[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

K-T Extinctions: Other Models

Full content for this article includes illustration and table.
                                                                           
  
   Source:  Science, Dec 13, 1985 v230 p1292(4).
                                                                           
  
    Title:  Cretaceous-Tertiary extinctions: alternative models.
   Author:  Jan Smit, Frank T. Kyte, Bevan M. French, Charles B. Officer
and 
            Charles L. Drake
                                                                           
  
 Subjects:  Mass extinction theory - Research
            Sedimentation and deposition - Research
            Paleontology - Research
            Iridium - Research
            Tektite - Research
            Extinct animals - Causes of
                                                                           
  
Electronic Collection:  A4059347
                   RN:  A4059347
                                                                           
  

Full Text COPYRIGHT American Association for the Advancement of Science
1985

Officer and Drake (1) appear to present a valid alternative to the large
impact model for the Cretaceous-Tertiary extinctions. Their basic premise
is
that the distribution of iridium (Ir) in some sections can only be
explained
by an event of relatively long duration. This proposal is supported by a
model which suggests that Ir distribution extending beyond an interval
expected by bioturbation requires a noninstantaneous source for the
siderophiles. They further suggest that evidence favors a mantle (volcanic)
origin for these elements. We find that (i) the bioturbation model is
inaccurately applied and inadequately explains possible sedimentary effects
for any given section, (ii) there is no evidence of prolonged Ir
sedimentation at any site, and (iii) the volcanic model, although not
positively excluded by the data, is not easily reconciled with the data and
remains at best a very low probability alternative to the impact
hypothesis.

The bioturbation interval of 11 cm (5 to 6 cm after compaction) used by
Officer and Drake applies only to the homogenization interval for surface
sediments. As deposition (and burial) proceeds, older sediments continue to
be mixed upward. Sediments that are deposited instantaneously (like
microtektites or ash) can be spread over tens of centimeters (2). Table 1
shows the observed distribution of microtektites (3), which are spread over
a minimum of 35 cm and an average of 59 cm. Bioturbation is only one process
that will affect Ir distribution after a presumed impact. Other potentially
important processes include diagenetic mobilization, secondary deposition
after transport by bottom currents, and delayed deposition of siderophiles
in solution, which have relatively long residence times.

Because there is no independent evidence for prolonged deposition of
IR-rich sediment at the four sites cited by Officer and Drake, the
bioturbation
argument may not be relevant. 

1) Site 465A was grossly disturbed by drilling, and Cretaceous and
Tertiary
sediments are mixed over an interval of at least 100 cm (4). The published
profile for this locality does not in any way reflect the original
stratigraphy, which cannot be determined in this section.

2) Anomalously high concentrations of Ir have been reported in only 10 cm
of the core at site 524 (5); the 43 cm cited by Officer and Drake is the
distance to the first background analysis.

3) High Ir concentrations below the fish clay in Stevns Klint, Denmark,
rely
on correction for more than 99 percent CaCo3, but the significance of the
procedure is not clear. The highest concentrations measured outside the
fish clay are approximately 0.1 ng/g Ir and probably amount to only a few
percent of the total Ir in the section at best. It is reasonable to expect
some
diffusion of siderophiles out of the clay during diagenesis, and there is
no
evidence for prolonged deposition of large amounts of siderophiles.
Moreover, Stevns Klint is a prime example of a lithologic discontinuity,
which Officer and Drake state "preclude[s] precise geologic time
discrimination." In fact, every known K/T boundary section has a lithologic
discontinuity, and that, if we fellow the criteria of Officer and Drake,
invalidates their time interval estimates.

4) The Brazos River, Texas, section shows irregular peaks of significant
amount of Ir over about 45 cm on top of a thick turbidite-like sediment at
the
K/T boundary (6). This is a shallow shelf environment where lateral
transport, reworking, and winnowing of sediments by storm waves is common.
The Ir distribution is easily explained in terms of these mechanisms and by
bioturbation.

With regard to characteristics of the event that suggest a mantle rather
than
meteoritic source, we make the following points.

1) The discovery of Ir-bearing particulates from Kilauea (7) is important.
However, such particulates would be deposited near the source; we cannot
imagine a volcanic event capable of worldwide distribution of the
spheroidal
material common to KT sediments (8).

2) Isotopic systematics (9) of the boundary clay that indicate mantle
affinities are likely to represent terrestrial impact ejecta. The impact
hypothesis predicts an exotic terrestrial component. We not that only the
base layer (<3 mm) at Caravaca, Spain, is highly enriched in this exotic
isotopic component.

3) Spherules found by Vannucci et al. (10) outside the boundary at Gubbio,
Italy, are texturally and compositionally different from those in the
boundary
clay. In particular the siderophile-rich spherules with skeletal
magnetitie
(11) described in Italy and the North Pacific are characteristics of the
KT
boundary.

4) The lamellar quartz can easily be distinguished from coesite, which
forms
under high static pressures.

5) Iridium concentrations do not correlate with clay concentrations.
Measurements within the boundary clay (12) at Stevns Klint and Caravaca
show a decrease in the ratio of Ir to clay by a factor of 3 to 5.
Extrapolations into the nearly pure carbonates at Stevns Klint are not
appropriate unless one first demonstrates that the Ir actually resides in
the silicate fraction. In
other localities (for example, Caravaca), where backgrounds contain a
substantial clay fraction, Ir is not observed.

6) We question the use of the data of Wezel et al. (10) on Ir in clays
below
the K/T boundary from Gubbio. Why was the data of Alvarez et al. (13) on
the same clays, which indicates no excess Ir below the K/T boundary,
ignored?

We agree that the available data do not exclude a volcanic source for the
siderophiles or the boundary clay. However, the shocked quartz and the
global spherule distribution would appear to limit it to extremely violent
events. A statistically probable impact in a more likely explanation than a
poorly defined mantle event resulting in worldwide simultaneous volcanic
eruptions of an unprecedented magnitude.

The statement by C.B. Officer and C. L. Drake (1, p. 1164) that "the
presence of lamellar quartz features does not in and of itself demonstrate a
meteor impact origin" for the shock-metamorphosed quartz grains discovered
at several widespread locations in the Cretaceous-Tertiary (K/T) boundary
layer (2) gives a misleading impression of the relative evidence for the
impact and nonimpact interpretations of such features. The impact
interpretation rests on a solid and extensive base of theoretical,
experimental, and observational studies, whereas such unspecified
alternative mechanisms as "intense volcanic or tectonic overpressure events"
(1, p. 1164) are not supported by field evidence or by the establishment of
a testable mechanism.

The use of these lamellar features (shock lamellae) in quartz, together
with
other shock-deformation effects, to indicate meteorite impact has been
justified in studies of the eart (3-5) and other planets (6).

1) It is generally accepted, even by some proponents of nonimpact
mechanisms (7) that certain unusual deformation features in rocks and
minerals are produced only by intense transient shock waves that have peak
pressures as high as 5 to 100 GPa (50 kbar to 1 Mbar). These features
include (8): (i) distinctive shock lamellae in quartz and other minerals,
which are clearly different in both appearance and orientation within the
host grain from conventional metamorphic deformation lamellae; (ii)
amorphous (diaplectic) glassy forms of minerals; (iii) shatter cones; (iv)
high-pressure minerals such as coesite and stishovite, when found in
low-pressure, high-level crustal rocks. The formation of these features by
shock waves have been recorded in laboratory experiments (9) and after
chemical and nuclear explosions (10). Other distinctive features, such as
extensive brecciation and the melting of refractory minerals such as quartz
and zircon at temperatures above 1700 deg.C, are produced by shock waves,
but they may also be produced in nonshock environments.

2) Hypervelocity meteorite impacts on the surface of the earth can and do
generate shock pressures adequate to form these distinctive features.
Evidence for this view comes from both theoretical studies of the impact
process (4,11) and from the occurrence of shock-metamorphic features in
young structures of undoubted meteorite impact origin, for example, Meteor
Crater, Arizona (12).

3) The resulting hypothesis, that any geological structure which exhabits
shock-metamorphic effects has been formed by meteorite impact, has not been
disproved in the more than 20 years since it was first proposed.
Shock-metamorphic effects have been found only in so-called
"cryptoexplosion" structures that are circular (or nearly so), localized,
and characterized by sudden and extensive deformation--generally consistent
with the expected effects of large meteorite impacts.

4) No shock-metamorphic effects have been observed in undisputed volcanic
or tectonic structures. The coesite found in deep-seated rocks (1)
apparently formed stably at high lithostatic pressures and is not
accompanied by any characteristic shock effects.

5) Despite much speculation (13, 14), no mechanism to explain the
generation of shock waves within the earth has been developed to a point
that permits critical evaluation and prediction.

Officer and Drake (1) correctly point out that ambiguities and
controversies
exist in the intepretation of shock-metamorphic effects in large and
complex
structures like those in Vredefort, South Africa (15, 16), and Sudbury,
Canada (14, 17). However, there is evidence for an impact origin of both
structures (15, 17), which is consistent with observations at other impact
structures. In view of the evidence for the impact mechanism, it is probable
that these ambiguities are caused by our ignorance about the details of how
impact structures with diameters of 100 km are formed rather than by flaws
in the impact theory itself.

Suggestions that the K/T extinctions have a volcanic origin should include
evidence that the shock-metamorphosed quartz grains are unrelated to the
K/T event or that there was volcanism that could have produced both global
iridium anomalies and shock-metamorphic features.

We welcome the commentary by Smit and Kyte (1) and by French (2) related to
our article (3). Their criticisms fall into five categories: (i)
bioturbation;
(ii) temporal extent of the iridium signature; (iii) iridium enhancements
in
the clays at Gubbio, Italy; (iv) microspherules; and (v) shock deformation
features.

Bioturbation. The statement by Smit and Kyte (1) that "the bioturbation
interval of 11 cm (5 to 6 cm after compaction) used by Officer and Drake
applies only to the homogenization interval for surface sediments" is
incorrect. They refer to an article on microtektites by Glass (4), but do
not
note that the result quoted in our article (3) of a mixing depth of 11
[plus-or-minus] cm for microtektite, ash, and pumice distributions was
based
on determinations by Officer and Lynch (5) from 16 piston cores, six from
the
same article by Glass (4), one reported by glass et al. (6), and nine more
recent observations by Ruddiman et al. (7). The analyses followed analytic
and computer parameter optimization procedures developed by Officer and
Lynch (8) and Lynch and Officer (9). All showed a characteristic
bioturbation
distribution expected for an instantaneous flux input. Glass (4) reported
two
additional cores, RC8-52 and RC8-53, for which the microtektite
distribution
did not follow this pattern, but showed an irregular and erratic
distribution
over a depth interval of 40 to 80 cm. As we noted in our article (3), this
is
to be expected at some locations where there can be extreme physical
distrubance of deep sea sediments (10). Glass (11) also furnished
microtektite distributions for 17 other cores. We have analyzed them for the
purpose herein. Sixteen have a characteristic bioturbation distribution with
an average mixing depth of 12 [plus-or-minus] 3 cm (table 1). The other
core, V19-171, shows an irregular distribution over 40 cm. In summary, of 35
cores with microtektite, ash, and pumice distributions, 32 show a
well-defined bioturbation characteristic with a mixed layer thickness of 11
to 12 cm; three show a distribution reflecting more extreme physical mixing
processes.

For compacted K/T sediments the bioturbation interval will be reduced to
about 5 to 6 cm (3). It is necessary to examine each K/T section to
determine the extent of bioturbation smearing. For example, at DSDP site
516F in the southwest Atlantic there is essentially no bioturbation in the
K/T core (12), and at DSDP sites 525 and 529 along the Walvis Ridge
bioturbation smearing varies from 1 to 5 cm (13).

Temporal extent of the iridium signature. Smit and Kyte (1) comment on the
drilling disturbance at DSDP site 465A, which we also noted (3).
Specifically, for the K/T core at this site there is a "white stringer of
Cretaceous material which has spread into the Paleocene section during the
drilling process" (14, figure 6). We consider that the observed distribution
of iridium and other associated elements over an extended depth interval
corresponding to the pyrite portion of the Paleocene gray ooze (3, figure 1)
correctly represents K/T conditions at this site. Smit and Kyte (1) also
criticize our citation of data from Stevns Klint, Denmark (15), because it
is corrected for carbonate. In most models the iridium is associated with
the noncarbonate fraction and is not related to the nannofossil and
microfossil remains of biogenic origin in these sediments. For sections in
which there are substantial variations in the fraction of bulk sediment that
these remains
represent, it is appropriate to make comparisons on a carbonate-free
basis.

For DSDP site 557B in the northwest Pacific, there is no clay layer and
little
variation in the carbonate fraction through the K/T transition. A broad
peak
for iridium and other associated elements including iron occurs over 50 cm
at the K/T transition corresponding to a time interval of approximately
50,000 years (16). In addition there are satellite peaks at 40 and 230 cm
below the main peak and 270 cm above the main peak. The satellite peaks have
concentrations of iron and associated elements comparable to those of the
K/T peak but lower concentrations of iridium. In particular the Ir/Fe ratio
shows a gradual increase (by two orders of magnitude) to the K/T peak
followed by a gradual decrease over a total core depth interval of 300 cm.
It is difficult to explain the extended iridium distribution for the K/T
peak in terms of a single asteroid impact. It is equally difficult to
explain the gradual change in the Ir/Fe ratio in terms of a series of
impacts. We suggests that these variations are best explained in terms of
volcanic activity and that the
change in the Ir/Fe ratio represents a deeper mantle plume source of
volcanism.

Iridium enhancements in the clays at Gubbio, Italy. Smith and Kyte (1)
question our use of the data from Wezel et al. (17) and Vannucci et al.
(18)
on iridium in the clays below the K/T boundary clay at Gubbio and ask why
we ignored the Alvarez et al. (19) data on the same clays which indicate no
excess iridium. One, we did not use the Wezel et al. (17) and Vannucci et
al. (18), data, although their reports prompted our own investigations. The
data presented in table 2 of our article (3) were from samples collected by
G. D. Johnson and analyzed by J. H. Crocket (20). Two, there are no data on
iridium determinations for the Gubbio clays in the article by Alvarez et al.
(19).

Wezel et al. (17) and Vannucci et al. (18) reported iridium anomalies in
clay
layers other than the K/T as well as an anomaly of around 10 ppb at the
Bonarelli level, of Turonian age, approximately 240 m below the K/T
boundary layer clay. Alvarez et al. (19) report that they find no iridium
enhancement in the Bonarelli layer, but do not mention the other Gubbio clay
layers. Crocket et al. (20) also find no iridium anomaly in the Bonarelli
layer, but do find iridium enhancements above background in clay layers and
the alternating limestones extending 2 m on both sides of the anomaly in the
reference K/T clay.

Microspherules. Smith and Kyte (1) state that we have essentially ignored
the significance of microspherules at the boundary. This is not the case.
Microspherules are, indeed, a feature of the K/T transition, and it is
generally agreed that their present composition is secondary (21). Naslund
et al. (22) have shown that while the microspherules are 10 to 50 times
more
abundant in the K/T layer, they also occur in clay layers at Gubbio
extending over an age span of 22 million years. Many of the microspherules
are hollow with a smooth outer surface, and the size and gross morphology of
these hollow spherules is similar to that reported for silicate glass
spherules formed during volcanism (23).

Shock deformation features. French (2) argues that the shock metamorphic
features of lamellar quartz or shatter cones, or both, that are associated
with the Vredefort and Sudbury intrusives and some of the cryptovolcanic
structures must be impact origin. Recent studies at Vredefort and Sudbury
contradict this conclusion. For the Vredefort and Sudbury intrusives one
hypothesis assumes that a single asteroid impact triggered the intrusion,
in
which case the shock metamorphic features must be contemporaneous with or
predate the intrusion deformation and static (high-temperature)
metamorphic
features. The other overpressure event, or events, associated with the
intrusion, in which case the shock metamorphic features will postdate the
onset of the intrusion. Schreyer (24) has shown that dynamic, or shock,
metamorphic events at Vredefort postdate the onset of the static, or
high-temperature, metamorphic events; the investigations by Lilly (25)
have
shown that there were two shock metamorphic events; and the investigations
by Simpson (26) have shown that the shatter cone features cut across the
postdate indurated fault breccia associated with the structure. All three of
these findings are not in accord with an impact origin for the Vredefort
dome. At Sudbury, Fleet (27) has shown that the shatter cones postdate the
emplacement of the nickel intrusive, again contrary to an impact origin.

French (2) also states that "no shock-metamorphic effects have been
observed in undisputed volcanic or tectonic structures." In a recent
investigations Carter et al. (28) have found shock metamorphic features
including lamellar quartz in rock samples associated with the Toba eruption.
Toba is the largest known volcanic eruption of the recent past, occurring
about 75,000 years ago with a total eruption magnitude of 400 times that of
Krakatoa (29). Microstructures in the Toba rocks record shock stress levels
greater than 10 GPa. Carter et al. (28) conclude that peak shock stresses
from explosive volcanism at K/T time could account for the microstructures
observed (30).

Smit and Kyte (1) conclude that "a statistically probably impact is a more
likely explanation than a poorly defined mantle event resulting in
worldwide
simultaneous volcanic eruptions of an unprecedented magnitude." We submit
that evidence for the impact is thin and the crater has yet to be found,
while massive volcanism is indicated by the geological record. In any event,
the question is not whether impacts occurred, it is weather they are related
to extinctions. The selective pattern of extinctions is not the stuff of a
global dust cloud.
                                                                           
  
                                -- End --

LOUIS VARRICCHIO
 Environmental Information Specialist &
 Producer/Writer, "Our Changing Planet"
  (Visit OCP-TV on the Web at: www.umac.org/ocp)
  Upper Midwest Aerospace Consortium
  Odegard School of Aerospace Sciences
  University of North Dakota
  Grand Forks, N.D. 58202-9007
    Phone: 701-777-2482
    Fax: 701-777-2940
    E-mail: varricch@umac.org (in N.D.); morbius@together.net (in Vt.)

"Behind every man alive stand thirty ghosts, for that is the ratio by
which the dead outnumber the living. Since the dawn of time, a hundred
billion human beings have walked the planet Earth." -- Arthur C. Clarke

----------
Archives located at:
http://www.meteoritecentral.com/list_best.html

For help, FAQ's and sub. info. visit:
http://www.meteoritecentral.com/mailing_list.html
----------