Author Topic: Traces of Cat.  (Read 518 times)

Admin

  • Administrator
  • Hero Member
  • *****
  • Posts: 555
    • View Profile
Traces of Cat.
« on: January 29, 2017, 04:46:25 am »
Contents
Chapter 1: LANDSCAPES WITH CRATERS: METEORITE IMPACTS,
EARTH, AND THE SOLAR SYSTEM..... 1
1.1. The New Geology: Meteorite Impacts on the Earth ..... 1
1.2. The Planetary Perspective..... 4
1.3. A Peculiar Process: Why Impacts are Different ..... 6
1.3.1. Rarity ..... 7
1.3.2. Immense Energy ..... 7
1.3.3. Instant Effects ..... 7
1.3.4. Concentrated Energy Release ..... 8
1.3.5. Extreme Physical Conditions ..... 9
1.3.6. Unique Deformation Effects ..... 10
Chapter 2: TARGET EARTH: PRESENT, PAST, AND FUTURE .....11
2.1. Comets and Asteroids: The Killer Neighbors? ..... 11
2.1.1. Asteroids ..... 11
2.1.2. Comets ..... 11
2.1.3. Close Encounters ..... 12
2.2. In Our Time: Small Catastrophes ..... 12
2.3. The Problems of Prediction: How Big, How Often? ..... 12
2.3.1. Ingredients of Catastrophe ..... 12
2.3.2. Uncertain Estimates ..... 13
2.3.3. An Uncertain Future? ..... 16
Chapter 3: FORMATION OF IMPACT CRATERS .....17
3.1. Shock Waves and Crater Formation ..... 17
3.1.1. Contact/Compression Stage ..... 18
3.1.2. Excavation Stage: The Transient Crater ..... 20
3.1.3. Modification Stage ..... 23
3.2. Simple and Complex Impact Structures ..... 23
3.2.1. Simple Craters ..... 23
3.2.2. Complex Craters ..... 24
3.2.3. Multiring Basins ..... 27
3.3. Subsequent Development of Impact Structures ..... 28
Chapter 4: SHOCK-METAMORPHIC EFFECTS IN ROCKS AND MINERALS .....31
4.1. Formation Conditions and General Characteristics ..... 31
4.2. Stages of Shock Metamorphism ..... 36
4.3. Megascopic Shock-Deformation Features: Shatter Cones ..... 36
4.4 High-Pressure Mineral Polymorphs ..... 40
4.5. Planar Microstructures in Quartz ..... 42
4.5.1. Planar Fractures ..... 42
4.5.2. Planar Deformation Features (PDFs) ..... 42
4.5.3. PDF Orientations ..... 49
4.5.4. PDFs in Sedimentary Rocks ..... 52
4.6. Planar Microstructures in Feldspar and Other Minerals ..... 53
4.7. Shock Isotropization and Diaplectic Glasses ..... 55
4.8. Selective Mineral Melting ..... 57
vi
Chapter 5: SHOCK-METAMORPHOSED ROCKS (IMPACTITES) IN
IMPACT STRUCTURES .....61
5.1. Rock Types in the Final Impact Structure ..... 61
5.2. Classification of Impactites ..... 62
5.3. Subcrater Rocks ..... 62
5.3.1. Formation Conditions ..... 62
5.3.2. In-Place Shock-Metamorphosed Rocks ..... 63
5.3.3. Lithic Breccias (Parautochthonous) ..... 64
5.3.4. Cross-Cutting (Allogenic) Breccias ..... 64
5.3.5. Pseudotachylite ..... 65
5.4. Crater Interior: Crater-Fill Deposits (Breccias and Melt Rocks) ..... 69
5.4.1. Formation Conditions ..... 69
5.4.2. Lithic Breccias (Allogenic) ..... 71
5.4.3. Melt-Fragment Breccias (Allogenic) (Suevites) ..... 71
5.4.4. Melt-Matrix Breccias (Impact-Melt Breccias) ..... 74
5.5. Crater Rim Zone and Proximal Ejecta Deposits ..... 74
5.6. Distal Ejecta ..... 78
Chapter 6: IMPACT MELTS .....79
6.1. Formation Conditions ..... 79
6.2. Impact Melt Volumes and Crater Size ..... 81
6.3. Impact Melt Varieties in the Near-Crater Environment..... 82
6.3.1. Small Glassy Bodies ..... 82
6.3.2. Impact Melt Breccias ..... 82
6.3.3. Large Crystalline Bodies (Dikes and Sills) ..... 86
6.4. Impact Melt in Distal Ejecta ..... 87
6.4.1. Spherules ..... 88
6.4.2. Tektites and Microtektites ..... 89
6.4.3. Miscellaneous Impact Glasses ..... 90
6.5. Recognition of Impact Melt Rocks ..... 90
Chapter 7: HOW TO FIND IMPACT STRUCTURES .....97
7.1. Reasons for the Search ..... 97
7.2. Detection of Candidate Impact Sites ..... 97
7.2.1. Geological Features ..... 98
7.2.2. Geophysical Features ..... 98
7.3. Verification of Impact Structures ..... 99
Chapter 8: WHAT NEXT? CURRENT PROBLEMS AND
FUTURE INVESTIGATIONS ..... 101
8.1. Identification of New Impact Structures ..... 101
8.2. Impact Events and Extinctions ..... 101
8.3. Distal Impact Ejecta ..... 102
8.4. Carbon Chemistry in the Impact Environment ..... 102
8.5. Postimpact Processes and Effects ..... 103
8.6. Petrogenesis of Igneous Rocks: Impact Melts ..... 103
8.7. Impacts and the Early Earth ..... 104
Appendix ..... 107
References ..... 111

1.3. A PECULIAR PROCESS: WHY IMPACTS
ARE DIFFERENT
Large impact events differ in many ways from more familiar
geological processes like volcanic explosions, earthquakes,
and the slow movements of plate tectonics. Much of
the past confusion and controversy about meteorite impact
on Earth has arisen from the fact that the chief features
of large impact events are unfamiliar to geologists and the
public alike.
1.3.1. Rarity
Unlike other geological processes, large meteorite impacts
are rare, even over geological timescales, and there have been
(fortunately) no historical examples of such events. For most
people, the impact process involves only the occasional falls
of small meteorites, which produce excitement and public
interest, but only occasional minor damage. This lack of direct
human experience with large impact events sets them
apart from more familiar recurrent geological “catastrophes”
such as floods, earthquakes, and volcanic eruptions and makes
them harder to appreciate.
1.3.2. Immense Energy
Large impact events release energies that are almost incomprehensibly
large by the more familiar standards of earthquakes
and volcanic explosions. The energy of an impact
event is derived from the kinetic energy of the impacting
projectile and is equal to 1/2 mv2, where m is the projectile
mass and v its velocity. Because velocities of impacting objects
are high, typically tens of kilometers per second, kinetic
energies are also large, even for small objects (for details,
see below and Table 2.1). An object only a few meters across
carries the kinetic energy of an atomic bomb, and its impact
could devastate a large city. Furthermore, unlike earthquakes
and volcanic explosions, where the properties of Earth itself
provide some upper bounds to the energy release, the impact
energy is limited only by the mass and velocity of the
projectile. The impact of an object only a few kilometers
across (still smaller than many known asteroids and comets)
can release more energy in seconds than the whole Earth
releases (through volcanism, earthquakes, tectonic processes,
and heat flow) in hundreds or thousands of years.
1.3.3. Instant Effects
Another critical difference between impacts and other
geological processes is that the energy release in an impact
event — and the formation of the resulting crater — is vir-
tually instantaneous. At the instant of impact, the object’s
kinetic energy is converted into intense high-pressure shock
waves, which radiate rapidly outward from the impact point
through the target rocks at velocities of a few kilometers per
second (see e.g., Melosh, 1989, Chapters 3–5). Large volumes
of target rock are shattered, deformed, melted, and even
vaporized in a few seconds, and even large impact structures
form in only minutes. A 1-km-diameter crater [about the
size of Barringer Meteor Crater (Arizona)] forms in a few
seconds. A 200-km-diameter structure [like Sudbury
(Canada) or Vredefort (South Africa)] forms in less than
10 minutes, although subsequent geological adjustments,
largely driven by gravity, will continue for many years.
1.3.4. Concentrated Energy Release
Most forms of internal terrestrial energy (heat flow, seismic
waves) are released over large areas that are subcontinental
to global in extent. By contrast, the energy of an impact
event is released instantly, at virtually a single point on Earth’s
surface. Most of the energy passes, directly and rapidly, into
the near-surface target rocks, the atmosphere, and the biosphere,
where it can produce immediate and catastrophic
changes.
A small impact, releasing the energy of only a few million
tons of TNT (approximately the amount released by a
hydrogen bomb), is similar in total energy to a severe earthquake
or volcanic explosion, and its effects will be largely
local (e.g., Kring, 1997). But a large impact transmits so much
energy into the target that an impact structure tens or hundreds
of kilometers in diameter is formed, accompanied by
catastrophic environmental effects on a continental or global
scale.
The near-surface release of impact energy, and the transfer
of much of the energy directly into the biosphere, makes
large impact events especially effective in causing devastating
and widespread biological extinctions. Current impactrelated
models for the major Cretaceous-Tertiary (K/T)
extinction (e.g., Silver and Schultz, 1982; Sharpton and Ward,
1990; Kring, 1993; Ryder et al., 1996) indicate that, during
the impact that formed the Chicxulub crater at 65 Ma, as
much as 25–50% of the projectile’s original kinetic energy
was converted into heat. This heat not only vaporized the
projectile itself, but also melted and vaporized large volumes
of the near-surface sedimentary target rocks, releasing large
amounts of CO2 (from carbonates) and SO2 (from evaporites).
Introduced into Earth’s atmosphere, together with large
quantities of impact-produced dust, these gases and their
reaction products could produce major environmental effects:
immediate darkening and cooling, subsequent global warming,
and deluges of acid rain. Any of these consequences, or
a combination of them, could have produced the resulting
widespread extinction.

1.3.5. Extreme Physical Conditions
The mechanism by which impacts do their work — generation
and transmission of intense shock waves through the
target rocks — is also unfamiliar to many geologists. Under
normal conditions, rocks in Earth’s crust and upper mantle
are subjected to static load pressures produced by the weight
of overlying rocks. These pressures are less than a few
gigapascals (GPa) (1 GPa, a standard unit of pressure, equals
104 bar or about 104 atm). Normal geological stresses within
Earth generate relatively low strain rates (typically10–3/s
to 10–6/s), and rocks either deform slowly at lower pressures
or fracture at higher pressures when their yield strengths (a
few GPa) are exceeded. The general tendency of terrestrial
rocks to fracture when the pressure gets too high, thus releasing
the pressure, limits the pressure buildup in ordinary
geological processes (e.g., earthquakes) to a few GPa.
These “normal” conditions do not exist in impact events.
The rapid release of large amounts of energy in such events
puts too much sudden stress on the target rocks for them to
respond in the normal way. Typical impact velocities of tens
of kilometers per second far exceed the velocities of sound in
the target rocks (typically 5–8 km/s). The resulting impactproduced
shock waves travel through the target rocks at supersonic
velocities, and they impose intense stresses on the
rocks without giving them time to give way by normal deformation.
In the shock-wave environment, transient pres-
sures may exceed 500 GPa at the impact point and may be
as high as 10–50 GPa throughout large volumes of the
surrounding target rock. Transient strain rates may reach
104/s –106/s, orders of magnitude higher than those in ordinary
geological processes. At the higher shock pressures
(>60 GPa), shock-produced temperatures can exceed
2000°C, and rapid, large-scale melting occurs immediately
after the shock wave has passed.
1.3.6. Unique Deformation Effects
The extreme physical conditions of pressure, temperature,
and strain imposed by transient shock waves produce
unique effects (e.g., mineral deformation, melting) in the
rocks and mineral grains through which they pass. These
shock-metamorphic effects are distinct from features produced
by normal geological deformation, and they are now
generally accepted as unique products of impact events (for
reviews and references, see French and Short, 1968; Stöffler,
1972, 1974; Stöffler and Langenhorst, 1994; Grieve et al.,
1996).

Shock-metamorphic effects (or “shock effects”) have been
crucial in establishing the importance of extraterrestrial impact
events. Preserved meteorites around an impact crater
can provide definite evidence of an impact origin, but only a
small fraction of terrestrial impact structures (about a dozen)
have actual preserved meteorites associated with them. These
structures are all relatively small and geologically young. The
Barringer Meteor Crater (Arizona), 1.2 km in diameter and
about 50,000 years old (Fig. 1.1), is the largest member of
this group.
The absence of meteorite fragments around older impact
craters results from two causes: (1) the projectile itself is also
subjected to the intense shock waves generated by the impact,
and it is almost completely melted and vaporized;
and (2) all meteorites are partly to completely composed of
nickel-iron metal, and even surviving fragments of the projectile
tend to be rapidly destroyed by surface weathering,
except in the driest desert regions or on polar icecaps.
The rapid destruction of meteorites means that other lines
of evidence must be used to identify older or deeply eroded
terrestrial impact structures. Shock-metamorphic effects can
be preserved in rocks for periods of 106–109 years, and they
provide a unique means of identifying impact structures, especially
ones that are old, deeply eroded, or both (French
and Short, 1968). The great majority of currently known
impact structures (currently over 150) have no preserved meteorites,
but have been identified by the discovery of shockmetamorphic
effects in their rocks (Grieve, 1991; Grieve et
al., 1995; Grieve and Pesonen, 1992, 1996).

« Last Edit: January 29, 2017, 05:46:41 am by Admin »

Share on Facebook Share on Twitter


Admin

  • Administrator
  • Hero Member
  • *****
  • Posts: 555
    • View Profile
Re: Traces of Cat.
« Reply #1 on: January 29, 2017, 04:48:49 am »
Formation of Impact Craters 17
17
The processes by which large impact craters form, and
the sudden releases of huge quantities of energy involved,
cannot be duplicated in the laboratory, and, fortunately, no
such structure has formed during recorded human history.
All our knowledge about large impact structures is therefore
indirect, and it has come from combining several areas of
once-separate research: theoretical and experimental studies
of shock waves (for reviews and literature, see Melosh,
1989), experimental production of small craters (e.g., Gault
et al., 1968; Gault, 1973; Holsapple and Schmidt, 1982, 1987;
papers in Roddy et al., 1977), and geological studies of larger
terrestrial impact structures (Shoemaker, 1963; Dence, 1968;
Dence et al., 1977; Grieve and Cintala, 1981; Grieve et al.,
1981; Schultz and Merrill, 1981; Stöffler et al., 1988). The
cratering process is complex, many details are still uncertain,
and neither calculations nor predictions can be made with
firm confidence. But these studies provide the essential basis
for understanding how impact craters form and for deciphering
the geological features they display.
3.1. SHOCK WAVES AND
CRATER FORMATION
The general term “impact crater” is used here to designate
a hypervelocity impact crater, the structure formed by
a cosmic projectile that is large enough and coherent enough
to penetrate Earth’s atmosphere with little or no deceleration
and to strike the ground at virtually its original cosmic
velocity (>11 km/s). Such projectiles tend to be relatively
large, perhaps >50 m in diameter for a stony object and >20 m
for a more coherent iron one.
Smaller projectiles, typically a few meters or less in size,
behave differently in passing through the atmosphere. They
lose most or all of their original velocity and kinetic energy
in the atmosphere through disintegration and ablation, and
they strike the ground at speeds of no more than a few hundred
meters per second. In such a low-velocity impact, the
projectile penetrates only a short distance into the target (depending
on its velocity and the nature of the target material),
and the projectile’s momentum excavates a pit that is
slightly larger than the projectile itself. The projectile survives,
more or less intact, and much of it is found in the
bottom of the pit. Such pits, sometimes called penetration
craters or penetration funnels, are typically less than a few
tens of meters in diameter.
Examples of these features include Brenham (Kansas),
the many small pits made by the Sikhote-Alin (Russia) meteorite
shower in 1947, and the pit dug by the largest piece
of the Kirin (China) meteorite fall in 1976. The process of
excavation is strictly a mechanical one, and high-pressure
shock waves are not produced.
In sharp contrast, a hypervelocity impact crater starts to
form at the instant that an extraterrestrial object strikes the
ground surface at its original cosmic velocity. These impact
velocities are much greater than the speed of sound in the
target rocks, and the crater is produced by intense shock
waves that are generated at the point of impact and radiate
outward through the target rocks. Shock waves are intense,
transient, high-pressure stress waves that are not produced
by ordinary geological processes (for details, see Melosh, 1989,
Chapter 3 and references therein). Peak shock pressures produced
at typical cosmic encounter velocities may reach several
hundred GPa. These pressure are far above the stress
levels (~1 GPa) at which terrestrial rocks undergo normal
elastic and plastic deformation, and the shock waves produce
unique and permanent deformation effects in the rocks
through which they pass.
The shock waves radiate from the impact point at high
velocities that may exceed 10 km/s, much greater than the
speed of sound in the target rocks. As they expand, they interact
with the original ground surface to set a large volume
of the target rock into motion, thus excavating the impact
crater. The formation of an impact crater by shock waves,
Formation of Impact Craters
18 Traces of Catastrophe
and the immediate modification of the newly formed crater
by gravity and rock mechanics, is a complex and continuous
process. However, it is convenient to divide this process,
somewhat arbitrarily, into three distinct stages, each dominated
by different forces and mechanisms: contact and compression,
excavation, and modification (Gault et al., 1968;
see also Melosh, 1989, Chapters 4, 5, and 8).
3.1.1. Contact/Compression Stage
This stage begins at the instant that the leading edge of
the moving projectile makes contact with the ground surface.
If the target is solid rock, the projectile is stopped in a
fraction of a second and penetrates no more than 1–2× its
own diameter (Fig. 3.1) before its immense kinetic energy is
transferred to the target rocks by shock waves generated at
the interface between projectile and target (Kieffer and
Simonds, 1980; O’Keefe and Ahrens, 1982, 1993; Melosh, 1989,
Chapter 4).
The general features of this conversion of kinetic energy
into shock waves have been determined from experiments
and theoretical studies (e.g., O’Keefe and Ahrens, 1975, 1977,
1993; Ahrens and O’Keefe, 1977; papers in Roddy et al., 1977;
Melosh, 1989, Chapter 4), although many details are still not
well understood. One clear result is that, as one set of shock
waves is transmitted outward from the interface into the target
rocks, a complementary shock wave is reflected back into
the projectile (Fig. 3.1) (Melosh, 1989, Chapter 4; O’Keefe
and Ahrens, 1993).
The shock waves transmitted into the target rocks lose
energy rapidly as they travel away from the impact point.
Two factors are involved in this energy loss: (1) the expanding
shock front covers an increasingly larger hemispherical
area with increasing radial distance, thus reducing the overall
energy density; (2) additional energy is lost to the target
rocks through heating, deformation, and acceleration. The
peak pressures of the shock waves therefore also drop rapidly
with distance from the impact point. Theoretical models
(Melosh, 1989, pp. 60–66) and geological studies of
shock-metamorphosed rocks in individual structures (Dence,
1968; Robertson, 1975; Grieve and Robertson, 1976; Dence et
al., 1977; Robertson and Grieve, 1977; Dressler et al., 1998)
indicate that the peak shock-wave pressure (Ps) drops exponentially
with the distance R from the impact point according
to an equation of the form Ps a R–n. Various field and
laboratory studies indicate a dependence of R–2 to R–4.5; the
exact value of the exponent depends on projectile size and
impact velocity (Ahrens and O’Keefe, 1977).
On the basis of these studies, it is possible to regard the
impact point as surrounded by a series of concentric, roughly
hemispherical shock zones, each zone distinguished by a
certain range of peak shock pressure (Fig. 3.2) and characterized
by a unique suite of shock-metamorphic effects produced
in the rocks. At the impact point, peak shock-wave
pressures may exceed 100 GPa (= 1000 kbar or 1 Mbar) for
typical cosmic encounter velocities, producing total melting,
if not vaporization, of the projectile and a large volume of
surrounding target rock. Further outward, pressures of 10–
50 GPa may exist over distances of many kilometers from
the impact point, producing distinctive shock-deformation
effects in large volumes of unmelted target rock.
At even greater distances from the impact point, the peak
shock-wave pressures eventually drop to about 1–2 GPa
(Kieffer and Simonds, 1980). At this point, near the eventual
crater rim, the shock waves become regular elastic waves or
seismic waves, and their velocity drops to that of the velocity
of sound in the target rocks (typically 5–8 km/s). These seismic
waves can be transmitted throughout the entire Earth,
like similar waves generated by earthquakes and volcanic
eruptions. Because of their low pressures, they do not produce
any permanent deformation of the rocks through which
they pass. However, seismic waves may produce fracturing,
brecciation, faulting, and (near the surface) landslides, and
the results may be difficult to distinguish from those of normal
geological processes.
The duration of the contact/compression stage is determined
by the behavior of the shock wave that was reflected
back into the projectile from the projectile/target interface
(Fig. 3.1) (Melosh, 1989, pp. 57–59). When this shock wave
reaches the back end of the projectile, it is reflected forward
into the projectile as a rarefaction or tensional wave (also
Fig. 3.1. Contact/compression stage: shock-wave generation
and projectile deformation. Theoretical cross-section showing
calculated conditions immediately after the impact of a large,
originally spherical, projectile (stippled) onto a uniform target. The
projectile has penetrated about half its diameter into the target,
and intense shock waves (pressures in GPa) are radiating outward
into the target from the interface. The projectile itself has become
intensely compressed, and similar shock waves from the interface
are spreading toward the rear of the projectile. When this shock
wave reaches the rear of the projectile, it will be reflected forward
as a tensional wave or rarefaction, unloading the projectile and
allowing it to transform, virtually instantaneously, into melt and
vapor. The original model, developed for large lunar impact events
(O’Keefe and Ahrens, 1975), represents conditions about 1 s after
the impact of a 46-km-diameter anorthosite projectile at 15 km/s
onto a gabbroic anorthosite target, but similar conditions will be
produced by smaller impacts and other material compositions.
(Modified from Melosh, 1989, Fig. 4.1a, p. 47.)
Formation of Impact Craters 19
called a release wave). As the release wave passes through
the projectile from back to front, it unloads the projectile
from the high shock pressures it had experienced. Because
the shock pressures, and the associated temperatures, have
been so high, this release results in the virtually complete
melting and vaporization of the projectile. At the instant at
which the release wave reaches the front end of the projectile,
the whole projectile is unloaded, and the release wave
continues forward into the target and begins to decompress
it as well. This point, at which the release wave reaches the
front of the projectile and begins to enter the adjacent compressed
target, is taken as the end of the complete contact/
compression stage.
The contact/compression stage lasts no more than a few
seconds, even for impacts of very large objects. The time
required for the shock wave to travel from the projectile/
target interface to the rear edge of the projectile is approximately
equal to the time it takes the projectile to travel the
distance of one diameter at its original velocity. Even for
large projectiles, this time is short: 2 s for a 50-km-diameter
projectile traveling at 25 km/s, and less than 0.01 s for a
100-m-diameter object traveling at the same speed. The
additional time required for the release wave to travel from
the rear to the front edge will be no more than a few times
this value, depending on the properties of projectile and target
rock (Melosh, 1989, pp. 48 and 58). For most impact
events, the entire contact/compression stage is over in less
than a second.
After the release wave has reached the front end of the
projectile and unloaded it completely, the projectile itself plays
no further role in the formation of the impact crater, and the
actual excavation of the crater is carried out by the shock
waves expanding through the target rocks. The vaporized
portion of the projectile may expand out of the crater as part
􀀀􀀀
􀀀􀀀
􀀀􀀀􀀀
􀀀􀀀􀀀
􀀀􀀀􀀀
􀀀
Fig. 3.2. Contact/compression stage: initial shock-wave pressures and excavation flow lines around impact point. Schematic crosssection
showing peak shock pressure isobars (pressures in GPa) developed in the target around the impact point near the end of the
contact/compression stage. The originally spherical projectile, after penetrating about two diameters into the target, has been almost
completely destroyed and converted to melt and vapor. Shock waves radiating from the projectile-target interface decline rapidly outward
in peak pressure (isobars in GPa on left side of cavity), creating concentric, approximately hemispherical zones of distinctive shock effects
(right side of cavity). From the original interface outward, these zones involve: (1) melting (>50 GPa) and formation of a large melt unit;
(2) shock-deformation effects (5–50 GPa); (3) fracturing and brecciation (1–5 GPa). The subsequent excavation stage involves two
processes: (1) upward ejection (spalling) of large near-surface fragments and smaller ejecta (ejecta curtain) (upward-pointing arrows
above ground surface); (2) subsurface flow of target material to form the transient crater (arrow paths crossing isobars at left side).
(Modified from Melosh, 1989, Fig. 5.4, p. 64.)
20 Traces of Catastrophe
of a vapor plume (Melosh, 1989, pp. 68–71), and the remainder,
virtually all melted, may be violently mixed into the
melted and brecciated target rocks.
3.1.2. Excavation Stage: The Transient Crater
The brief contact/compression stage grades immediately
into a longer excavation stage, during which the actual impact
crater is opened up by complex interactions between
the expanding shock waves and the original ground surface
(Fig. 3.3) (Melosh, 1989, Chapter 5; Grieve, 1991). As the
contact/compression stage ends, the projectile is surrounded
by a roughly hemispherical envelope of shock waves that
expand rapidly through the target rock. Because the projectile
has penetrated a finite distance into the target, the center
of this hemisphere actually lies within the original target
rock at a point below the original ground surface.
Within this hemispherical envelope, the shock waves that
travel upward and intersect the original ground surface are
reflected downward as rarefactions (release waves). In a nearsurface
region where the stresses in the tensional release
wave exceed the mechanical strength of the target rocks,
the release wave is accompanied by fracturing and shattering
of the target rock (Fig. 3.2). This reflection process also
converts some of the initial shock-wave energy to kinetic
energy, and the rock involved is accelerated outward, much
of it as individual fragments traveling at high velocities
(Fig. 3.4).
These complex processes drive the target rock outward
from the impact point, producing a symmetric excavation
flow around the center of the developing structure. Exact
flow directions vary with location within the target rocks
(Fig. 3.4). In the upper levels, target material moves dominantly
upward and outward. At lower levels, target material
moves dominantly downward and outward. These movements
quickly produce a bowl-shaped depression (the transient
cavity or transient crater) in the target rocks (Maxwell,
1977; Grieve at al., 1977; Grieve and Cintala, 1981; Melosh,
1989, pp. 74–78).
The transient crater is divided into approximately equal
upper and lower zones (Figs. 3.4 and 3.5). Within the upper
ejection zone, velocities imparted to the target rocks may be
as high as several kilometers per second, high enough to excavate
the fragmented material and to eject it beyond the
rim of the final crater (Grieve et al., 1977; Dence et al., 1977;
Fig. 3.3. Development of a simple impact structure. Series of cross-section diagrams showing progressive development of a small,
bowl-shaped simple impact structure in a horizontally layered target: (a) contact/compression stage: initial penetration of projectile,
outward radiation of shock waves; (b) start of excavation stage: continued expansion of shock wave into target; development of tensional
wave (rarefaction or release wave) behind shock wave as the near-surface part of original shock wave is reflected downward from ground
surface; interaction of rarefaction wave with ground surface to accelerate near-surface material upward and outward; (c) middle of
excavation stage: continued expansion of shock wave and rarefaction wave; development of melt lining in expanding transient cavity;
well-developed outward ejecta flow (ejecta curtain) from the opening crater; (d) end of excavation stage: transient cavity reaches maximum
extent to form melt-lined transient crater; near-surface ejecta curtain reaches maximum extent, and uplifted crater rim develops; (e) start
of modification stage: oversteepened walls of transient crater collapse back into cavity, accompanied by near-crater ejecta, to form
deposit of mixed breccia (breccia lens) within crater; (f ) final simple crater: a bowl-shaped depression, partially filled with complex
breccias and bodies of impact melt. Times involved are a few seconds to form the transient crater (a)–(d), and minutes to hours for the
final crater (e)–(f ). Subsequent changes reflect the normal geological processes of erosion and infilling.
Kieffer and Simonds, 1980; Melosh, 1989, pp. 74–76). Even
at significant distances from the impact point, shock pressures
and the resulting ejection velocities remain high enough
(>100 m/s) to eject material. For this reason, the diameter of
the final crater is many times larger (typically 20–30×) than
the diameter of the projectile itself.
At deeper levels, tensional stresses in the release waves
are lower. As a result, fracturing is less pronounced, excavation
flow velocities are lower, and the excavation flow lines
themselves are not oriented to eject material beyond the crater
rim (Fig. 3.4). This region forms a displaced zone in
which material is driven downward and outward more or
less coherently.
Both zones in the transient crater continue to expand,
accompanied by the uplift of near-surface rocks to form the
transient crater rim, as long as the expanding shock waves
and release waves are strong enough to eject or displace material
from the developing cavity. However, these waves continually
lose energy by deforming and ejecting the target rocks
through which they pass. Eventually, a point is reached at
which the shock and release waves can no longer excavate or
displace target rock. At that point the growth of the transient
crater ceases. There is an instant of theoretical balance
in which the energies of the shock wave no longer act, and
the waiting forces of gravity and rock mechanics have not
yet reasserted themselves. At this instant, the transient crater
reaches its maximum extent, the excavation stage ends,
and the subsequent modification stage begins immediately.
The excavation stage, although longer than the contact/
compression stage, is still brief by geological standards. If
the near-surface excavation flow has a minimum average velocity
of 1 km/s, then a 200-km-diameter transient crater
can be excavated in less than 2 min. More detailed calculations
(Melosh, 1989, p. 123) indicate that excavation of a
l-km-diameter crater (e.g., Barringer Meteor Crater [Arizona])
will occur in about 6 s, while a 200-km-diameter
crater requires only about 90 s.
The concept of the transient crater has been developed
from a combination of theoretical studies (Melosh, 1989,
Chapter 5) and geological investigations (Dence, 1968; Grieve
and Cintala, 1981; Grieve et al., 1981). The ideal transient
crater is a bowl-shaped depression with a structurally uplifted
rim (Figs. 3.4 and 3.5). Its shape is approximately
hemispherical but is actually a paraboloid of revolution
Formation of Impact Craters 21
􀀀􀀀􀀀
􀀀􀀀􀀀
􀀀􀀀􀀀
22 Traces of Catastrophe
􀀀􀀀
􀀀􀀀􀀀
􀀀􀀀􀀀
􀀀􀀀
Fig. 3.4. Excavation stage: formation of transient crater. Theoretical cross section showing development of the transient crater
immediately after the contact/compression stage. Original peak shock pressures (units in GPa) around the impact point are shown for
simplicity as hemispherical isobars (for details, see Fig. 3.2). Complex interactions of the shock wave, the ground surface, and the
subsequent rarefaction wave produce an outward excavation flow (dashed arrows) that opens up the transient crater. In the upper part of
this region (excavated zone; ruled area), target material is fractured, excavated, and ejected beyond the transient crater rim. In the lower
region (displaced zone), target material is driven downward and outward, more or less coherently, and does not reach the surface. This
model yields two important geological results: (1) ejected material is derived only from the upper part (approximately the top one-third
to one-half ) of the transient cavity; (2) because the excavation flow lines in the excavated zone cut across the initially hemispherical shock
isobars, ejected material will reflect a wide range of original shock pressures and deformation effects, ranging from simple fracturing to
complete melting and vaporization. (Modified from Grieve, 1987, Fig. 5; Hörz et al., 1991, Fig. 4.3a, p. 67.)
Fig. 3.5. Transient crater: locations of shock-metamorphosed materials. Cross section through a theoretical transient crater, showing
discrete zones from which various shock-metamorphosed materials are derived. The “vaporized” zone closest to the original impact point
(stippled) contains a mixture of vaporized target rock and projectile, which expands upward and outward into the atmosphere as a vapor
plume. The adjacent “melt” zone (solid black) consists of melt that moves downward and then outward along the floor of the final
transient cavity (for details, see Fig. 6.2). Material in the upper “ejected” zones on either side of the melt zone, which contains a range of
shock-metamorphic effects, is ejected outward to and beyond the transient crater rim. The lower “displaced” zone moves downward and
outward to form the zone of parautochthonous rocks below the floor of the final transient crater. Hat = the final transient crater depth;
Hexc = the depth of excavation, which is significantly less than the total depth. (From Melosh, 1989, Fig. 5.13, p. 78.)
􀀀􀀀
􀀀􀀀
Formation of Impact Craters 23
(Dence, 1973). Its maximum depth is approximately onethird
its diameter, and this proportion seems to remain
approximately constant for craters of widely different size
(Maxwell, 1977; Croft, 1985).
The theoretical instant of ideal overall balance in a transient
crater at the end of the excavation stage may not be
actually attained during formation of a real crater. For example,
in these models, the maximum diameter is normally
attained after the maximum depth is reached. Subsequent
modification of one part of an actual transient crater might
therefore begin while other parts are still being excavated.
Even so, the transient crater is a key concept in models of
crater formation. All impact structures, regardless of their
final size or the complexity of their subsequent development,
are assumed to pass through the transient-crater stage, making
this stage of critical importance in comparing impact
structures of different sizes or on different planets. Defining
the transient crater is also an essential step in determining
critical characteristics of an impact structure: its original
(pre-erosion) diameter and depth, the energy of impact, the
size and velocity of the projectile, the distribution of shock
pressures and shock effects within the crater, the amount of
material melted and ejected during formation of the crater,
the amount of structural uplift during formation of the central
peak of complex impact structures, and the depth from
which excavated materials were derived.
3.1.3. Modification Stage
The excavation stage ends when the transient crater has
grown to its maximum size, and the subsequent modification
stage begins immediately. The expanding shock waves
have now decayed to low-pressure elastic stress waves beyond
the crater rim, and they play no further part in the
crater development. Instead, the transient crater is immediately
modified by more conventional factors like gravity and
rock mechanics.
The immediate part of the modification stage, during
which the major impact-related changes occur, lasts only
slightly longer than the excavation stage: less than a minute
for a small structure, a few minutes for a large one (Melosh,
1989, Chapter 8, pp. 141–142). (One simple definition is
that the modification stage ends “when things stop falling.”)
However, the modification stage has no clearly marked end,
and the modification processes of uplift and collapse merge
gradually into the normal processes of geological mass movement,
isostatic uplift, erosion, and sedimentation.
3.2. SIMPLE AND COMPLEX
IMPACT STRUCTURES
The extent to which the transient crater is altered during
the modification stage depends on its size and (to a lesser
extent) on the structure and properties of the target rock.
Small transient craters are altered chiefly by the collapse of
their upper walls, and the shape of the final crater is little
changed from that of the original transient crater. In larger
structures, modification may involve major structural
changes: uplift of the central part of the floor and major
peripheral collapse around the rim. Depending on the extent
to which the transient crater is modified, three distinct
types of impact structures can be formed: simple craters,
complex craters, and multiring basins.
3.2.1. Simple Craters
The smallest impact structures occur as bowl-shaped depressions
(simple craters) less than a few kilometers across,
which help to preserve the shape and dimensions of the original
transient cavity (Figs. 1.1 and 3.6). In evolving to a simple
crater, the transient crater is modified only by minor collapse
of the steep upper walls into the crater cavity and by
redeposition of a minor amount of ejected material in the
crater. As a result, the crater diameter may increase by as
much as 20%, but the original transient crater depth remains
largely unaffected (Fig. 3.7) (Melosh, 1989, p. 129).
During modification, the simple crater is immediately
filled, to perhaps half its original depth, by a mixture of redeposited
(fallback) ejecta and debris slumped in from the
walls and rim (Fig. 3.7). This crater-filling unit, variously
called the breccia lens or crater-fill breccia, is a mixture of
rock fragments, both shocked and unshocked, together with
fragments or lenses of shock-melted rock (impact melt).
Fig. 3.6. A simple lunar impact crater. This small, well-preserved
crater (Moltke: D = 7 km) shows features typical of simple impact
craters: a circular outline, a bowl-like shape, an uplifted rim, and
hummocky deposits of ejecta around the rim. In the relatively low
gravity of the Moon, this structure formed as a simple crater; a
terrestrial structure of the same diameter, formed under Earth’s
higher gravity, would have formed as a complex crater with a central
uplift. (Apollo 10 image AS10-29-4324.)
24 Traces of Catastrophe
Fig. 3.7. Simple impact structure: locations of impactite types. Schematic cross section of a typical simple impact structure, showing
the simple bowl shape and the locations of various types of impactites in and around the structure. The parautochthonous rocks below
the true crater floor are fractured and brecciated but generally show no distinctive shock effects, except in a small zone (fine vertical
ruling) in the center of the structure. The crater is filled, to approximately half its original height, with a variety of allogenic breccias and
impact melts, which forms the crater-fill units or the breccia lens. A thinner layer of ejected material (fallout ejecta) overlies the uplifted
crater rim and surrounds the crater. This unit is easily eroded and is present only in the youngest and best-preserved structures. D = final
crater diameter, which is 10–20% greater than the diameter of the original, premodification transient crater; dt = true depth of the final
crater, which is approximately the depth of the original transient crater; da = apparent depth of the crater, or the depth from the final rim
to the top of the crater-fill units. The diagram represents the state of the final crater before any subsequent geological effects, e.g., erosion,
infilling. The model is based on drilling studies at Barringer Meteor Crater (Arizona) (Roddy et al., 1975; Roddy, 1978), Brent Crater
(Canada) (Dence, 1968; Grieve and Cintala, 1981), and similar structures (e.g., Masaitis et al., 1980; Gurov and Gurova, 1991). (From
Grieve, 1987, Fig. 1.)
Depending on the subsequent geological history, the breccia
lens may be eroded or may be covered and preserved by a
cap of later sedimentary fill.
3.2.2. Complex Craters
The bowl-shaped form of simple craters appears only in
relatively small structures less than a few kilometers across.
Larger impact structures (complex craters) display a different
and more complicated form, characterized by a centrally
uplifted region, a generally flat floor, and extensive inward
collapse around the rim (Figs. 1.3, 3.8, and 3.9) (Dence, 1968;
Grieve et al., 1977, 1981; Grieve, 1991). For terrestrial structures,
the transition between simple and complex craters
occurs at a diameter of about 4 km in massive crystalline
rocks, but at only about 2 km in sediments. (However, these
values apply only to Earth. The transition diameter varies
inversely with gravitational acceleration, and it is different
on different planets.) The larger impact events that form
complex craters apparently release enough energy to overcome
the fundamental strength of the target rocks over a
large volume beneath the large transient crater. As a result,
late-stage modification involves complex interactions between
shock-wave effects, gravity, and the strength and structure
of the target rocks, and the modification is characterized
by outward, inward, and upward movements of large volumes
of the subcrater rocks.
The details of these interactions are uncertain, but the
general result is that the original bowl-shaped transient crater
is immediately modified as deep-seated rocks beneath
the center of the transient crater rise to form a central uplift
(Dence, 1968; Grieve et al., 1981). At the same time, rocks
around the periphery of the transient crater collapse downward
and inward along concentric faults to form one or more
depressed rings (ring grabens) and a series of terraces along
the outer margins of the final structure (Fig. 3.10). [A simple
model of the formation of a complex crater and its central
uplift is presented by the familiar slow-motion movies of a
drop of liquid hitting a liquid surface (e.g., Melosh, 1989,
p. 148; Taylor, 1992, p. 168). There is the same initial cavity
formation, the same outward and downward ejection of target
material, the same upward rebound of the central cavity
floor, and the same collapse of the periphery back into the
cavity. However, in impact events, these processes take place
in solid rock and may operate over distances of tens to hundreds
of kilometers.]
The idea that such rapid deformation and subsequent
uplift can occur in large volumes of crustal rocks has been
difficult for many geologists to appreciate. Key evidence has
come from studies of impact structures formed in sedimentary
rocks, in which the actual uplift of key stratigraphic
markers has been established beyond question through drilling
and geophysical studies (e.g., Milton et al., 1972, 1996a,b;
Formation of Impact Craters 25
Grieve et al., 1981; Grieve and Pilkington, 1996). Geological
studies have also established that the amount of actual stratigraphic
uplift (SU) in impact structures is about one-tenth
the final diameter (D) of the structure. A detailed statistical
relation derived from studies of well-constrained complex
impact structures (Grieve et al., 1981, p. 44) is SU = 0.06 D1.1
(both SU and D are in kilometers). A subsequent analysis,
using more craters (Grieve and Pilkington, 1996, p. 404),
gave SU = 0.086 D1.03. The two equations are virtually identical,
and a value of SU = 0.1 D is a reasonable approximation
to either. For large (D = 100–200 km) impact structures,
these relations imply that the crustal rocks beneath the structure
are uplifted vertically by 10–20 km during the impact
event. An uplift of this magnitude has been estimated for
the Vredefort (South Africa) structure on geological grounds
(Reimold and Gibson, 1996; Therriault et al., 1997; Turtle and
Pierazzo, 1998).
Both theoretical and field studies indicate that central
uplifts form in only a few minutes, almost instantaneously
by geological standards, even in the largest structures (Melosh,
1989, pp. 129 and 141–142). Theoretical studies also suggest
that the central uplifts of structures 200–300 km in
Fig. 3.8. A complex lunar crater. This relatively young crater
(Theophilus: D = 100 km) displays well-preserved features that
are typical of complex impact structures: a central uplift, a scalloped
circular outline, ruggedly terraced walls with possible landslide
deposits inside the rim, and hummocky ejecta deposits just outside
the rim. This view also indicates the continuing nature of
lunar cratering; an older impact crater (upper right) has been partly
destroyed by Theophilus, while a younger small crater has formed
within Theophilus itself (near rim, lower right). The flat dark
area in the background (upper left) is made up of lava flows covering
part of Mare Nectaris. The spiral-like rod at left center is an
instrument boom on the Apollo 16 spacecraft, from which this
orbital picture was taken. (Apollo 16 image AS16-M-0692.)
Fig. 3.9. A complex impact basin on Venus. A large, wellpreserved
multiring impact basin on the surface of Venus
(Meitner: D = 150 km) is revealed beneath the planet’s opaque
atmosphere by the imaging radar system of the Magellan spacecraft.
Meitner, the third-largest impact structure identified on
Venus, shows a flat smooth (dark-colored) interior, two circular
rings, and a rough, irregular blanket of lobate ejecta (light-colored).
The crater was formed on a surface of smooth plains, possibly
underlain by lava flows and cut by abundant parallel fractures
(white lines). (Magellan image F-MIDRP .55S319;201.)
diameter, such as Vredefort (South Africa), formed in less
than 15 minutes (Melosh, 1989, pp. 141–142; Turtle and
Pierazzo, 1998).
Despite the extensive evidence that central uplifts do form
in large impact structures, the details of the process are still
the subject of continuing uncertainty and active debate
(Dence, 1968; Grieve et al., 1981; Melosh, 1989, Chapter 8;
Hörz et al., 1991; Spudis, 1993). Even so fundamental a quantity
as the ratio between the diameter of the initial transient
crater and the diameter of the final complex impact structure
has not been well established; values estimated by various
workers, using both theoretical and geological studies,
range from about 0.5 to 0.7 (see, e.g., Therriault et al., 1997,
Table 2).
At larger crater diameters, the resulting structures, and
especially the centrally uplifted area, become even more complicated.
As the crater size increases the character of the central
uplift changes, and the single central peak is progressively
replaced by a more complex series of concentric rings and
basins. At least three types of complex impact structures can
be distinguished with increasing crater diameter: centralpeak
structures, central-peak-basin structures, and peak26
Traces of Catastrophe
Fig. 3.10. Development of a complex impact structure. Series of cross sections showing progressive development of a large, complex
impact structure in a horizontally layered target: (a) formation of a large transient crater by the excavation process is virtually identical to
transient crater formation in smaller structures (compare with Fig. 3.3a–d); (b) initial development of central uplift during the subsequent
modification stage; (c) start of peripheral collapse, accompanied by continuing development of the central uplift and the thinning and
draping of the original melt layer (black) over the uplifted rocks; (d) final structure, which is of the central-uplift type, consists of a central
uplift of deeper rocks, surrounded by a relatively flat plain and by a terraced rim produced by inward movement along stepped normal
faults. The central uplift is surrounded by an annular deposit of allogenic breccias and impact melt (black), which may be absent from the
central peak itself. An ejecta layer (stippled) covers the target rocks around the structure. The diameter of the final structure, measured at
the outer rim beyond the outermost fault, may be 1.5–2× the diameter of the original transient crater. This central-peak morphology is
observed in terrestrial structures ranging from about 2–25 km in diameter; larger structures tend to develop one or more concentric rings
within the crater (for details, see text).
Formation of Impact Craters 27
ring basin structures (Grieve at al., 1981; Melosh, 1989,
Chapter 8; Spudis, 1993). As the terms suggest, these structures
are characterized by the initial development of a basin
in the central peak and eventually by the complete conversion
of the central peak area to a ring structure (Figs. 1.3,
3.9, and 3.11).
These distinctions, and the transition diameters at which
they occur, have been most clearly established on airless bodies
like the Moon, where even large ancient structures have
been well preserved (Figs. 3.6, 3.8, and 3.11) (e.g., Taylor,
1982, 1992; Melosh, 1989, pp. 131–135; Spudis, 1993). Classification
of large terrestrial structures (e.g., papers in Schultz
and Merrill, 1981; Spudis, 1993, pp. 24–41) is more difficult
and uncertain, because the impact structures, especially their
critical upper parts, tend to be removed by erosion or buried
by later sediments. Furthermore, the critical diameters at
which one form changes to another depend inversely on the
planetary gravity, making it difficult to apply data from structures
on other planets to terrestrial features. For example,
the transition between simple and complex craters occurs at
about 20 km diameter on the Moon but at only 2–4 km on
Fig. 3.11. A lunar impact basin. This large impact structure
(Schrödinger: D = 320 km) is located on the lunar farside near
the Moon’s South Pole. Although ancient and highly degraded,
it still preserves features distinctive of larger complex impact
structures: a central uplift and terraced walls. However, in this
large structure, the central uplift appears as an interior peak ring
about 150 km in diameter (arrows), in sharp contrast to the simpler
central peak formed in smaller complex structures. (Lunar Orbiter
image LO-IV-8M.)
Earth. The subsequent transition between a central-peakbasin
structure to a peak-ring structure occurs at about 150–
200 km on the Moon, but at only about 20–25 km on Earth.
Despite the various difficulties, it has been possible to
establish rough boundaries for different types of terrestrial
complex structures (Grieve et al., 1981, p. 42, Fig. 2). These
limits, and some typical examples, are: central-peak structures
(D = 4–22 km) [Steinheim (Germany), Sierra Madera
(Texas)]; central-peak-basin structures (D = 22–30 km)
[Mistastin (Canada)]; peak-ring-basin structures (D = 30–
62 km) [West Clearwater (Canada); Fig. 1.3]. These values
are only approximations, and they will almost certainly change
as more structures are studied in detail and as the formation
of complex craters is better understood.
3.2.3. Multiring Basins
The largest planetary impact structures so far identified
have diameters of a few hundred kilometers to more than
1000 km (e.g., papers in Schultz and Merrill, 1981; Melosh,
1989, Chapter 9; Spudis, 1993). In contrast to smaller impact
structures, they appear as huge geological bulls-eyes,
composed of multiple concentric uplifted rings and intervening
down-faulted valleys (ring grabens) (Fig. 3.12). These
features, designated multiring basins, are defined as structures
that have two or more interior rings in addition to the
outer rim of the structure.
Multiring impact basins have been produced by the impact
of projectiles tens to hundreds of kilometers in diameter,
and they date mainly from an early period in the solar
system (>3.9 Ga), when such large objects were more abundant
and collisions were more frequent. The best multiring
basins are best observed on planets with well-preserved ancient
surfaces, such as the Moon, Mercury, parts of Mars,
and some of the moons of Jupiter. Mare Orientale, on the
Moon, with a diameter of at least 900 km, is one of the most
prominent and best-known multiring basins (Fig. 3.12),
but even larger features exist, such as the Valhalla Basin
(D ~4000 km) on Jupiter’s icy moon Callisto. In addition,
there are numerous large basins in the solar system that
do not display a pronounced multiring structure, possibly
because they have been deeply eroded since they formed.
These include the Caloris Basin (Mercury; D = 1300 km),
the Argyre Basin (Mars; D > 900 km) (Fig. 1.9), and the
recently identified South Pole-Aitken Basin on the Moon
(D ~2500 km).
On the Moon, the transition to multiring basins occurs
at diameters of about 400–600 km. Because the transition
diameters for different crater forms vary inversely with planetary
gravity, this observation implies that multiring basins
should begin to form on Earth at crater diameters greater
than about 100 km. Because the few terrestrial impact structures
in this size range have been deeply eroded or buried
(e.g., Fig. 1.4), it has not yet been possible to demonstrate
clearly that any multiring basins exist on the Earth. The
few possible candidates (and their current estimated diameters)
are Manicouagan (Canada, 100 km), Popigai (Russia,
100 km), Vredefort (South Africa, >200 km), Sudbury
28 Traces of Catastrophe
Fig. 3.12. A lunar multiring impact basin. One of the largest,
freshest, youngest, and best-known multiring impact basins in the
solar system, Mare Orientale (D = 930 km) lies on the boundary
between the Earth-facing lunar nearside (right) and the lunar
farside. The structure, formed at about 3.8 Ga, is bounded by an
outer ring about 930 km in diameter (Cordillera Mountains), and
inner rings with diameters of 620, 480, and 320 km can be
distinguished. Mare Orientale is surrounded by radial features
(especially at lower right) that may have been produced by the
low-angle ejection of large blocks of excavated material. The
postimpact history of the structure is also complex, and much of
the area inside the rings has been modified by later volcanic activity.
The flat dark areas at upper right are the younger lava flows that
cover Oceanus Procellarum. (Lunar Orbiter image LO-IV-187M.)
(Canada, >200 km), and Chicxulub (Mexico, >180 km). It
has not proved possible to establish beyond question the
multiring character of these structures for various reasons,
including deep erosion, postcrater deformation, or insufficient
geological study. The strongest current candidate for a
terrestrial multiring structure is Chicxulub, which, although
buried, appears well preserved (Sharpton et al., 1993, 1996b;
Morgan et al., 1997).
Multiring basins represent the most energetic and catastrophic
impact events in the solar system, and the postimpact
movements — upward, downward, and inward — of
the target rock that modify the transient crater are far more
complex and widespread than in smaller structures. It is therefore
not surprising that the formation of multiring basins is
even more uncertain and hotly debated than is the origin of
smaller complex impact structures (e.g., papers in Schultz
and Merrill, 1981; Melosh, 1989, Chapter 9; Spudis, 1993).
For example, it is not clear whether the transition between
smaller impact structures and multiring basins is a
natural development with increasing crater diameter (Herrick
et al., 1997), or whether multiring basins only form when
special conditions are present within the target, e.g., a crustmantle
structure with a weak layer (asthenosphere) at depth
within the planet (see Melosh, 1989, pp. 176–180). Nor is it
understood why some planetary features in the 1000–2000-
km-diameter range have a pronounced multiring form
(Fig. 3.12) and others do not (Fig. 1.9). Finally, it is not yet
established whether multiring impact structures — ancient
or modern — do exist on Earth and which large structures
they may be.
3.3. SUBSEQUENT DEVELOPMENT OF
IMPACT STRUCTURES
When the crater formation process ends, the resulting
circular structure, whether simple or complex, consists of
deformed subcrater rocks covered by an ejecta blanket outside
the crater and with crater-fill deposits (usually a mixture
of breccias and bodies of impact melt) within it (Figs. 3.7
and 3.13). This assemblage of distinctive near-surface rocks
is immediately subject to more normal geological processes:
erosion, burial, and tectonic deformation. If the crater
forms on land and remains exposed after formation,
erosion will quickly remove the surface ejecta blanket and
destroy any surviving meteorite fragments. At the same time,
however, a lake may form in the crater depression, covering
the crater-fill material with a preserving cap of sediments,
e.g., as at Brent (Canada) (Dence, 1968; Grieve, 1978) and
the Ries Crater (Germany) (von Engelhardt, 1990).
If the original impact site is covered by water, the formation
and subsequent history of the resulting crater may be
more complex. At the moment of impact, the overlying layer
of water will be excavated with the underlying bedrock, and
the development of the crater and the deposition of the impact-
produced rock units will be modified by the immediate
and violent resurge of this displaced water back into the crater
cavity (Therriault and Lindström, 1995; Lindström et al.,
1996). If the crater remains below the water level, it will
immediately begin to fill with sediments, and its subsequent
history will depend on whether it remains below water level
(continuous sediment filling) or is uplifted at some future
time (beginning of erosion). A number of such submarine
impact structures have now been recognized; some have subsequently
been raised above sea level [e.g., Lockne (Sweden)
(Therriault and Lindström, 1995; Lindström et al., 1996)]
and others still remain buried [e.g., Montagnais (Canada)
(Jansa and Pe-Piper, 1987); the Chesapeake Bay Crater
(USA) (Poag, 1996, 1997); and the recently discovered
Mjølnir structure (Norway) in the Barents Sea (Dypvik et
al., 1996)].
Formation of Impact Craters 29
Fig. 3.13. Complex impact structure: locations of impactite types. Schematic radial cross section across a complex impact structure of
the central-uplift type, from the central uplift (right) to the outer, downfaulted rim (left). (Vertical scale is exaggerated.) The subcrater
parautochthonous rocks, exposed in the central uplift, are highly fractured and brecciated and may contain distinctive shock features such
as shatter cones. These rocks may also contain widespread pseudotachylite breccias and dike-like intrusive bodies of allogenic breccias
and impact melts. Larger and thicker subhorizontal units of allogenic breccias and melts occur as an annular unit of crater-fill material
that covers the parautochthonous rocks between the central uplift and the rim. The bulk of these crater-fill deposits consist of melt-free
lithic breccias, with lesser amounts of melt-bearing suevite breccias. The melt component in the crater-fill deposits becomes more
abundant toward the center and upward, and a discrete layer of impact melt (solid black) may occur at or toward the top of the crater fill.
(Modified from Stöffler et al., 1988, Fig. 12, p. 290.)
Because impact is a near-surface process, the deformation
associated with impact structures dies away rapidly
with depth. Typical impact structures are relatively shallow,
and impact-produced rocks form comparatively thin units.
The distinctive rock types and shock effects in a structure
tens of kilometers in diameter may extend only a few kilometers
below the original ground surface. Impact structures
are therefore especially vulnerable to erosion. Initial erosion
will preferentially remove the near-surface ejecta deposits
and the distinctively shocked and melted materials they contain,
thus rapidly destroying the most convincing evidence
for impact. Deeper erosion over longer periods of time will
eventually produce major destructive changes in the crater.
The breccias and melt units that fill the crater, and the distinctive
shocked materials they contain, together with any
protecting cap of sediments, will be reduced to small remnants
or completely removed. The original circular outline
will disappear. Eventually, all trace of the crater will be removed
except for the weakly shocked subcrater rocks. If
erosion continues long enough, the whole impact structure
will be erased.
Impact structures that are not destroyed by erosion may
be entirely filled and buried by younger sediments, so that
their detection depends on geophysical methods and drilling
rather than on surface field geology. About one-third of
the presently known impact structures are subsurface (Grieve,
1991, 1997; Grieve and Masaitis, 1994; Grieve et al., 1995);
they were first discovered during geophysical explorations,
and their impact origin has been verified by the discovery of
shocked rocks in drill core samples. This group includes several
continental structures that are actual or potential petroleum
producers [Ames (Oklahoma); Avak (Alaska); Marquez
(Texas); Red Wing Creek (North Dakota)] (Donofrio, 1997),
as well as a few submarine impact structures [e.g., Montagnais
(Canada) (Jansa and Pe-Piper, 1987)]. Several large and relatively
young buried impact structures have also been identified
by geophysical techniques: the 90-km-diameter
Chesapeake Bay Crater (USA) (Poag, 1996, 1997); the larger
(>180-km diameter) Chicxulub structure (Mexico), which
is associated with the K/T event (Hildebrand et al., 1991;
Sharpton et al., 1992; papers in Ryder et al., 1996); and the
large (>70 km?) Morokweng structure (South Africa) (Corner
et al., 1997; Koeberl et al., 1997a). Many more impact
structures remain to be found, and the evidence for their
existence may already be sitting unrecognized in existing
drill cores and geophysical records around the world.
Impact structures may also be caught up in subsequent
tectonic deformation, with varying results. Horizontal compression
may deform the original circular shape, making study
and interpretation more difficult [as at Sudbury (Canada)].
30 Traces of Catastrophe
Tectonism can also break up regions of original shocked rocks
and disperse them as large discrete areas across the geological
landscape [e.g., the Beaverhead (Idaho) structure
(Hargraves et al., 1990; Fiske et al., 1994)]. Sufficient tectonism
and metamorphism could destroy even large impact
structures or make them totally unrecognizable.
Geologists must therefore be prepared to recognize impact
structures in all states of preservation, from young, fresh,
well-exposed circular structures filled with distinctive shocked
breccias to older features in which distinctive shock effects
are scattered, barely recognizable, or deeply buried. It is essential
to be able to recognize the variety of distinctive shock
effects associated with impact structures and to understand
where different types of shock effects may be located in the
original crater.
« Last Edit: January 29, 2017, 05:55:46 am by Admin »

Admin

  • Administrator
  • Hero Member
  • *****
  • Posts: 555
    • View Profile
Re: Traces of Cat.
« Reply #2 on: January 29, 2017, 06:00:55 am »
4.1. FORMATION CONDITIONS AND
GENERAL CHARACTERISTICS
The growing recognition since the 1960s of the geological
importance of meteorite impact events, and the large
number of impact structures still preserved on Earth, is largely
the result of two related discoveries: (1) The extreme physical
conditions that are imposed by intense shock waves on
the rocks through which they pass produce unique, recognizable,
and durable shock-metamorphic effects; (2) such
shock waves are produced naturally only by the hypervelocity
impacts of extraterrestrial objects (French, 1968a, 1990b;
French and Short, 1968). Shock-metamorphic effects (also
called “shock effects” or “shock features”) have been critical
to the identification of terrestrial impact structures because
of their uniqueness, wide distribution, ease of identification,
and especially their ability to survive over long periods of
geologic time.
With the acceptance of shock effects as a criterion for
impact, the record of terrestrial impact events is no longer
limited to small young structures that still preserve definite
meteorite fragments. Equally convincing evidence for impact
can now be provided by a wide variety of distinctive
deformation effects in the rocks themselves, and it has become
possible to identify numerous old impact structures
from which weathering and erosion have removed all physical
traces of the projectiles that formed them. The recognition
of preserved shock effects has been the main factor
behind the steady increase in the number of recognized impact
structures since the 1960s (Grieve, 1991; Grieve et al.,
1995; Grieve and Pesonen, 1992, 1996; for historical reviews,
see Hoyt, 1987; Mark, 1987).
The approximate physical conditions that produce shockdeformation
effects in natural rocks have been established
by a combination of theoretical studies, artificial explosions
(both chemical and nuclear), and experiments with laboratory
shock-wave devices (for details, see papers in French
and Short, 1968 and Roddy et al., 1977; also Stöffler, 1972;
Kieffer and Simonds, 1980; Melosh, 1989; Stöffler and
Langenhorst, 1994). Peak shock pressures produced in an
impact event range from >2 GPa ( >20 kbar) near the final
crater rim to >100 GPa (>1000 kbar) near the impact point.
These pressures, and the resulting shock-deformation effects,
reflect conditions that are far outside the range of normal
geological processes (Fig. 4.1, Table 4.1). In ordinary
geological environments, pressures equivalent to those of
typical shock waves are attained only under static conditions
at depths of 75–1000 km within Earth, well below the shallow-
crustal regions in which impact structures are formed.
Shock-wave pressures differ in other important ways from
pressures produced by more normal geological processes. The
application of shock-wave pressures is both sudden and brief.
A shock wave traveling at several kilometers per second will
traverse the volume of a mineral grain or a rock sample in
microseconds, and both the onset and release of pressure are
highly transient. Shock-deformation effects therefore reflect
transient stress conditions, high strain rates, and rapid
quenching that are inconsistent with the rates of normal
geological processes (Table 4.1). In addition, shock waves
deposit energy in the materials through which they pass. A
particular shock pressure will produce a specific postshock
temperature, which depends chiefly on the nature of the target
material. These postshock temperatures increase with increasing
shock pressure (see the P-T curve labeled “Shock
metamorphism” in Fig. 4.1). For large shock pressures, the
resulting temperatures are high enough to produce melting
and even vaporization within the target.
The unique conditions of shock-wave environments produce
unique effects in the affected rocks. The nature and
intensity of the changes depend on the shock pressures
Shock-Metamorphic Effects in
Rocks and Minerals
32 Traces of Catastrophe
Fig. 4.1. Conditions of shock-metamorphism. Pressure-temperature plot showing comparative conditions for shock metamorphism
and normal crustal metamorphism. [Note that the pressure axis (X-axis, in GPa) is logarithmic.] Shaded region at lower left (P < 5 GPa,
T < 1000°C) encloses the conventional facies (labeled) for crustal metamorphism. Shock-metamorphic conditions (at right) extend from
~7 to >100 GPa and are clearly distinct from normal metamorphic conditions. Approximate formation conditions for specific shock
effects (labeled) are indicated by vertical dashed lines, and the exponential curve (“Shock metamorphism”) indicates the approximate
postshock temperatures produced by specific shock pressures in granitic crystalline rocks. Relatively high shock pressures (>50 GPa)
produce extreme temperatures, accompanied by unique mineral decomposition reactions (at left, near temperature axis). Stability curves
for high-pressure minerals (coesite, diamond, stishovite) are shown for static equilibrium conditions; formation ranges under shock
conditions may vary widely. (Adapted from Stöffler, 1971, Fig. 1; Grieve, 1990, p. 72; Grieve and Pesonen, 1992, Fig. 9.)
TABLE 4.1. Shock metamorphism: Distinction from other geological processes.
Regional and Contact Metamorphism;
Characteristic Igneous Petrogenesis Shock Metamorphism
Geological setting Widespread horizontal and vertical regions Surface or near-surface regions of Earth’s crust
of Earth’s crust, typically to depths of 10–50 km
Pressures Typically <1–3 GPa 100–400 GPa near impact point; 10–60 GPa in large
volumes of surrounding rock
Temperatures Generally >1000°C Up to 10,000°C near impact point (vaporization);
typically from 500° to 3000°C in much of
surrounding rock
Strain rates 10–3/s to 10–6/s 104/s to 106/s
Time for completion From 105–107 yr “Instantaneous”: Shock-wave passage through 10-cm
of process distance, <10–5 s; formation of large (100-kmdiameter)
structure <1 hr
Reaction times Slow; minerals closely approach equilibrium Rapid; abundant quenching and preservation of
metastable minerals and glasses
Shock-Metamorphic Effects in Rocks and Minerals 33
(Table 4.2). Lower shock pressures (~2–10 GPa) produce
distinctive megascopic shatter cones in the target rocks
(Milton, 1977; Roddy and Davis, 1977). Higher pressures
(>10–45 GPa) produce distinctive high-pressure mineral
polymorphs as well as unusual microscopic deformation features
in such minerals as quartz and feldspar (Stöffler, 1972).
Even higher pressures (>50 GPa) produce partial to complete
melting and even vaporization (>100 GPa) of large
volumes of the target rocks.
An especially distinctive and convincing form of evidence
for meteorite impact is the suite of unique microscopic deformation
features produced within individual minerals by
higher-pressure (~10–45 GPa) shock waves. During the
impact event, such pressures develop in target rocks near the
center of the crater, and most of these rocks are immediately
broken up and incorporated into the excavation flow that is
being initiated by the expanding release wave (Figs. 3.4 and
3.5). As a result, these shock effects are found chiefly in individual
target rock fragments in the breccias that fill the
crater or in the ejecta deposited beyond the rim.
A wide variety of shock-produced microscopic deformation
features has been identified in the minerals of shockmetamorphosed
rocks (for reviews, see Chao, 1967; papers
in French and Short, 1968; Stöffler, 1972, 1974; Stöffler
and Langenhorst, 1994; Grieve et al., 1996). These include
(1) kink bands in micas and (more rarely) in olivine and
pyroxene; (2) several types of distinctive planar microstructures
and related deformation effects in quartz, feldspar, and
other minerals; (3) isotropic mineral glasses (diaplectic or
thetomorphic glasses) produced selectively, most commonly
from quartz and feldspar, without actual melting; (4) selective
melting of individual minerals. Kink bands, although
common in impact environments (Fig. 4.2), can also be produced
by normal tectonic deformation; they are not a unique
criterion for shock metamorphism, and they will not be discussed
further. The other effects, particularly the distinctive
planar microstructures in quartz and the diaplectic glasses,
are now generally accepted as unique criteria for shock waves
and meteorite impact events.
These shock-produced microscopic deformation features
have several distinctive general characteristics. They are
pervasive, and usually occur throughout a centimeter-sized
rock sample, although they may be more erratically developed
over larger distances (meters or tens of meters). They
TABLE 4.2. Shock pressures and effects.
Approximate Estimated
Shock Pressure Postshock Effects
(GPa) Temperature (°C)*
2–6 <100 Rock fracturing; breccia formation
Shatter cones
5–7 100 Mineral fracturing: (0001) and {1011}
in quartz
8–10 100 Basal Brazil twins (0001)
10 100*
Quartz with PDFs {1013}
12–15 150 Quartz ® stishovite
13 150 Graphite ® cubic diamond
20 170*
Quartz with PDFs {1012}, etc.
Quartz, feldspar with reduced refractive
indexes, lowered birefringence
>30 275 Quartz ® coesite
35 300
Diaplectic quartz, feldspar glasses
45 900 Normal (melted) feldspar glass (vesiculated)
60 >1500 Rock glasses, crystallized melt rocks (quenched
from liquids)
80–100 >2500 Rock glasses (condensed from vapor)
* For dense nonporous rocks. For porous rocks (e.g., sandstones), postshock temperatures = 700°C
(P = 10 GPa) and 1560°C (P = 20 GPa). Data from Stöffler (1984), Table 3; Melosh (1989),
Table 3.2; Stöffler and Langenhorst (1994), Table 8, p. 175.
34 Traces of Catastrophe
Fig. 4.2. Kink-banding; in biotite. Large biotite grain in basement granitic gneisses, northeast side of Sudbury structure (Canada),
showing two sets of kink-banding at high angles to original cleavage (horizontal). Associated quartz (upper and lower left) and feldspar
show no shock-deformation effects. Sample CSF-68-67 (cross-polarized light).
Fig. 4.3. Progressive shock metamorphism in sandstone (I). Unshocked Coconino Sandstone from the Barringer Meteor Crater
(Arizona) is composed of well-sorted quartz grains with minor carbonate cement and pore space. The quartz grains are rounded to
angular, clear, and undeformed; some grains display secondary overgrowths. (Black dots are bubbles in thin section mounting medium.)
Ejecta sample from rim of crater. Sample MCF-64-4 (plane-polarized light).
0.5 mm
0.5 mm
Shock-Metamorphic Effects in Rocks and Minerals 35
Fig. 4.4. Progressive shock metamorphism in sandstone (II). Moderately shocked Coconino Sandstone from the Barringer Meteor
Crater (Arizona). The quartz grains are highly fractured and show numerous sets of subparallel fractures along cleavage planes. The
original interstitial pore space has been collapsed and heated during passing of the shock wave, producing a filling of dark glass that
frequently contains coesite. Ejecta sample from ground surface outside crater. Sample MCF-65-15-4 (plane-polarized light).
Fig. 4.5. Progressive shock metamorphism in sandstone (III). Highly shocked, melted, and vesiculated Coconino Sandstone from the
Barringer Meteor Crater (Arizona). The original sandstone has been converted to a light, frothy, highly vesicular pumice-like material
composed dominantly of nearly pure silica glass (lechatelierite). The vesicular glass contains a few remnant quartz grains (e.g., upper
center, arrow) that are highly fractured and show development of distinctive PDFs in addition to the open cleavage cracks. Ejecta sample
from ground surface outside crater. Sample MCF-65-11-2 (plane-polarized light).
0.5 mm
0.5 mm
36 Traces of Catastrophe
are mineralogically selective; a given effect (e.g., isotropization)
will occur in grains of a single mineral (e.g., quartz or
feldspar), but not in grains of other minerals, even adjacent
ones. Shock metamorphism is also characterized by a progressive
destruction of original textures with increasing shock
pressure, a process that eventually leads to complete melting
or vaporization of the target rock (Figs. 4.3, 4.4, and 4.5).
4.2. STAGES OF SHOCK METAMORPHISM
The fact that different shock pressures produce a variety
of distinctive shock features (Table 4.2) has made it possible
to recognize different levels or stages of shock metamorphism
(Chao, 1967; Stöffler, 1966, 1971, 1984; von Engelhardt
and Stöffler, 1968; Stöffler and Langenhorst, 1994). These stages
are not equivalent to the different facies recognized in normal
metamorphism, because shock metamorphism is a rapid
and nonequilibrium process and many of the most distinctive
features produced by shock waves (e.g., high-pressure
minerals and diaplectic glasses) are metastable under normal
geological conditions. Nevertheless, key shock features
occur frequently and consistently in natural impact structures,
and the production of the same features in experimental
studies has made approximate pressure and temperature
calibrations possible. As a result, the stages of shock metamorphism
have become an important concept for field studies
of impact structures and for using certain features as approximate
shock-wave barometers.
Current classifications of shock-metamorphic stages are
based almost entirely on features developed in nonporous,
quartz-bearing, crystalline igneous and metamorphic rocks.
These lithologies are abundant in many of the impact structures
studied so far, and they develop a varied suite of shock
features over a wide range of shock pressures. Individual classifications
of shock-metamorphic stages in these rocks differ
in details, but the following summary of distinctive shock
features and their approximate shock pressures (based largely
on Stöffler, 1966, 1971, 1984; Stöffler and Langenhorst, 1994)
provides a useful classification based on field and petrographic
characteristics. [Other effects observed with increasing shock
pressure include decreases in refractive index and increasing
structural disorder (shock mosaicism) in mineral grains; for
details, see Stöffler, 1972, 1974; Stöffler and Langenhorst,
1994).] It should be remembered that estimated pressures
are only approximate, and that the formation of a given shock
effect will also reflect such individual factors as rock type,
grain size, and other structural features. The shock effects
observed, and the inferred stages of shock metamorphism,
will be different for other rock types, especially for carbonates,
basaltic rocks, and porous rocks of any type.
For nonporous crystalline rocks, the following stages have
been distinguished (see Table 4.2):
<2 GPa
Fracturing and brecciation, without development of
unique shock features (see Chapter 5).
>2 GPa to <30? GPa
Shatter cones. At lower pressures (2 to <10 GPa), occurring
without distinctive microscopic deformation features.
At higher pressures (10 to >30 GPa), shatter cones may also
contain distinctive microdeformation features.
~8 GPa to 25 GPa
Microscopic planar deformation features in individual
minerals, especially quartz and feldspar. It has been possible
to subdivide this zone on the basis of different fabrics of
deformation features in quartz (Robertson et al., 1968; Stöffler
and Langenhorst, 1994).
>25 GPa to 40 GPa
Transformation of individual minerals to amorphous
phases (diaplectic glasses) without melting. These glasses
are often accompanied by the formation of high-pressure
mineral polymorphs.
>35 GPa to 60 GPa
Selective partial melting of individual minerals, typically
feldspars. Increasing destruction of original textures.
>60 GPa to 100 GPa
Complete melting of all minerals to form a superheated
rock melt (see Chapter 6).
>100 GPa
Complete rock vaporization. No preserved materials
formed at this stage (e.g., by vaporization and subsequent
condensation to glassy materials) have been definitely identified
so far.
4.3. MEGASCOPIC SHOCK-DEFORMATION
FEATURES: SHATTER CONES
Shatter cones are the only distinctive and unique shockdeformation
feature that develops on a megascopic (hand
specimen to outcrop) scale. Most accepted shock-metamorphic
features are microscopic deformations produced at relatively
high shock pressures (>10 GPa). Lower shock pressures
(1–5 GPa) produce a variety of unusual fractured and brecciated
rocks, but such rocks are so similar to rocks formed by
normal tectonic or volcanic processes that their presence cannot
be used as definite evidence for an impact event. However,
such low shock pressures also generate distinctive conical
fracturing patterns in the target rocks, and the resulting shatter
cones have proven to be a reliable field criterion for identifying
and studying impact structures (Dietz, 1947, 1959,
1963, 1968; Milton et al., 1972, 1996a; Roddy and Davis,
1977; Sharpton et al., 1996a; Dressler and Sharpton, 1997).
Shatter cones are distinctive curved, striated fractures that
typically form partial to complete cones (Figs. 4.6 and 4.7).
They are generally found in place in the rocks below the
crater floor, usually in the central uplifts of complex impact
structures, but they are also rarely observed in isolated rock
Shock-Metamorphic Effects in Rocks and Minerals 37
Fig. 4.6. Shatter cones; small, well-developed. Small, finely sculptured shatter cones, developed in fine-grained limestone from the
Haughton structure (Canada). The cone surfaces show the typical divergence of striae away from the cone apex (“horsetailing”). Photograph
courtesy of R. A. F. Grieve.
fragments in breccia units. Shatter cones occur as individuals
or composite groups, and individual cones may range from
millimeters to meters in length (Figs. 4.7, 4.8, and 4.9) (Dietz,
1968; Sharpton et al., 1996a). Far more common, however,
are partial cones or slightly curved surfaces with distinctive
radiating striations (“horsetailing”) on them (Fig. 4.10).
The details of shatter cone morphology are also distinctive.
Smaller secondary (“parasitic”) cones commonly occur
on the surfaces of both complete and partial shatter cones,
forming a unique composite or “nested” texture. The surfaces
of shatter cones, and the striations on them, are definite
positive/negative features. The striations are also
directional; they appear to branch and radiate along the surface
of the cone, forming a distinctive pattern in which the
acute angle of the intersection points toward the apex of the
cone (Figs. 4.6, 4.8, and 4.10).
Shatter cones form in all kinds of target rocks: sandstones,
shales, carbonates, and crystalline igneous and metamorphic
rocks. The most delicate and well-formed cones form in finegrained
rocks, especially carbonates (Fig. 4.6). In coarser
rocks, shatter cones are cruder, and their striations are larger,
making the cones more difficult to recognize and distinguish
from nonshock deformational features such as slickensides
(Figs. 4.8 and 4.10).
Shatter cones, especially well-formed examples, are
easy to distinguish from similar nonimpact features (see
Table 4.3). Some shatter cone occurrences may superficially
resemble the “cone-in-cone” structures produced during
38 Traces of Catastrophe
lithification of carbonate-bearing clastic sediments. However,
the cones in cone-in-cone features have their axes normal
to the bedding of the host rocks and their apexes pointing
down. Shatter cones generally point upward, and their axes
may lie at any angle to the original bedding, depending on
the preimpact orientation of the target rock and its location
relative to the impact point. Furthermore, the occurrence of
shatter cones in a variety of rock types, especially nonsedimentary
ones, is a good indication of an impact origin.
The horsetailing striations on shatter cone surfaces sometimes
resemble slickensides formed on faults, especially when
the surfaces are approximately flat (Figs. 4.8 and 4.10). However,
unlike slickensides, shatter cone striations are nonparallel
and often show strong radiation and directionality, so
that it is easy to determine the direction of the cone apex.
Shatter cones are now generally accepted as unique indicators
of shock pressures and meteorite impact. They are
especially valuable in this role because they form at relatively
low shock pressures (typically 2–10 GPa, but perhaps as
high as 30 GPa) and therefore develop throughout a large
volume of target rock below the crater floor. They are typically
widely and intensely developed in exposed central uplifts
of large structures. Shatter cones form in a wide range
of rock types, they are resistant to subsequent metamorphism,
and (when well developed) they can be easily and immediately
recognized in the field. Frequently, an initial discovery
of shatter cones has spurred the search for, and discovery of,
a range of equally definite shock effects produced at even
higher pressures.
For well-developed shatter cones, it is possible to measure
the orientation of the cone axes and to statistically determine
the varying orientations of shatter cones throughout
an impact structure. Such measurements (e.g., Manton,
1965; Guy-Bray et al., 1966; Milton et al., 1972, 1996a)
have provided strong support for the use of shatter cones
Fig 4.7. Shatter cones; large. Large shatter cone and crudely
conical striated surfaces in Mississagi Quartzite from the South
Range (Kelley Lake) of the Sudbury structure (Canada). Cone
axes point upward and into the Sudbury Basin (toward viewer) at
a high angle. Cone axes are nearly parallel to the original bedding
in the quartzite, which dips steeply back and to the right.
Fig. 4.8. Shatter cone; huge, wellstriated.
A large shatter cone, 2–
3 m long, in quartzite in the central
uplift of the Gosses Bluff structure
(Australia). The cone axis plunges
gently to the left, nearly normal to
the original bedding in the quartzite,
which appears as parallel joints dipping
steeply to the right. Despite the
crudeness of the large cone, the direction
of the apex (right), parasitic
cones, and distinctive horsetailing are
all visible. Scale rule (at top) is 15 cm
long.
Shock-Metamorphic Effects in Rocks and Minerals 39
Fig. 4.9. Shatter cone; huge. Unusually large shatter cone (megacone) (light-colored area, center) exposed in a cliff along a wave-cut
shoreline on Patterson Island, one of the islands in the Slate Islands impact structure, Lake Superior (Canada). The huge cone, developed
in Archean felsic metavolcanic rocks, points nearly straight up and is at least 10 m in length. At the exposed base, the exposed surface of
the cone is at least 7 m wide. Only ~25° of the cone’s basal perimeter is exposed, indicating that the true width of the feature may exceed
20 m at its base. Horsetail striations and parasitic cones cover all exposed surfaces. Several other large, conical features are obvious on the
near-vertical cliff, but because of the steep scree-covered slopes these features have not yet been examined in detail. Photograph courtesy
of V. L. Sharpton.
Fig. 4.10. Shatter cones; crude,
striated surfaces. Poorly developed
shatter cones in Serpent Quartzite,
Sudbury (Canada). The cones are only
partially developed, appearing as
curved and striated surfaces. Divergence
of the striae indicates that the
cone apexes are to the right. Pen (at
center) is 12 cm long.
40 Traces of Catastrophe
as a criterion for impact. In several impact structures that
formed in originally flat-lying sediments, the apexes of shatter
cones in the rocks point inward and upward when the rocks
are graphically restored to their original horizontal preimpact
position, indicating that the source of the shock wave
that produced the shatter cones was located above the original
ground surface (Guy-Bray et al., 1966; Dietz, 1968;
Manton, 1965; Howard and Offield, 1968; Wilshire et al., 1972;
Milton et al., 1972, 1996a). More recently, shatter cones in
the Beaverhead (Idaho) structure (Hargraves et al., 1990)
have been used to reconstruct the original shape and size
of a large, ancient impact structure that was subsequently
dissected and redistributed by major faulting during the
Laramide Orogeny.
The use of shatter cones to identify impact structures requires
caution, especially in cases where no other shock effects
can be identified. Poorly developed shatter cones
(Figs. 4.8 and 4.10) can be easily confused with normal fractures
and slickensides, and the latter may be misidentified
as shatter cones. Even in well-established impact structures,
shatter cones may be entirely absent or poorly developed, or
their orientations may be locally diverse and ambiguous
(Fig. 4.11). Detailed studies of shatter cone orientations need
to be done at more impact structures where they are well
developed, but such studies need to be done with care (see,
e.g., Manton, 1965; Milton et al., 1972, 1996a).
It is a paradox that, even though shatter cones are a proven
and valuable indicator of shock metamorphism and impact
structures, the exact mechanisms by which the radiating
TABLE 4.3. Shatter cones: Distinction from other geological features.
Cone-in-Cone Shatter Cones
Conical secondary growth features formed during Conical fracture features formed by transient shock waves (P ~2 to
diagenesis; found in undisturbed sedimentary rocks. >10 GPa) and found in meteorite impact structures, typically in uplifted
central rocks.
Restricted to carbonate-bearing rocks (limestones, Found in all rock types (sedimentary, igneous, metamorphic). Best
limy shales); associated with secondary carbonate. developed in fine-grained rocks, especially limestones.
Cone axes normal to bedding planes. Cone axes oriented at any angle to bedding, depending on orientation of
rock at time of impact and on postimpact movements.
Cones oriented point-down. Cones originally form pointing in direction of source of shock wave, i.e.,
inward and upward. Orientation varies over structure. Orientation further
modified by development of central uplift or later postcrater deformation.
When beds restored to original horizontal position, cones point toward a
focus above original surface, indicating external source of shock wave.
Striations along cone surface generally continuous, Striations along cone surface typically show development of divergent
uniform. radiations (“horsetailing”) along surface. Development of secondary
(parasitic) cones on main cone is typical.
Cone surfaces are growth surfaces against other cones Cone surfaces are actual fracture surfaces; rock splits into new shatteror
fine matrix in rock. coned surfaces along cone boundaries. Unlike slickensides, striated cone
surfaces show no relative motion, fit together without displacement.
Rocks typically show no deformation, metamorphism. Frequently contain kink-banded micas or quartz (coarser grains) with
shock-produced planar deformation features (PDFs).
shock wave interacts with the target rock to generate shatter
cones have not been studied in great detail and are still not
understood (e.g., Dietz, 1968; Gash, 1971; Milton, 1977;
Sharpton et al., 1996a). A further complication in shatter
cone formation is the evidence that, although the cones themselves
form at relatively low shock pressures, localized melting
and glass formation can occur along the cone surfaces,
probably as the result of a complex combination of shock
and frictional mechanisms (Gay, 1976; Gay et al., 1978;
Gibson and Spray, 1998). Combined theoretical, experimental,
and field studies to understand the exact conditions of
shatter cone formation are a major challenge for the future.
4.4. HIGH-PRESSURE MINERAL
POLYMORPHS
When subjected to impact-produced shock waves, some
minerals in target rocks (e.g., quartz, graphite) may transform
to high-pressure minerals, just as they do under high
static pressures produced in laboratory experiments or deep
in Earth’s crust. Graphite (C) can be converted to diamond.
Quartz can be converted to stishovite at shock pressures
of >12–15 GPa and to coesite at >30 GPa (Stöffler and
Langenhorst, 1994). [These numbers illustrate one of the
many differences between shock processes and normal geological
deformation. Under conditions of static equilibrium,
where reaction rates are slower and kinetic factors less imShock-
Metamorphic Effects in Rocks and Minerals 41
Fig. 4.11. Shatter cones; small, diversely oriented. This specimen shows a group of small, well-developed shatter cones, formed in a
sample of Precambrian crystalline target rock at the Slate Islands structure (Canada). The cones show two distinct orientations, and cone
axes appear to diverge above and below the coin. This type of diverse orientation may reflect small-scale nonuniformities in the shock
waves, produced by local heterogeneities (bedding planes, joints, etc.) in the rock sample. Coin is about 2 cm in diameter. Photograph
courtesy of V. L. Sharpton.
Fig. 4.12. Diaplectic quartz glass; with coesite. Diaplectic quartz glass (clear), with strings of small, high-relief crystals of coesite (“C”).
From biotite granite inclusion in suevite breccia, Aufhausen, Ries Crater (Germany). Photograph courtesy of W. von Engelhardt (planepolarized
light).
0.1 mm
42 Traces of Catastrophe
portant, coesite forms from quartz at lower pressures
(>2 GPa) than does stishovite (10–15 GPa).]
The identification of coesite and stishovite at several sites
in the early 1960s provided one of the earliest criteria for
establishing the impact origin of several structures, most
notably the Ries Crater (Germany) (Chao et al., 1960;
Shoemaker and Chao, 1961) (Fig. 4.12). Most subsequent
identifications of impact structures have been based on
shock-produced planar deformation features (PDFs) in
quartz, which are more widely distributed and simpler to
identify. However, the discovery of both coesite and stishovite
in the ancient Vredefort structure (South Africa) (Martini,
1991) was an important step in the growing acceptance of
this structure as an impact site. Diamond and other highpressure
carbon compounds [e.g., lonsdaleite (hexagonal diamond)]
produced from graphite in the shocked target rocks
have also been identified at an increasing number of impact
structures (Masaitis, 1998; Masaitis et al., 1972; Hough et al.,
1995; Koeberl et al., 1997c).
Coesite, stishovite, and diamond, when they are found in
near-surface rocks, are unique and reliable indicators of meteorite
impact. None of these minerals has been identified,
for example, as the result of explosive volcanic eruptions. The
use of coesite and diamond as impact criteria does require
some care, however, because both minerals also occur naturally
in deep-seated (depth >60 km) terrestrial rocks, where
they have formed in stable equilibrium at the high static pressures
(>2 GPa) present at these depths. Both minerals may
then be transported to Earth’s surface: coesite by tectonic
processes and diamond in fragments carried up by unusual
mafic (kimberlite) volcanic eruptions. However, stishovite,
formed only at pressures >10 GPa, has never been identified
in a nonimpact setting. Such static pressures could be produced
only at depths of 300–400 km within Earth. Furthermore,
the occurrence of such high-pressure minerals as
coesite, stishovite, or diamond in near-surface crustal rocks
[e.g., coesite and stishovite in sandstone at Barringer Meteor
Crater (Arizona)], particularly when they occur as a disequilibrium
assemblage with other chemically equivalent
minerals (e.g., coesite + stishovite + silica glass + quartz), is
definite evidence for meteorite impact.
4.5. PLANAR MICROSTRUCTURES
IN QUARTZ
Shock waves produce a variety of unusual microscopic
planar features in quartz, feldspar, and other minerals. These
features typically occur as sets of parallel deformation planes
within individual crystals. The recognition and interpretation
of these features, particularly those in quartz, as unique
products of meteorite impact has been a critical factor in
identifying most new impact structures, in recognizing the
impact origin of large, ancient, or deeply eroded structures,
and in demonstrating the role of meteorite impact in the
K/T extinction event.
Distinctive planar features in quartz (SiO2) have been one
of the most widely applied criteria for recognizing impact
structures (for reviews, details, and literature references, see
papers in French and Short, 1968; also von Engelhardt and
Bertsch, 1969; Stöffler and Langenhorst, 1994; Grieve et al.,
1996). Quartz is an ideal mineral for this purpose. It is abundant
in a wide range of sedimentary and crystalline rocks. It
is stable over long periods of geologic time, and it resists
change by alteration and metamorphism. It is an optically
simple (uniaxial) mineral to study and to analyze on the Universal
Stage (U-stage). In particular, it displays a variety of
different planar features whose development can be correlated
with shock pressure (Table 4.2) (Hörz, 1968; Robertson
et al., 1968; Stöffler and Langenhorst, 1994), and can thus be
used as a shock barometer to reconstruct the shock-pressure
distribution that existed within an impact structure during
the impact event (Robertson, 1975; Grieve and Robertson,
1976; Robertson and Grieve, 1977; Grieve et al., 1996; Dressler
et al., 1998).
The production and properties of planar microstructures
in quartz have been studied intensely since the early 1960s
by geological investigations, shock-wave experiments, and
both optical and electron microscopy (papers in French and
Short, 1968; also Stöffler and Langenhorst, 1994). It is now
recognized that shock waves produce several kinds of planar
microstructures in quartz, and their detailed characterization
and interpretation has been — and still is — an active
and much-debated problem (e.g., Alexopoulos et al., 1988;
Sharpton and Grieve, 1990). At present, two basic types of
planar features can be recognized, planar fractures and planar
deformation features (PDFs) (Table 4.2).
4.5.1. Planar Fractures
Planar fractures are parallel sets of multiple planar cracks
or cleavages in the quartz grain; they develop at the lowest
pressures characteristic of shock waves (~5–8 GPa)
(Figs. 4.13 and 4.14). The fractures are typically 5–10 μm
wide and spaced 15–20 μm or more apart in individual quartz
grains. Similar cleavage also occurs rarely in quartz from nonimpact
settings, and therefore planar fractures cannot be used
independently as a unique criterion for meteorite impact.
However, the development of intense, widespread, and closely
spaced planar fractures (Fig. 4.15) is strongly suggestive of
shock, and such fractures are frequently accompanied in
impact structures by other features clearly formed at higher
shock pressures (Robertson et al., 1968; Stöffler and
Langenhorst, 1994; Grieve et al., 1996; French et al., 1997).
4.5.2. Planar Deformation Features (PDFs)
Planar deformation features (PDFs) is the designation
currently used for the distinctive and long-studied shockproduced
microstructures that were formerly given a variety
of names (e.g., “planar features,” “shock lamellae”). In contrast
to planar fractures, with which they may occur, PDFs
are not open cracks. Instead, they occur as multiple sets of
closed, extremely narrow, parallel planar regions (Fig. 4.16).
Individual PDFs are both narrow (typically <2–3 μm) and
more closely spaced (typically 2–10 μm) than planar fractures
(Figs. 4.17 and 4.18). Detailed optical and TEM studies
have shown that, within individual PDFs, the atomic
Shock-Metamorphic Effects in Rocks and Minerals 43
Fig. 4.13. Quartz; cleavage and PDFs. High-magnification view of relict deformed quartz grain in highly shocked and vesiculated
Coconino Sandstone [Barringer Meteor Crater (Arizona)]. The quartz grain shows irregular, subparallel fractures (dark, near-vertical),
combined with shorter cross-cutting light-and-dark planar features, possibly PDFs (upper right/lower left). Note the irregular extinction
in the grain. Sample MCF-65-15-3 (cross-polarized light).
0.05 mm
Fig. 4.14. Quartz; cleavage. Quartz grain in moderately shocked Coconino Sandstone from Barringer Meteor Crater (Arizona), showing
irregular extinction and multiple sets of cleavage fractures parallel to c(0001), m{1010}, r{1011}, and r'. c-axis direction (arrow) and
directions of cleavage traces indicated in inset. Photograph courtesy of T. E. Bunch (cross-polarized light).
0.1 mm
44 Traces of Catastrophe
Fig. 4.15. Quartz; fractured, in quartzite. Intense fracturing of quartz in a coarse-grained metamorphosed orthoquartzite target rock
from the Gardnos structure (Norway). The large quartz grain (right) grades into a finer-grained recrystallized shear zone (left). The
quartz grain is cut by numerous subparallel planar fractures (longer, dark, subhorizontal lines) and by much shorter planar features (short,
dark, near-vertical lines) that originate along the fracture planes. These latter features may be relicts of actual PDFs or of Brazil twins
parallel to the base (0001). Within the Gardnos structure, the originally white quartzite is dark gray to black and highly fractured, and the
fractures within the quartz grains contain carbonaceous material. Sample NG-94-17B (cross-polarized light).
and the Ries Crater (Germany) (age 15 Ma) (Fig. 4.16).
However, preservation of fresh, continuous PDFs depends
on geological circumstances, including cooling rate and
postimpact temperatures. Fresh, well-preserved PDFs are
also present in older structures, e.g., Sierra Madera (Texas)
(age <100 Ma) (Fig. 4.19) and Gardnos (Norway) (age
>400 Ma) (Fig. 4.20). The occurrence of striking fresh PDFs
in quartz exactly at the K/T boundary, a worldwide layer of
ejecta from the Chicxulub structure (Mexico) (age 65 Ma)
(Figs. 4.17 and 4.18), provided some of the most important
initial evidence that a large meteorite impact event had
occurred at that time.
In altered, geologically old, or metamorphosed samples,
PDFs have an equally distinctive but discontinuous character.
The original amorphous material in the PDF planes is
recrystallized back to quartz, and in the process, arrays of
small (typically 1–2 μm) fluid inclusions (“decorations”)
develop along the original planes (Figs. 4.21 and 4.22). The
resulting features, called decorated PDFs (Robertson et al.,
1968; Stöffler and Langenhorst, 1994) preserve the orientation
of the original PDFs, and the distinctive shock-produced
fabric can still be recognized in old rocks that have
even undergone metamorphism [e.g., greenschist facies at
Sudbury (Canada); Fig. 4.23]. More intense recrystallization
produces arrays of small mosaic quartz crystals
(subgrains), especially along PDFs originally parallel to the
base c(0001) of the quartz grain (Leroux et al., 1994).
A second type of PDF, oriented parallel to the base
c(0001), has recently been identified, chiefly by studies of
1 mm
structure of the original crystalline quartz is severely deformed,
so that the quartz has been transformed into a distinct
amorphous phase (Müller, 1969; Kieffer et al., 1976a;
Goltrant et al., 1991, 1992).
The importance of PDFs arises from the fact that they
are clearly distinct from deformation features produced in
quartz by nonimpact processes, e.g., cleavage or tectonic
(metamorphic) deformation lamellae (Böhm lamellae)
(Carter, 1965, 1968; Alexopoulos et al., 1988; Stöffler and
Langenhorst, 1994). Cleavages are open fractures; they tend
to be relatively thick (~10 μm) and widely spaced (>20 μm).
Deformation lamellae consist of bands of quartz typically
10–20 μm thick and >10 μm apart that are optically distinct
and slightly misoriented relative to the host grain. In
contrast to these features, shock-produced PDFs are narrow
(<2–3 μm) straight planes consisting of highly deformed or
amorphous quartz, and they are generally oriented parallel
to specific rational crystallographic planes in the host quartz
crystal, especially to the base c(0001) or to low-index rhombohedral
planes such as w{1013}, p{1012}, and r{1011}
(Table 4.4).
The presence of well-developed PDFs produces a striking
and distinctive appearance in thin section. Unaltered
PDFs form multiple sets of continuous planes that extend
across most or all of the host grain (Figs. 4.16, 4.17, and
4.18). These fresh, continuous PDFs tend to be observed
only in unaltered material from shock-wave experiments
and from younger, well-preserved impact structures, e.g.,
Barringer Meteor Crater (Arizona) (age 50 ka) (Fig. 4.13)
Shock-Metamorphic Effects in Rocks and Minerals 45
Fig. 4.17. Quartz; multiple PDFs, fresh. Small quartz grain
(0.20 mm long) from K/T boundary ejecta layer, showing two
prominent sets of fresh (undecorated) PDFs. (Small dots with
halos are artifacts.) Specimen from Starkville South, a few kilometers
south of Trinidad, Colorado. Photograph courtesy of
G. A. Izett. Spindle stage mount (plane-polarized light).
Fig. 4.18. Quartz; multiple PDFs, fresh. Small quartz grain
(0.36 mm long) from K/T boundary ejecta layer, containing one
opaque inclusion and multiple (3–5?) prominent sets of fresh
(undecorated) PDFs. Specimen from Clear Creek North, a few
kilometers south of Trinidad, Colorado. Photograph courtesy of
G. A. Izett. Spindle stage mount (plane-polarized light).
0.1 mm
Fig. 4.16. Quartz; multiple PDFs, fresh. Striking multiple sets of PDFs developed in a quartz grain from a shocked granite inclusion
in suevite from the Ries Crater (Germany). “A” indicates PDFs parallel to {1013} or {0113}; “B” indicates PDFs parallel to {1011} or
{0111}. Note the irregular mottled extinction within the quartz grain. From von Engelhardt and Stöffler (1965), Fig. 1. Photograph
courtesy of W. von Engelhardt (cross-polarized light).
46 Traces of Catastrophe
Fig. 4.19. Quartz; multiple PDFs, fresh. Shocked quartz grain containing multiple sets of fresh PDFs. The grain is included with rare
sandstone fragments in a carbonate breccia dike that cuts the deformed basement rocks at Sierra Madera (Texas), an impact structure
developed in a target composed dominantly of carbonate rocks. The closely spaced PDFs give a distinctive darkened, yellowish appearance
to the quartz grain. Sample SMF-65-2-2 (plane-polarized light).
0.1 mm
TABLE 4.4. Typical crystallographic orientations of planar
microstructures in shocked quartz (modified from
Stöffler and Langenhorst, 1994, Table 3, p. 164).
Polar Angle
(Angle Between Pole to Plane
Symbol Miller Indexes and Quartz c-axis)
c * (0001) 0°
w, w' * {1013},{0113} 23°
p, p' * {1012},{0112} 32°
r, z * {1011},{0111} 52°
m {1010} 90°
x {1122},{2112} 48°
s {1121},{2111} 66°
a {1120},{2110} 90°
* {2241},{4221} 77°
t {4041},{0441} 79°
k {5160},{6150} 90°
x {5161},{6511} 82°
{6151},{1561}
— {3141},{4311} 78°
{4131},{1341}
— {2131},{3211} 74°
{3121},{1231}
*Prominent planes in typical shock fabrics.
Shock-Metamorphic Effects in Rocks and Minerals 47
Fig. 4.21. Quartz; multiple PDFs, decorated. Large compound quartz grain from shocked basement rock inclusion in suevite breccia
from Rochechouart (France), showing two prominent sets of partially decorated PDFs (north-northeast/south-southwest; northeast/
southwest). Original, partly continuous PDF traces are still recognizable from the location of small fluid inclusions (black dots) along the
original PDF planes. Sample FRF-69-16 (cross-polarized light).
0.1 mm
0.1 mm
Fig. 4.20. Quartz; multiple PDFs, slightly decorated. Quartz grain in a carbon-bearing crater-fill breccia from Gardnos (Norway),
showing two well-developed sets of {1013} PDFs. In places, the normally continuous PDFs break down into a string of small fluid
inclusions (small black dots) that follow the original trace of the PDFs. This process, by which the originally glassy material in the PDFs
is recrystallized and replaced by fluid inclusions, has produced decorated PDFs, in which the original PDFs are visible only by the arrays
of fluid inclusions that reproduce their original orientations. Sample NG-94-31 (plane-polarized light).
48 Traces of Catastrophe
Fig. 4.22. Quartz; multiple PDFs, decorated. Compound quartz grain showing two prominent sets of decorated PDFs (north/south;
northwest/southeast). The original PDF planes are now largely replaced by arrays of small fluid inclusions that preserve the original PDF
orientations. Sample from Precambrian basement gneiss in the central uplift of the Carswell Lake structure (Canada). Photograph
courtesy of M. R. Dence. Sample DCR-11-63B (cross-polarized light).
Fig. 4.23. Quartz; multiple PDFs, decorated. High-magnification view of shocked quartz from ejecta block in metamorphosed suevite,
showing multiple sets of recrystallized PDFs (northwest/southeast; east/west) now expressed by arrays of small fluid inclusions
(black dots). Quartz grain also contains numerous random larger fluid inclusions scattered through the grain. Sample from a small
granitic gneiss inclusion in the Onaping Formation “Black Member,” from the type locality, Onaping Falls (Highway 144, Dowling
Township), northwestern corner of the Sudbury structure (Canada). Photograph courtesy of N. M. Short. Sample CSF-66-39 (crosspolarized
light).
0.1 mm
0.1 mm
Shock-Metamorphic Effects in Rocks and Minerals 49
Fig. 4.24. Quartz; basal PDFs. Large irregular quartz grain associated with sericitized feldspar (dark) in footwall granitic rocks on
North Range of Sudbury structure (Canada), together with shatter cones and pseudotachylite. Grain shows one well-developed set of
PDFs (upper left/lower right), which appear as linear arrays of small fluid inclusions parallel to the base (0001) of the quartz grain.
Sample CSF-67-55-2 (cross-polarized light).
shocked quartz with transmission election microscopy
(TEM), as Brazil twins (Fig. 4.24) (Leroux et al., 1994; Joreau
et al., 1996). This form of twinning also occurs in natural
unshocked quartz, but it has never been observed parallel to
the base in such samples. Experimental formation of basaloriented
Brazil twins in quartz requires high stresses (about
8 GPa) and high strain rates, and it seems probable that such
features in natural quartz can also be regarded as unique
impact indicators (Stöffler and Langenhorst, 1994).
4.5.3. PDF Orientations
Despite the distinctive appearance of PDFs in thin section,
appearance alone is not adequate to distinguish them
from nonshock features or to argue that they are impact
produced. An additional and definitive characteristic of PDFs
is their tendency to form along specific planes in the quartz
crystal lattice. Measurements of PDF orientations within the
host quartz grain therefore provide a simple and reliable
method to distinguish them from planar structures produced
by nonshock processes. PDF orientations can be measured
using standard petrofabric procedures on a U-stage (for details,
measurement techniques, and specific studies, see Carter,
1965, 1968; Robertson et al., 1968; von Engelhardt and Bertsch,
1969; Alexopoulos et al., 1988; Stöffler and Langenhorst, 1994)
or on the related spindle stage (Bloss, 1981; Medenbach, 1985;
Bohor et al., 1984, 1987; Izett, 1990).
The procedures involve measuring, in a single quartz grain,
both the orientation of the pole (normal) to each set of PDFs
0.2 mm
and the orientation of the c-axis (= optic axis) of the grain.
The measurement data are then plotted on a standard
stereonet, and the results are expressed as the location of the
pole to the PDFs relative to the c-axis. If a large number of
PDF measurements can be made on a sample, a convenient,
although not entirely rigorous, method to present comparative
results is to plot a frequency diagram (histogram) of the
angles between the c-axis and the pole to each set of PDFs.
Because shock-produced PDFs in a given quartz grain
are parallel to only a few specific crystallographic planes, the
angles measured between the quartz c-axis and the poles to
the PDFs tend to concentrate at a few specific values. In a
histogram plot, the poles appear as sharp concentrations at
specific angles, each of which corresponds to a particular
plane (Figs. 4.25 and 4.26).
This sharply peaked pattern of PDF orientations, typically
characterized by peaks at c(0001) (0°), w{1013} (23°),
and p{1012} (32°), is one of the most useful and most-used
indicators of meteorite impact. Such plots clearly demonstrate
the great difference between PDF distributions
(Figs. 4.25a–c) and the more widely distributed, bell-shaped
distribution characteristic of metamorphic deformation
lamellae (Fig. 4.25e). Such plots are also used to distinguish
different shock-produced fabrics that reflect different shock
pressures (Fig. 4.26).
Experimental and geological studies have demonstrated
that PDFs form in quartz at pressures of ~7–35 GPa, or at
the lower end of the range of shock-metamorphic pressures
50 Traces of Catastrophe
Fig. 4.25. Quartz; PDF orientations. Comparative histograms showing orientations of shock-produced PDFs and other planar
deformation features in quartz (from Carter, 1965). In each diagram, the angle between the quartz c-axis and the pole to the planar
feature is plotted on the x-axis; y-axis indicates frequency for each given angle. Shock-produced fabrics are characterized by strong
orientations parallel to a few specific crystallographic planes. (a) and (b) Basal-oriented sets of deformation lamellae in shocked sandstones
from the Vredefort (South Africa) and Barringer Meteor Crater (Arizona) structures; (c) distinctive PDFs showing the distinctive
concentration parallel to w{1013} [shocked crystalline rocks; Clearwater Lakes (Canada)]; (d) low-angle, near-basal fabric of deformation
lamellae generated under high-strain experimental conditions; (e) broad distribution of metamorphic deformation lamellae (Böhm lamellae)
produced by normal metamorphic conditions. The distinctive differences between shock-produced fabrics (a), (b), and (c) and those of
normal metamorphism (e) have been one of the strongest arguments for the meteorite impact origin of suspected impact structures.
Shock-Metamorphic Effects in Rocks and Minerals 51
Fig. 4.26. Quartz; PDF orientations. Comparative histograms showing different fabrics displayed by PDFs in quartz produced at
different shock pressures, based on measurements of shocked crystalline rocks from several Canadian impact structures (from Robertson
et al., 1968). With increasing shock pressures, both the total number of PDFs and the number of different orientations increase. The
following fabrics, and the minimum shock pressures estimated to form them (Grieve and Robertson, 1976, pp. 39–40), can be recognized:
type A (P > 7.5 GPa): basal PDFs only; type B (P > 10 GPa), appearance of w{1013} planes, typically with basal planes; type C
(P > 14 GPa), appearance of {2241} planes with others; type D (P > 16 GPa), appearance of p{1012} planes with others. These fabrics
have been used as shock barometers to measure the intensity and distribution of shock pressures in several structures (Grieve and Robertson,
1976; Robertson and Grieve, 1977; Dressler and Sharpton, 1997). From Carter (1965).
52 Traces of Catastrophe
been observed in sedimentary rocks from several impact
structures (Kieffer, 1971, Kieffer et al., 1976a; Grieve et al.,
1996).
Despite these similarities, a growing amount of data now
indicates that sedimentary rocks, especially porous ones, respond
differently to shock waves than do nonporous crystalline
rocks. One indication of significant differences is that
PDF fabrics measured in sediments show a large proportion
of PDFs whose poles are oriented at high angles (>45°) to
the quartz c-axis (Grieve et al., 1996; Gostin and Therriault,
1997). Other possible differences are that PDFs may first
appear, or a particular PDF fabric may develop, at different
shock pressures in sedimentary rocks than in crystalline rocks.
A more important difference between porous and nonporous
rocks is that a shock wave passing through porous
sediments will generate more heat than in passing through
crystalline rocks, chiefly because more of the shock-wave
energy is absorbed by the numerous grain interfaces and pore
spaces in the sediment (Kieffer, 1971; Kieffer et al., 1976a;
Kieffer and Simonds, 1980; Stöffler, 1984). As a result, extensive
melting will occur at lower shock pressures in sediments
than in crystalline rocks, i.e., at about 15–20 GPa in sandstone
vs. 50–60 GPa in crystalline rocks (Stöffler, 1972, 1984).
Therefore, the higher-pressure fabrics of quartz PDFs, which
form at 20–30 GPa in crystalline rocks, may not be found in
sediments, either because they did not form or because they
Fig. 4.27. Quartz; multiple PDFs, fresh. Photomicrograph showing at least four sets of fresh PDFs in a shocked quartz grain from
crystalline target rocks at the Lake St. Martin impact structure, Manitoba (Canada). Two prominent PDF sets (northwest/southeast and
west-northwest/east-southeast) are accompanied by less obvious sets oriented approximately north/south and east/west. Petrofabric
measurements with a U-stage show that the PDFs are oriented parallel to both w{1013} and p{1012}, indicating moderately high shock
pressures (>15 GPa). Patches of diaplectic glass, associated with the shocked quartz, appear as dark zones (e.g., upper right). Width of
field is ~100 μm. Photograph courtesy of V. L. Sharpton (cross-polarized light).
(e.g., Hörz, 1968; Stöffler and Langenhorst, 1994). However,
the relative abundance of different PDF orientations varies
significantly with shock pressure. Basal Brazil twins, although
little studied so far, appear restricted to shock pressures below
10 GPa. PDFs parallel to w{1013} develop at about >7–
10 GPa, and PDFs parallel to p{1012} at about >20 GPa.
At higher pressures, e.g., 20–35 GPa, the total number of
PDF sets increases, and additional orientations appear
(Fig. 4.26). The PDFs formed at these higher levels tend to
be intensely developed and very closely spaced within the
quartz grains (Figs. 4.16, 4.18, and 4.27).
4.5.4. PDFs in Sedimentary Rocks
Although PDFs and their orientations can be reliably used
as indicators of shock and impact events, it is becoming clear
that our current knowledge about such features is incomplete
and unrepresentative. Nearly all our information to date
has come from impact structures formed in dense, coherent,
quartz-bearing crystalline rocks. There is relatively little information
about the effects of shock deformation in other
kinds of quartz-bearing rocks, e.g., porous sandstones or finegrained
shales.
Several studies have demonstrated that shocked sandstones
and shales also develop PDFs in quartz, and even
diaplectic quartz and feldspar glasses, similar to those observed
in shocked crystalline rocks, and these features have
Shock-Metamorphic Effects in Rocks and Minerals 53
were immediately destroyed by postshock melting. The
unique shock effects observed in sedimentary rocks can still
provide conclusive evidence for an impact origin [e.g., at
Barringer Meteor Crater (Arizona) (Kieffer, 1971)], but the
details of such occurrences cannot be accurately interpreted
on the basis of results from shocked nonporous crystalline
rocks (Grieve et al., 1996).
4.6. PLANAR MICROSTRUCTURES IN
FELDSPAR AND OTHER MINERALS
Similar planar microstructures are produced by shock in
many other minerals (e.g., Stöffler, 1972, 1974), but such
features have been less used as indicators of meteorite impact.
Feldspars of all kinds (both alkali varieties and plagioclase)
display various shock-produced planar microstructures:
fractures, deformation bands, kink bands, and actual PDFs.
Frequently, short and closely spaced PDFs may be combined
with longer and more widely spaced features (deformation
bands or albite twinning) to produce a distinctive ladder texture
(Figs. 4.28, 4.29, and 4.30).
Although several studies have been made of shock-produced
planar features in feldspars (e.g., Stöffler, 1967, 1972;
papers in French and Short, 1968), these features have been
less studied and less well characterized than those in quartz.
There are several reasons for this: the greater diversity and
complexity of such features, the greater optical complexity
(biaxial) of feldspars, and the common secondary alteration
of the feldspar and its planar features to clays, iron
oxides, etc. (Figs. 4.29 and 4.30). Another factor in studies
focused on identifying new impact structures is the fact
that shocked feldspar in crystalline rocks is generally associated
with shocked quartz, whose features (especially PDFs)
provide a quicker and simpler method for establishing an
impact origin.
Planar microstructures, both planar fractures and true
PDFs, have also been observed in other minerals, including
pyroxene, amphiboles, and several accessory phases (apatite,
sillimanite, cordierite, garnet, scapolite, and zircon) (Stöffler,
1972). Less is known about PDF formation and orientations
in these minerals, because appropriate rocks are less
abundant in most impact structures, and because the specific
minerals have not been studied in detail. However, recognition
of shock-produced PDFs in zircon has been
especially important in applying U-Th-Pb dating methods
to individual zircons in shocked target rocks to determine
the ages of impact structures (e.g., Krogh et al., 1984, 1993;
Kamo and Krogh, 1995).
The development of distinctive shock-metamorphic features
such as PDFs in denser mafic minerals like amphibole,
pyroxene, and olivine apparently occurs at higher pressures
and over a more limited pressure range than in quartz and
feldspar. At pressures <30 GPa, sufficient to form PDFs in
both quartz and feldspar, the most common shock effects
observed in mafic minerals are planar fractures, mechanical
twins, and general comminution (Stöffler, 1972); features
Fig. 4.28. Feldspar; multiple PDFs and diaplectic glass
(maskelynite). Shocked plagioclase feldspar grain from the Ries
Crater (Germany), showing development of multiple sets of PDFs
(lower right) and gradational conversion of the same crystal to
diaplectic glass (maskelynite) (upper left). Original polysynthetic
albite twin lamellae (northwest/southeast) are still preserved in
part of the crystal (lower right), but alternate twin lamellae have
either been converted to maskelynite (clear) or are crosscut by short,
closely spaced PDFs to form a distinctive “ladder” structure.
Elsewhere in the crystal (upper left), both the original twins and
the subsequent shock-produced PDFs disappear, and the whole
crystal consists of maskelynite. Sample from a moderately shocked
amphibolite fragment in suevite breccia. From Stöffler (1966), Fig. 4
(plane-polarized light).
resembling true PDFs are only rarely observed. At higher
pressures, mafic minerals in naturally and experimentally
shocked basalts generally show only extreme comminution,
accompanied by the melting and flow of associated feldspar
(Kieffer et al., 1976b; Schaal and Hörz, 1977). PDFs are therefore
unlikely to be observed in mafic minerals in impact structures.
The higher pressures apparently required for their
formation imply that they will form in a correspondingly
smaller volume of shocked rock in the structure. Furthermore,
the higher shock pressures required are closer to pressures
that produce partial to complete melting of the rock,
so that PDFs, even if formed, would not survive any subsequent
melting episode.
0.1 mm
54 Traces of Catastrophe
Fig. 4.29. Feldspar; multiple PDFs, “ladder” texture. Shocked K-feldspar, showing multiple sets of altered PDFs. Two types of planar
deformation features are present: (1) long, thicker, widely spaced planes (clear areas, approximately east/west) that may be deformation
bands or kink bands; (2) short, narrower, closely spaced features (northeast/southwest and north-northwest/south-southeast) that combine
with the first type to form a distinctive “ladder” texture. The planar features have a brownish-red color, possibly caused by alteration of the
feldspar to clay minerals and iron oxides. Sample from a small granitic gneiss inclusion in the Onaping Formation “Black Member” from
the type locality, Onaping Falls (Highway 144, Dowling Township), northwestern corner of the Sudbury structure (Canada). Photograph
courtesy of N. M. Short. Sample CSF-66-39 (cross-polarized light).
Fig. 4.30. Feldspar; twinning and PDFs. Large deformed feldspar crystal (microcline?) in granitic fragment in suevite breccia. Original
twinning in the feldspar (light/dark pattern, northwest/southeast) is deformed and faulted along multiple parallel fractures (east-northeast/
west-southwest). Elsewhere, the feldspar is cut by a single set of short, narrow, closely spaced planar features (northeast/southwest) that
may be actual PDFs. Sample from a small block of granitic gneiss from the Onaping Formation “Black Member,” Sudbury (Canada).
Sample CSF-67-73 (cross-polarized light).
0.1 mm
0.1 mm
Shock-Metamorphic Effects in Rocks and Minerals 55
4.7. SHOCK ISOTROPIZATION AND
DIAPLECTIC GLASSES
Planar microstructures form at relatively low shock pressures
(>7–35 GPa) (Table 4.2) (Stöffler and Langenhorst,
1994) and involve only partial and localized deformation of
the host crystal. PDFs, which develop in the upper part of
this range (10–35 GPa), involve actual conversion of the
quartz crystal structure to an amorphous phase within the
individual planes. Higher shock pressures (35–45 GPa),
which transmit more energy into the crystal, do not form
PDFs. Instead, the shock waves convert the entire crystal to
an amorphous (glassy) phase.
This shock-produced diaplectic glass (also called thetomorphic
glass) (Stöffler, 1966, 1967, 1972, 1984; Chao, 1967;
papers in French and Short, 1968) is completely different from
conventional glasses produced by melting a mineral to a liquid
at temperatures above its melting point. Diaplectic glasses
do not melt or flow; they preserve the original textures of the
crystal and the original fabric of the mineral in the rock. In
addition, although diaplectic glasses are optically isotropic
(i.e., they show no birefringence when examined petrographically
under crossed polarizers), studies of quartz and feldspar
diaplectic glasses by X-ray diffraction and infrared
spectrometry have shown that they retain much of the ordered
atomic structure of the original crystal (e.g., Bunch et
al., 1967, 1968; Stöffler, 1974, 1984; Arndt et al., 1982).
Samples of diaplectic feldspar glasses have also been experimentally
annealed by heating at ambient pressure to produce
original single crystals (Bunch et al., 1967, 1968; Arndt
et al., 1982) or microcrystalline aggregates that preserve the
shapes of the original feldspar crystals (Arndt et al., 1982;
Ostertag and Stöffler, 1982).
Quartz and feldspar are the most common examples of
minerals converted to diaplectic glasses by shock waves.
Diaplectic plagioclase feldspar glass, called maskelynite, was
in fact observed in meteorites more than a century before it
was discovered in shocked terrestrial rocks. The same material,
often well preserved, is also observed at several impact
structures where highly shocked rocks are preserved, e.g.,
the Ries Crater (Germany) (Figs. 4.28, 4.32, and 4.33) and
Manicouagan (Canada) (Fig. 4.31).
In these occurrences, the unique textures of the diaplectic
glasses clearly indicate formation without melting to the liquid
state. The overall grain fabric of the rock is unchanged,
and the diaplectic glasses preserve the shapes of the original
quartz and feldspar grains. In some grains, the transformation
to diaplectic glass is incomplete, and areas of relict birefringence
remain in the otherwise isotropic material
(Figs. 4.28 and 4.31). In some shocked plagioclase grains,
one set of alternating albite twins is converted to maskelynite,
while the twins of the other set remain birefringent. Other
minerals (e.g., amphibole, garnet, micas), associated with (or
even in contact with) grains of diaplectic glass, show little
Fig. 4.31. Feldspar; diaplectic glass (maskelynite). Shocked plagioclase feldspar, partially converted to isotropic diaplectic feldspar
glass (maskelynite). Parts of the original coarse feldspar grains remain crystalline and birefringent (light areas); these regions grade into
adjoining areas of maskelynite (dark). Drill-core sample from coarse-grained basement anorthosite, exposed in the central uplift of the
Manicouagan structure (Canada). Photograph courtesy of M. R. Dence. Sample DMM-73-63B (cross-polarized light).
0.05 mm
56 Traces of Catastrophe
Fig. 4.32. Feldspar and quartz; diaplectic glasses. Biotite gneiss containing diaplectic feldspar glass (maskelynite) (clear, low relief;
e.g., upper right) and diaplectic quartz glass (clear, higher relief, e.g., lower right). The associated biotite crystals (dark) have retained
their original shape and have remained crystalline and birefringent, despite the complete transformation of adjacent quartz and plagioclase
into glassy phases (compare with Fig. 4.33). Biotite gneiss inclusion in suevite breccia, Otting, Ries Crater (Germany). From Stöffler
(1967), Fig. 12a. Photograph courtesy of D. Stöffler (plane-polarized light).
Fig. 4.33. Feldspar and quartz; diaplectic glasses. Biotite gneiss containing diaplectic feldspar glass (maskelynite) and diaplectic quartz
glass (compare with Fig. 4.32). Both phases are isotropic (dark) under crossed polarizers. The associated biotite crystals have retained
their original shape and have remained crystalline and birefringent, despite the complete transformation of adjacent quartz and plagioclase
into glassy phases. Biotite gneiss inclusion in suevite breccia, Otting, Ries Crater (Germany). From Stöffler (1967), Fig. 12b. Photograph
courtesy of D. Stöffler (cross-polarized light).
0.1 mm
0.1 mm
Shock-Metamorphic Effects in Rocks and Minerals 57
deformation and retain their original form (Figs. 4.32 and
4.33), although they may show reduced birefringence and
reddening produced by the formation of hematite (e.g.,
Feldman, 1994) and cordierite (Stähle, 1973).
Diaplectic glasses formed from other minerals (e.g., scapolite)
have rarely been observed. Mafic minerals (e.g., pyroxene,
amphibole, and biotite) do not seem to form diaplectic
glasses, probably because the pressures required are higher
than those for quartz and feldspar, high enough so that shockproduced
melting occurs instead.
Diaplectic quartz and feldspar glasses are metastable. They
apparently do not survive if they are exposed to even relatively
mild postimpact thermal effects. Diaplectic glasses are
not observed in impact structures that have been even slightly
metamorphosed, even though decorated PDFs may still be
preserved in associated quartz. In such settings, instead of
diaplectic glasses, one observes quartz and feldspar grains
that are recrystallized to microcrystalline aggregates that replace
the original crystal (Figs. 4.34, 4.35, and 4.36). Textures
in the altered feldspars sometimes suggest intense plastic
deformation and flow within the original grain. These features
are often accompanied by the development of plumose
or spherulitic microcrystalline textures that may reflect significant
thermal effects as well. Such grains of quartz and
Fig. 4.34. Feldspar; possible diaplectic glass, recrystallized. Large, highly deformed and recrystallized feldspar clast in suevite breccia,
surrounded by finer fragments in an opaque carbon-bearing matrix. The feldspar shows deformation and recrystallization throughout, as
indicated by the intensely mosaic extinction. The crystal is subdivided by thin irregular zones of nearly isotropic material, possibly
original melt. Plastic behavior of the fragment is also suggested by indentations of the matrix into the clast (e.g., at top). This clast can be
interpreted as a fragment of diaplectic feldspar glass that has subsequently been recrystallized to form a fine-grained microcrystalline
texture that is still similar to the original crystal. Similar reactions have been produced in experimentally annealed maskelynite. Another
possibility is that the fragment was shock-heated above its melting point, but was rapidly quenched (perhaps during deposition) before
extensive flow could occur. In any case, the unusual texture has been preserved despite subsequent metamorphism of the unit in which it
occurs. Fragment in Onaping Formation “Black Member” from type locality, Onaping Falls (Highway 144, Dowling Township),
northwestern corner of Sudbury structure (Canada). Sample CSF-66-37-2 (cross-polarized light).
feldspar have been tentatively interpreted as original diaplectic
glasses that have been annealed and recrystallized,
either by immediate postshock thermal effects or by subsequent
metamorphism (McIntyre, 1968; French, 1968b,
pp. 401–404).
4.8. SELECTIVE MINERAL MELTING
The high-pressure (35–45 GPa) shock waves that produce
diaplectic glasses also generate significant and sudden
postshock temperature rises of several hundred degrees
Celsius in the rocks and minerals through which they pass
(Fig. 4.1). In the region of diaplectic glass formation,
postshock temperatures are still low enough (300°–900°C)
that virtually no actual melting occurs, and rapidly quenched
samples of diaplectic glasses suffer no further immediate
alteration. However, at slightly higher shock pressures
(~45–50 GPa), the higher postshock temperatures (>1000°C)
begin to exceed the melting points of typical rock-forming
minerals, and distinctive localized melting effects appear in
the affected rocks.
This shock-produced selective mineral melting differs
significantly from normal equilibrium melting. Under nor-
0.5 mm
58 Traces of Catastrophe
Fig. 4.35. Feldspar; possible diaplectic glass, recrystallized.
Shock-deformed and recrystallized feldspar and quartz from a
coarse-grained granitic fragment in suevite breccia. Large original
quartz grains (lower center; gray, higher relief ) are recrystallized
to finely crystalline mosaic quartz. Original feldspar grains (clear,
lower relief ) are generally finely recrystallized and virtually isotropic
in some areas (compare with Fig. 4.36), although some areas of
original feldspar crystals are preserved. From granitic inclusion in
Onaping Formation “Black Member” at type locality, Onaping
Falls (Highway 144, Dowling Township), northwestern corner of
Sudbury structure (Canada). Sample CSF-66-50-13 (planepolarized
light).
Fig. 4.36. Feldspar; possible diaplectic glass, recrystallized.
Shock-deformed and recrystallized feldspar and quartz from a
granitic fragment in suevite breccia. Large original quartz grains
are recrystallized to finely crystalline mosaic quartz. Original
feldspar grains are generally finely recrystallized and virtually
isotropic in some areas, although some areas of original feldspar
crystals are preserved (compare with Fig. 4.35). In one such area
(right center), a plagioclase crystal has been plastically deformed,
bending the original polysynthetic albite twinning (light/dark
bands) through a large angle. Despite the intense deformation of
quartz and feldspar, a single apatite grain (lower right) shows no
deformation. Sample from granitic inclusion in Onaping Formation
“Black Member” at type locality, Onaping Falls (Highway 144,
Dowling Township), northwestern corner of Sudbury structure
(Canada). Sample CSF-66-50-13 (cross-polarized light).
mal conditions of increasing overall temperature, melting
occurs first at the boundaries between different mineral
grains. Two or more different minerals are involved, and the
resulting eutectic melt has a composition intermediate between
that of the adjacent minerals and forms at a temperature
well below that of their individual melting points. In a
shock-wave environment, each mineral grain is instantaneously
raised to a postshock temperature that depends on
the shock-wave pressure and on the density and compressibility
of the mineral itself. If the postshock temperature produced
in a mineral exceeds its normal melting temperature,
each grain of that mineral in the rock will melt, immediately
and independently, after the shock wave has passed. The melt
will have approximately the same composition as the original
mineral before any flow or mixing takes place, and the
melt regions will initially be distributed through the rock in
the same pattern as the original mineral grains.
Selective melting therefore produces unusual textures in
which one or more minerals in a rock show typical melting
features while others — even immediately adjacent ones —
do not. Shocked granitic inclusions from the Ries Crater
(Germany) frequently show a texture in which feldspar has
melted, flowed, and vesiculated, but the adjacent quartz remains
in the form of unmelted diaplectic glass (Chao, 1967;
Stöffler, 1972, 1984). Similar textures can be preserved even
in subsequently metamorphosed rocks, in which flowed and
recrystallized feldspar is accompanied by recrystallized but
undeformed grains of quartz (Fig. 4.37).
0.5 mm 0.5 mm
Shock-Metamorphic Effects in Rocks and Minerals 59
Fig. 4.37. High-temperature effects; plastic deformation, grain-boundary melting. Highly shocked and recrystallized quartzofeldspathic
inclusion in metamorphosed suevite breccia, showing extreme deformation of quartz and feldspar. Quartz (gray, higher relief, lower right)
is recrystallized to a fine mosaic of small quartz grains. Feldspar (clear, lower relief, top) shows intense, contorted flow structure, indicating
either incipient melting or extreme plastic flow. Definite incipient melting has occurred at the grain boundaries, forming a brown melt
(dark) with lath-like microlites (white; feldspar?). (Circular feature at center is a bubble in the thin section.) Coarse-grained granitic
inclusion in Onaping Formation “Black Member,” Sudbury structure (Canada). Sample CSF-67-67 (plane-polarized light).
At higher shock pressures, where temperatures are higher
and cooling times may be longer, these selective melting textures
may be complicated by the effects of normal eutectic
melting at grain boundaries (Fig. 4.37). In some shocked
rocks, postshock temperatures may exceed the melting points
of all the minerals present, and the rock will melt to a mixture
of heterogeneous glasses that may preserve (depending
on the amount of subsequent flow and mixing) the original
shapes and mineral compositions. If such rocks are quenched
before flow and mixing can occur, the chemically diverse
glasses can survive and be recognized, even after significant
metamorphism (Fig. 4.38) (Peredery, 1972).
Such distinctive selective melting textures are relatively
uncommon in rocks from impact structures. The region of
shock pressures that produces them is relatively narrow (~45–
55 GPa), and their preservation, once formed, requires rapid
quenching, most commonly as small inclusions in crater-fill
breccias. At progressively higher shock pressures (>55 GPa),
postshock temperatures increase rapidly, melting becomes
complete, flow and mixing processes become dominant in
the melted rock, and more chemically homogeneous bodies
of impact melt are produced (see Chapter 6).
0.5 mm
60 Traces of Catastrophe
Fig. 4.38. High-temperature effects; complete melting. Highly shocked, melted, and recrystallized rock inclusion in metamorphosed
suevite breccia. Postshock temperatures apparently exceeded the melting points of all component minerals, converting the originally
crystalline rock into an initially heterogeneous glass that developed limited flow textures before it was quenched. The inclusion was
subsequently recrystallized to secondary minerals such as quartz, feldspar, amphibole, and chlorite, but the original mineralogy and the
character of the shock-formed heterogeneous glass are still detectable in the distribution and chemical variations in the secondary mineral
assemblage. Inclusion in Onaping Formation “Black Member” at the type locality, Onaping Falls (Highway 144, Dowling Township),
northwestern corner of Sudbury structure (Canada). Sample CSF-66-50-3 (plane-polarized light).
0.5 mm

Admin

  • Administrator
  • Hero Member
  • *****
  • Posts: 555
    • View Profile
Re: Traces of Cat.
« Reply #3 on: January 29, 2017, 06:03:53 am »
Shock-Metamorphosed Rocks (Impactites) in Impact Structures 61
61
5.1. ROCK TYPES IN THE FINAL
IMPACT STRUCTURE
A wide variety of distinctive rock types — breccias, melts,
and shock-metamorphosed target rocks — are produced
during formation of impact structures. The classification
of these complex and diverse rocks is an active and muchdebated
activity (see below). However, the general term
impactite is used here as a convenient overall designation
for all rocks affected by, or produced by, the shock waves
and other processes generated by hypervelocity meteorite
impact events.
Different varieties of impactites are produced at different
times during the impact process, and they occur in different
locations beneath, within, and around the final impact structure.
The diverse features of impactites reflect, in varying
ways, different aspects of the impact event itself: (1) the
initial shock-wave distribution around the impact point;
(2) the subsequent excavation flow, formation of the transient
crater, and ejection of material from it; (3) the crater
modification processes. The general model described below
will be modified, in actual impact structures, by such individual
factors as the target lithology, stratigraphy, and the
nature and impact angle of the projectile, but the model provides
a general basis for the identification and classification
of impactites (see also Dence, 1968; Grieve, 1991; Stöffler et
al., 1988).
The basic distribution of shock-wave pressures around
the impact point is largely established by the end of the contact/
compression stage. The expanding shock waves deposit
energy continuously in the target rocks through which they
pass, and both their peak pressures and the resulting postshock
temperatures drop rapidly with distance from the impact
point. As the contact/compression stage ends, and the
transient crater begins to form, the zones of shock pressure
form a series of approximately hemispherical shells around
the impact point, with the peak shock pressure decreasing
rapidly outward (Fig. 3.2).
During the subsequent excavation stage and formation
of the transient crater, virtually all the target rock exposed to
shock pressures of >25–30 GPa, which now consists of a
mixture of vapor, superheated rock melt, and coherent but
highly shocked target rock, is broken up and accelerated
outward (Dence, 1968; Dence et al., 1977; Grieve and Cintala,
1981). Because the excavation flow lines cut across the originally
hemispherical shock-pressure zones (Fig. 3.4), the excavated
material will consist of a mixture of target rocks
subjected to widely differing shock pressures and showing a
wide range of shock effects. A melt-rich portion flows downward
and outward from the center to form a coating along
the floor and walls of the growing crater (Grieve et al., 1977).
The remainder, a mixture of rock fragments and smaller bodies
of melt, is impelled outward from the center of the cavity.
Much of this material may be entirely ejected from the transient
crater; some may remain within the crater as a unit of
mixed rubble and melt above the fractured crater floor.
The subcrater rocks beneath the zone of excavation are
subjected to lower shock pressures ( >30 GPa), and the dominant
effects produced are shatter cones, brecciation, and inplace
fracturing. As the upper part of the target rocks are
excavated from the transient crater, these rocks are displaced
downward, more or less coherently, to form the floor of the
transient crater and the zone of parautochthonous rocks
beneath it.
The final modification of the transient crater into a simple
or complex impact structure involves several distinct gravity-
related processes that influence the distribution of
impactite units: (1) rapid relative movements of large blocks
of subcrater target rocks downward, inward, and upward
along relatively narrow faults; (2) collapse of oversteepened
Shock-Metamorphosed Rocks (Impactites)
in Impact Structures
62 Traces of Catastrophe
crater walls into the crater cavity; (3) deposition of a minor
amount of ejected material within the crater. The first process
may create additional breccias and related rock types
beneath the crater. The other two processes produce a large
portion of the crater-fill deposits, which are characterized
by a generally fragmental character and the presence of shockmetamorphic
effects that range from simple fracturing to
complete melting.
5.2. CLASSIFICATION OF IMPACTITES
The definition and classification of impact-produced
materials, both individual rock fragments and large formations,
is a complex, longstanding, and difficult subject (for
details, see Stöffler, 1971; Stöffler et al., 1979; Taylor et al.,
1991; Stöffler and Grieve, 1994, 1996; Reimold, 1995). No
attempt will be made here to develop a complete and unanimously
acceptable system. The simplified system presented
here emphasizes field and petrologic characteristics and is
based, as far as possible, on objective features that are observable
in outcrop, hand specimen, and thin section. This
classification also uses, as much as possible, traditional terms
already applied to equivalent rocks (e.g., breccias, melt rocks)
formed by common geological processes. Although this system
is generally consistent with more detailed classifications
(e.g., Stöffler and Grieve, 1994), it is restricted to terrestrial
rock types produced in single impact events and does not
consider the special complexities of cratering on other planets,
including the effects of multiple impacts or the absence
of an atmosphere (see Taylor et al., 1991; Stöffler and Grieve,
1994, 1996).
The term impactite is used here to designate all rocks
produced during an impact event, including shock-metamorphosed
(but still recognizable) target rocks (both in place
and as fragments in breccias), breccias, and impact melts.
Under this umbrella, the classification and terminology of
impactite formations are based on a few key features: location
with respect to the crater, source(s) of component materials,
and lithologic characteristics (Table 5.1).
More detailed discriminators, used in other classifications,
include (1) particle sizes and size ranges; (2) relative percentages
of components in breccias, e.g., ratios of fragments/
matrix, and lithic/glassy fragments; (3) shock-metamorphic
effects in individual breccia fragments (both the shock level
in individual fragments and the range of shock effects in
multiple fragments); and (4) textures and crystallinity of
melt rocks.
In earlier discussions of impactites and the cratering process
(Dence, 1965, 1968; Grieve, 1991), a fundamental and
useful distinction has been made between the parautochthonous
rocks beneath the crater floor and the allogenic (or
allochthonous) units (breccias and melt rocks) that fill the
crater (crater-fill units) and form the units of ejecta outside
it (Figs. 3.7 and 3.13). The observed characteristics of these
different rock types are frequently distinctive enough that
they can be distinguished, even in isolated hand specimens
or outcrops.
The parautochthonous rocks beneath the crater have
remained relatively coherent during crater formation, although
they have been deformed and displaced. These rocks,
which correspond to the lower displaced zone of the transient
crater, are subjected to relatively lower shock pressures,
and observed shock-deformation effects are generally
limited to fracturing, brecciation, and the formation of shatter
cones, although higher-pressure mineral-deformation
features may be developed in a relatively small volume beneath
the crater floor. The allogenic rocks, chiefly breccias
and melts, that fill the crater and make up the ejecta beyond
the crater rim, are characterized by a more diverse lithology,
a fragmental or melted character, and a wide range
of observed shock effects. In particular, the crater-fill breccias
are a complex mixture of materials with different histories
of shock pressures and transport: unshocked rocks
derived from the distant parts of the crater rim and walls,
more highly shocked and melted fragments excavated from
the transient crater and redeposited, and large and small
bodies of impact-generated melt.
The following sections discuss impactites on the basis of
location with respect to the impact structure: (1) subcrater:
parautochthonous rocks, cross-cutting allogenic
units, and pseudotachylite; (2) crater interior: allogenic
crater-fill deposits (lithic breccias, suevite breccias, and impact
melt breccias); (3) crater rim region: proximal ejecta
deposits; (4) distant from crater: distal ejecta. A detailed
discussion of impact melt rocks in these different environments
is provided in Chapter 7.
5.3. SUBCRATER ROCKS
5.3.1. Formation Conditions
During formation of the transient crater, the rocks located
in the displaced zone below the zone of excavation are
driven downward and outward, more or less coherently
(Fig. 3.4), but they are not completely broken up or excavated.
Instead, they are deformed, thinned, and moved downward
and outward as the transient crater forms, and then (in
the central parts of larger structures) rapidly elevated as the
central uplift forms (Dence, 1968; Dence et al., 1977; Kieffer
and Simonds, 1980; Grieve and Cintala, 1981; Grieve et al.,
1981; Stöffler et al., 1988).
During these movements, the subcrater rocks are generally
displaced as large individual blocks typically tens to hundreds
of meters (or even larger) in size. However, adjacent
regions within this zone may display little displacement relative
to each other, and original stratigraphy and structural
features may be well preserved within individual blocks. The
term parautochthonous has therefore been applied to these
rocks to indicate their general relative coherence.
The shock pressures imposed on the parautochthonous
rocks vary widely because of the complex relationship beShock-
Metamorphosed Rocks (Impactites) in Impact Structures 63
tween the original shock-wave distribution and the subsequent
crater modification. Shock pressures in the parautochthonous
rocks are therefore highest near the center of
the structure and decrease rapidly outward toward the margin.
Along the floor of the transient cavity (approximately
the floor of the final crater), shock pressures may exceed 25–
30 GPa in the center, decreasing to >2 GPa at the rim, the
minimum pressure needed to excavate material from the transient
crater (Grieve and Robertson, 1976; Robertson and
Grieve, 1977; Kieffer and Simonds, 1980; Dressler et al., 1998).
Shock pressures also drop off rapidly with increasing depth
below the crater floor. In the center, pressures typically drop
from about 25–30 GPa to a few GPa over distances of less
than a few hundred meters in small structures (Dence et al.,
1977; Grieve et al., 1981) and over no more than a few kilometers
in larger ones (Stöffler et al., 1988).
5.3.2. In-Place Shock-Metamorphosed Rocks
The shock effects preserved in the parautochthonous
subcrater rocks therefore reflect a wide range of shock pressures.
In a small region immediately below the central part
of the crater floor (i.e., at the base of the excavation zone),
TABLE 5.1. Criteria for impactite classification.
1. Location with respect to crater (Rc = crater radius)
Crater Floor and Subcrater Within Crater Crater Rim and Near-Surface
Parauthochtonous rocks: Allogenic rocks: Allogenic rocks:
target rocks (coherent) Crater-fill deposits Ejecta:
lithic breccias (= crater-fill breccias) proximal (<5 Rc)
(= “breccia lens”) distal (>5 Rc)
Allogenic rocks (cross-cutting) lithic breccias
breccia dikes melt-bearing breccias
impact melt dikes suevites
impact melt breccias
Pseudotachylite (= melt-matrix breccias)
impact melt rocks
2. Sources of component materials
Parautochthonous rocks Allogenic rocks
Approximately in place (local). Original stratigraphy Derived from single or multiple sources elsewhere.
and structure (largely) preserved.
3. Breccia characteristics
a. Fragment character Lithic breccia Suevite (breccia)
Rock/mineral fragments only Melt/glass fragments present
Rock/mineral fragments
b. Fragment lithology Monomict (breccia) Polymict (breccia)
Single rock type Multiple rock types
c. Matrix character Clastic-matrix (breccia) Impact melt breccia (= melt-matrix breccia)
Discrete fragments Coherent melt (glassy or crystalline)
4. Melt rock character (standard geological terms)
Holohyaline (glassy) For grain size, texture, etc., use other standard igneous rock
Hypocrystalline (mixed glassy/crystalline) discriminators, e.g.:
Holocrystalline (completely crystalline) Microcrystalline
Porphyritic
Trachytic, etc.
64 Traces of Catastrophe
pressures of 10–30 GPa produce distinctive microscopic
deformation effects in quartz and feldspar, while creating
postshock temperatures of >300°C. In smaller impact structures,
this zone of identifiably high shock pressures is less
than a few hundred meters thick, partly because of the rapid
decay of the original shock wave with distance from the impact
point, and partly because of the subsequent compression,
thinning, and displacement of the subcrater rocks during
transient crater formation (Dence et al., 1977; Grieve and
Cintala, 1981). Beneath this zone, lower shock pressures
(possibly 2–6 GPa) produce distinctive megascopic deformation
features (shatter cones) in a deeper region near the
center of the crater.
Shock pressures over most of the zone of parautochthonous
rocks are too low ( >2 GPa) to produce distinctive
shock-deformation effects, but they are high enough to exceed
the yield strengths of near-surface crustal rocks (typically
<1–2 GPa; Kieffer and Simonds, 1980). As a result, large
volumes of rock beneath the crater floor are broken and
crushed during the early stages of crater formation, producing
units of in-place lithic breccia that generally lack distinctive
high-pressure shock-metamorphic effects. At the
same time, and subsequently, larger fractures that develop in
this zone may be intruded by allogenic materials (rock fragments
and/or melt) to form cross-cutting dike-like bodies
(e.g., Lambert, 1981; Bischoff and Oskierski, 1987; Dressler
and Sharpton, 1997).
The parautochthonous rocks below the crater may also
be strongly affected by subsequent large-scale movements
during the crater modification stage. Such movements may
produce equally striking but different breccias. In large structures,
where modification involves the development of a central
uplift, deep-seated parautochthonous rocks may be
suddenly uplifted for distances of hundreds of meters to several
kilometers. This uplift may bring distinctively shocked
rocks (e.g., containing shatter cones) to the surface, where
they may provide definite evidence for the impact origin of a
large structure. However, these rapid movements may also
generate additional varieties of breccias and destroy the original
spatial relations of the parautochthonous rocks to each
other, making the geology and history of the structure more
difficult to decipher.
Understanding the variety of breccias in subcrater rocks
is complicated by several factors (e.g., Lambert, 1981; Bischoff
and Oskierski, 1987; Dressler and Sharpton, 1997). Breccias
may form at various stages in the cratering process: (1) during
the initial shock-wave expansion and transient crater
formation; (2) during the subsequent modification of the
transient crater, including (in large structures) movements
associated with the rise of the central uplift and peripheral
collapse around the rim. Even within the brief formation
time of an impact crater, it is possible for multiple generations
of breccia to develop and to produce distinctive crosscutting
relations, even though the time between one breccia
generation and the next may be measured in seconds or minutes
(Lambert, 1981; Bischoff and Oskierski, 1987; Dressler
and Sharpton, 1997). Another problem is melt formation;
rocks can be shock-melted by the initial impact and then
distributed as melts or melt-bearing breccias throughout the
crater basement, but rocks can also be melted subsequently
by friction generated during the rapid movements of large
volumes of rock involved in crater modification and central
uplift formation.
5.3.3. Lithic Breccias (Parautochthonous)
Impactite breccias that form by the shattering and pulverizing
of target rock essentially in place (autoclastic) typically
form irregular bodies tens to hundreds of meters in size,
which show gradational contacts against areas of similar
but more coherent target rocks. These lithic breccias are
composed entirely of rock and mineral fragments in a clastic
matrix of smaller, but similar, fragments. Fragments tend
to be angular to sharp, although fragments of softer rocks
like carbonates and shales may be well rounded. The breccias
themselves tend to be poorly sorted. The fragments are
derived from local target rocks, and the breccias may be
monomict or polymict, depending on the lithologic variety
present in the nearby target rocks. Distinctive shock-metamorphic
effects (e.g., PDFs in quartz) are generally absent
in the fragments. The breccias show no evidence of significant
transport, and they contain no exotic fragments or
glassy material.
These rocks often resemble breccias formed by more normal
geological mechanisms such as volcanic explosions or
tectonic movements, and their identification as impact products
is often difficult and uncertain. In general, the subcrater
regions of impact structures display highly localized and variable
deformation over short distances, a close association of
different kinds of breccias developed from basement rocks,
and the presence of allochthonous dike-like bodies of breccia
and melt. This variability in deformation and rock types
contrasts with the more uniform or gradational effects produced
by endogenic mechanisms. Even so, identification of
these rocks as impact breccias can generally not be done
directly, but depends on demonstrating their association with
more highly shocked rocks whose impact origin is clear (e.g.,
French et al., 1997).
5.3.4. Cross-Cutting (Allogenic) Breccias
Other bodies of breccia in the subcrater rocks contain
significant amounts of material that have clearly been introduced
into them from elsewhere, and they are therefore considered
here as allogenic breccias. These bodies tend to have
more regular shapes and to show sharp contacts and clear
cross-cutting relations against the subcrater rocks. Such breccias
often occur as distinctive breccia dikes, which typically
range from less than a meter to tens of meters in width and
may be as much as a kilometer long (Lambert, 1981; Bischoff
and Oskierski, 1987; Dressler and Sharpton, 1997). These bodies
contain fragments of target rock that are angular to
rounded and range in size from <1 mm to several meters.
These breccias tend to be polymict, with lithologically diverse
fragments, indicating mixing over distances of at least
several hundred meters. In addition, they frequently contain
Shock-Metamorphosed Rocks (Impactites) in Impact Structures 65
significant amounts of allogenic material, such as fragments
from even more distant rock units. This allogenic material is
frequently derived from more central regions of the crater,
often from above the present location of the dike, and it often
consists of distinctive highly shocked rock fragments
or melt.
A wide variety of such cross-cutting breccias has been
reported from several impact structures (Lambert, 1981;
Bischoff and Oskierski, 1987; Dressler and Sharpton, 1997):
(1) melt-free, typically polymict, lithic breccias with a clastic
matrix; (2) melt-fragment breccias containing fragments
of heterogeneous glass, rocks, and minerals in a clastic matrix;
(3) melt-matrix breccias (impact melt breccias), composed
of rock and mineral fragments in a matrix of glassy or
crystalline melt; (4) impact melt rocks, composed of glassy
or crystalline melt with few or no inclusions (e.g., Dence,
1971). Many of these dikes are similar to units of breccia or
melt in the crater-fill units above the crater floor, and they
may in fact be continuous with them (e.g., Lambert, 1981).
Subcrater breccia dikes often contain materials (e.g., rock
fragments or melt) that were originally located at higher
stratigraphic levels closer to the impact point, indicating that
the materials in the dikes have been emplaced downward
and/or outward into fractures that opened in the crater floor
during formation and modification of the crater. In many
structures, more than one generation of dikes occurs, with
later ones cutting earlier ones (Lambert, 1981; Dressler and
Sharpton, 1997). These relations indicate that, even during
the brief duration (seconds to minutes) of crater formation
and modification, a variety of distinct breccia types can be
generated and emplaced. However, in the crater environment,
cross-cutting relations between breccia bodies do not
imply the passage of significant amounts of time between
emplacements, a conclusion supported by the fact that the
cross-cutting relations between different types of breccia may
not be consistent from place to place within the whole structure
(Dressler and Sharpton, 1997).
5.3.5. Pseudotachylite
Pseudotachylite is an unusual, much-studied, and longdebated
type of impactite breccia that occurs in the parautochthonous
rocks of large impact structures (for recent
reviews, see Reimold, 1991, 1995; Spray, 1995). Pseudotachylite
is most strikingly developed at two large, ancient
impact structures: Vredefort (South Africa) (Shand, 1916;
Reimold, 1991; Reimold and Colliston, 1994) and Sudbury
(Canada) (Fairbairn and Robson, 1941; Speers, 1957; Dressler,
1984; Thompson and Spray, 1994; Spray and Thompson, 1995),
where it forms striking and extensive exposures (Figs. 5.1
and 5.2). The Vredefort pseudotachylite, first described more
than 80 years ago (Shand, 1916), typically occurs as abundant
irregular, anastomosing, and dike-like bodies that
contain numerous large and small rounded inclusions of
target rock set in a dense, aphanitic or crystalline matrix that
is generally black to blackish-green in color. Similar breccias,
although developed on a much smaller scale, have been
observed in other impact structures, e.g., Rochechouart
(France) (Reimold et al., 1987), Manicouagan (Canada)
(Dressler, 1990), and Slate Islands (Canada) (Dressler and
Sharpton, 1997).
At Sudbury and Vredefort, pseudotachylite is extensive.
Pseudotachylite exposures at Sudbury cover as much as 100–
200 km2, or a few percent of the total area of the structure.
Individual pseudotachylite bodies can also be large; the largest
body so far recognized at Sudbury is more than 11 km
long, more than 400 m wide, and contains discrete fragments
that are hundreds of meters in size (Dressler, 1984). In smaller
impact structures, pseudotachylite bodies are smaller and less
abundant; the material typically occurs as irregular dike-like
bodies less than a meter across.
The individual pseudotachylite bodies in impact structures
are not uniform over long distances and may change
size and shape radically within meters or tens of meters.
The more elongate dike-like bodies show little or no preferred
orientation in direction. The fragment/matrix ratio in
Fig. 5.1. Pseudotachylite in granitic gneisses. Pseudotachylite
exposure, showing rounded gneiss inclusions from a few centimeters
up to a few meters in size in a dense black matrix. The inclusions
show a significant amount of rotation relative to each other. Southwest
sector of the Vredefort structure (South Africa) (farm Samaria
484). Black pen on large inclusion in center (arrow) is 15 cm long;
inclusion itself is about 50 cm long. From Reimold and Colliston
(1994); photograph courtesy of W. U. Reimold.
66 Traces of Catastrophe
pseudotachylite bodies also varies significantly over short
distances, and some pseudotachylite breccias consist only of
fractured target rocks cut by thin veins of black matrix less
than a few millimeters wide. (The descriptive term “cobweb
breccias” has been used as a convenient field label for such
occurrences.)
Contacts between pseudotachylite bodies and the enclosing
target rock are irregular and generally not parallel on
opposite sides. Offsets of wallrock along pseudotachylite
bodies are uncommon, and observed displacements are minor
(e.g., <100 m). In very large pseudotachylite bodies with
large inclusions, the boundary between the breccia body and
the unbrecciated wallrock may not be clear. In such occurrences,
e.g., at Sudbury, the exact boundaries between breccia
and undisturbed wallrock may be difficult to establish
(Dressler, 1984).
Inclusions in pseudotachylite range from submicroscopic
to hundreds of meters in size. They invariably consist of local
bedrock, and there is generally no evidence for significant
long-distance (>100 m) transport of fragments during
formation. The inclusions are irregularly oriented, and outcrops
of the breccia give the strong impression of an overall
tensional or explosive environment (Figs. 5.1 and 5.2), rather
than the narrower compressional/shear environment that is
characteristic of zones of major thrust faulting (Philpotts,
1964; Sibson, 1975; Spray, 1995). Larger inclusions (>1 cm)
are generally rounded, while smaller ones tend to be angular
or sharp. Contacts between both large and small inclusions
and the surrounding matrix are generally sharp. However,
some inclusions may be deformed at the rims, forming a
flow structure that can be observed, both megascopically
and microscopically, to grade into the surrounding matrix
(Fig. 5.3).
The matrix between larger rock fragments is dense and
coherent. In hand specimen, the matrix often shows a conchoidal
or hackly texture on broken surfaces. The color is
commonly black to blackish green on fresh surfaces, although
the color may vary slightly with the host rock involved. The
matrix occurs in a wide variety of forms. It may cover large
(meter-sized) areas of inclusion-poor material, or it may form
tiny submillimeter filaments that penetrate bedrock and inclusions
and often terminate within them. In hand specimen
and thin section, the matrix is commonly structureless
(Fig. 5.4), but flow-banding is often observed, especially in
thin section (Fig. 5.3). This flow-banding may involve inclusions
that have been plastically deformed and possibly
melted (Fig. 5.5).
The matrix, generally aphanitic in hand specimen, is extremely
fine-grained and difficult to characterize, even in
thin section. In some samples, the matrix shows definite mi-
Fig. 5.2. Pseudotachylite; metamorphosed, in quartzite. Dark pseudotachylite (“Sudbury Breccia”) in Mississagi Quartzite on South
Range of Sudbury structure (Canada). Exposure shows large rounded blocks of quartzite in a pervasive black matrix (note penetration of
matrix into large quartzite block at lower right). Hammer (upper right) gives scale. Photograph courtesy of W. Peredery.
Shock-Metamorphosed Rocks (Impactites) in Impact Structures 67
Fig. 5.3. Pseudotachylite; flow-banded texture. Pseudotachylite (“Levack breccia”) in granitic gneisses from the North Range of the
Sudbury structure (Canada). In thin section, the black pseudotachylite matrix material consists of small irregular rock and mineral
inclusions in a dark microcrystalline to aphanitic groundmass. Numerous inclusions (white) show plastic deformation and alignment to
form a flow structure; note concentric deformation of the flow structure around larger inclusions (e.g., top right). Thin vertical white lines
are filled hairline fractures in the specimen. Sample CSF-67-53 (plane-polarized light).
Fig. 5.4. Pseudotachylite; structureless matrix. Pseudotachylite from Vredefort (South Africa), showing typical irregular to rounded
inclusions, ranging in size from <100 μm to several millimeters, in a dark aphanitic groundmass. Inclusions, which are rock and mineral
fragments from granitic gneisses, show sharp contacts with the matrix. In this pseudotachylite sample, the matrix is structureless, and the
inclusions show no deformation, preferred orientation, or other flow structures. Sample AV-81-53 (plane-polarized light).
1 mm
1 mm
68 Traces of Catastrophe
crocrystalline melt textures at SEM or microscopic scales
(Fig. 5.6). This characteristic, i.e., a matrix of igneous melt,
has been proposed (but not unanimously accepted) as a distinguishing
feature of pseudotachylite breccias (Spray, 1995).
In other samples, the matrix appears to consist of small fragments
in a cataclastic texture, and distinguishing between
the two types is a difficult process with important implications
for both classification and origin (Reimold, 1995).
Chemical studies of pseudotachylites (e.g., Dressler, 1984;
Reimold, 1991) have shown that they correspond closely to
Fig. 5.5. Pseudotachylite; extensive melting and flow. Pseudotachylite (“Levack Breccia”) from granitic gneisses in the North Range
of the Sudbury structure (Canada). The pseudotachylite consists of a heterogeneous mixture of plastically deformed and possibly melted
wallrock fragments (light-colored), mixed with discontinuous areas of more typical pseudotachylite material (dark) consisting of small
rock and mineral fragments in a fine black matrix. Sample CSF-88-2A (plane-polarized light).
Fig. 5.6. Pseudotachylite; igneous matrix with microlites. Black pseudotachylite developed in central granitic gneisses at Vredefort
structure (South Africa), consisting of small, irregular, generally rounded rock and mineral fragments in a black, finely crystalline matrix.
Matrix shows igneous flow-banding, expressed by alignment of small feldspar microlites typically 50–100 μm long. The microlites are
often concentrically aligned around larger inclusions. Sample AV81-52A (plane-polarized light).
1 mm
0.5 mm
Shock-Metamorphosed Rocks (Impactites) in Impact Structures 69
the adjacent host rocks, indicating that they have formed
essentially in place by locally generated cataclastic milling
and/or frictional melting processes.
Controversy and debate over the characteristics, terminology,
and origin of pseudotachylite has existed ever since
the term was first used (Shand, 1916) and continues actively
today (e.g., Spray, 1995; Reimold, 1995). Shand (1916,
pp. 188–189) deliberately coined the word “pseudotachylite”
to distinguish the Vredefort material from tachylite (basaltic
glass) and also from highly crushed and melted materials
formed tectonically along major faults (“flinty crush-rock,”
ultramylonite, hyalomylonite, etc.). Unfortunately, Shand’s
term has since been widely applied to the latter type of
material, so that it now designates similar glassy breccias
that are clearly tectonic in origin (Philpotts, 1964; Sibson,
1975; Reimold, 1995). Such breccias form in entirely different
environments and are the results of intense deformation
(including frictional melting) of rocks along the linear
trends of faults. They form in a compressional /shear regime,
but they can resemble impact-produced pseudotachylite,
including the presence of melted material in the matrix
(Philpotts, 1964).
Recently, some workers have suggested that impactproduced
pseudotachylites are formed in the same way as
tectonic ones, i.e., by frictional heating during the rapid
movements of late-stage crater development and modification
(e.g., Thompson and Spray, 1994; Spray, 1995, 1997; Spray
and Thompson, 1995). In this view, impact-produced
pseudotachylites have essentially the same frictional-melt
origin as tectonic ones. One possible way to distinguish between
them may be size. Bodies of tectonic pseudotachylite
tend to be linear and less than a few meters wide (Sibson,
1975; Spray, 1995). Impact-produced pseudotachylites, at
least at Sudbury and Vredefort, form more irregular bodies,
some of which may reach tens to hundreds of meters in size
(Thompson and Spray, 1994; Spray and Thompson, 1995).
Another problem, even within the study of impact-produced
breccias, is that the term “pseudotachylite” has been
used to designate different types of impact-produced breccias
formed at different stages (and possibly by different
mechanisms) during crater formation (Martini, 1991;
Reimold, 1995; Dressler and Sharpton, 1997). One suggestion
(Martini, 1991) is to use the term “type A pseudotachylite”
to designate relatively rare, small, glassy veins,
typically less than a centimeter wide, that contain fragments
in a matrix of melted material, often accompanied by shockproduced
high-pressure mineral polymorphs such as coesite
and stishovite (Martini, 1991). Such veins are believed to
form during the early, higher-pressure, compressive stages
of shock-wave expansion. In contrast, the more abundant,
widespread, and more intensely studied material (called “type
B pseudotachylite”) is thought (Martini, 1991) to form later,
during crater modification and central uplift formation, probably
by friction generated by the rapid movement of large
volumes of target rock below the crater.
Pseudotachylite breccias (especially the more familiar “type
B” variety) are distinctive and recognizable at Vredefort and
Sudbury, but their wider use as unique indicators of impact
is complicated by several factors. First, since they form below
the original crater floor, they are found only in impact
structures that have been deeply enough eroded to expose
target rocks originally located beneath the crater, and
pseudotachylites are usually restricted to the central-uplift
regions of larger structures. Second, pseudotachylites resemble
rocks formed by nonimpact processes, and the
distinction is difficult unless definite preserved shockmetamorphic
effects can be found. The current confusion in
terminology and formation mechanisms, combined with the
scarcity of distinctive shock effects in many impact-produced
pseudotachylites, makes it difficult to use pseudotachylites
by themselves as unique indicators of impact structures.
Despite these problems, well-developed pseudotachylites
may still be a useful field tool for identifying possible
impact structures for more detailed study. Pseudotachylites
can be widespread in impact structures, and their distinctive
appearance can survive even high-grade metamorphism
(Fig. 5.2). The striking irregular and anastomosing character
of pseudotachylite bodies, their rounded inclusions (often
altered at the rims), their development over large areas,
and the frequent absence of a regular shape or of compressional
effects typical of similar fault-related breccias make
them a valuable field indicator of a possible impact structure,
and their discovery should be followed up with an intensive
search for more definite shock effects. In addition, melt-rich
pseudotachylite breccias in established impact structures
have proven valuable for determining the formation ages
of the structures themselves (Spray et al., 1995; Kelley and
Spray, 1997).
5.4. CRATER INTERIOR: CRATER-FILL
DEPOSITS (BRECCIAS AND
MELT ROCKS)
5.4.1. Formation Conditions
During the modification stage, material excavated from
various locations in the growing transient crater is deposited
within the final crater to form crater-fill deposits of breccia
and melt rock. These allogenic units consist of four main
components: (1) material ejected ballistically on steep or
near-vertical trajectories that impacts within the final crater;
(2) large and small bodies of impact melt that do not travel
beyond the rim of the final crater; (3) large and small fragments
of unshocked target rock that collapse from the
oversteepened walls and rim of the original transient crater;
(4) ejecta originally deposited near the transient crater rim
and caught up in the subsequent collapse.
As a result of these processes, the final crater is partially
filled with a complex mixture of rock fragments (shocked
and unshocked) together with bodies of impact melt. These
deposits consist mostly of crater-fill breccias, often accompanied
by discrete units of impact melt rocks. In small, bowlshaped,
simple craters, the various components tend to be
mixed together, and the final deposit may fill the crater to
70 Traces of Catastrophe
about half its depth. [This crater-fill unit is also called the
breccia lens because of its shape (Fig. 3.7).] In larger complex
structures, particularly those formed in crystalline target
rocks, the crater-fill rocks typically contain discrete units
of breccias and impact melts that form a large annular deposit
around the central uplift (Fig. 3.13).
Subsequent to formation of the crater and the deposition
of impact-produced crater-fill breccias, the structure may
be filled, and the breccias buried, by younger crater-fill sediments
deposited more slowly by the conventional processes
of erosion, transport, and deposition. These sediments not
only preserve the underlying impact-produced breccias, but,
because of their circular outcrop pattern and often anomalous
character, they may call attention to previously unsuspected
impact structures. In this section, the discussion and
the term “crater-fill deposits” are limited only to the impactproduced
breccias that fill the crater during and immediately
after formation and do not include any ordinary
sediments that may also be present.
Many of the individual fragments in the crater-fill deposits
have been derived from within the zone of crater excavation
(Fig. 3.4) and may be highly shocked. Much of the
target rock within the excavation zone is subjected to relatively
high shock pressures of about 5 GPa to >100 GPa.
The lowest pressures in this range are sufficient to shatter
and brecciate the target rocks extensively; at higher pressures,
the rocks are deformed and melted as well. Shocked
Fig. 5.7. Crater-fill breccias. Recent drill coring along the southern
flank of the Chicxulub structure (Mexico), has recovered impact
breccias and melt rocks only shallowly buried beneath the younger
carbonate sediments. This mosaic shows the sequence of diverse
crater-fill breccias retrieved from the UNAM-5 drill core located
near the village of Santa Elena in southern Yucatán, ~112 km from
the center of the basin. The core pieces are arranged so that each
represents 10 m of core. The top of the impact sequence (top of
picture) occurs at a depth of ~330 m below the surface and is
characterized by a 30-m interval of highly vesicular and pulverized
impact melt rock (M). The melt rock horizon is almost completely
altered to clay but contains abundant clasts of the target rock
assemblage. Below this horizon is a varicolored continuous unit of
suevite breccia (SB). As is typical of suevites, this unit has a clastic
matrix containing a substantial proportion of highly shocked and
melted clasts derived from lithologies that were originally deep
within the target assemblage. The upper 50 m of the UNAM-5
suevite (SB1) is characterized by abundant, centimeter-scale clasts
of vesicular melt rock, similar to that of the overlying melt horizon
but less altered. The middle 50 m of the suevite (SB2) is dominated
by larger clasts of shocked to partially melted silicate basement
rock showing abundant evidence of shock deformation. The matrix
of the lower section of suevite (SB3) is more melt-rich and contains
a greater proportion of centimeter-scale silicate clasts. Total depth
was reached at the UNAM-5 well while still in the suevite. Coin is
~3 cm in diameter. Photograph courtesy of V. L. Sharpton.
SB3 SB2 SB1 M
330 m
Shock-Metamorphosed Rocks (Impactites) in Impact Structures 71
rock fragments, derived from this zone and deposited in the
crater-fill breccias, have provided the best evidence for the
impact origin of numerous structures.
The crater-filling process is both rapid and chaotic, and
mixing of the different components is not complete. The
crater-fill deposits therefore contain a variety of distinctive
allogenic breccias and melt rocks (Fig. 5.7). The simple classification
used below is based on (1) fragment lithologies
(lithic vs. melt-fragment breccias; (2) nature of the matrix
(clastic vs. melt-matrix). (For more detailed discussions and
classifications, see, e.g., Stöffler et al., 1979; Taylor et al., 1991;
Stöffler and Grieve, 1994, 1996.)
5.4.2. Lithic Breccias (Allogenic)
Melt-free breccias (lithic breccias) form a common and
distinct lithology in both large and small impact structures
(Figs. 3.7 and 3.13). In small impact structures, e.g., Brent
(Canada) (Dence, 1968; Grieve and Cintala, 1981), lithic breccias
may form units hundreds of meters thick that extend
over much of the final crater. At the larger Ries Crater
(Germany), a distinctive allogenic polymict lithic breccia
[the Bunte (“colored”) Breccia] occurs beneath the overlying
melt-bearing suevite breccias both inside and outside
the crater (Hörz, 1982; Hörz et al., 1983), with a sharp contact
between the two units. In some impact structures, especially
those formed in carbonate target rocks, lithic breccias
may be the only type of crater-fill material present (Roddy,
1968; Reiff, 1977).
Lithic breccias consist of rock and mineral fragments in a
clastic matrix of finer-grained similar material (Fig. 5.8). The
breccias are poorly sorted; fragment sizes generally range from
<1 mm to tens of meters. Fragments are typically sharp to
angular in appearance. Unlike the lithic breccias found in
parautochthonous rocks, crater-fill lithic breccias are more
apt to be polymict because their fragments have been derived
from a wider region of the original target rocks. Because
most of the material in lithic breccias is derived from
less-shocked regions around the walls and rim of the transient
crater, distinctive shock effects are only rarely observed
in the fragments.
Within the crater-fill deposits, lithic breccias are often
associated, both horizontally and vertically, with units that
contain a melt component as discrete fragments or as a matrix
for lithic fragments. Breccias with a few percent or more
of a melt component are regarded as melt-bearing breccias,
but the transition between these breccia types appears continuous,
and no formal boundary has been established. Such
melt-bearing breccias typically form a smaller proportion of
the crater fill, perhaps 10–25 vol%, and the amount of melt
component they contain varies from a few percent to
>90 vol% (e.g., Hörz, 1982; Masaitis, 1983; von Engelhardt,
1990, 1997).
Two basically different types of melt-bearing breccias can
be distinguished. In melt-fragment breccias (suevites), the
melt component occurs as large (centimeter-sized) discrete
bodies; in melt-matrix breccias (impact melt breccias), the
melt forms a matrix for rock and mineral fragments (Stöffler
and Grieve, 1994, 1996).
5.4.3. Melt-Fragment Breccias (Allogenic) (Suevites)
Melt-fragment breccias (suevites, pronounced “SWAYvites”)
are composed of discrete fragments of rocks and minerals,
together with bodies of melt, in a clastic matrix of similar
but finer-grained materials. Many of the rock and mineral
Fig. 5.8. Crater-fill breccia; lithic breccia. Poorly sorted crater-fill lithic breccia composed of angular to sharp fragments of granitic
rocks and constituent minerals (quartz, feldspar, etc.) in a finer clastic matrix. Drill core sample from the Brent Crater (Canada). Photograph
courtesy of R. A. F. Grieve (cross-polarized light).
0.1 mm
72 Traces of Catastrophe
Fig. 5.9. Crater-fill breccia; suevite. Large hand specimen, about 45 cm long, of typical fresh suevite from the Ries Crater (Germany)
(Otting quarry). The specimen consists of irregular and contorted individual fragments of glass (dark), which show a roughly parallel
elongation, and crystalline rock fragments (light) in a fine clastic matrix. The glass fragments, which range up to 5 cm in size, are
composed of a mixture of rock and mineral fragments in heterogeneous, flow-banded glass. Photograph courtesy of D. Stöffler.
Fig. 5.10. Crater-fill breccia; suevite. Suevite breccia from Nicholson Lake (Canada), containing glass fragments (dark) with rock and
mineral clasts in a finer fragmental matrix. The glass fragments are heterogeneous mixtures of mineral clasts (light) in dark, flow-banded
glass. Photograph courtesy of M. R. Dence (plane-polarized light).
5 mm
Shock-Metamorphosed Rocks (Impactites) in Impact Structures 73
fragments are highly shocked, and these breccias often provide
the most distinctive evidence for a meteorite impact
origin of the structures in which they are found.
The term suevite was originally applied to melt-fragment
breccias from the type occurrence at the Ries Crater (Germany),
a relatively young (15 Ma) and well-preserved structure
24 km across, in which well-exposed suevites and other
impactites have been extensively studied and drilled (for reviews,
see von Engelhardt et al., 1969; von Engelhardt and
Graup, 1984; von Engelhardt, 1990, 1997). Suevite breccias
are found both inside the structure (crater suevite or fallback
suevite) and as preserved ejecta deposits (ejecta or fallout
suevite) as far as 40 km from the center of the Ries structure.
Suevite breccias from the Ries Crater and other impact
structures typically consist of large (centimeter-sized)
and smaller glassy bodies (typically 5–15 vol%), together
with rock and mineral clasts in a matrix of finer fragments
(Figs. 5.9 and 5.10). Glass-rich suevites are also known, in
which the glass fragments may make up >50 vol% of the
rock (Masaitis, 1994). Individual rock and glass fragments
typically range from a maximum size of 10–20 cm down to
submillimeter dimensions.
The glassy bodies in the fallout suevite beyond the Ries
Crater rim typically show irregular to contorted shapes and
textures (Hörz, 1965). These bodies are typically heterogeneous,
consisting of a polymict mixture of rock and mineral
clasts (frequently highly shocked or partially melted) in a
matrix of glass that may be compositionally heterogeneous
and often shows well-developed flow structure (Fig. 5.11).
At the Ries Crater, the larger (5–20 cm) glassy fragments
in the ejecta deposits outside the structure, called Fladen,
show a grooved and lobate flow structure that is evidence of
aerodynamic sculpturing during their flight through the
atmosphere (Hörz, 1965). These bodies also show brittle fractures
developed on landing, implying that they were solid
when they struck the ground. In contrast, glass bodies in the
crater suevite are smaller (normally <5 cm) and lack distinctive
sculpturing, implying that they did not travel through
the atmosphere for any significant length of time (Fig. 5.12)
(von Engelhardt and Graup, 1984; von Engelhardt, 1990).
Although the Ries suevites are the best-known and most
intensely studied examples of this rock type, impressive
suevite breccias have been recognized in many other impact
structures. However, in many of these structures, erosion has
largely removed the ejecta deposits outside the crater, and
the suevites occur only as crater-fill units, where they are
associated with, and often interbedded with, lithic breccias
and impact-melt rocks. Examples include Brent (Canada)
(Dence, 1965, 1968; Grieve, 1978); Rochechouart (France)
(Kraut and French, 1971); Popigai (Russia) (Masaitis et al.,
1980; Masaitis, 1994); Manson (Iowa) (Koeberl and Anderson,
1996a; Koeberl et al., 1996b); Gardnos (Norway) (French
et al., 1997); Slate Islands (Canada) (Dressler and Sharpton,
1997); and Roter Kamm (Namibia) (Reimold et al., 1997a).
The Onaping Formation, a complex and metamorphosed
Fig. 5.11. Crater-fill breccia; suevite; glassy inclusion. Heterogeneous, fragment-rich glassy fragment (Fladen) in suevite breccia from
Lake Mien (Sweden), showing complex, multiple layering with varying amounts of rock and mineral inclusions. The mineral inclusions
are typically sharp to angular and do not show the phenocryst shapes that are typically observed in glassy volcanic rocks. The generally
laminar flow-banding is emphasized by a sharp difference in clast content and by dark streaks that may represent decomposed and melted
opaque minerals. Note that flow-banding in the clast-rich layers (e.g., top) is more highly contorted. Sample NBS-61-0487 (planepolarized
light).
1 mm
74 Traces of Catastrophe
Fig. 5.12. Crater-fill breccia; suevite. Typical poorly sorted suevite
breccia in a core sample from the Nördlingen deep drill hole
(369.9 m depth), Ries Crater (Germany). The unit contains crystalline
rock fragments (light-colored) and glassy fragments (Fladen)
(dark) in a fine clastic matrix. Inclusion at upper left contains a
rock fragment (core) surrounded by a rim of flow-banded glass.
Specimen is 10 cm wide. Photograph courtesy of H. Newsom.
crater-fill unit at the 1.85-Ga Sudbury (Canada) impact
structure, contains the oldest suevite unit identified so far
(Fig. 5.13) (French, 1968b; Muir and Peredery, 1984;
Avermann, 1994).
Because of their high melt content and the occurrence of
individual glassy bodies, suevite breccias resemble conventional
volcanic breccias, and the suevite from the Ries Crater
was considered to be a volcanic tuff for nearly two
centuries. However, suevites differ from volcanic breccias in
several ways, both in hand specimen and microscopically.
Fragments in suevites show no volcanic textures; such typical
volcanic features as feldspar phenocrysts or corroded
quartz phenocrysts are absent (Figs. 5.10, 5.11, 5.14, and
5.15). Rock fragments in suevites are not deep-seated volcanic
xenoliths but are derived entirely from the underlying
shallow target rocks. Suevites often contain cored inclusions,
composite fragments in which a rim of glass is wrapped
around a fragment of basement rock, indicating that both
rock and melt were ejected into the air at the same time
(Figs. 5.16, 5.17, and 5.18). Most convincing is the presence
of unique high-pressure shock-metamorphic effects
(such as PDFs in quartz or the high-pressure minerals coesite
and stishovite), in rock and mineral inclusions in the suevite.
High-temperature melting effects, e.g., the formation of silica
glass (lechatelierite) from quartz, may also be present in the
glass fragments in suevite.
Despite their widespread distribution, suevite breccias are
not found in all meteorite impact structures. In some cases,
their absence is probably due to erosion, which has removed
these near-surface deposits from the structure. However, the
nature of the target rocks also seems important in determining
whether suevites are formed (Kieffer and Simonds, 1980;
Grieve and Cintala, 1992). Suevites have so far been observed
only in impact structures formed largely or entirely in crystalline
silicate rocks, possibly because these rocks melt to
produce cohererent and durable bodies of glass. No suevite
deposits have yet been found in impact structures formed in
carbonate rocks, in which decarbonation and volatile loss,
rather than melting, would be important.
5.4.4. Melt-Matrix Breccias (Impact-Melt Breccias)
Suevites inside the crater are closely associated with a different
type of melt-bearing breccia: melt-matrix breccias
or impact-melt breccias. In these units, the melt occurs, not
as individual fragments, but as a matrix that typically makes
up 25–75 vol% of the rock and may range from glassy material
to completely crystalline igneous rock. The fragments,
which consist of target rocks and minerals, are frequently
shocked or melted.
Impact-melt breccias form distinct bodies of widely varying
size, from small glassy inclusions in suevite breccias to
distinct dike-like and sill-like units tens to hundreds of meters
thick. As the melt component increases, impact-melt breccias
grade into impact melt rocks (see Chapter 6), in which
the melt component is dominant and the included fragments
are minor or entirely absent. These rocks often have the appearance
of conventional igneous rocks.
5.5. CRATER RIM ZONE AND PROXIMAL
EJECTA DEPOSITS
The region near the rim of the transient crater is subjected
to relatively low shock pressures (typically <1–2 GPa
in smaller structures; Fig. 3.4) (Kieffer and Simonds, 1980).
These pressures are high enough to fracture and brecciate
target rocks but are too low to produce unique shock-deformation
features in them. The dominant effects in this region
are related to the excavation of the crater and the ejection
of material from it. In simple craters, which are only slightly
larger than the original transient crater, the rim is characterized
by structural uplift (and even overturning) of the target
rocks that occurs during development of the original transient
crater (Fig. 3.3). Even though much of this original
transient crater rim may collapse into the final crater during
modification, significant uplift may be preserved, especially
in smaller and younger craters (e.g., Shoemaker, 1963; Roddy
et al., 1975; Roddy, 1978). Such rim uplift and overturnShock-
Metamorphosed Rocks (Impactites) in Impact Structures 75
Fig. 5.13. Crater-fill breccia; suevite, metamorphosed. Typical exposure of Onaping Formation “Black Member,” showing centimetersized
fragments of rock fragments and contorted recrystallized glassy inclusions in a black fragmental matrix. Despite color differences,
the unit has a strong resemblance to fresh suevite from the Ries Crater (Germany) (see Fig. 5.9). Exposure located at “Black Member”
type locality at Onaping Falls (Highway 144, Dowling Township) in the northwestern part of the Sudbury structure (Canada). Diameter
of coin near large glassy inclusion is about 2 cm. Photograph courtesy of J. Guy-Bray.
Fig. 5.14. Crater-fill breccia; suevite, heterogeneous glasses. Complex heterogeneous glassy breccia from West Clearwater Lake
(Canada), composed of distinct areas of light- and dark-colored mixed glasses, which show short-range turbulent flow and mixing.
The glassy areas contain abundant small rock and mineral fragments. Photograph courtesy of M. R. Dence (plane-polarized light).
0.1 mm
76 Traces of Catastrophe
Fig. 5.15. Crater-fill breccia; suevite, metamorphosed. Heterogeneous glassy breccia consisting of fragments of recrystallized glass,
together with rock and mineral fragments, in a fine opaque carbon-bearing matrix. Despite greenschist-level metamorphism, the glassy
fragments still preserve original melt textures such as flow banding and vesicles (now filled with chlorite; gray). Many of the fragments
display sharp crosscutting fractures, indicating that they were cool and brittle when deposited. The rock and mineral clasts represent
broken basement (target) rocks; no typical volcanic textures (phenocrysts, etc.) are observed. Discrete fragments as small as 5 μm across
can be distinguished in the opaque matrix. Onaping Formation “Black Member,” from type locality at Onaping Falls (Highway 144,
Dowling Township), northwestern corner of Sudbury structure (Canada). Sample CSF-66-36-1 (plane-polarized light).
Fig. 5.16. Crater-fill breccia; suevite, “cored” inclusion. Large flow-banded fragment (about 15 cm long) from a larger glassy inclusion
in the suevite unit of the Ries Crater (Germany) (Bollstadt quarry). The specimen is a composite or “cored” inclusion containing a large
block of shocked and fractured crystalline rock (light) surrounded by dark, flow-banded glass. Photograph courtesy of F. Hörz.
0.5 mm
Shock-Metamorphosed Rocks (Impactites) in Impact Structures 77
Fig. 5.17. Crater-fill breccia; suevite, “cored” inclusion. Composite (cored) inclusion in Onaping Formation “Black Member” in
northwestern corner of Sudbury structure (Canada). Inclusion consists of a core fragment of crystalline granitic rock (light-colored)
surrounded by flow-banded glassy material, now recrystallized. Similar inclusions are observed in fresher suevite deposits, e.g., at the Ries
Crater (Germany) (see Figs. 5.12 and 5.16). A separate angular granitic fragment appears at lower right. Coin at left of inclusion is about
2 cm in diameter. Exposure located at “Black Member” type locality at Onaping Falls (Highway 144, Dowling Township). Photograph
courtesy of J. Guy-Bray.
Fig. 5.18. Crater-fill breccia; suevite, “cored” inclusion. Composite rock fragment in metamorphosed suevite unit. The fragment contains
a core of fine-grained granitic basement rock surrounded by a rim of microcrystalline recrystallized glass. The fragment is associated
with smaller individual clasts of glassy material and rock and mineral fragments in a black, opaque, carbon-bearing matrix. Onaping
Formation “Black Member,” from type locality at Onaping Falls (Highway 144, Dowling Township), northwestern corner of Sudbury
structure (Canada). Sample CSF-66-36-2 (cross-polarized light).
0.1 mm
78 Traces of Catastrophe
ing are only rarely observed in volcanic explosion structures
such as maars and diatremes, and the presence of such rim
deformation provides a strong indication of an impact origin
for a structure.
In a newly formed crater the rim and the surrounding
region are generally covered with allogenic ejecta ejected from
the growing transient crater (Melosh, 1989; Chapter 6). Two
kinds of ejecta deposits can be distinguished: those deposited
near the crater (proximal ejecta) and those distant from
the crater (distal ejecta).
Most of the material ejected beyond the crater rim is deposited
near the crater (Melosh, 1989, p. 90). In terms of
crater radius (Rc, the distance from the center of the crater
to the final rim), approximately half the ejecta is deposited
within 2 Rc from the center (or 1 Rc from the rim) to form a
continuous ejecta blanket that may be tens to hundreds of
meters thick, depending on the size of the crater. At greater
distances, the ejecta unit becomes thinner and increasingly
discontinuous; most of the ejecta (>90%) is deposited within
about 5 Rc. (This value may serve as an arbitrary boundary
between proximal and distal ejecta.) Because many of the
fragments in the ejecta deposits were originally close to the
impact point, they are often distinctively shocked and melted.
Ejecta blankets, where they are preserved, may therefore
provide the best and most accessible evidence for an impact
origin of the structure.
Ejecta deposits around impact craters are not homogeneous,
but are made up of distinct lithologic units derived
from different regions of the transient crater and transported
by different mechanisms to the site of deposition. Mixing
during the ejection and deposition process is not complete,
and the ejecta deposits that surround a crater contain the
same diversity of rock types that are found as crater fill within
the structure: lithic breccias, suevites, and impact melt rocks.
In large impact structures, the ejecta deposits preserved outside
the crater contain a recognizable sequence of different
lithologies. The sequence at the Ries Crater (Germany) (see
von Engelhardt, 1990, 1997, and references therein) contains
a lower unit of polymict lithic melt-free breccia (Bunte Breccia)
overlain by melt-bearing breccia (suevite). Some of the
ejecta at the Ries also occurs as large (tens to hundreds of
meters in size) limestone blocks ejected intact from the crater
and skidded for many kilometers across the surrounding
ground surface (von Engelhardt, 1990, pp. 264–265).
In impact structures formed on land, the near-surface
regions are quickly removed by erosion, and the distinctive
rim uplift and ejecta deposits are observed only at relatively
young structures such as the Barringer Meteor Crater (Arizona)
(age 50 ka) (Shoemaker, 1963) and the Ries Crater
(Germany) (age 15 Ma) (von Engelhardt, 1990). At older
structures (e.g., Dence, 1965, 1968), distinctively shocked
rocks tend to be preserved in only two areas: in the target
rocks immediately beneath the crater floor, and in the breccia
and melt deposits that fill the crater itself.
5.6. DISTAL EJECTA
Although most of the material (about 90 vol%) ejected
from the crater is deposited relatively close (<5 Rc) to the
crater (Melosh, 1989, p. 90), a significant amount (about
10 vol%) may travel to even greater distances (>5 Rc) to form
deposits of distal ejecta. Where an atmosphere is present, as
in terrestrial impact events, a combination of disruption of
the atmosphere by the impact fireball, ballistic ejection from
the crater, and subsequent atmospheric transport can distribute
the smaller ejecta particles (typically >1 mm) to regional
or even global distances (Alvarez et al., 1995). The
resulting deposits, usually less than a few centimeters thick,
may contain distinctive evidence for impact: shocked rock
and mineral fragments, distinctive chemical and isotopic signatures,
and unusual glassy objects. It has thus become possible
to recognize debris from a given impact structure over
a large area of Earth, and even to establish the existence of a
major impact event from a globally distributed ejecta layer
before the structure itself could be located.
Although few layers of distal ejecta have been identified,
they have been critical to recognizing large impact structures
and determining their age. Coarse ejecta (millimeterto
centimeter-sized fragments) from the Acraman structure
(Australia) (D = 90 km) has been recognized as a discrete
layer several centimeters thick at distances of 300–400 km
from the site (Gostin et al., 1986; Williams, 1986). Ejecta
from the Manson structure (Iowa) (D = 36 km) has been
recognized more than 250 km away (Izett et al., 1993). The
most striking and best-known example of distal ejecta is the
thin layer of material ejected from the Chicxulub structure
(Mexico) and distributed worldwide to form the K/T boundary
layer (Alvarez et al., 1980; papers in Sharpton and Ward,
1990, and in Ryder et al., 1996). The occurrence in this layer
of shocked quartz grains and small spherules of melted target
rock, accompanied by an anomalously high content of
the element iridium (derived from the projectile), provided
conclusive evidence that a large meteorite impact had occurred
at the end of the Cretaceous Period, even before the
Chicxulub impact structure itself was identified. The layer
also provided key geochemical and geochronological evidence
to demonstrate that the Chicxulub structure was identical in
age to the K/T boundary and that it was also the source for
the global ejecta layer itself.
Generally, ejecta found at greater distances from the crater
displays a higher level of shock effects, and much distal
ejecta consists of small fragments of melted target rock. One
peculiar and much-studied variety of distal ejecta is tektites
and microtektites, small (centimeter- to millimeter-sized)
bodies of pure glass that have been ejected from a few impact
structures and spread over areas (strewnfields) that may
be thousands of kilometers in extent (see Chapter 6).