Planet. Space Sci.

Pergamon

Vol. 46, No. 213, pp. 271-281, 1998

0 1998 Elsevier Science Ltd

All rights reserved. Printed in Great Britain
0032-0633/98
$19.00+0.00
PII: SOO32-0633(97)00130-X

Earth cratering record and impact energy flux in the last 150 Ma
A. Montanari,,* A. Camp0 Bagatin3 and P. Farinella3
Osservatorio Geologic0 di Coldigioco, 62020 Frontale di Apiro, Italy
Ecole des Mines de Paris, Paris, France
3Gruppo di Meccanica Spaziale, Dipardmento di Matematica, Universita di Pisa, Via Buonarroti 2, 56127 Pisa, Italy
Received 20 September 1996; revised 2 June 1997; accepted 4 June 1997

1. Introduction
The Tunguska
explosion which occurred in 1908 in the
Siberian sky, is the outcome of one of the relatively small
and frequent
impact events of 50-100m
sized interplanetary
objects, which in most cases as at Tunguska,
may not even form detectable
craters on the Earths

Correspondence to: A. Montanari

surface. However, the population of potential projectiles

extends to much larger sizes (10 km or larger) with a quasi
power-law distribution, and today it is well known that
very large and energetic impacts have also occurred over
the history of our planet, and have strongly influenced the
evolution of the biosphere on both local and global scales.
Such impacts did form large craters, and studying these
craters can tell us much about their frequency and time
distribution, the properties of the impacting bodies, as
well as about their consequences on global ecological
equilibria.
In this paper, we provide a brief review of the Earths
cratering record and the biological effects of the impacts
undergone by our planet in the geologically recent past.
In particular, we attempt to estimate the energy flux
associated with the greatest cratering events that happened on Earth in the last 150 Ma, as a function of time,
to infer from the available cratering record any possible
evidence for periodicities or other non-random features
in the distribution of such events, and to analyze their
correlation
with global environmental
catastrophes
inferred from the geological and paleontological record.
We have thus compiled a list of the largest impact
cratering events that occurred in the last 150 Ma by considering all the craters observed on Earth whose diameter
is greater than 5 km. Actually, there are 47 known craters

A. Montanari

212

et al.: Earth cratering

record and impact energy flux in the last 150 Ma

Table 1. List of impact craters on Earth younger than 150 Ma and larger than 5 km in diameter
Diameter
(km)

Crater name and country

1 Aorounga (Chad)
Bosumtwi (Ghana)
Zhamanshin
(Kazakhstan)
Kara-Ku1 (Tajikistan)
Elgygytgyn (Russia)
Bigach (Kazakhstan)
Karla (Russia)
Ries (Germany)
Haughton (Canada)
10 Chesapeake (U.S.A.)
11 Popigai (Russia)
12 Mistastin (Canada)
13 Logoisk (Russia)
14 Chiyly (Kazakhstan)
15 Kamensk (Russia)
(Canada)
16 Montagnais
17 Ragozinka (Russia)
18 Wanapitei (Canada)
19 Marquez (U.S.A.)
20 Chicxulub (Mexico)
21 Kara (Russia)
22 Ust Kara (Russia)
23 Manson (U.S.A.)
24 Lappajarvi (Finland)
25 Boltysh (Ukraine)
26 Dellen (Sweden)
27 Steen River (Canada)
28 Avak (U.S.A.)
29 Carswell (Canada)
30 Mien Lake (Sweden)
(Australia)
31 Tookoonooka
32 Gosses Bluff (Australia)
33 Morokweng (South Africa)

2
3
4
5
6
7
8
9

Non-refereed

information

12.6
10.5
13.5
52
18
7
12
24
24
90
100
28
17
5.5
25
45
9
7.5
22
180
65
25
37
23
24
15
25
12
39
9
55
22
340

Age (Ma)

<0.004
1.1
1.09
2.5
3.5
6
10
15.1
23.4
35.3
35.7
38
40
46
49.2
50.5
55
57
58.3
64.98
73
73
73.8
77.3
88
89.0
95
100
115
121.0
212.8
142.5
144.7

taken from the compilation

Error
(Ma)

Dating
method

Reference

historical
40Ar/39Ar
fission track
stratigraphic
K/Ar
stratigraphic
stratigraphic
40Ar/39Ar
40Ar/39Ar
stratigraphic
40Ar/39Ar
40Ar/39Ar
stratigraphic
?
40Ar/39Ar
40Ar/39Ar
stratigraphic
40Ar/39Ar
fission track
40Ar/39Ar
40Ar/39Ar
40Ar/39Ar
40Ar/39Ar
40Ar/39Ar
K/Ar
87Rb/87Sr
K/Ar
stratigraphic
40Ar/3gAr
40Ar/39Ar
stratigraphic
40Ar/3gAr
*08Pb/232Th

Grieve et al., 1985

Grieve, 1991
Koeberl and Storzer, 1988
Grieve et al., 1985
Gurov and Gurova, 1980
Kilesev and Korotushenko,
1986
Masaitis et al., 1980
Staudacjer et al., 1982
Jessberger, 1988
Koeberl et al., 1996
Bottomley et al., 1993
Mak et al., 1976
Masaitis et al., 1980
Grieve et al., 1995
Izett et al., 1994
Bottomley and York, 1988
Vishnevsky and Lagutenko, 1986
Bottomley et al., 1979
McHone and Sorkhabi, 1994
Swisher et al., 1992
Koeberl et al., 1990
Koeberl et al., 1990
Izett et al., 1993
Jessberger and Reimold, 1980
Boiko et al., 1985
Deutsch et al., 1992
Carrigy, 1968
Kirschner et al., 1992
Bottomley et al., 1989
Bottomley et al., 1989
Gorter et al., 1989
Milton and Sutter, 1987
Koeberl et al., 1997

0.2
0.05
2.5
0.5
3
10
1.0
1.0
0.2
0.8
4
5
7
0.2
0.8
5
2
3.1
0.05
3
3
0.3
3
3
2.7
7
5
10
2.3
5
0.5
1.9

of Grieve et al. (1995)

of this size (see Grieve et al., 1995) and with an estimated

age less than 150 Ma, but we have taken into account only
33 of them because, as we are interested in their time
distribution,
for our analysis we needed fairly precise and
accurate estimates of crater ages with 20 errors < 10 Ma.
This selection results in the record of craters listed in Table
1.
Figure 1 shows the distribution
of the selected craters
plotted on a chronostratigraphic
time scale, and correlated with the curves of extinction of marine organism
families
(Raup
and
Sepkoski,
1986), and
genera
(Sepkoski, 1990), and the signatures of impacts including
iridium
anomalies,
shocked
quartz,
spherules
(i.e.,
microtektites
and microcrystites),
megawave
deposits,
and anoxic crises recorded in sedimentary
sequences. In
the next section, we will examine the reliability
of the

complex set of data upon which our subsequent

energy flux analysis is based.

impact

2. The cratering record and its uncertainties

The most crucial aspect of Fig. 1 is the geochronologic
time frame on which the correlation between stratigraphic
signatures, paleontological
record, and impact craters is
based. The distal ejecta of an impact found in a sedimentary layer, and a biological signature which may be
caused by an impact, have to be dated in terms of numerical age in Ma in order to be correlated with a known
impact crater. The most reliable way to assign a precise
age to a crater is to date its impact melt rock with a

Fig. 1. Impact craters and extinctions plotted against the stratigraphic column of the Umbria-Marche
sequence. The main extinctions are : MM/LM = Middle/Late Miocene boundary ; LE = Late Eocene
(E/O boundary) ; K/T = Cretaceous/Tertiary
(K/T) b oundary ; C/T = Cenomanian/Turonian
boundary ; EA = Early Aptian ; J/K = Jurassic/Cretaceous
boundary. Extinction intensities are after
Raup and __.
Sepkoski
_ (1986) and Sepkoski (1990). References on the size and age of impact craters are
shown m lable 1

et al.: Earth cratering

A. Montanari

MARINE EXTINCTIONS

CHRONCSTRATIGRAPHY MAIN EVENTS

m genera

40

:
:

2) LE

:
:
:
.:

:
:
:
::
::
:
::
::
::

P
:
:
:
:
:
:
:
:
:
:
:
:

4) cfr

:
:
:

looA

120 -

9
$
$
5
b
>r
%
W

130.

5) EA
B

B
150 II

:
:
:
:
:
:
:
:
:
:

140 II

Jurassic

5-20 km
20-50 km
50-l 00 km

+ shocked quartz

:
:
:

6) J/K

IMPACT CRATERS
1
2
3
4
5
6
7
8
9
10

Aorounga
Bosumtwi
Zhamanshin
Kara-Kul
Bfgygytgyn
Bigach
Karla
Ries
Haughton
Chesapeake

0 microspherules

441

31

:
:

CRATER DIAMETER

:
:
:::
::
:
::
::
::

:
:
:
:
:
:

09

::
::
::
:

:
:
:

3) Kfr

e
:
::
:
::

:
:
:
:
:
:
:
:
:
:
:
:
:
:
:.
:
:
:
:

1) MMLM

SIGNATURES

IMPACT CRATERS

families

30
t:

110 9

273

record and impact energy flux in the last 150 Ma

O32

:
.
:

33

:
:

-L-

11
12
13
14
15
16
17
18
19
20
21

Popigai
Mastastin
Logoisk
Chyly
Kamensk
Montagnais
Ragozinka
Wanapitei
Marquez
Chicxulub
Kara

megawave deposit

22
23
24
25
26
27
28
29
30
31
32
33

Ust-Kara
Manson
Lappajarvi
Boltysh
Dellen
Steen River
Avak
Carswell
Mien Lake
Tookoonooka
Gosses Bluff
Morokweng

-- Ir iridium anomaly

274

A. Montanari

et al.: Earth cratering record and impact energy flux in the last 150 Ma

radioisotopic
method. In fact, the majority of craters are
found on land and are buried or filled by incomplete
sequences of terrestrial sediments which do not permit a
direct, and accurate stratigraphic
correlation with marine,
fossiliferous sediments. In other cases, impact craters are
filled with marine sediments with a disturbed (i.e., mixed
or reworked) paleontological
record that is not conducive
to precise and accurate numerical
dating of the impact
event.
On the other hand, marine sedimentary
sequences are
dated by interpolation
of few radioisotopic
ages, obtained
from interbedded
volcanic
ashes in distant
sections
around the world, and tied to paleontological,
geochemical, and geophysical signatures which are assumed to be
contemporaneous
worldwide. Therefore, the strength of
a correlation
between an impact signature found in the
sedimentary
record and an apparently
contemporaneous
crater is based solely on the accuracy and precision of
the geochronologic
calibration
of the time scale, and the
reliability of the geochronologic
date of that crater.
For the calibration
of the Tertiary time scale (from the
top of the Cretaceous
to the Pliocene, see Fig. l), we
used the remarkable record of the Umbria-Marche
pelagic
sequence, which is extensively
exposed throughout
the
northeastern
Appennines
of Italy. This is a unique situation, where a continuous
and complete
sequence of
fossiliferous
marine sediments contains
numerous
volcanic ashes which permit direct radioisotopic
calibration
of the chronostratigraphic
time scale (e.g. Montanari
et
al., 1985, 1988, 1991 ; Odin et al., 1991 ; Oberli and Meier,
1991; Odin et al., 1997). Moreover, the Umbria-Marche
sequence contains the signatures of impacts (i.e., spherules, shocked quartz, Ir anomalies,
biological crises) at
the Cretaceous (K/T) boundary (e.g. Alvarez et al., 1980 ;
Montanari
et al., 1983, 1991) and in the Late Eocene
(e.g., Asaro and Montanari,
1988 ; Montanari
et al., 1993 ;
Clymer et al., 1996; Langenhorst,
1996), which appear to
be correlatable
in numerous
other sections worldwide.
As for the geochronologic
calibration
of the Cretaceous
Period, we have used the time scale of Harland
et al.
(1990).
A remarkable aspect of Fig. 1 is the obvious scarcity of
impact signatures known at present in the stratigraphic
record, compared with the number of large craters on the
surface of the Earth. This is mainly because the search for
signatures, such as Ir anomalies,
spherules, and shocked
minerals,
has been mainly focused on those few short
stratigraphic
intervals
where major
extinctions
are
recorded, or which were known to cover the time of major
impacts. The work of the stratigrapher
searching for a
millimetric impact layer in a sedimentary
sequence hundreds of meters thick is comparable
with the proverbial
search for the needle in the haystack.
It has to be stressed that the record shown in Table 1
and Fig. 1, or any other that can be derived from published
compilations
such as that by Grieve et al. (1995), does not
represent all the impact events that occurred on Earth in
the past 150 Ma, but only the few that have left their
signature on the Earths surface and have been discovered
up to now. There are at least three types of problems
which limit the accuracy for a detailed statistical study of
the impact record through Earths history. Here is a brief
review of them.

A first source of analytical inaccuracy lies in the actual

incompleteness
of the cratering record because the Earths
surface is extremely dynamic on a geological time scale :
many craters may have been completely
obliterated
by
erosion or tectonic deformation,
or buried under orogens,
or subducted
under tectonic plates. Moreover, the size
versus frequency relationship
of impacts follows approximately a power law : small events (craters 20 km in diameter or less) are much more frequent than medium size
events (20-100 km in diameter) which, in turn, are much
more frequent than giant impacts producing craters larger
than 100 km. On the other hand, the preservation
probability of impact craters grows with their size : small structures may be erased from the record much more easily
than large ones. Consequently,
the record in Fig. 1 and
Table 1 shows an abnormal
abundance
of medium size
craters (averaging 25 km in diameter), and an anomalous
scarcity of small size craters which, according to a power
law distribution,
should be much more numerous.
This observation
also suggests that the probability
that
a giant (Z 100 km) crater has been obliterated and is thus
unrecorded in the time interval examined is relatively low.
Nevertheless,
medium or large impact events are those
that count more in the general study of impact flux and
possible cause-and-effect
relationships
between impacts
and biological crises. Large impacts are those which may
leave a world-wide signature in the geological record, and
may have climatic, environmental,
and biological effects
on a global scale.
A second problem lies in the fact that many impact
craters have been discovered recently, at a rate of approximately l-2 new ones per year. This is because there are
on Earth large areas that have not yet been thoroughly
explored for impact structures. Moreover, about 2/3 of
the Earths surface is made of oceanic crust, and at present
all the known impact craters are located on continental
crust.
Just as an example of how critical the incompleteness
of the record may be, let us consider the pioneering work
by Alvarez and Muller (1984), which analyzed the cratering periodicity in the last 250 Ma. This study was based
on a record of only 13 craters larger than 10 km in diameter. In the present work, we have analyzed the distribution
of 28 craters with diameters larger than 10 km,
and over a much shorter geological time span (150 Ma)
than that analyzed by Alvarez and Muller. The recent
flux of new entries in cratering records may explain, by
statistical arguments,
the discrepancies
between different
investigations
(for instance on periodicities)
that have
been carried out in the course of the past decade (e.g.,
Alvarez and Muller, 1984 ; Grieve, 1984 ; Pohl, 1987 ;
Baksi, 1990; Stothers and Rampino,
1990; Grieve et al.,
1995). A study with a few tens of craters which seem to fit
well a periodic age distribution
may be invalidated merely
by introducing
a number of new craters into the record.
Finally,
there are still basic geochronologic
uncertainties in the age estimate of impact craters. The present
work, as well as preceding studies of this kind, is strongly
dependent
on the reliability
of radioisotopic
dates of
craters, and the chronostratigraphic
time scale which, as
geochronology
goes, improves with the progress of sophistication
of analytical
instruments
and methods.
For
instance, for a long time the medium size Manson crater

A. Montanari

275

et al.: Earth cratering record and impact energy flux in the last 150 Ma

was believed to have an age close to the K/T boundary

(i.e. about 66Ma; Hartung et al., 1990). Recently, the
melt rock of this crater has been re-dated yielding an
age about 7Ma older (Izett et al., 1993; Zeitler, 1990).
Similarly Kamensk, another medium size crater (see Table
l), was for a long time considered a K/T boundary event
and was only recently re-dated at 49.2Ma (i.e. 16Ma
younger ; Izett et al., 1994).
Another relevant case is the Kara-Ust Kara double
structure, which was dated by a Russian team using the
traditional K/Ar technique at around 66 Ma (Kolesnikov
et al., 1988), and in an American geochronology laboratory using the 40Ar/3gAr technique at about 74Ma
(Koeberl et al., 1990). Further geochronologic dating by
Trieloff and Jessberger (1992) led to an age estimate
between 69 and 71 Ma. A compromise (conservative) age
of 13 + 3 Ma was finally suggested by Grieve et al. (1995)
and is the one we use in this paper.
Uncertainties also exist in the estimate of the actual size
of some craters. Often these structures are buried and
definable only upon geophysical prospections, remote
sensing, and deep borehole drilling, or they are deeply
eroded and their original sizes inferred from geological
surveying, topographic and hydrographic regional analyses, and modeling. This may leave some room for
interpretation and inference which, in some cases, lead
to significantly different estimates of the original crater
size.
For example, the diameter of the Kara crater, which is
entirely located on land, is established at 65 km in the
compilation of Grieve et al. (1995), whereas the size of the
Ust Kara, submerged under the Kara Sea, is still uncertain
with early estimates indicating a diameter of about 25 km.
Geological studies by Nazarov et al. (1989) suggested a
diameter between 70 and 155 km for the Kara, whereas
analyses of geophysical profiles by Koeberl et al. (1990)
led to a diameter estimate of 80 km. Here we have adopted
the smallest, most conservative estimates, but this may
turn out to be wrong in the future, also considering that
Nazarov et al. (1991) argued that there is not enough
evidence for two Kara craters, and that it may well be a
single, > 120 km, structure.
Another case of size uncertainty is represented by the
second largest crater in our analyzed record, the Chicxulub structure buried under the Yucatan Peninsula. Early
estimates based on gravity anomalies and aeromagnetic
data indicated a % 180-200 km basin (Hildebrand et al.,
1991) but further reprocessing of the same data led to an
estimate of about 300 km (Sharpton et al., 1993, 1996).
For our analysis, we have used a conservative diameter
size for Chicxulub of 180 km.
In summary, we would like to stress that this paper,
along with the record in Table 1 and Fig. lj is not the
ultimate compilation of exact sizes and ages of terrestrial
impact craters but an experimental study mainly to analyze as simply and objectively as possible the time distribution of the 33 best known large craters on Earth
throughout the best known interval of Earth history (i.e.,
the past 150 Ma). Moreover, the compilation we use is a
limited number of craters out of an inferred total population of perhaps more than 200 impact events which are
unknown to us because they are not yet discovered or
which have been completely obliterated by geological pro-

cesses. Nevertheless, although the overall record will certainly be improved with future interdisciplinary geological
research, from this partial record we can deduce the energy
distribution versus time of the impacts that caused the
known craters, and test the possibility of periodicity or
non-random distribution of these events.

3. Impact energy flux analysis

3.1. Cratering energy scaling
Given the craters listed in Table 1, we now have to estimate
the energy associated with any single cratering impact,
namely the kinetic energy of the corresponding extraterrestrial projectiles (either comets or asteroids). As largescale impacts have never been directly observed on Earth,
and the evidence provided by experiments (including
nuclear tests) is limited to energies lower than those of
kilometer-sized extraterrestrial impactors by many orders
of magnitude, we have to resort to suitable scaling laws.
These are extrapolating
relationships inferred from
experiments through dimensional analysis, which allow
one to relate the outcome of an impact (e.g., the characteristics of the crater) to the physical properties of the
projectile.
A number of such relationships have been proposed
over the years for cratering impacts (see Melosh, 1989,
Sec. 7.8). We have adopted the scaling law of Gault (1974)
to be applied to lunar craters larger than 1 km across :
D,, = 0.27p~i6p;12~28(sin

19)~

(1)

where D,, stands for the apparent crater diameter, as

shown in Fig. 2, pp and pt are the projectiles and targets
densities, respectively, W is the projectiles kinetic energy
(all these quantities being expressed in SI units), and 6 is
the impact angle with respect to the local zenith.
We have chosen this scaling law from among the several
laws discussed by Melosh (1989) essentially for simplicity,
as it does not involve additional unknown parameters
such as the projectiles size. As emphasized by Melosh,
owing to the differences between the predictions of different scaling laws, the determination of W from the crater
diameter is rather imprecise, with about one order of
magnitude of uncertainty for large craters. However, since
we are interested only in determining relative (rather than
absolute) impact energies over a limited range of crater
sizes, and since there are other significant sources of uncertainty (to be discussed below), our results are not really
sensitive to the chosen scaling relationship.
In order to apply equation (1) to the observed diameters
of terrestrial craters, two corrections are needed. First, we
have to account for the weaker lunar gravity. According
to Gault, the craters diameter scales with the -0.165
power of the surface gravity acceleration, so the same
projectile
will generate on the Earth a crater
(9.8/1.67)-.65 = 0.75 times as large as on the Moon.
Second, the apparent diameter does not correspond to
the visible (i.e. measurable) rim-to-vim diameter (see Fig.
2). For this, we shall adopt the estimate of Holsapple
(1993) that the measurable crater diameter Do is

et al.: Earth cratering record and impact energy flux in the last 150Ma

A. Montanari

276

Fig. 2. Simplified sketch showing the difference between the apparent diameter (D,Jand the true rimto-rim diameter (Do)of an impact crater

g 1.3 x D,,. Applying these two corrections

equation (1) to obtain W, we have :
W123p;5g5p:.786(sin

0)-1-19 Dt

and inverting
(SI units)

(2)

This is the scaling relationship

that we will use in the
following
calculations
to derive impact energies from
crater diameters.
As for the other parameters appearing in equation (2),
we
adopted
the
following
numerical
values :
,o.*.= 2500 kg rnp3, in agreement with the estimated densities of asteroids, pt = 3500 kg me3 for the Earths crust,
and 0 = 60 (the median value for an isotropic flux of
impactors). Owing to the different exponents of the power
laws making up equation (2), it is easy to show that a 10%
uncertainty
in the crater diameters corresponds to a larger
error in the estimated energy than that due to the expected
uncertainty
in the densities of about 20% (assuming that
only a minor fraction of craters are due to comet impacts).
Taking into account the unknown impact angles, it is clear
that for individual cratering events our estimates may well
be wrong by a factor of 5. However, this unavoidable
uncertainty
is not likely to affect the main qualitative
conclusions
we are going to draw from the present analysis.
Of course all the steps of this procedure are somewhat
arbitrary, and may be replaced by alternative ones, yielding quantitatively
different results. However, given all the
sources of uncertainty
we discussed earlier, we are interested mainly in the qualitative features of the curve shown
in Fig. 3. These features, to be discussed in Section 4,
are not really sensitive to the details of the derivation
procedure.

3.2. EnergyJEux
In order to get an idea of how the impacting
bodies
deposited their energy as a function
of time, we have
chosen the following methodology.
For every impact we
considered
the diameter of the observed crater, its age
estimate and its 20 error. Then we associated with every
crater a gaussian (bell-like) distribution
centered on its
estimated age, with a half-width equal to 0 (half of the
assumed error on the age estimate), and such that its
total area is normalized to the estimated energy W of the

corresponding
cratering event, according to the scaling
law described in Section 3.1.
Then, we summed all the gaussian curves and represented the resulting diagram on a semilogarithmic
scale
for energy flux versus time (the energy deposition
rate
is measured in MtonMa-,
where 1 Mton is the energy
released
by lo9 kg of TNT
explosive
(i.e. about
4.2 x 1015J). The logarithmic energy scale was more suitable than a linear one because of the large differences (up
to 6 orders of magnitude) between the energies associated
with craters of different sizes. To smooth out the sharpest
peaks of the curve generated in this way, we computed
running averages over a succession of 4 Ma interval, and
plotted the resulting smoothed curve in Fig. 3.

3.3. Distribution of impacts times

The time distribution
of the cratering events included in
our sample can be also analyzed, in order to look for
possible non-random
features. To check for possible
periodicities,
we applied a simple test based on the distribution
of crater age differences, devised by S. Perlmutter and applied by Alvarez and Muller (1984) with a
more limited set of data. This test consists in computing
all the time differences between any pair of dated events,
and then representing
the frequency with which any given
value occurs in the data, as a function of the time difference itself.
Every difference is plotted as a gaussian with the two
errors of the corresponding
ages combined quadratically,
and all the gaussians are superimposed.
In this way we get
a curve with a number of peaks corresponding
to the most
frequent time differences. It is clear that a periodic signal
in the data would yield peaks at time differences equal to
the period and its multiples. A more refined Fourier analysis is warranted if there is some hint for periodicity from
this test (cf. Grieve et al., 1985).
The results of the test in the case of the crater age data
listed in Table 1 are shown in Fig. 4(A) (full line). Since
some peaks are apparent in this curve, we tested for their
significance
by applying
the same procedure
with 10
sequences of randomly generated sets of time differences
with the same total number of events. In Fig. 4(A), these
random
curves are represented
by dotted lines. The

A. Montanari

277

et al.: Earth cratering record and impact energy flux in the last 150 Ma

EARLY CRETACEOUS

20

40

60

80

100

120

140

160

Time (Ma)
Fig. 3. A semi-logarithmic

diagram of impact energy flux vs. time over the last 150 Ma (see text).
Major biological crises (as in Fig. 1) are indicated by arrows

peak at about 64 Ma and possibly that at about 33 Ma

appear to exceed the highest random peaks, and are therefore not likely to be caused just by random fluctuations
due to small-number
statistics.
However,
we do not
believe that they correspond
to real periodicities
in our
data set. Rather, they are an artifact of Perlmutters
test,
which yields a high peak whenever a pair of well-determined crater ages is considered : this is, for instance, the
case for the first (youngest) three craters listed in Table 1,
and the Chicxulub crater which combine to produce the
64 Ma top peak in Fig. 4(A).
To check this conclusion, we have slightly modified the
original procedure
of the test by associating
with each
crater a pair of non-normalized
gaussians
of constant
height. As shown in Fig. 4(B), in this case no peak in the
real-data (full) curve exceeds those of the random cases.
Therefore, we conclude that the currently available crater
age data do not provide any significant empirical evidence
for a periodic component
of the Earth impact flux, confirming a number of previous analyses of the same problem (for recent reviews, see e.g. Grieve and Shoemaker,
1994 and Grieve and Pesonen, 1996). On the other hand,
it is clear from the previous discussion that such analyses
on a fairly limited set of data are a delicate matter, and
that one should always evaluate the results of statistical
tests with a critical eye, lest subtle artifacts lead to mistaken (or premature) conclusions.
The fact that no periodicity is present in the data does
not necessarily mean that they are completely random,
and that all the impacts are to be seen as uncorrelated
events distributed
according to Poisson statistics. Actually, from Fig. 1 some clusters of craters closely spaced

in time are quite apparent, and one may wonder whether

they are statistically significant or just random flukes. To
answer this question,
one can apply a simple Poisson
test. According to the Poisson distribution,
for a random
sequence of events the probability
of finding N events in
a time interval where N are expected on average is :
P(N) = (We-)/N!

(3)

Applying this formula to the clusters of craters 1-5,

10-13, 15-19, and 21-24 (numbers as in Fig. 1 and Table
I), assuming average rates of 15/50 = 0.30, 12/50 = 0.24
and 6/50 = 0.12 observed events per Ma in the three intervals between 0 and 50, 50 and 100, 100 and 150 Ma ago,
respectively (see Table l), we obtain the result that all the
four clusters had an a priori probability
of occurrence
< 10% if the process was a random one. Taking into
account that the discovery of the youngest craters 1-5
has been favored by an obvious positive bias, the most
interesting
cluster is probably
that formed by the Late
Eocene craters
10-13, concentrated
between
35 and
40 Ma, two of which (Chesapeake
Bay and Popigai) are
fairly large (so that this cluster produces a prominent peak
in Fig. 3). Thus we have some evidence for correlating
groups of impacts occurring over intervals of several Ma.
In other words, the Earth has probably undergone a significantly variable cratering rate, with enhancements
up
to a factor of 4 for projectile sizes of 1 km and larger. This
finding is not surprising, given our current understanding
of the physical and dynamical
evolution of comets and
asteroids : an enhanced Earth cratering rate over several
Ma could result either from discrete collisions in the main

et al.: Earth cratering record and impact energy flux in the last 150 Ma

A. Montanari

278

20

40

60
Time Intervals (Ma)

20

40

60

SO

100

120

140

160

Time Intervals (Ma)

Fig. 4. (A) Perlmutter test curves for all the pairs of cratering events : the full line corresponds to real
data from Table 1, the dotted ones to 10 curves obtained from fictitious random sequences (see text).
(B) Modified Pelmutter test curves, obtained by using non-normalized gaussians (see text). No peak
in the real-data full curves stands out any longer with respect to the dotted random sequences curves

asteroid belt, injecting swarms of kilometer-sized

fragments into fast-track
chaotic routes to the Earth-crossing region (Farinella et al., 1993, 1994; Menichella et al.,
1996), or from the disintegration
of sizeable comet nuclei,
possibly after a Sun-grazing
passage (Bailey et al., 1994),
or from comet showers triggered by gravitational
perturbations
of the Oort cloud (Hut et al., 1987).

4. Discussion
The incompleteness
of the impact record as a whole, and
the scarcity of impact signatures in sedimentary sequences
limit, perhaps inevitably, an accurate estimate of the overall flux of energy delivered to the Earth by colliding extraterrestrial
objects. Nevertheless,
a few conclusions
can

A. Montanari

et al.: Earth cratering record and impact energy flux in the last 150 Ma

be drawn from this study. For instance, no evidence for

periodicity emerges from the processing of the data considered in this work. As stated above, the set of data we
have processed may represent less than 10% of all the
impact craters > 5 km in diameter that were formed on
Earth in the past 150 Ma. There is no way to infer the
time distribution
of hundreds
of other missing events.
However, it can be said that the record of large impacts
capable of excavating giant craters, which are more resistant to geological obliteration,
is closer to the unknown
total number of this category of events than the few small
craters (< 20 km in diameter) known in the interval of
geological time studied. In any case, because of the way
impact energy scales with crater diameter, the energy flux
versus time diagram
shown in Fig. 3 gives a disproportionate
weight to large craters, and no obvious
periodic feature is seen in this diagram too.
As far as the correlation
of paleontological
extinctions
with the energy distribution
associated
with extraterrestrial
impacts is concerned,
encouraging
but not
decisive arguments can be deduced by comparing
Fig. 1
with Fig. 3. As shown in Fig. 3, the energy flux associated
with the Chicxulub
giant crater (180 km in diameter
according to Hildebrand
et al., 1991, or possibly 300 km
in diameter according
to Sharpton
et al., 1993, 1996),
which has been accurately
and precisely
dated
at
64.98+0.05Ma
(Swisher et al., 1992) and was probably
the cause of the mass extinction at the K/T boundary (the
most prominent
in the time interval
analyzed),
corresponds to the second highest peak.
The highest energy peak in Fig. 3 is given by the recently
discovered Morokweng
crater in South Africa (Koeberl
et al., 1997) which represents
the largest impact crater
known on Earth. This impact event occurred near the
Jurassic/Cretaceous
(J/K) boundary which coincides with
the second largest peak in marine genera extinction in the
150 Ma record examined (Sepkoski, 1990).
The third energy peak shown in Fig. 3, almost two
orders of magnitude lower than the J/K boundary,
is the
result of five closely spaced impacts which occurred in
the Late Eocene, and correlates well with another prominent extinction
peak. Among the Late Eocene impacts,
Chesapeake
(90 km in diameter) and Popigai (100 km in
diameter) are, after Morokweng and Chicxulub, the largest
craters in the record analyzed.
Note that the interplanetary population,
in both its asteroidal and cometary
components,
includes potential
impactors
much bigger
than that at Morokweng, but such impacts probably occur
only on a time scale of 1000 Ma (e.g. Jansa et al., 1990 ;
Bailey et al., 1994; Michel et al., 1996).
As for the rest of the 150 Ma record, there seems to be
a weak or no relation between other extinction maxima
and impact energy peaks, which are anyway orders of
magnitude
smaller than the J/K, K/T and Late Eocene
events. The sequence of closely spaced small and medium
size impacts in the Early Eocene actually corresponds
to
a minimum in extinction rate (see Fig. 1). The Kara-Ust
Kara doublet crater, which may represent a very energetic
and still uncertain event in the Late Cretaceous (Koeberl
et al., 1990; Nazarov
et al., 1991) did not leave a significant signature in the paleontological
record. Moreover, it is worth pointing out that Ir anomalies, or other
impact signatures
in the stratigraphic
record, do not

279

necessarily indicate large impacts capable of affecting the

global biota. The small Ir anomaly found in marine sediments near the Middle/Late
Miocene boundary (Asaro et
al., 1988) is approximately
correlated with a prominent
extinction peak, but there is no known large crater associated with these signatures. On the other hand, the group
of closely spaced small impacts in the Pliocene are not
associated with any relevant biological crisis.
These observations
lead to the view that one single giant
impact, such as those at Morokweng
or Chicxulub, may
cause a significant global biological crisis, whereas smaller
impacts, even if distributed over a geologically short span
of time, may have no relevant effects on the global
environment
and biota recordable
in the stratigraphic
sequence. In other words, there may be an actual lower
impact threshold (in terms of impactor size and energy
release), above which a global environmental
catastrophe
may occur. Such a threshold effect was proposed by Jansa
et al. (1990) who deduced
as zero extinction
threshold
limit the impact caused by a 3 km object (i.e. the
Montagnais
crater with a diameter of 45 km), and about
50 % marine genera extinction with a 10 km bolide impact
(i.e., the Chicxulub
crater with a diameter of at least
180 km and a 35% genera extinction as reported by Sepkoski, 1990). Our results and observations
reported in this
paper agree to some extent with this empirical inference.
As for the time distribution
of impact events, our analysis supports previous results on the absence of any convincing evidence for periodicities.
On the other hand, the
Late Eocene and possibly a few other groups of several
craters appear more closely spaced in time than in a purely
random distribution,
pointing to the intriguing possibility
that discrete breakup events of sizeable parent comets or
asteroids, or comet showers, generated swarms of Earthcrossing objects, yielding significant enhancements
of the
Earth impact rate over intervals of 5-10Ma.
Note that
according to Menichella et al. (1996), at least for the main
belt asteroid source, these enhancements
of the impact
flux would be much more pronounced
for smaller (i.e.,
Tunguska-sized)
events.
In conclusion,
there is still considerable
work to be
carried out for an acceptable understanding
of the causeeffect relationship
between large impacts and global biological crises. This can be developed both by searching for
new craters hidden somewhere throughout
the Earths
surface, and with further systematic high-resolution
interdisciplinary
studies of the stratigraphic
impact record in
continuous
and complete marine sedimentary
sequences.
In doing this, the causal relationship
between impacts
and biological crises can be taken as a simple working
hypothesis,
and high-resolution
stratigraphic
analyses
should be focused across those geological time intervals
where large impacts are known to have occurred, and/or
where major extinction events are in some way recorded.
However, it is worth noting that the effects of extraterrestrial impacts on the biota may depend on the state
of the evolving global ecosystem. Slowly changing climates and environments
controlled
by global tectonics
may produce (select) biological
populations
which are
more or less sensitive to the stress produced by small or
medium size impacts, and some evolving populations
may
either be favored or abruptly ended by such events. Thus,
in the general issue of evolution versus impact, we have

280

A. Montanari

et al.: Earth cratering record and impact energy flux in the last 150 Ma

to take into account the historical

contingency
of each
individual event. As an example, the Tunguska event, too
small to produce a crater, might have caused a global
catastrophe
had it occurred a half a century or so later (a
blink compared with the immensity
of geological time),
and a few tens of degrees more to the west (an infinitesimal
in the vastness of planetary distances), to destroy Moscow
during the Bay of the Pigs crisis. In that case we would
probably not have written this paper, and you would not
be reading it.
Acknowledgements.
This work was partially supported by a grant
from Armines (Paris), and the Coldigioco Research Fund. We
would like to thank Christian Koeberl and an unknown second
P&SS reviewer for the useful comments and suggestions on the
original manuscript.

References
Alvarez, L. W., Alvarez, W., Asaro, F. and Michel, H. V. (1980)
Extraterrestrial cause for the Cretaceous-Tertiary
extinction.
Science 208, 1095-l 108.
Alvarez, W. and Muller, R. A. (1984) Evidence from crater ages
for periodic impacts on Earth. Nature 308, 718-720.
Asaro, F., Alvarez, W., Michel, H. V., Alvarez, L. W., Anders,
M. H., Montanari,
A. and Kennett, J. P. (1988) Possible
world-wide
Middle
Miocene
iridium
anomaly
and its
relationship to periodicity of impacts and extinctions. Lunar
Planet. Inst. 673, 6-7
Asaro, F. and Montanari, A. (1988) Small Late Eocene iridium
anomalies
in the Contessa
Highway III section. In The
Eocene-Oligocene
Boundary in the Marche-Umbria
Basin
(Italy), eds I. Premoli Silva, R. Coccioni and A. Montanari,
pp. 187-188. IUGS Special Publication,
F.lli Aniballi Publishers, Ancona.
Bailey, M. E., Clube, S. V. M., Hahn, G., Napier, W. M. and
Valsecchi, G. B. (1994) Hazards due to giant comets : Climate
and short-term catastrophism.
In Hazards due to Comets and
Asteroids, ed. T. Gehrels, pp. 479-533. University of Arizona
Press, Tucson, AZ.
Baksi, A. K. (1990) Search for periodicity in global events in the
geologic record : Quo vadimus? Geology l&983-986.
Boiko, A. K., Vailler, A. A. and Vishnyak, M. M. (1985) On the
age of the Boltish depression (in Russian). Geol. Zh. 45, 8690.
Bottomley, R. J. and York, D. (1988) Age measurement
of the
submarine Montagnais impact crater. Geophys. Res. Lett. 15,
1409-1412.
Bottomley, R. J., York, D. and Grieve, R. A. F. (1979) Possible
source craters for the North American
tektites:
A geochronological
investigation.
Eos Trans. AGU 60, 309.
Bottomley, R. J., York, D. and Grieve, R. A. F. (1989) 40Ar39Ar dating of impact craters. Lunar PEanet. Sci. Co& 20,
421431.
Bottomley, R. J., York, D. and Grieve, R. A. F. (1993) Age of
the Popigai impact event using the 40Ar/3gAr method. Lunar
Planet. Sci. Conf. 24, 161.
Carrigy, M. A. (I 968) Evidence of shock metamorphism
in rocks
from the Steen River structure, Alberta. In Shock Metamorphism in Natural Material, eds. B.M. French and M.N.
Short, pp. 367-378. Mono Book Corporation,
Baltimore,
MD.
Clymer, A. K., Bite, D. M. and Montanari,
A. (1996) Shocked
quartz
from the late Eocene:
Impact
evidence
from
Massignano,
Italy. Geology 24,483-486.
Deutsch, A., Buhll, D. and Langenhorst,
F. (1992) On the significance of crater ages : new ages from Dellen (Sweden) and
Araguainha (Brasil). Tectonophysics 216, 205-218.

Farinella,
P., Gonczi, R., Froeschlir, Ch. and FroeschlC, C.
(1993) The injection of asteroid fragments into resonances.
Icarus 101, 174-187.
Farinella, P., FroeschlC, Ch., FroeschlC, C., Gonczi, R., Hahn,
G., Morbidelli,
A. and Valsecchi, G. B. (1994) Asteroids
falling into the Sun. Nature 371, 3 14-3 17.
Gault, D. E. (1974) Impact cratering. In A Primer in Lunar
Geology , eds. R. Greeley and P. H. Schultz, pp. 137-175.
NASA Ames, Moffett Field, CA.
Gorter, J. D., Gostin, V. A. and Plummer, P. S. (1989) The
enigmatic sub-surface Tookoonooka
complex in south-west
Queensland:
Its impact origin and implications
for hydrocarbon accumulations.
In The Cooper and Eromanga Basins,
Australia, ed. B. J. ONeil, pp. 441456. Society of Petroleum
Engineers, Adelaide.
Grieve, R. A. F. (1984) The impact cratering rate in recent time.
J. Geophys. Res. 89, B403-B408.
Grieve, R. A. F. (1991) Terrestrial impact: the record in the
rocks. Meteoritics 26, 174-194.
Grieve, R. A. F. and Pesonen, L. J. (1996) Earth Moon Planets
72,357-376.
Grieve, R., Rupert, J., Smith, J. and Therriault, A. (1995) The
record of terrestrial impact record. GSA Today 5, 189-196.
Grieve, R. A. F., Sharpton, V. L., Goodacre, A. K. and Garvin,
J. B. (1985) A prospective
on the evidence for periodic
cometary impacts. Earth Planet. Sci. Lett. 76, 1-9.
Grieve, R. A. F. and Shoemaker, E. M. (1994) The record of
past impacts on Earth. In Hazards due to Comets and
Asteroids, ed. T. Gehrels, pp. 417462. University of Arizona
Press, Tucson, AZ.
Gurov, E. P. and Gurova, E. P. (1980) Shocked metamorphosed
rocks from the Elgygytgyn meteorite
crater in Chukchi
National Okrug (in Russian). Meteoritika 39, 102-109.
Harland, B. W., Armstrong,
R. L., Cox. A. V., Craig, L. E.,
Smith, A. G. and Smith, D. G. (1990) A Geologic Time Scale
1989. pp. l-223. Cambridge University Press, Cambridge.
Hartung, J. B., Kunk, M. J. and Anderson, R. R. (1990) Global
Catastrophes in Earth History, eds. V. L. Sharpton and P. D.
Ward, Geol. Sot. Am., Spec. Pap. 247, pp. 207-222.
Hildebrand,
A. R., Penrield, G. T., Kring, D. A., Pilkington,
M., Camargo, Z. A., Jacobsen, S. B. and Boynton, W. V.
(1991) Chicxulub crater : A possible Cretaceous-Tertiary
boundary crater on the Yucatan peninsula, Mexico. Geology
19,867-871.
Holsapple,
K. A. (1993) The scaling of impact processes in
planetary sciences. Annu. Rev. Earth Planet. Sci. 21,333-373.
Hut, P., Alvarez, A., Elder, W. P., Hansen, T., Kaufman, E. G.,
Keller, G., Shoemaker, E. M. and Weissman, P. R. (1987)
Comet showers as a cause of mass extinctions. Nature 329,
118-126.
Izett, G. A., Cobban, W. A., Obradovich,
J. D. and Kunk, M.
J. (1993) The Manson impact structure: Ar/39Ar age and
its distal impact ejecta in the Pierre Shale in southwestern
South Dakota. Science 262,729-732.
Izett, G. A., Masaitis, V. L., Shoemaker, E. M., Dalrymple, G.
B. and Steiner, M. B. (1994) Eocene age of the Kamensk
buried crater of Russia. LPI Contr. 825, 55.
Jansa, L. F., Aubry, M. P. and Gradstein, F. M. (1990) Comets
and extinctions ; Cause and effects? In Global Catastrophes in
Earth History, eds. V. L. Sharpton and P. D. Ward, Geol.
Sot. Am., Spec. Pap. 247, pp. 223-232.
Jessberger,
E. K. (1988) 40Ar-39Ar dating og the Haughton
impact structure. Meteoritics 23,233-234.
Jessberger, E. K. and Reimold, W. U. (1980) A Late Cretaceous
40Ar-3gAr age for the Lappajgrvi impact Crater. Finland. J.
Geophys. 48, 57-59.
Kilesev, N. P., Korotushenko,
Yu. G. (1986) The Bigach astrobleme in eastern Kazakhstan.
Meteoritika 45, 119-121 (in
Russian).
Kirschner, C. E., Grantz, A. and Mullen, M. W. (1992) Impact
origin of the Avak structure, Arctic Alaska and genesis of

A. Montanari

et al.: Earth cratering

record and impact energy flux in the last 150 Ma

the Barrow gas fields. Am. Assoc. Petrol. Geol. Bull. 76,651678.
Koeberl, C., Poag, C. W., Reimold, W. U. and Brandt, D. (1996)
Impact origin of the Chesapeake Bay structure and source of
the North American tektites. Science 271, 1263-1266.
Koeberl, C., Armstrong,
R. A. and Reimold, W. U. (1997)
Morokweng,
South Africa: A large impact structure
of
Jurassic-Cretaceous
boundary age. Geology 8,73 l-734.
Koeberl,
C., Sharpton,
V., Harrison,
T. M., Sandwell, D.,
Murali, A. V. and Burke, K. (1990) The Kara/Ust-Kara
twin
structure ; A large-scale impact event in the late Cretaceous.
In Global Catastrophes in Earth History, eds. V. L. Sharpton
and P. D. Ward, Geol. Sot. Am., Spec. Pap. 247, pp. 233238.
Koeberl, C. and Storzer, D. (1988) Chemical composition
and
fission track age of Zhamanshin
crater glass. In Proc. 2nd
Int. ConJ on Natural Glasses, Prague, pp. 203-205.
Kolesnikov, E. M., Nazarov, M. A., Badjukov, D. D. and Shukulov, Y. A. (1988) The Karsidy craters are the probable
record of catastrophe
at the Cretaceous-Tertiary
boundary.
LPI Contr. 673,99-100.
Langenhorst,
F. (1996) Characteristics
of shocked quartz in
late Eocene impact ejecta from Massignano (Ancona, Italy) :
clues to shock conditions and source crater. Geology 24,487490.
Mak, E. K., York, D., Grieve, R. A. F. and Dence, M. R. (1976)
The age of the Mistastin Lake crater, Labrador,
Canada.
Earth Planet. Sci. Lett. 31, 345-357.
Masaitis, V. L., Danilin, A. N., Mastchak, M. S., Raikhlin, A.
I., Selivanovskaya,
T. V. and Shadenkov,
E. M. (1980) In
Geology of Astroblems, pp. l-231 (in Russian). Nedra Press,
Moscow.
McHone, J. F. and Sorkhabi, R. B. (1994) Apatite fission track
age of the Marquez Dome impact structure. Texas. Lunar
Planet. Sci. Conf. 25,88 1.
Melosh, H. J. (1989) Impact Cratering : A Geologic Process, pp.
l-245. Clarendon Press, New York.
Menichella, M., Paolicchi, P. and Farinella, P. (1996) The main
belt as a source of near-Earth asteroids. Earth Moon Planets
72, 133-149.
Michel, P., Farinella, P. and Froeschle, Ch. (1996) The orbital
evolution of the asteroid Eros and implications for collision
with the Earth. Nature 380, 689-69 1.
Milton, D. J. and Sutter, J. F. (1987) Revised age for the Gosses
Bluff impact structure, Northern Territory, Australia, based
on 40Ar/39Ar dating. Meteoritics 22, 281-289.
Montanari,
A. (1991) Authigenesis
of impact spheroids in the
K/T boundary clay from Italy: new constraints
for highresolution
stratigraphy
of terminal Cretaceous
events. J.
Sediment. Petrol. 61,315339.
Montanari,
A., Asaro, F., Kennett, J. P. and Michel, E. (1993)
Iridium anomalies of Late Eocene age at Massignano (Italy)
and in ODP Site 689B (Maud Rise, Antarctica).
Palaios 8,
420437.
Montanari,
A., Deino, A., Drake, R., Turrin, B. D., De Faolo,
D. J., Odin, S. G., Curtis, G. H., Alvarez, W. and Bite,
D. M. (1988) Radioisotopic
dating of the Eocene-Oligocene
boundary in the pelagic sequence of the Northern Apennines.
In The Eocene-Oligocene
Boundary in the Marche-Umbria
Basin (Italy), eds. I. Premoli Silva, R. Coccioni and A. Montanari, pp. 195-208. IUGS Spec. Publ., Aniballi Publ.,
Ancona.
Montanari,
A., Deino, A., Coccioni, R., Langenheim,
V. E.,
Capo, R. and Monechi, S. (1991) Geochronology,
Sr isotope
stratigraphy,
magnetostratigraphy
and plankton
stratigraphy across the Oligocene-Miocene
boundary in the Contessa section (Gubbio, Italy). News Strut. 23, 151-180.
Montanari,
A., Drake, R., Bite, M. D., Alvarez, W., Curtis, G.
H., Turrin, B. D. and DePaolo, D. J. (1985) Radiometric
time scale for the upper Eocene and Oligocene based on

281

K/Ar and Rh/Sr dating of volcanic biotites from the pelagic

sequence of Gubbio, Italy. Geology 13, 596-599.
Montanari,
A., Hay, R. L., Alvarez, W., Asaro, F., Michel, H.
V., Alvarez, L. W. and Smit, J. (1983) Spheroids at the
Cretaceous-Tertiary
boundary are altered impact droplets of
basaltic composition.
Geology 11, 6688671.
Nazarov,
M. A., Kolesnikov,
E. M., Badjukov, D. D. and
Masaitis, V. L. (1989) Reconstruction
of the original morphology of the Kara impact structure and its relevance to the
K/T boundary event. Lunar Planet. Sci. 22,959-960.
Nazarov, M. A., Badjukov, D. D., Barsukova, L. D. and Alekseev, A. S. (1991) Reconstruction
of original morphology
of the Kara impact structure and its relevance to the K/T
boundary event. Lunar Planet. Sci. ConJ 22,959-960.
Oberli, F. and Meier, M. (1991) Age of Eocene-Oligocene
boundary in the Marche-Umbria
basin, Italy, by high resolution U-Th-Pb
dating. Terra 3,286.
Odin, G. S., Montanari,
A. and Coccioni, R. (1997) Chronostratigraphy of Miocene stages : A proposal for the definition
of precise boundaries.
In Miocene Stratigraphy : An integrated approach, eds. A. Montanari,
G. S. Odin and R.
Coccioni. Developments in Paleontology and Stratigraphy,
Vol. 15, pp. 597629. Elsevier, Amsterdam.
Odin, G. S., Montanari,
A., Deino, A., Drake, R., Guise, P.,
Kreuzer, H. and Rex, D. C. (1991) Reliability of volcanosedimentary
biotite
ages around
the Eocene/Oligocene
boundary. Chem. Geol. (Isotope Geosci.) 86,203-224.
Pohl, J. (1987) Research in terrestrial impact structures : Introduction. In Research in Terrestrial Impact Structures, ed. J.
Pohl, pp. 14. Vieweg, Wiesbaden.
Raup, D. M., Sepkoski, J. 1. Jr. and (1986) Periodic extinctions
of families and genera. Science 231, 833-836.
Sepkoski, J. J., Jr. (1990) The taxonomic structure of periodic
extinction. In Global Catastrophes in Earth History, eds. V.
L. Sharpton and P. D. Ward, Geol. Sot. Am., Spec. Pap.
247, pp. 3344.
Sharpton, V. L., et al. (1993) Chicxulub multi-ring impact basin :
Size and other characteristics
derived from gravity analyses.
Science 261, 1564-l 567.
Sharpton, V. L., Marin, L. E., Carney, J. L., Scott, L., Ryder,
G., Schuraytz, B. C., Sikora, P. and Spudis, P. D. (1996) A
model of the Chicxulub impact basin based on evaluation of
geophysical data, well logs and drill core samples. In The
Cretaceous-Tertiary
Event and other Catastrophes in Earth
History, eds. G. Ryder and D. Fastovsky, Geol. Sot. of Am.
Spec. Publ. 307, pp. 55-74.
Staudacjer, T., Jessberger, E. K., Dominik, B., Kirsten, T. and
Schaeffer, 0. A. (1982) 40Ar-39Ar ages of rocks and glasses
from the Ndrdlinger Ries crater and the temperature history
of impact breccias. J. Geophys. 51, l-l 1.
Stothers, R. B. and Rampino, M. R. (1990) Periodicity in flood
basalts, mass extinctions and impacts : A statistical view and
a model. In Global Catastrophes in Earth History, eds. V. L.
Sharpton and P. D. Ward, GSA Special Paper 247, pp. 9-19.
Swisher, C. C. III Grajales, J. M., Montanari, A., Cedillo-Pardo,
E., Margolis, S. V., Clayes, P., Alvarez, W., Smit, J., Renne,
P. and Maurasse, F. (1992) Chicxulub crater and K-T boundary tektites from Mexico and Haiti yield 40Ar/39Ar ages of
65 Ma. Science 257, 954958.
Trieloff, M. and Jessberger, E. K. (1992) 40Ar/39Ar ages of the
large impact structures Kara and Manicougan and their relative relevance to the Cretaceous-Tertiary
and the TriassicJurassic boundary. Meteoritics 27, 299-300.
Vishnevsky, S. A. and Lagutenko, V. N. (1986) The Ragozinka
astrobleme:
An Eocene crater in the central Urals. Dokl.
Akad. Nauk SSSR 14, l-42 (in Russian).
Zeitler, P. (1990) 40Ar/39Ar thermochronology
of shocked feldspars from the Manson impact structure. In Global Catastrophes in Earth History, eds. V. L. Sharpton and P. D. Ward,
Geol. Sot. Am., Spec. Pap. 247, pp. 283-296.