Charcoal & Isotopes

From
Date
Reply-To
I think the absence of the sediments that should have been derived from the
burning off of all the plant cover is far more telling than the presence or
absence of charcoal, but, since it has been brought up, let's talk about
it.  First, soot and charcoal are two different things.  As Tim Jones has
pointed out, charcoal is by no means rare in the fossil record.  I'm not
sure about soot but I suspect that, being microscopic, we don't really know
how common it may be.  If you are going to talk about a soot (or charcoal)
anomaly, however, you are going to have to demonstrate that the anomaly
occurs in temporally complete successions and is not being magnified via
stratigraphic condensation.  To my knowledge, these conditions have not
been met.  Several of the sections Tom mentions are demonstrably
incomplete.  If a soot anomaly fails to show up in demonstrably complete
sections (e.g., Brazos River, El Kef, Site 738, Nye Klov) I think you're
soot anomaly and it's proposed link to an impact has a problem.  That's not
to say that an impact couldn't account for the soot anomaly, just that it
is also consistent with a wide variety of other causal scenarios.

A more serious problem is present in the d13C data of Ivany and Salawitch.
All of the deep-sea sections from which they recorded their anomaly lack
the lowermost Danian planktonic foraminiferal biozone (Zone P0) and any
good evidence for the lowermost part of the overlying Zone (P1a, see Keller
and MacLeod's comment on the Ivany and Salawitch paper in Geology, Dec.
1993).  Half of the 8 sections they analyzed don't even exhibit an Ir
anomaly (so their placement of the boundary seems to be based solely on the
d13C spike - not really an independent source of stratigraphic data).  In
response to our comment Ivany and Salawitch made some general criticisms of
graphic correlation which were both irrelevant (graphic correlation was not
used to obtain our completeness estimates) and incorrect (MacLeod in press,
also the MacLeod and Keller's [1991] graphic correlation for the Brazos
River sections were recently confirmed by Stein et al., 1994; AMOCO Field
Trip Guidebook).  Moreover, Barrera and Keller (1990) demonstrated that the
d13C excursion at Brazos River was gradual and extending throughout the
uppermost Maastrichtian and first 40 k.y. of the Danian.  This pattern has
subsequently been confirmed at Nye Klov (see Schmitz et al., 1993, P3;
Keller et al., 1994, GSA Bulletin).  These authors link the progressive
d13C excursion to terrestrially-induced climate changes linked to the
eustatic sea-level fall, not global wildfires.

I'm not familiar with the Turkmenistan section but I doubt that the
stratigraphy is known in much detail.  I do know that the Stvens Klint
section is another temporally incomplete K-T boundary sequence (see Schmitz
et al., 1993).  As for the pollen spike, see Sweet, et al. (1990, GSA
Special Paper 247) for an alternative (non-catastrophic) interpretation.


Norm MacLeod



----------------------------------------------------------------------------
Norman MacLeod
Senior Scientific Officer
N.MacLeod@nhm.ac.uk (Internet)
N.MacLeod@uk.ac.nhm (Janet)

Address: Dept. of Palaeontology, The Natural History Museum,
         Cromwell Road, London, SW7 5BD

Office Phone: 071-938-9006
Dept. FAX:  071-938-9277
----------------------------------------------------------------------------