event. However, according to the onset and distinct distribution of the peak abundance of trilete spores, the base of the ED [event deposit] may actually reflect the Chicxulub impact, which probably contributed significantly to K/Pg boundary crisis within the biosphere."

    Unless I'm reading this wrong, he's saying that the impact did indeed contribute to the extinction -- just that it wasn't the sole cause.

    I think this is an important aspect of this debate. All too often single-factor explanations are pitted against each other.

  5. Isn't the argument over relative timing? I was under the impression that it's generally accepted there was some degree of climatic deterioration at the end of the Cretaceous, which contributed to the extinction: but whereas the more popular chronology is deterioration, then the Chicxulub impact marking the K-Pg boundary, the Keller group propose impact-deterioration-boundary (or possibly deterioration-impact-deterioration-boundary).

  6. I also get somewhat confused when reading these papers. Anyone know of a good review of all the plausible hypotheses? Something that doesn't necessarily conclude one way or another? Might be a good project for a science writer/blogger.

  7. Brian's reading is correct. Keller's group has maintained that Chicxulub was one of the factors in the extinction but not the sole one.

    the Keller group propose impact-deterioration-boundary (or possibly deterioration-impact-deterioration-boundary).

    or maybe deterioration-impact-deterioration-impact (boundary). the Ir anomaly at the boundary has to be explained. if the Brazos stratigraphic measurements and interpretations of Keller are correct and there is really no pronounced Ir anomaly associated with the Chicxulub impact what caused the Ir anomaly seen at the K-Pg boundary?