Re: K/T boundary section in TX

[John Alroy <jack@homebrew.uchicago.edu>]

Thanks for the references, guys.

I believe that Dr. Schimmer has misrepresented the Lehman (1990)  
article on Big Bend. First, the earlier article by Schiebout et al.  
(1987) made no claim concerning an uncomformity at the K-T in this  
area; this is attributed by Lehman to a much earlier paper (Maxwell  
et al. 1967).
Second, the fact remains that even though there is no major  
unconformity at the boundary, there IS a very significant lack of  
fossils across it. Lehman states the the unfossiliferous interval is  
31 m, Schiebout et al. give a similar figure of 35 m and clearly were  
talking about the same thing. However, Lehman shows a measured  
section indicating that the highest dinosaur locality is just at the  
base of this gap, i.e., in the paleosol immediately underlying the  
iridium spike and formational contact! In other words, unless the K-T  
is misplaced in this section, there IS no gap at all between the Ir  
spike and the last occurrence of dinosaurs. So much for the "gap"  
theory, unless, of course, you want to argue about the number of  
centimeteres WITHIN this paleosol that separate the dinos and the  
spike.
Third, Lehman does argue for a long-term climate change. However, 1)  
this is AT the K-T boundary and perfectly coincides with it, not  
BEFORE the boundary, so climate change in this section has nothing to  
do with the extinction; and 2) although Lehman does argue that the  
climate change was so long-term (order of 2 million years) that it  
couldn't have been a simple, immediate consequence of the impact,  
there is nothing new about this observation at all. A dramatic and  
long-term change in climate across the boundary has been known for  
years to everyone working in the North American terrestrial K-T, and  
Lehman cites multiple authors as having said this earlier (e.g.,  
Fastovsky, Retallack, Hickey, Wolfe). There is nothing discordant  
between an impact and a long-term post-impact climate change; in  
fact, I believe that pro-impact workers would predict just such a  
change.

Heward's two references were very interesting. The review of the  
Chicxulub crater (Pope et al. 1993) comes up with an estimate of 240  
km for the diameter of the crater, much higher than earlier  
estimates. Once again, this is far and away the largest known  
Phanerozoic crater. As in multiple other papers, the well data  
reinterpreted by Meyerhoff et al. are shown in a figure and no  
mention is made of the volcanism issue. I think the reason is that  
the lower volcanic interval discussed by Meyerhoff et al. is  
considered by Pope et al. and other authors to be Lower Cretaceous  
and to have nothing to do with the impact per se, underlying the  
impact-related strata. This interval includes the supposed Late  
Cretaceous microfossil assemblage I have discussed previously, but if  
I understand Pope et al. correctly the exact age of this interval  
within the Cretaceous is completely irrelevant because it underlies  
the impact crater. The remainder of Meyerhoff et al.'s argument rests  
on strictly geological interpretations of the impact-related strata,  
which I am not in a position to judge.

The Stothers paper on flood basalts and extinctions is very  
intriguing and at first glance makes a convincing case. However, if I  
understand his very truncated discussion of statistical methods (the  
paper is only four pages long), the significance tests for an  
association are incorrectly performed. Stothers has assumed as a null  
model that the flood basalt and extinction events act as "ordered,  
uniformly distributed random deviates," i.e., they can happen any  
time at all in the interval being analyzed (the Mesozoic and  
Cenozoic) with equal likelihood. But this model is fundamentally  
flawed because it already has been argued that extinctions are more  
widely spaced than one might expect at random, and if so a spurious  
correlation easily could result. Note that widely spaced extinctions  
could be purely an artifact of how we construct time scales and  
recognize extinctions in the first place, but that wouldn't affect my  
argument at all. Until the analysis is redone to take this into  
account the issue is unresolved. I really wish someone would look  
into this, but I don't have time for it.