The Stac Fada impact ejecta deposit and the Lairg Gravity Low: Evidence for a buried Precambrian impact crater in Scotland?

Abstract

The Stac Fada Member, an impact ejecta deposit within the Mesoproterozoic Stoer Group, is represented today by just a narrow outcrop, truncated by faulting and erosion, extending for 50 km north-south along the coast of north-west Scotland. It appears to represent a non-erosive Single Layer Ejecta deposit rather than the erosively emplaced Double Layer Ejecta deposits characteristic of terrestrial impact craters and it is unique in preserving spallation debris, ejected very early in the impact process, beneath the ejecta blanket. Various sedimentary structures associated with the Stac Fada Member, from ejecta intrusions along bedding planes immediately beneath it, to erosional troughs eroded into its top, consistently indicate emplacement from the east. No surface manifestation of an impact crater has been identified but there is a remarkable correspondence between its location, as inferred from these directional data, and the position of the Lairg Gravity Low, an ∼40 km diameter geophysical anomaly centred more than 50 km east of the closest point on the Stoer Group outcrop. Proximal-distal facies changes along the outcrop of the ejecta deposit are consistent with this inferred relationship between the ejecta deposit and the gravity low. Post-impact drainage reconfiguration suggests a regional isostatic doming in response to excavation of the crater that also appears to be centred on the Lairg Gravity Low. Comparison with gravity data from impact craters elsewhere suggests that the Lairg Gravity Low represents a complex crater at least 40 km in diameter that now lies buried beneath the Moine Thrust complex.

Full text

The
Stac
Fada
impact
ejecta
deposit
and
the
Lairg
Gravity
Low:
evidence
for
a
buried
Precambrian
impact
crater
in
Scotland?
Michael
J.
Simms
Department
of
Natural
Sciences,
National
Museums
Northern
Ireland,
Cultra,
BT18
0EU
Northern
Ireland,
UK
1.
Introduction
Almost
200
terrestrial
impact
craters
are
currently
known
worldwide
with
perhaps
twice
as
many
more
still
to
be
discovered
(Hergarten
and
Kenkmann,
2015),
yet
none
have
been
reported
from
the
onshore
UK
despite
its
exceptionally
diverse
geology
spanning
almost
three
billion
years
and
a
history
of
intensive
geological
research
stretching
back
more
than
two
centuries.
Less
direct
evidence
of
impacts,
in
the
form
of
ejecta
deposits,
has
been
identified
in
the
UK
only
recently
and
at
just
two
stratigraphic
levels.
A
thin,
and
geographically
highly
localized,
horizon
of
reworked
microtektites
occurs
in
the
late
Triassic
Mercia
Mudstone
Group
of
south-west
England
(Walkden
et
al.,
2002;
Kirkham,
2003)
but
these
are
distal
ejecta
that
demonstrably
originate
from
the
late
Triassic
Manicouagan
impact
in
eastern
Canada
(Thackrey
et
al.,
2009)
rather
than
from
a
more
proximal
source
within
the
UK.
The
second
example,
the
Stac
Fada
Member
in
the
Mesoproterozoic
Stoer
Group
of
north-west
Scotland,
has
been
known
for
a
considerably
longer
time
(Lawson,
1972)
but
was
regarded
as
volcaniclastic
in
origin
and
recognized
as
impact
ejecta
only
recently
(Amor
et
al.,
2008).
Just
a
narrow
strip
of
this
ancient
ejecta
blanket
survives
along
the
north-
west
coast
of
Scotland,
truncated
down-dip
to
the
west
by
faulting
and
up-dip
to
the
east
by
erosion.
It
is
underlain
by
undisturbed
strata
and
hence
must
lie
significantly
beyond
the
crater
rim
but
its
thickness
(up
to
12
m)
and
relative
continuity
along
a
50
km
outcrop
implies
a
proximal
source.
No
surface
manifestation
of
any
associated
impact
crater
has
been
identified,
suggesting
that
the
crater
itself
may
have
been
either
destroyed
by
erosion
or
buried
beneath
younger
strata.
Amor
et
al.
(2008,
2011)
inferred
from
features
within
the
Stac
Fada
Member
that
the
crater
was
located
to
the
west
of
the
present
outcrop,
deeply
buried
beneath
younger
strata
in
the
Minch
Basin,
but
they
provided
no
direct
evidence
in
support
of
this
location.
However,
new
field
observations
of
the
Stoer
Group
described
here
and
a
re-evaluation
of
published
geological
and
geophysical
data
indicates
instead
that
the
impact
was
to
the
east
of
the
Stoer
Group
outcrop
and
suggests
that
the
crater
may
still
survive
beneath
northern
Scotland.
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
A
R
T
I
C
L
E
I
N
F
O
Article
history:
Received
8
December
2014
Received
in
revised
form
27
August
2015
Accepted
28
August
2015
Available
online
2
October
2015
Keywords:
Mesoproterozoic
Torridonian
Meteorite
impact
Bouguer
anomaly
Moine
Thrust
A
B
S
T
R
A
C
T
The
Stac
Fada
Member,
an
impact
ejecta
deposit
within
the
Mesoproterozoic
Stoer
Group,
is
represented
today
by
just
a
narrow
outcrop,
truncated
by
faulting
and
erosion,
extending
for
50
km
north-south
along
the
coast
of
north-west
Scotland.
It
appears
to
represent
a
non-erosive
Single
Layer
Ejecta
deposit
rather
than
the
erosively
emplaced
Double
Layer
Ejecta
deposits
characteristic
of
terrestrial
impact
craters
and
it
is
unique
in
preserving
spallation
debris,
ejected
very
early
in
the
impact
process,
beneath
the
ejecta
blanket.
Various
sedimentary
structures
associated
with
the
Stac
Fada
Member,
from
ejecta
intrusions
along
bedding
planes
immediately
beneath
it,
to
erosional
troughs
eroded
into
its
top,
consistently
indicate
emplacement
from
the
east.
No
surface
manifestation
of
an
impact
crater
has
been
identified
but
there
is
a
remarkable
correspondence
between
its
location,
as
inferred
from
these
directional
data,
and
the
position
of
the
Lairg
Gravity
Low,
an
40
km
diameter
geophysical
anomaly
centred
more
than
50
km
east
of
the
closest
point
on
the
Stoer
Group
outcrop.
Proximal-distal
facies
changes
along
the
outcrop
of
the
ejecta
deposit
are
consistent
with
this
inferred
relationship
between
the
ejecta
deposit
and
the
gravity
low.
Post-impact
drainage
reconfiguration
suggests
a
regional
isostatic
doming
in
response
to
excavation
of
the
crater
that
also
appears
to
be
centred
on
the
Lairg
Gravity
Low.
Comparison
with
gravity
data
from
impact
craters
elsewhere
suggests
that
the
Lairg
Gravity
Low
represents
a
complex
crater
at
least
40
km
in
diameter
that
now
lies
buried
beneath
the
Moine
Thrust
complex.
ß
2015
The
Geologists’
Association.
Published
by
Elsevier
Ltd.
All
rights
reserved.
E-mail
address:
michael.simms@nmni.com.
Contents
lists
available
at
ScienceDirect
Proceedings
of
the
Geologists’
Association
jo
ur
n
al
ho
m
ep
ag
e:
www
.els
evier
.c
om
/lo
cat
e/p
g
eo
la
http://dx.doi.org/10.1016/j.pgeola.2015.08.010
0016-7878/ß
2015
The
Geologists’
Association.
Published
by
Elsevier
Ltd.
All
rights
reserved.

2.
The
Stoer
Group
The
Stac
Fada
Member
lies
within
the
Stoer
Group,
which
today
is
restricted
to
a
narrow
outcrop
just
a
few
kilometres
wide
but
extending
north-south
for
50
km
along
the
north-west
coast
of
Scotland
(Fig.
1).
Despite
their
great
age
of
1.2
Ga
(Turnbull
et
al.,
1996;
Darabi
and
Piper,
2004;
Parnell
et
al.,
2011),
these
non-
marine
clastic
rocks
have
experienced
remarkably
little
deforma-
tion,
with
a
fairly
uniform
dip
of
20–308
to
the
west
and
a
minimal
degree
of
metamorphism
(Stewart,
2002).
The
Stoer
Group
is
up
to
2
km
thick
and
today
occupies
a
north-south
half-graben
bounded
to
the
west
by
the
Coigach
Fault
and
truncated
by
erosion
up-dip
to
the
east,
where
it
is
overlain
unconformably
by
the
Torridon
Group
of
Neoproterozoic
age
(Fig.
2).
Sediment-filled
fractures
in
the
underlying
basement
indicate
synsedimentary
faulting
during
deposition
of
the
Stoer
Group
(Beacom
et
al.,
1999;
Dulin
et
al.,
2005)
and
it
has
been
inferred
that
during
Mesoproterozoic
times
the
Stoer
Group
was
deposited
in
an
active
Fig.
2.
Broad-scale
lithostratigraphy
and
palaeocurrent
data
along
the
outcrop
of
the
Stoer
Group
(adapted
from
Stewart,
2002,
figs
4
and
17).
Location
key
as
for
Fig.
1.
CF,
Clachtoll
Formation;
SFM,
Stac
Fada
Member;
PMM,
Poll
a
`Mhuilt
Member;
MDF,
Meall
Dearg
Formation.
The
dashed
line
within
the
Clachtoll
Formation
represents
the
boundary
drawn
by
Stewart
(2002)
between
his
facies
Ct1–Ct8
and
facies
BS1,
formerly
part
of
the
Bay
of
Stoer
Formation.
Fig.
1.
Regional
geology
of
north-west
Scotland.
Locations
referred
to
in
the
text
are:
CB,
Cnoc
Breac;
St,
Stoer;
EB,
Enard
Bay;
Ac,
Achiltibuie;
CH,
Cailleach
Head;
SP,
Stattic
Point
(3
sites);
SC,
Second
Coast;
LT,
Loch
Thurnaig
(2
sites);
Pe,
Poolewe;
BL,
Bac
an
Leth-choin.
Inset
map
shows
the
location
(starred)
of
the
region
encompassed
by
the
main
figure.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
743

rift
valley
(Stewart,
2002),
but
the
evidence
for
active
rifting
is
equivocal
as
no
basin-bounding
extensional
faults
have
yet
been
identified
at
outcrop
(Stewart,
2002;
Kinnaird
et
al.,
2007).
The
basin
fill
is
dominated
by
fluvial
facies
with
subordinate
lacustrine
units
and
overlies
Lewisian
(Archean)
basement
with
a
topographic
relief
of
several
hundred
metres.
Breccias
and
conglomerates
derived
from
the
local
rocks
occur
adjacent
to
the
basement
subcrop
(Fig.
3).
Stewart
(2002)
recognized
three
main
lithostratigraphic
units
within
the
Stoer
Group;
the
Clachtoll,
Bay
of
Stoer
and
Meall
Dearg
formations.
He
further
divided
the
Bay
of
Stoer
Formation
into
three
members;
a
lower
un-named
member,
the
Stac
Fada
Member,
and
the
Poll
a
`Mhuilt
Member
at
the
top.
The
marginal
facies
of
the
Clachtoll
Formation
is
very
variable,
encompassing
breccias,
conglomerates
and
sandstones
(facies
Ct1
to
Ct5
of
Stewart,
2002).
The
breccias
and
conglom-
erates
invariably
are
framework-supported,
with
clasts
mainly
subrounded
in
shape
and
seldom
more
than
half
a
metre
across.
Decimetre
to
metre-scale
blocks
are
rarely
found
more
than
a
few
metres
from
basement
outcrops
(Stewart,
2002).
These
facies
pass
upwards
and
laterally
into
more
persistent
planar
and
trough
cross-bedded
sandstones
with
minor
mudstone
and
aeolian
sand
facies
(facies
Ct6
to
Ct8)
that
also
encompasses
the
un-named
lower
member
of
the
Bay
of
Stoer
Formation
(facies
BS1).
Stewart
(2002)
interpreted
the
marginal
facies
of
the
Clachtoll
Formation
as
having
been
deposited
in,
or
adjacent
to,
valleys
incised
into
the
Lewisian
basement.
Accordingly
the
palaeocurrent
data
are
quite
variable
and
reflect
local
topography
rather
than
a
regional
drainage
pattern
(Fig.
2).
The
more
laterally
extensive
basinal
facies,
which
Stewart
(2002)
assigned
to
the
lower
part
of
the
Bay
of
Stoer
Formation,
were
deposited
largely
on
low
gradient
fluvial
plains.
Palaeocurrent
azimuths
in
these
basinal
facies
are
more
uniform;
mainly
to
the
northwest
in
the
southern
part
of
the
outcrop
and
predominantly
eastwards
at
its
northern
end
(Fig.
2).
The
marginal
facies
of
the
Clachtoll
Formation
(facies
Ct1–Ct5)
contain
pebbles
that
are
locally
derived,
mainly
felsic
gneiss
and
minor
mafics
from
the
Scourie
dykes,
but
the
trough
cross-bedded
sandstones
in
the
lower
member
of
the
Bay
of
Stoer
Formation
(facies
BS1)
also
contain
well-rounded
pebbles
of
other
durable
lithologies
including
metaquartzites,
silica-cemented
arkoses,
and
fuchsite-bearing
sandstones.
Stewart
(2002)
noted
that
this
was
the
only
significant
difference
between
facies
Ct5
and
facies
BS1.
The
Stac
Fada
Member
is
markedly
different
from
the
strata
above
and
below
and
is
the
most
distinctive
unit
within
the
entire
Stoer
Group
succession.
Its
thickness
varies
from
as
little
as
2
m
up
to
about
12
m
(Fig.
3)
and
can
vary
laterally
by
a
factor
of
almost
two
over
distances
of
just
a
few
tens
of
metres,
but
otherwise
it
is
remarkably
constant
along
the
outcrop
compared
with
contiguous
strata.
There
is
some
lateral
variation
in
facies
but
typically
it
is
dominated
by
a
massive
or
indistinctly
bedded,
red-brown,
poorly
sorted,
muddy
arkosic
sand
matrix
containing
typically
20–30%
by
volume
of
centimetre-scale,
angular
and
vesicular,
green
to
grey,
matrix-supported,
chloritic
clasts
of
devitrified
mafic
melt.
The
matrix
is
a
mixture
of
grains
of
quartz,
feldspar
(plagioclase,
orthoclase
and
microcline),
epidote,
mica,
chlorite
and
opaques
(Sanders
and
Johnston,
1989),
together
with
hematite-coated
clay
and
silt,
and
fragments
of
Lewisian
gneiss
and
occasional
mafics
(Lawson,
1972).
These
mineral
grains
vary
from
fresh
and
angular
to
rounded
and
weathered
but
their
composition
is
broadly
consistent
with
a
provenance
directly
from
the
Lewisian
basement
and
associated
mafic
dykes,
or
from
sediments
similar
to
those
of
the
Stoer
Group.
Larger
extraformational
clasts,
of
Lewisian
gneiss
and
more
rarely
quartzite,
and
intraformational
clasts
of
sand-
stone,
are
locally
common.
The
melt
clasts,
described
in
detail
by
Lawson
(1972),
Sanders
and
Johnston
(1989),
Stewart
(2002)
and
Young
(2002),
have
an
overall
geochemistry
indicating
a
mafic
source
comparable
with
basalt
(Lawson,
1972;
Stewart,
2002)
or
gabbro
(Young,
2002).
Spherical,
ovoid
or
tubular
vesicles
are
common
in
most
of
the
melt
fragments,
with
some
melt
clasts
even
described
by
Lawson
(1972)
as
pumiceous.
Most
melt
clasts
contain
abundant
crystal
inclusions,
from
a
few
m
m
to
several
mm
across,
that
may
comprise
more
than
50%
of
the
volume
of
a
melt
clast.
These
inclusions,
many
with
corroded
edges,
are
predomi-
nantly
of
quartz,
but
orthoclase,
plagioclase
and
microcline
inclusions
are
also
common.
There
is
a
conspicuous
absence
of
any
mafic
mineral
inclusions
despite
the
mafic
composition
of
the
melt
itself.
Pink
to
orange
authigenic
K-feldspar
is
ubiquitous
lining
or
filling
cavities
and
pore
spaces
within
the
Stac
Fada
Member.
Feldspar-lined
sub-vertical
elutriation
pipes,
sometimes
several
metres
in
length,
are
present
at
most
sites
along
the
outcrop
and
are
particularly
conspicuous
at
Stac
Fada
and
Stattic
Point
(Lawson,
1972;
Sanders
and
Johnston,
1989;
Amor
et
al.,
2008).
Fluid
inclusion
analysis
of
the
feldspar
indicates
a
consistent
deposition
temperature
for
the
feldspar
of
200
8C
along
the
entire
outcrop
of
the
Stac
Fada
Member
while
Ar–Ar
dating
gives
a
radiometric
age
for
it
of
1178

9
Ma
(Parnell
et
al.,
2011).
Accretionary
lapilli
occur
in
the
upper
part
of
the
member
at
three
northern
sites
but
are
absent
from
any
of
the
sites
south
of
Enard
Bay.
The
Stac
Fada
Member
is
overlain
at
most
sites
by
the
Poll
a
`
Mhuilt
Member,
which
forms
the
upper
part
of
the
Bay
of
Stoer
Formation
and
comprises
up
to
100
m
of
planar-bedded
lacustrine
sandstones,
mudstones
and
a
few
thin
limestones.
It
commonly
contains
small
reworked
melt
clasts
in
its
lower
few
metres.
The
Bay
of
Stoer
Formation
is
succeeded
by
several
hundred
metres
of
predominantly
planar
cross-bedded,
largely
pebble-free,
sandstones
of
the
Meall
Dearg
Formation.
Palaeocurrent
data
from
the
Meall
Dearg
Formation
indicate
drainage
broadly
from
east
to
west
along
the
entire
outcrop
(Fig.
2).
The
sandstones
of
the
Meall
Dearg
Formation
are
geochemically
similar
to
those
of
facies
BS1
that
underlie
the
Stac
Fada
Member
(Stewart,
2002)
but,
except
in
the
lowest
few
metres,
they
lack
the
non-gneiss
pebbles
seen
in
facies
BS1.
The
Meall
Dearg
Formation
appears
to
contain
no
reworked
detritus
from
the
Stac
Fada
Member.
Along
the
Stoer
Group
outcrop
the
Stac
Fada
Member
rests
either
on
various
facies
of
the
Clachtoll
Formation,
on
the
un-
named
member
of
the
Bay
of
Stoer
Formation
or,
at
one
site,
directly
on
an
inlier
of
Lewisian
basement
(Stewart,
2002).
I
agree
with
most
previous
authors,
other
than
Young
(2002),
that
the
melt-rich
breccia
of
the
Stac
Fada
Member
is
a
single
isochronous
unit.
However,
the
relationship
between
the
Stac
Fada
Member
and
the
various
facies
beneath
it
(Fig.
2)
necessarily
implies
that
the
boundary
between
the
Clachtoll
Formation
and
the
Bay
of
Stoer
Formation,
as
defined
by
Stewart
(2002),
is
markedly
diachronous.
Consequently,
I
have
chosen
to
simplify
the
lithostratigraphy
by
incorporating
the
un-named
member
of
the
Bay
of
Stoer
Formation,
which
lies
below
the
Stac
Fada
Member
at
several
sites
and
closely
resembles
certain
facies
of
the
Clachtoll
Formation,
into
an
amended
definition
of
the
Clachtoll
Formation.
In
this
account
the
Clachtoll
Formation
encompasses
all
strata
below
the
Stac
Fada
Member
while
the
Bay
of
Stoer
Formation
is
used
in
a
more
restricted
sense
that
encompasses
only
the
Stac
Fada
Member
and
the
Poll
a
`Mhuilt
Member
directly
above
it.
3.
The
Stac
Fada
Member:
an
impact
ejecta
deposit
The
melt-rich,
matrix-supported,
breccia
of
the
Stac
Fada
Member
has
intrigued
geologists
for
decades.
Understandably
it
was
interpreted
as
volcaniclastic
in
origin
for
most
of
this
time.
Lawson
(1972)
regarded
it
as
an
ash
flow
arising
from
a
phreatomagmatic
eruption.
Sanders
and
Johnston
(1989,
1990)
suggested
it
was
a
fluidised
peperite
generated
by
magma
rising
through
wet
sediment.
Young
(2002)
interpreted
it
as
the
product
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
744

Fig. 3. Representative sections through the Stac Fada Member and contiguous strata at key sites, and their relationship to the Lewisian basement. Thicknesses of strata other than the Stac Fada Member are indicative.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
745

of
at
least
three
separate
gravity-induced
volcanic
debris
flows
rather
than
a
single
isochronous
event,
and
proposed
that
these
were
sourced
from
two
different
centres
that
may
have
been
just
a
few
kilometres
to
the
west
of
the
present
outcrop.
Stewart
(2002)
considered
the
Stac
Fada
Member
to
have
formed
largely
through
remobilization
of
rain-soaked
tephra
deposits
from
a
group
of
maar
volcanoes
to
the
east,
but
he
invoked
a
subsequent
event
to
account
for
the
deposition
of
accretionary
lapilli
in
the
upper
part
of
the
member.
All
recognized
the
mafic
nature
of
the
melt
clasts,
variously
compared
with
basalt
(Lawson,
1972;
Stewart,
2002)
or
gabbro
(Young,
2002).
Evidence
for
volcanism
would
not
be
unexpected
within
the
Stoer
Group
(Lawson,
1972)
and
there
is
extensive
evidence
of
mafic
magmatism
at
this
time
in
the
Gardar
Rift
of
southern
Greenland
(Upton
et
al.,
2003),
an
area
that
may
well
have
been
adjacent
to
north-west
Scotland
at
this
time
(Darabi
and
Piper,
2004).
However,
the
considerable
thickness
and
extraordinary
lateral
extent
of
the
Stac
Fada
Member
within
a
succession
that
otherwise
lacks
any
substantial
evidence
for
magmatic
activity
intrigued
Lawson
(1972),
and
others
subsequently
(Stewart,
2002),
and
remained
an
issue
that
was
difficult
to
resolve.
A
thin
crystal
tuff
has
since
been
discovered
several
tens
of
metres
below
the
Stac
Fada
Member
(Batchelor
and
Prave,
2010),
but
its
geochemistry
is
distinct
from
that
of
the
Stac
Fada
Member
and
it
might
easily
represent
a
windblown
distal
ashfall
from
an
eruption
that
was
perhaps
many
hundreds
of
kilometres
away.
It
was
the
discovery
within
the
Stac
Fada
Member
of
shocked
quartz
grains
together
with
anomalous
levels
of
iridium,
nickel
(also
noted
by
Stewart,
2002,
and
Young,
2002)
and
chromium-53
that
led
to
its
reinterpretation
as
a
meteorite
impact
ejecta
blanket
(Amor
et
al.,
2008,
2009).
Iridium
is
widely
known
as
an
indicator
of
impact
deposits
through
its
association
with
the
Chicxulub
impact
and
Cretaceous-Palaeogene
boundary
layer
(Alvarez
et
al.,
1980)
but
mafic
and
ultramafic
rocks
may
also
be
relatively
enriched
in
nickel
and
platinum
group
elements
(PGEs)
(Naldrett
and
Cabri,
1976).
Consequently
the
presence
of
these
element
anomalies
does
not,
in
itself,
constitute
proof
of
an
extraterrestrial
origin.
With
chromium-53
there
has
been
an
assumption
that
terrestrial
Cr
53
/Cr
52
ratios
do
not
vary
because
planetary
differen-
tiation
was
completed
long
after
all
of
the
Mn
53
parent
isotope
had
decayed,
and
hence
any
chromium
anomaly
can
be
used
as
an
unambiguous
indicator
of
a
meteoritic
source
(Koeberl
et
al.,
2012;
Goderis
et
al.,
2013).
However,
terrestrial
fractionation
processes
may
have
produced
significant
variation
in
Cr
53
/Cr
52
ratios
during
the
Proterozoic
(Frei
et
al.,
2009)
such
that
the
quoted
values
of
Amor
et
al.
(2008,
2009)
cannot
be
taken
as
conclusive
proof
of
an
impact
origin
for
the
Stac
Fada
Member.
The
evidence
from
element
anomalies
for
an
impact
origin
for
the
Stac
Fada
Member
may
appear,
on
face
value,
to
be
ambiguous.
However,
the
presence
of
planar
deformation
fabrics
(PDFs)
in
quartz
grains
(Amor
et
al.,
2008,
2009;
Osinski
et
al.,
2011a)
and
shocked
zircons
with
lamellae
of
reidite,
a
high
pressure
polymorph
of
zircon
(Reddy
et
al.,
2015),
provides
conclusive
evidence
for
an
impact
origin
and,
in
so
doing,
strengthens
the
case
for
an
extraterrestrial
cause
of
these
element
anomalies.
Multiple
planar
deformation
fabrics
(PDFs)
in
mineral
grains
are
uniquely
diagnostic
of
large
hypervelocity
impact
events
as
they
are
formed
in
an
extreme
pressure-temperature
regime
that
is
entirely
separate
from
that
experienced
during
normal
crustal
metamorphism
or
tectonic
processes
(French,
1998;
Reimold
and
Jourdan,
2012).
More
specifically,
reidite
provides
unambiguous
evidence
of
shock
pressures
in
excess
of
30
GPa.
Despite
this
remarkable
discovery
of
the
first
substantial
impact
deposit
in
the
UK,
little
has
been
published
on
the
deposit
since
2008
other
than
two
conference
abstracts
(Amor
et
al.,
2011;
Osinski
et
al.,
2011a)
and
articles
focusing
on
the
lapilli-bearing
facies
of
the
Stac
Fada
Member
(Branney
and
Brown,
2011),
the
radiometric
dating
of
the
Stoer
Group
and
Stac
Fada
Member
(Parnell
et
al.,
2011),
and
the
discovery
of
shocked
zircons
and
reidite
(Reddy
et
al.,
2015).
A
guidebook
to
the
region
has
also
been
published
(Goodenough
and
Krabbendam,
2011),
but
these
authors
maintain
that
the
contrast
between
the
felsic
composition
of
the
Stac
Fada
Member
matrix
and
the
mafic
chemistry
of
the
melt
points
to
very
different
sources
for
them
that,
in
their
view,
indicates
a
volcanic
rather
than
an
impact
origin
for
the
melt.
Amor
et
al.
(2008,
2009)
interpreted
the
Stac
Fada
Member
as
a
primary
ejecta
deposit
composed
predominantly
of
a
rock
type
called
suevite.
This
term
is
now
considered
specific
to
the
Ries
impact
in
Germany,
with
the
more
general
term
of
‘impact
melt-
bearing
breccia’
being
favoured
(Osinski,
2013).
It
can
be
defined
as
a
‘‘polymict
impact
breccia
with
clastic
matrix
containing
lithic
and
mineral
clasts
in
various
stages
of
shock
metamorphism,
including
cogenetic
impact
melt
particles
which
are
in
a
glassy
or
recrystallised
state’’
(Sto
¨ffler,
1977).
Compared
with
typical
impact
melt-bearing
breccia
(suevite)
as
found
at
Ries
and
other
impact
sites,
the
Stac
Fada
Member
contains
significantly
less
melt
material
(<50%
vs.
>85%;
Amor
et
al.,
2008;
Sto
¨ffler
et
al.,
2013)
and
was
emplaced
at
a
substantially
lower
temperature
(200
8C
vs.
>600
8C;
Parnell
et
al.,
2011;
Sto
¨ffler
et
al.,
2013).
Osinski
et
al.
(2011a)
noted
this
relatively
low
abundance
of
melt
in
the
Stac
Fada
Member
and
also
found
that
shocked
quartz
grains
were
an
order
of
magnitude
less
abundant
than
in
the
Ries
suevite.
From
this
he
concluded
that
the
Stac
Fada
Member
was
not
a
primary
impact
deposit
but
the
product
of
reworking
and
transport
by
water
or
wind.
Stewart
(2002)
similarly
had
invoked
reworking,
in
this
instance
of
volcanic
tephra
remobilised
as
a
mudflow.
However,
the
high
emplacement
temperature
of
the
deposit
precludes
either
scenario
and
it
is
considered
here
to
have
been
generated
directly
by
a
giant
meteorite
impact.
Although
the
evidence
of
an
impact
origin
for
the
Stac
Fada
Member
is
now
unequivocal
(Amor
et
al.,
2008,
2009;
Osinski
et
al.,
2011a;
Reddy
et
al.,
2015),
so
little
of
this
ejecta
deposit
has
survived
to
the
present
day
that
its
outcrop
offers
no
immediate
clues
as
to
the
original
extent
of
the
ejecta
blanket
or
the
source
crater’s
location,
and
no
surface
manifestation
of
an
impact
crater
is
evident
anywhere
in
northern
Scotland.
These
issues
can
potentially
be
resolved
through
careful
field
observations
to
identify
the
direction
in
which
the
ejecta
was
moving,
and
through
analysis
of
geophysical
data
from
across
the
region
to
identify
anomalies
consistent
with
a
buried
impact
crater
from
which
the
Stac
Fada
Member
might
have
originated.
Fracturing
and
breccia-
tion
of
dense
target
rocks
may
extend
to
considerable
depths
beneath
large
impacts
creating
a
density
contrast
between
the
impact
structure
and
surrounding
unfractured
rock
(French,
1998).
This
may
be
accentuated
by
infilling
of
the
crater
with
more
porous,
i.e.
less
dense,
breccias
and
sediments.
Such
density
reductions
produce
a
characteristic
negative
gravity
anomaly,
as
observed
at
Ries
(Pohl
et
al.,
2010)
but
also
evident
for
buried
impact
craters
such
as
Chicxulub
and
Chesapeake
(Plescia
et
al.,
2009).
Hence
even
if
the
impact
crater
responsible
for
the
Stac
Fada
Member
is
deeply
buried,
it
may
still
manifest
itself
as
a
deep
negative
gravity
anomaly.
The
magnitude
of
the
gravity
anomaly
generally
increases
with
crater
diameter
to
a
maximum
of
20–30
mGal
towards
the
centre
of
a
30
km
diameter
impact
crater
(Pilkington
and
Grieve,
1992).
Craters
with
diameters
greater
than
30
km
tend
to
have
more
complex
structures
with
a
central
peak
and
one
or
more
concentric
rings
(Kenkmann
et
al.,
2013)
producing
a
correspondingly
more
complex
gravity
anomaly
that
may
be
more
evident
in
local
gravity
gradient
variations
than
in
the
actual
magnitude
of
the
anomaly.
Gravity
surveys
of
northern
Scotland
have
revealed
several
significant
negative
anomalies
(Fig.
4).
One
of
these,
descending
to
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
746

about
18
mGal,
lies
to
the
west
of
the
Stoer
Group
outcrop
in
the
Minch
Basin
between
Stoer
and
Stornaway.
This
is
approximately
where
Amor
et
al.
(2008,
2011)
inferred
that
the
impact
site
might
be
located,
based
on
their
field
observations
and
subsequent
AMS
(Anisotropy
of
Magnetic
Susceptibility)
analysis.
A
second
gravity
low,
descending
to
23
mGal,
is
located
onshore
and
centred
on
the
town
of
Lairg
about
60
km
to
the
east
of
the
Stoer
Group
outcrop.
Still
further
east
is
a
much
larger
and
deeper
(to
45
mGal)
gravity
anomaly
occupying
the
Moray
Firth.
Of
these
the
Moray
Firth
anomaly
is
too
distant
from
the
Stoer
Group
outcrop
to
be
a
realistic
candidate
for
the
impact
crater,
and
its
configuration
is
entirely
consistent
with
the
structural
framework
of
this
largely
post-Palaeozoic
sedimentary
basin
(Pilkington
et
al.,
1995;
Roberts
et
al.,
2009).
However,
might
one
of
the
other
two
gravity
anomalies
represent,
at
least
in
part,
a
buried
impact
crater?
4.
Terrestrial
impact
processes
and
products
No
kilometre-scale
impact
has
ever
been
witnessed
on
Earth
and
all
of
the
large
terrestrial
impact
structures
currently
known
are
many
thousands,
or
more
often
millions,
of
years
old.
In
the
absence
of
any
direct
observations
uniformitarian
principles
are
difficult
to
apply
to
the
interpretation
of
these
ultra-extreme
events.
Instead
it
is
the
field
and
laboratory-based
analysis
of
ancient
impact
structures
and
their
associated
deposits,
combined
with
both
actualistic
and
computer
modeling
of
hypervelocity
impacts,
that
contributes
to
understanding
the
processes
and
products
of
these
events
(French,
1998;
Osinski
and
Pierazzo,
2013).
The
best
preserved
of
all
terrestrial
impact
structures
is
the
24
km
diameter
Ries
crater
and
its
associated
deposits
in
southern
Germany,
which
has
been
the
focus
of
detailed
research
for
decades
(Sto
¨ffler
et
al.,
2013).
At
Ries
a
distinct
sequence
of
impact
processes
has
been
identified
and
modeled
(Sto
¨ffler
et
al.,
2013;
Artemieva
et
al.,
2013)
that
is
broadly
applicable
to
other
well-documented
terrestrial
impact
sites
such
as
Chicxulub
and
Bosumtwi
(Osinski
et
al.,
2013).
Initial
compression
of
the
impactor
and
target
rocks
in
a
narrow
zone
around
the
point
of
contact
causes
ejection
at
high
velocity
of
incandescent
melt,
cooling
to
form
tektites,
to
distances
as
far
as
tens
of
crater
radii
from
the
impact
(McCall,
2001;
Melosh,
2013).
This
is
immediately
followed
by
ejection
of
a
relatively
small
volume
of
lightly
shocked
material
at
high
velocities
(up
to
5
km
s
1
)
from
a
shallow
(<50
m
depth)
spallation
zone
in
a
narrow
ring
immediately
around
the
expanding
crater.
This
ejecta
debris
travels
on
ballistic
trajectories
to
distances
of
up
to
15
crater
radii
(Melosh,
1989;
Osinski
et
al.,
2011b).
In
the
case
of
the
Ries
impact
this
spallation
debris,
the
so-called
‘Reutersche
Blo
¨cke’
of
the
North
Alpine
Foreland
Basin,
is
found
more
than
100
km
from
the
impact
site
(Buchner
et
al.,
2007),
but
modeling
suggests
that
some
could
occur
more
than
200
km
away
(Artemieva
et
al.,
2013).
Spallation
ejecta
are
not
evenly
distributed
around
an
impact
and
with
low
angle
impacts
(308)
can
form
distinct
ray
patterns
(Tornabene
et
al.,
2006)
in
which
clusters
of
small
secondary
craters
are
abundant
(McEwen
et
al.,
2005).
Within
seconds
of
this
initial
phase
of
contact
the
impactor
penetrates
the
target
rock
to
a
depth
of
one
to
two
times
impactor
diameter
and
comes
to
a
halt,
forming
a
transient
crater
and
transferring
kinetic
energy
to
the
target
rocks
as
shock
waves.
Seismic
shock
waves
radiating
out
from
the
impact
through
the
target
rock
may
generate
seismites
in
suitable
pre-impact
sediments
even
considerable
distances
from
the
crater
(Terry
et
al.,
2001).
At
Ries
the
depth
of
this
transient
crater
is
estimated
at
4.3
km
(Artemieva
et
al.,
2013),
with
fracturing
extending
to
still
greater
depths.
Transient
peak
pressures
at
the
point
of
impact
may
exceed
100
GPa
and
generate
temperatures
of
more
than
10,000
8C.
Almost
immediate
decompression
of
impactor
and
target
then
causes
complete
melting
and
even
vapourisation
of
the
impactor,
and
melting
of
a
significant
volume
of
target
rock
(French,
1998;
Osinski
et
al.,
2013).
Impact
melting
is
unlike
thermal
melting
in
magmatic
systems,
which
are
governed
by
thermodynamic
relations
such
as
eutectic
mixtures
and
partial
melting
(Osinski
et
al.,
2013).
At
pressures
60
GPa,
near
the
impact
point,
there
is
sufficient
energy
released
to
completely
melt
a
large
volume
of
target
rock
within
seconds.
Whole
rock
melting
of
mafic
rocks
requires
shock
pressures
>75
GPa
while
felsic
rocks
melt
at
lower
pressures
of
60
GPa
(Sto
¨ffler
et
al.,
2006).
In
large
impact
structures
(D
>
25
km)
tens
to
hundreds
of
km
3
of
impact
melt
might
be
generated
by
complete
melting
of
target
rock,
with
a
significant
volume
of
this
being
ejected
during
the
excavation
stage
and
the
remainder
forming
a
melt
pool
within
the
crater.
Some
of
the
most
unambiguous
evidence
of
impact
is
formed
at
lower
pressures.
Shatter
cones,
conical
striated
fracture
surfaces
formed
at
pressures
of
2–10
GPa
(French,
1998),
are
the
only
macroscopic
features
and
sometimes
developed
on
a
metre
scale,
but
they
are
confined
to
the
floor
and
central
uplift
of
craters
(Ferrie
´re
and
Osinski,
2013).
At
a
microscopic
scale
the
most
unambiguous
evidence
of
impact
are
Planar
Deformation
Features
(PDFs)
that
form
in
quartz
and
other
mineral
grains
at
pressures
of
5–35
GPa.
Shocked
mineral
grains
may
be
ejected
considerable
distances
and
are
a
ubiquitous,
and
diagnostic,
component
of
impact
ejecta
deposits.
As
crater
excavation
progresses
beyond
the
initial
spallation
phase
large
volumes
of
cold,
weakly
shocked,
material
are
ejected
Fig.
4.
Bouguer
Residual
Gravity
field
for
northern
Scotland
(from
Rollin,
2009).
The
dark
areas
are
gravity
lows
deeper
than
-15
mGal.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
747

from
relatively
shallow
levels
(depths
of
just
a
few
hundred
metres)
around
the
target
region
at
velocities
up
to
hundreds
of
metres
per
second.
Larger
blocks
cause
secondary
cratering
upon
landing,
generating
a
mix
of
primary
and
secondary
ejecta
that
moves
radially
outwards
as
an
erosive
ground-hugging
flow.
At
Ries
this
is
represented
by
a
near
continuous
sheet
of
melt-free
lithic
breccia,
termed
Bunte
Breccia
(Ho
¨rz
et
al.,
1983),
extending
out
for
several
crater
radii
from
the
impact.
Comparable
lithic
breccias
are
found
in
association
with
other
terrestrial
impact
structures
(Osinski
et
al.,
2013)
but
the
tektites
and
spallation
debris
ejected
during
the
initial
phase
of
impact
are
preserved
only
beyond
the
outer
limits
of
these
erosively
emplaced
breccias.
The
melt-free
lithic
breccia
is
succeeded
with
sharp
contact,
and
presumed
temporal
discontinuity,
by
a
melt-rich
breccia
that,
at
Ries,
has
been
termed
suevite
and
was
emplaced
as
a
hot
(>600
8C)
ground-hugging
flow
forming
a
patchy
to
near
continuous
blanket
within
two
crater
radii
of
the
impact
(Artemieva
et
al.,
2013;
Osinski
et
al.,
2011b;
Sturm
et
al.,
2013).
Modeling
by
Artemieva
et
al.
(2013)
suggests
that
relatively
shallow-sourced
solid
fragments
from
the
upper
few
hundred
metres
of
the
target
are
ejected
first
while
deeper-sourced
material,
both
solid
and
molten,
is
ejected
20–30
seconds
later.
Particles
from
a
few
millimetres
to
tens
of
metres
across
sourced
from
relatively
shallow
lithified
sediments
are
ejected
ballistically
to
form
the
lithic
breccia.
Smaller
particles
may
remain
suspended
in
a
developing
hot
plume
above
the
crater
to
be
joined
by
the
melt-rich
material
that,
being
ejected
slightly
later,
contributes
to
a
melt-rich
breccia
layer
(suevite)
above
the
lithic
breccia.
This
creates
the
characteristic
Double
Layer
Ejecta
structure
of
terrestrial
impact
deposits.
Accretionary
lapilli
occur
in
the
upper
part
of
ejecta
deposits
associated
with
the
impact
craters
at
Sudbury,
Chicxulub
(Grieve
and
Therriault,
2013),
Ries
(Artemieva
et
al.,
2013)
and
elsewhere.
Modeling
suggests
that
the
relatively
late-stage
accretionary
lapilli,
and
indeed
much
of
the
suevite,
may
actually
have
been
generated
not
by
the
impact
itself
but
by
violent
melt–coolant
interactions
arising
from
the
post-impact
inflow
of
water
and/or
wet
sediment
into
the
crater
(Artemieva
et
al.,
2013).
5.
Site
descriptions
Thirteen
separate
exposures
of
the
Stac
Fada
Member
and
contiguous
strata
were
visited
along
the
outcrop,
from
Cnoc
Breac
in
the
north
(Grid
ref.
NC
036317)
to
Bac
an
Leth-choin
in
the
south
(Grid
ref.
NG
773893)
(Fig.
1).
All
have
been
described
previously
by
Stewart
(2002).
Three
distinct
facies
associations
of
the
Stac
Fada
Member
were
recognized
in
the
field,
corresponding
to
the
exposures
at
Stoer,
at
Enard
Bay,
and
at
the
remaining
sites
from
Achiltibuie
to
Bac
an
Leth-choin
(Fig.
3).
5.1.
Stac
Fada,
Stoer
The
Stac
Fada
Member
at
its
type
locality
near
Stoer
(NC
033284)
is
12
m
thick
and
dominated
by
a
virtually
structureless
unit
of
very
hard
melt-rich
breccia
(Fig.
3).
It
overlies
a
sequence
of
decimetre
to
metre-scale
muddy
or
trough
cross-bedded
sand-
stones
interbedded
with
thin
(dm-scale)
mudstones
(Stewart,
2002).
Virtually
no
bedding
is
discernible
in
the
Stac
Fada
Member
except
for
a
major
planar
break
halfway
up
and
some
rather
indistinct
bedding
in
the
uppermost
metre.
Melt
clasts,
up
to
several
centimetres
across,
are
randomly
orientated
within
the
matrix.
Irregular
feldspar-lined
elutriation
pipes
are
conspicuous
and
may
extend
upwards
through
the
Stac
Fada
Member
for
several
metres,
while
feldspar
also
fills
cavities
and
pore
spaces
throughout
the
rest
of
the
Stac
Fada
Member.
Large
rafts
of
sandstone
are
conspicuous
in
the
lower
part
of
the
member
exposed
on
the
shore
at
the
southern
end
of
the
outcrop
(Fig.
5)
but
also
are
evident
in
the
outcrop
for
several
hundred
metres
inland
to
the
north.
Some
of
the
rafts
exposed
on
the
shore
are
tens
of
square
metres
in
area,
a
metre
or
more
thick,
and
clearly
were
derived
from
the
underlying
fluvio-
lacustrine
sandstones.
Smaller,
decimetre-
to
metre-scale,
sand-
stone
blocks
are
also
present
in
the
upper
part
of
the
member.
Many
of
the
larger
sandstone
intraclasts
have
been
deformed
after
they
had
become
incorporated
into
the
ejecta
blanket
(Stewart,
2002,
fig.
60).
All
have
been
thermally
metamorphosed
to
quartzite
by
the
heat
of
the
ejecta,
but
have
random
palaeomagnetic
orientations
indicating
that
the
temperature
did
not
exceed
675
8C
(Irving
and
Runcorn,
1957;
Stewart,
2002).
Beneath
the
Stac
Fada
Member
at
Stoer
at
least
four
separate
wedges
of
melt-rich
breccia
extend
along
bedding
planes
in
the
underlying
planar-bedded
sandstones
(Fig.
5).
Stewart
(2002)
noted
that
these
intrusions
occur
along
thin
shale
beds
within
the
sandstone
succession.
The
sandstone
beds
directly
above
these
intrusions
of
melt-rich
breccia
dip
at
between
48
and
378
more
than
the
regional
dip
and
with
azimuths
that
pinch
out
broadly
to
the
west
(Fig.
6).
In
contrast,
fold
axes
of
the
detached
sandstone
rafts
in
the
Stac
Fada
Member
lack
any
consistent
orientation.
In
the
down-faulted
block
of
the
Stac
Fada
Member
immediately
east
of
the
main
outcrop
(referred
to
as
SF1
by
Young,
2002),
one
such
intrusive
wedge
is
up
to
five
metres
thick
while
just
a
short
distance
offshore
to
the
south
the
isolated
rock
of
Stac
Gruinn
represents
part
of
another
intrusive
wedge
that
is
at
least
eight
metres
thick.
Embedded
in
the
top
surface
of
several
of
the
sandstone
rafts
are
large
(up
to
tens
of
centimetres)
angular
fragments
and
cobbles
of
gneiss
and
schist
(Fig.
3).
These
clasts
are
substantially
larger
than
any
observed
in
the
Stoer
Group
sandstones
for
many
tens
of
metres
below
the
Stac
Fada
Member
at
this
site.
Two
large,
rounded,
gneiss
cobbles
also
occur
apparently
within
one
of
the
sandstone
rafts.
These
latter
clasts
are
associated
with
a
very
thin
seam
of
melt-rich
breccia
(Stewart,
2002;
Young,
2002)
and
hence
cannot
be
considered
as
original
components
of
the
sandstone
but
have
been
incorporated
into
it
subsequently
during
emplacement
of
the
melt-rich
breccia
(Stewart,
2002).
The
internal
stratigraphy
of
individual
rafts
suggests
that
all
of
the
oversize
clasts,
other
than
these
latter
examples,
lie
on
the
same
original
bedding
plane
surface
that
represents
the
top
of
the
pre-Stac
Fada
Member
succession.
The
top
1.5
m
of
the
Stac
Fada
Member
at
Stoer
is
weakly
stratified
and
contains
scattered
accretionary
lapilli
typically
5
mm
across
(Sanders
and
Johnston,
1989;
Amor
et
al.,
2008).
Near
the
base
of
the
succeeding
lacustrine
Poll
a
`Mhuilt
Member
several
metres
of
strata
are
disrupted
by
large-scale
slumping.
A
few
metres
above,
in
the
overlying
planar-bedded
strata,
a
series
of
metre-scale
cross-bed
foresets
dip
to
the
north-east
(approxi-
mately
0508)
while
Stewart
(2002)
noted
current
lineation
and
ripple
drift
lamination
towards
1008
in
the
lower
beds
of
the
Poll
a
`
Mhuilt
Member
that
indicate
palaeocurrents
from
west
to
east
(Fig.
6).
Only
the
uppermost
part
of
the
Stac
Fada
Member
is
exposed
in
the
small
quarry
at
Cnoc
Breac
(NC
036317),
about
three
kilometres
north
of
Stac
Fada
(Stewart,
2002).
It
shows
similar
melt-rich
breccia
with
small
accretionary
lapilli
as
is
seen
in
the
upper
part
of
the
member
at
Stac
Fada
itself.
5.2.
Camas
a
´Bhothain,
Enard
Bay
There
are
two
key
exposures
at
Camas
a
´Bhothain,
on
the
south
side
of
Enard
Bay
(Fig.
3).
That
to
the
east
(referred
to
as
Enard
Bay
East;
extending
westwards
from
NC
0318
1465
to
NC
0302
1459)
exposes
the
upper
part
of
the
Clachtoll
Formation,
the
Stac
Fada
Member
and
the
Poll
a
`Mhuilt
Member
while
about
150
m
to
the
west
(referred
to
as
Enard
Bay
West;
NC
02861467)
the
upper
part
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
748

of
the
Poll
a
`Mhuilt
Member
rests
directly
on
Lewisian
gneiss
basement
(Stewart,
2002).
There
are
conflicting
descriptions
of
the
thickness
of
the
Stac
Fada
Member
and
the
nature
of
the
subjacent
strata
at
Enard
Bay
East,
perhaps
a
consequence
of
the
complexity
of
the
exposure.
The
Stoer
Group
here
is
overlain
unconformably
by
the
Diabaig
Formation
of
the
Torridon
Group,
which
is
lithologically
similar
but
200
Ma
younger
(Gracie
and
Stewart,
1967),
and
localized
reworking
of
material
can
blur
the
distinction
between
the
Stac
Fada
Member
and
the
lower
part
of
the
Diabaig
Formation
in
places.
Furthermore,
both
the
angle
and
direction
of
the
dip
varies
across
the
outcrop,
perhaps
explaining
the
variation
in
thickness
estimates
for
the
Stac
Fada
Member
at
this
site
from
10.5
m
(Branney
and
Brown,
2011)
to
as
much
as
32
m
(Lawson,
1972).
However,
based
on
the
variable
and
shallow
dip
across
most
of
the
outcrop
I
estimate
a
thickness
of
about
12–15
m.
The
Stac
Fada
Member
here
rests
on
trough
cross-bedded
sandstones,
with
thin
pebbly
partings,
of
the
Clachtoll
Formation
that
extend
below
it
for
at
least
several
tens
of
metres.
However,
abundant
angular
blocks
of
gneiss,
both
mafic
and
felsic,
are
embedded
in
the
top
of
the
sandstone
immediately
beneath
the
Stac
Fada
Member.
In
places
these
angular
clasts
are
crowded
together
and
in
contact
but
the
breccia
is
predominantly
matrix-
supported
and
the
clasts,
the
largest
more
than
a
metre
across,
are
widely
spaced.
Where
clasts
are
crowded
together
some
are
broken
yet
their
component
fragments
are
separated
by
just
a
few
centimetres
(Fig.
7).
Gracie
and
Stewart
(1967)
recognized
that
the
Stac
Fada
Member
was
underlain
largely
by
sandstones
but
others
considered
the
breccia
horizon
to
represent
part
of
a
more
stratigraphically
extensive
breccia
unit
within
the
Clachtoll
Formation
(Stewart,
2002;
Young,
2002).
Impersistent
pebble
lenses
are
common
in
the
cross-bedded
sandstones
below
the
Stac
Fada
Member,
but
nothing
comparable
with
this
thin,
largely
matrix-supported,
breccia
horizon
is
seen
anywhere
else
along
the
extensive
Enard
Bay
outcrop
of
the
Stoer
Group.
Where
marginal
breccia
facies
are
developed
they
occur
within
a
few
metres
of
Lewisian
Gneiss
inliers
(Davison
and
Hambrey,
1996)
and
generally
are
thicker
and
clast-supported
(Stewart,
2002).
The
Stac
Fada
Member
at
this
site
differs
significantly
from
what
is
found
at
other
sites,
with
several
different
facies
forming
distinct
units
within
it.
The
lowest
part,
resting
on
the
irregular
breccia-strewn
surface
of
the
sandstone
below,
is
a
very
hard
red-
brown
melt-rich
breccia,
in
two
distinct
beds
about
1.6
m
thick,
containing
abundant
angular
clasts
of
gneiss
and
quartzite
(Fig.
3).
The
largest
gneiss
clasts
are
more
than
a
metre
across.
This
is
succeeded
by
5–6
m
of
massive,
intensely
hard,
melt-rich
breccia
of
which
only
the
lower
metre
or
so
resembles
the
typical
melt-rich
breccia
seen
at
other
sites
(cm-scale
green
melt
clasts
in
a
reddish
sandy
matrix).
As
at
Stac
Fada,
melt
clasts
are
randomly
orientated
within
the
matrix.
The
remainder
of
this
unit
comprises
a
massive,
structureless,
poorly-sorted
red-brown
sandstone
with
abundant
small
(<1
cm)
melt
clasts
and
mm-scale
feldspar
phenocrysts.
There
is
a
pervasive
feldspar
infill
of
cavities
and
pore
spaces
throughout
the
Stac
Fada
Member
at
this
site
but
the
elutriation
pipes
so
conspicuous
at
Stac
Fada
appear
to
be
absent.
Much
of
this
unit
has
a
brecciated
appearance
but
within
it
occur
large
sphaeroidal
masses
(3–5
m
across
and
1–2
m
thick)
of
more
coherent
material,
each
surrounded
by
a
zone
of
concentric
fractures
(Fig.
3).
The
upper
part
of
the
Stac
Fada
Member
here
comprises
about
5
m
of
more
distinctly
stratified
material
in
which
accretionary
lapilli,
commonly
10–15
mm
across,
are
abundant
together
with
Fig.
5.
The
Stoer
Group
on
the
eastern
side
of
Stac
Fada,
Stoer,
looking
north-west,
showing
the
relationship
of
the
Stac
Fada
Member
to
the
Clachtoll
Formation
beneath.
A
melt-rich
breccia
intrusion,
pinching
out
to
the
west-north-west,
separates
undisturbed
planar
bedded
sandstones
beneath
from
a
deformed
and
partially
rafted
sandstone
sheet
above.
Viewed
from
this
level,
the
sandstone
raft
obscures
the
lower
part
of
the
Stac
Fada
Member
behind.
The
Poll
a
`Mhuilt
Member
is
exposed
in
the
cliffs
visible
through
the
gap
eroded
in
the
main
body
of
the
Stac
Fada
Member.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
749

angular
cm-scale
melt
clasts
orientated
parallel
to
bedding.
Low
angle,
westward-dipping,
planar
cross-beds
are
present
in
a
broad
channel-like
body,
oriented
east-west
(about
1008),
in
the
upper
part
of
these
lapilli
beds
(Fig.
6).
Two
broad
troughs,
the
larger
>15
m
across
and
>2
m
deep,
separated
by
an
intervening
ridge,
cut
through
the
lapilli
beds
with
broadly
east-west
orientations
and
plunge
gently
to
the
west
(Fig.
6).
This
undulating
surface
is
draped
uniformally
with
a
thin
(8
cm)
dust
pellet
layer
(Fig.
8)
that
was
termed
Division
C
by
Branney
and
Brown
(2011)
and
is
taken
as
the
top
of
the
Stac
Fada
Member.
The
dust
pellets
are
unstructured
and
show
evidence
of
partial
disaggregation
and
soft
deformation
at
point
contacts.
Pore
space
in
this
dust
layer
is
filled
with
authigenic
feldspar
similar
to
that
in
the
rest
of
the
Stac
Fada
Member
here
and
at
other
sites
(Branney
and
Brown,
2011).
The
Stac
Fada
Member
at
Enard
Bay
East
is
succeeded
by
at
least
10
m
of
planar-bedded
sandstone,
siltstone
and
mudstone
assigned
to
the
succeeding
Poll
a
`Mhuilt
Member
(Fig.
3).
Near
its
base
these
decimetre-scale
beds
onlap
against
the
margins
of
the
troughs
without
any
erosion
or
modification
of
the
dust
pellet
layer
(Fig.
8).
The
sandstones
in
the
lower
part
of
the
member
include
thin
graded
beds
that
incorporate
small
reworked
melt
clasts
and
lapilli,
which
may
have
led
to
their
misinterpretation
as
Fig.
6.
Azimuthal
data
for
sedimentary
features
within
the
Stac
Fada
Member
and
contiguous
strata.
(a)
Cross-bedding
in
the
upper
part
of
the
Clachtoll
Formation
(from
Stewart,
2002).
(b)
Long
axis
orientations
of
angular
lithic
clasts
immediately
beneath
the
Stac
Fada
Member.
(c)
Dip
of
sandstone
beds
immediately
overlying
melt-breccia
intrusions
in
the
lower
part
of
the
Stac
Fada
Member.
(d)
Cross
beds
in
uppermost
lapilli
beds
of
the
Stac
Fada
Member.
(e)
Long
axis
orientations
of
lapilli
in
uppermost
lapilli
beds
of
the
Stac
Fada
Member.
(f)
Gently
plunging
troughs
incised
into
lapilli
beds
in
the
upper
part
of
the
Stac
Fada
Member.
(g)
Approximate
trend
normal
to
ogive
fractures
on
the
upper
surface
of
the
Stac
Fada
Member.
(h)
Cross-bed
foresets
in
lower
part
of
the
Poll
a
`Mhuilt
Member.
(i)
Ripple
cross
lamination
in
the
Poll
a
`Mhuilt
Member
(from
Stewart,
2002).
(j)
Cross-bedding
in
the
Meall
Dearg
Formation
(Stoer
and
Poolewe
from
Stewart,
2002;
Enard
Bay,
new
data).
Fig.
7.
Shattered
gneiss
clasts
embedded
in
the
top
of
the
Clachtoll
Formation
immediately
beneath
the
Stac
Fada
Member
at
Enard
Bay.
(a)
The
shattered
clast
on
the
right
partly
underlies
the
much
larger
clast,
more
than
a
metre
across,
on
the
left.
Hammer
is
37
cm
long.
(b)
Penknife
is
9
cm
long.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
750

part
of
the
Stac
Fada
Member
(Stewart,
2002,
p.
72).
There
is
a
conspicuous
laminated
limestone
about
10–12
m
above
the
base
that
has
been
correlated
both
with
similar
limestones
at
Stoer
and
with
a
limestone
that
directly
encrusts
a
gneiss
inlier
at
Enard
Bay
West
150
m
further
west
(Grid
ref.
NC
024147)
(Stewart,
2002)
(Fig.
3).
The
Stac
Fada
Member
is
absent
from
this
latter
site,
although
Stewart
(2002)
noted
minute
green
vitreous
tephra
within
the
limestone.
5.3.
Achiltibuie
to
Bac
an
Leth-choin
South
of
Enard
Bay
the
Stac
Fada
Member
is
thinner
than
at
the
northern
sites,
typically
4–8
m
thick,
and
usually
rests
on
sandstones
of
the
Clachtoll
Formation
that
are
bedded
at
a
metre-scale
without
any
interbedded
mudstone
or
siltstone
units
(Fig.
3).
At
Achiltibuie
(Grid
ref.
NC
022082)
the
Stac
Fada
Member
is
just
2
m
thick
and
rests
directly
on
Lewisian
basement
(Fig.
3).
The
most
informative
sections
are
those
at
Stattic
Point
(Grid
ref.
NG
97149588
to
NG
97239609),
its
continuation
to
south
of
Sro
`n
na
Fa
`ire
Mo
´ire
(Grid
ref.
NG
964943),
and
at
Second
Coast
(Grid
ref.
NG
925911)
referred
to
as
Gruinard
Bay
in
some
accounts.
At
Second
Coast
the
sandstone
surface
immediately
beneath
the
Stac
Fada
Member
is
strewn
with
angular
clasts,
mostly
of
felsic
gneiss,
that
are
markedly
discordant
in
size
from
the
surrounding
sandstone
matrix
(Fig.
9).
A
significant
proportion
of
these
are
more
than
20
cm
across
(Fig.
10a),
with
the
largest
having
a
maximum
dimension
of
>90
cm
and
an
estimated
mass
of
at
least
500
kg,
yet
very
few
of
the
clasts
are
in
contact
with
others.
No
preferred
orientation
of
clast
long
axes
is
evident
(Fig.
10b),
with
several
slab-like
clasts
actually
embedded
with
their
major
or
intermediate
axis
at
a
steep
angle
to
the
sandstone
surface
(Fig.
9).
Almost
70
clasts
more
than
10
cm
across
occur
scattered
over
an
area
of
250
m
2
at
Second
Coast
yet
fewer
than
ten
have
been
found
across
a
considerably
larger
area
of
the
equivalent
surface
at
Stattic
Point.
Thin
slabs
of
baked
sandstone,
together
with
cobbles
of
gneiss
and
occasionally
quartzite,
are
not
uncommon
in
the
lowest
few
decimetres
of
the
Stac
Fada
Member
which
throughout
this
southern
part
of
the
outcrop
is
developed
in
a
single
fairly
uniform
facies
of
green,
devitrified,
angular
melt
clasts
in
a
poorly
sorted
red
sandy
matrix.
Typically
the
melt
clasts
are
no
more
than
a
few
centimetres
across
but
at
Stattic
Point
and
Cailleach
Head
(Grid
ref.
NG
990973)
many
are
conspicuously
larger
and
may
be
more
than
20
cm
across
and
several
cm
thick.
A
high
proportion
of
the
melt
clasts,
particularly
the
larger
ones,
lie
sub-horizontally
within
the
matrix.
This
facies,
although
similar
to
that
seen
at
Stoer,
is
markedly
less
indurated
at
all
of
the
southern
sites
than
at
Stoer
or
Enard
Bay,
and
also
shows
a
much
more
clearly
developed
decimetre
to
metre-scale
sub-planar
fabric.
This
fabric
is
seen
most
clearly
at
Stattic
Point
where
planar,
gently
undulating
or
concave
surfaces
can
be
seen
to
cut
through
the
Stac
Fada
Member
(Fig.
11)
but
without
affecting
the
strata
above
or
below.
These
surfaces
are
separated
by
just
a
decimetre
or
two
in
the
upper
part
of
the
member
but
by
a
metre
or
more
lower
down.
They
are
unmineralised
but
are
marked
by
thin
seams,
no
more
than
1–
2
cm
thick,
of
fine
red
sandy
or
silty
material.
Feldspar-lined
elutriation
pipes
are
common,
typically
surrounded
by
vertically
aligned
melt
clasts,
but
they
are
truncated
or
displaced
by
these
sub-planar
surfaces.
At
Stattic
Point
the
Stac
Fada
Member
is
exposed
along
strike
for
about
250
m
and
shows
significant
thickness
variations,
from
just
over
4
m
to
about
6
m
over
distances
of
as
little
as
about
15
m.
There
appears
to
be
some
correlation
between
these
thickness
variations
and
undulations
in
the
internal
fabric
of
the
member
(Fig.
11).
Where
the
upper
Fig.
8.
Uppermost
Stac
Fada
Member
and
succeeding
Poll
a
`Mhuilt
Member
at
Enard
Bay,
looking
north-west.
The
planar-bedded
sandstones
of
the
Poll
a
`Mhuilt
Member
onlap
against
the
thin
dust
pellet
layer
(Division
C
of
Branney
and
Brown,
2011),
the
gently
curved
layer
that
drapes
a
gently
plunging
trough
incised
into
the
lapilli
beds
in
the
upper
part
of
the
Stac
Fada
Member.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
751

surface
is
well
preserved
some
of
this
relief
can
be
seen
to
be
due
to
broad
(10
m
wide)
lobate
features
which
are
crossed
by
gently
arcuate
fractures
that
are
convex
to
the
west
but
do
not
extend
up
into
the
base
of
the
overlying
Poll
a
`Mhuilt
Member
(Fig.
12).
The
uppermost
10
cm
of
the
Stac
Fada
Member
at
Stattic
Point
and
Second
Coast
is
softer,
and
the
melt
clasts
smaller
(mostly
<1
cm)
than
lower
in
the
member
but,
unlike
the
sections
at
Stoer
and
Enard
Bay,
accretionary
lapilli
have
not
been
seen
in
the
upper
part
of
the
Stac
Fada
Member
at
any
of
these
southern
locations.
The
Stac
Fada
Member
has
its
thinnest
development
in
a
small
outcrop
at
Achiltibuie,
where
just
2
m
of
typical
melt-rich
breccia
rests
directly
on
Lewisian
Gneiss
(Fig.
3).
The
melt-rich
breccia
here
has
a
more
thinly
bedded
foliose
fabric
than
is
typical
for
other
sites,
although
reminiscent
of
the
uppermost
part
of
the
member
at
Second
Coast
for
instance.
Despite
its
relative
proximity
to
Enard
Bay,
no
accretionary
lapilli
have
been
observed
at
Achiltibuie.
At
Achiltibuie,
Stattic
Point
and
Second
Coast
the
Stac
Fada
Member
is
succeeded
by
up
to
15
m
of
planar-bedded
sandstones,
siltstones
and
mudstones
that
are
assigned
to
the
Poll
a
`Mhuilt
Member
(Fig.
3).
At
Stattic
Point
the
lowest
two
or
three
beds
are
concordant
with
the
Stac
Fada
Member
beneath
but
they
are
succeeded
by
several
metres
that
display
intense
and
large-scale
soft-sediment
folding
and
deformation
(Fig.
12).
This
deformation
is
particularly
spectacular
at
Sro
`n
na
Fa
`ire
Mo
´ire,
south
of
Stattic
Point
(Grid
ref.
NG
964945).
Some
confusion
surrounds
the
stratigraphic
position
of
these
disturbed
strata
at
Stattic
Point.
Lawson
(1972)
regarded
them
as
representing
the
upper
6
m
of
the
Stac
Fada
Member
while
Stewart
(2002)
placed
them
within
the
Meall
Dearg
Formation
more
than
15
m
above
the
Stac
Fada
Member.
However,
at
each
of
these
sites
the
relationship
of
the
deformation
to
the
top
of
the
Stac
Fada
Member,
no
more
than
a
metre
or
two
below,
is
clearly
visible
(Fig.
12).
There
is
no
clear
preferred
orientation
of
the
folding
here,
or
at
any
other
site,
with
fold
axes
varying
from
0108
to
1108.
6.
Interpreting
the
Stac
Fada
impact
sequence
At
most
sites
the
Stac
Fada
Member
rests
on
non-marine
clastics,
predominantly
sandstones,
that
are
tens
to
hundreds
of
metres
thick.
These
strata
commonly
show
evidence
of
soft-sediment
Fig.
10.
Dimensions
and
azimuths
of
lithic
clasts
embedded
in
the
Clachtoll
Formation
immediately
beneath
the
Stac
Fada
Member
at
Second
Coast.
(a)
Major
vs.
minor
clast
dimensions.
(b)
Orientations
of
major
axes
for
47
lithic
clasts.
Fig.
9.
Angular
spallation
blocks
of
gneiss
embedded
in
the
sandstone
surface
immediately
below
the
Stac
Fada
Member,
Second
Coast.
The
slab
on
the
right
is
embedded
vertically.
Hammer
is
37
cm
long.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
752

deformation
(Stewart,
2002)
but
all
are
relatively
small-scale
localized
phenomena.
There
is
nothing
in
the
sediments
beneath
the
Stac
Fada
Member
that
might
be
interpreted
as
a
regionally
extensive
seismite
associated
with
an
impact
(Terry
et
al.,
2001),
indicating
that
the
present
outcrop
of
the
Stoer
Group
is
located
both
beyond
the
crater
rim
and
any
significant
seismic
effects
associated
with
the
impact.
The
seemingly
anomalous
occurrence
of
angular
gneiss
blocks
embedded
in
the
sandstone
surface
immediately
beneath
the
Stac
Fada
Member
at
Second
Coast
was
noted
by
Stewart
(2002,
p.
87).
He
attributed
their
presence
to
deposition
from
a
hypercon-
centrated
flow
but
the
matrix
surrounding
them
is
identical
to
the
underlying
sandstones
with
no
evidence
of
any
erosive
contact,
suggesting
that
the
sand
matrix
and
the
large
clasts
embedded
in
it
are
the
products
of
two
unrelated
events.
The
clasts
immediately
beneath
the
Stac
Fada
Member
are
predominantly
large
(Fig.
10a)
and
angular
and
occur
as
isolated
blocks
within
a
well-sorted
sand
matrix
(Fig.
9).
Both
breccias
and
thin
sheet-flood
deposits
occur
at
other
levels
in
the
Clachtoll
Formation
at
Second
Coast
(Stewart,
2002,
fig.
81),
but
they
are
sedimentologically
distinct
from
this
occurrence.
Breccias
derived
from
the
underlying
Lewisian
Gneiss
occur
within
the
lowest
10
m
of
the
Clachtoll
Formation
here
and
elsewhere
(Fig.
3)
but
these
have
a
framework-supported
rather
than
matrix-supported
fabric,
and
clasts
that
are
subangular
to
subrounded,
rather
than
almost
exclusively
angular,
and
seldom
more
than
40
cm
across
(Stewart,
2002).
At
higher
levels
in
the
Fig.
12.
Ogive
fractures,
gently
convex
to
the
west,
on
the
planar
top
of
the
Stac
Fada
Member,
Stattic
Point.
The
vertical
face
behind
(4
m
high)
exposes
disturbed
sandstones
of
the
Poll
a
`Mhuilt
Member.
The
curved
fractures
do
not
cut
the
sandstones
of
the
Poll
a
`Mhuilt
Member
above
nor
do
they
extend
more
than
a
metre
into
the
Stac
Fada
Member
below.
Fig.
11.
The
Stac
Fada
Member
and
contiguous
strata
at
Stattic
Point,
looking
west,
showing
undulating
shear
planes
within
it
and
their
relationship
to
thickness
variations.
Width
of
face
20
m.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
753

Clachtoll
Formation
there
are
just
thin
horizons
of
coarse
sheet-
flood
debris
composed
largely
of
well-rounded
pebbles
(Fig.
13)
mostly
less
than
10
cm
across.
Palaeorelief
on
the
Lewisian
basement
here
was
at
least
200
m
at
the
onset
of
deposition
of
the
Stoer
Group
(Stewart,
2002)
and
Lewisian
inliers
with
substantial
relief
are
located
within
a
few
hundred
metres
of
the
present
outcrop
of
the
Stac
Fada
Member
at
several
sites
(Second
Coast,
Stattic
Point,
Enard
Bay).
These
might
be
invoked
as
a
source
of
the
angular
debris
directly
beneath
the
Stac
Fada
Member
and
basement
inliers
are
indeed
a
proven
source
of
large
gneiss
blocks
within
the
Clachtoll
Formation
at
various
sites
and
the
Poll
a
`Mhuilt
Member
at
Enard
Bay
(Davison
and
Hambrey,
1996;
Stewart,
2002),
but
invariably
such
debris
occurs
no
more
than
about
10
m
either
vertically
or
laterally
from
basement
palaeo-outcrops
(Stewart,
1997,
2002).
However,
when
the
westward
dip
of
the
Stoer
Group
is
taken
into
account
it
becomes
evident
that
many
of
these
large
inliers
were
already
buried
or
greatly
reduced
in
extent
before
the
Stac
Fada
Member
was
emplaced,
thereby
requiring
transport
of
large
angular
blocks
for
hundreds
of
metres
across
a
virtually
flat
sand
surface.
Flow
sufficiently
powerful
to
achieve
this
might
be
expected
to
align
at
least
some
of
the
clasts
yet
their
long
axes
show
no
preferred
orientation
at
all
(Fig.
10b),
with
several
even
embedded
at
steep
angles
to
the
sandstone
surface
(Figs.
9
and
10a).
With
the
Stac
Fada
Member
now
reinterpreted
as
a
melt-rich
impact
ejecta
deposit
and
the
restriction
of
these
angular
clasts,
randomly
orientated
and
remote
from
any
basement
inlier,
to
the
surface
immediately
beneath
it,
a
genetic
link
with
the
impact
seems
distinctly
plausible.
The
spectacular
examples
at
Second
Coast
can
be
interpreted
as
spallation
ejecta,
launched
ballistically
during
the
initial
moments
of
the
impact,
rather
than
as
a
conventional
breccia
or
sheet
flood
deposit.
Similar,
more
scattered
clasts
occur
on
the
equivalent
surface
at
Stattic
Point,
and
others
embedded
in
the
upper
surface
of
some
of
the
sandstone
rafts
at
Stac
Fada
(Fig.
3)
may
be
interpreted
likewise.
At
Enard
Bay
these
putative
spallation
blocks
beneath
the
Stac
Fada
Member
locally
may
be
very
abundant,
in
some
cases
are
more
than
a
metre
across,
and
occur
only
about
150
m
away
from
a
basement
palaeo-outcrop.
However,
the
presence
of
shattered,
but
otherwise
undisturbed,
gneiss
clasts
within
the
breccia
layer
(Fig.
7)
is
incompatible
with
lateral
transport
of
basement
debris
over
such
distances
but
it
is
consistent
with
their
emplacement
as
ballistic
ejecta,
either
shattering
on
impact
or,
more
probably,
being
shattered
by
the
subsequent
impact
of
other
ballistic
ejecta
debris.
The
presence
of
occasional
well-rounded
cobbles
within
this
breccia
horizon
at
each
of
the
sites
does
not
conflict
with
this
hypothesis
since
spallation
ejecta
are
launched
from
surface
or
near
surface
locations,
where
any
pebbles
or
cobbles
that
are
present
will
constitute
ready-made
projectiles.
In
general
this
putative
spallation
debris
does
not
appear
to
show
any
consistent
pattern
of
relative
abundance
or
size
between
the
various
sites
that
might
be
interpreted
as
indicating
relative
proximity
to
the
impact
site.
Debris
is
particularly
abundant
at
Second
Coast
but
at
Stattic
Point
and
Stac
Fada
it
is
noticeably
more
sparse.
At
Enard
Bay
both
the
size
and
abundance
of
clasts
varies
significantly
across
quite
a
limited
area
of
exposure.
Clustering
of
secondary
craters,
formed
by
impact
ejecta,
is
common
around
large
impact
craters
on
the
Moon
and
other
planets
and
indicates
that
ejecta
distribution
in
general
is
not
uniform
(McEwen
et
al.,
2005).
This
may
provide
an
explanation
for
the
variation
in
abundance
of
spallation
debris
beneath
the
Stac
Fada
Member.
It
might
be
envisaged
that
spallation
blocks
ejected
from
the
periphery
of
an
impact
perhaps
several
tens
of
kilometres
away
might
be
lithologically
distinct
from
the
local
country
rocks,
but
this
need
not
be
the
case.
Thrust
slices
of
Lewisian
Gneiss
have,
in
some
cases,
been
displaced
westwards
by
several
tens
of
kilometres
within
the
Moine
Thrust
Belt
(Coward
et
al.,
1980)
yet
lithological
comparison
with
nearby
in
situ
Lewisian
has
gone
unremarked
suggesting
that
any
lithological
differences
that
might
exist
are
not
obvious.
It
would
seem
that
there
is
insufficient
systematic
lateral
variation
within
individual
Lewisian
terranes
(Kinny
et
al.,
2005;
Park,
2005)
to
infer
from
the
lithology
of
these
putative
spallation
clasts
the
distance
that
they
have
traveled.
However,
in
the
specific
instance
of
Second
Coast
it
is
perhaps
worth
noting
that
whereas
mafic
lithologies
form
a
significant
component
of
the
adjacent
Gruinard
Bay
Lewisian
outcrop
(Rollinson
and
Fowler,
1987),
the
spallation
clasts
at
Second
Coast
are
predominantly
felsic
in
nature
with
fewer
than
5%
composed
of
mafic
lithologies.
The
preservation
of
these
apparently
undisturbed
spallation
blocks
directly
beneath
the
Stac
Fada
Member
indicates
that
emplacement
of
the
Stac
Fada
Member
ejecta
sheet
itself
was
a
non-erosive
process
yet
the
entrainment
of
sediment
rafts
and
intrusion
of
melt-rich
breccia
into
underlying
strata
at
Stac
Fada
would
appear
to
contradict
this.
However,
several
lines
of
evidence
indicate
that
these
clearly
erosional
features
developed
after
an
initially
non-erosive
phase
of
emplacement.
Deformation
of
the
sandstone
rafts
indicates
that
they
were
composed
of
unlithified,
and
hence
easily
disaggregated,
sand.
Their
survival
as
coherent
slabs,
complete
with
spallation
ejecta
blocks
embedded
in
their
upper
surfaces,
suggests
that
they
were
completely
enveloped
by
melt-rich
breccia
during
entrainment
and
hence
the
unlithified
sand
must
already
have
been
buried
by
the
advancing
ejecta
blanket
before
these
rafts
became
detached.
The
relatively
high
emplacement
temperature
of
the
Stac
Fada
Member,
compared
with
lithic
breccias
in
the
lower
part
of
other
terrestrial
ejecta
deposits
(200
8C
vs.
<50
8C),
may
help
to
explain
its
initial
non-
erosive
emplacement,
the
subsequent
disruption
of
the
underlying
strata
at
Stac
Fada,
and
also
the
apparent
absence
of
comparable
intrusions
and
sediment
rafting
at
any
of
the
other
sites.
Surface
water,
for
which
there
is
abundant
evidence
in
the
Stoer
Group
(Stewart,
2002),
may
have
been
flash-heated
by
the
advancing
ejecta
blanket,
generating
a
cushion
of
superheated
steam
on
which
it
traveled
(Freundt,
2003).
Steam
generated
beneath
the
Fig.
13.
Relative
abundances
of
angular
and
rounded
lithic
clasts
at
two
horizons
within
the
Clachtoll
Formation
at
Second
Coast.
(a)
Bedding
surface
10
m
below
the
Stac
Fada
Member.
(b)
Bedding
surface
immediately
beneath
the
Stac
Fada
Member.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
754

surface
was
able
to
escape
upwards
through
the
porous
sands
at
most
sites
but
at
Stac
Fada
it
became
trapped
beneath
the
thin
interbedded
mudstones.
Its
subsequent
explosive
release
dis-
rupted
the
sands
above,
allowing
intrusion
of
melt-rich
breccia
along
bedding
planes
and
the
detachment
of
slabs
of
unlithified
sediment.
The
significant
thinning
of
the
Stac
Fada
Member
over
the
basement
inlier
at
Achiltibuie,
and
its
apparent
absence
across
the
inlier
at
Enard
Bay
West,
suggests
that
the
ejecta
blanket
was
emplaced
as
a
ground-hugging
flow
influenced
by
topography,
as
has
been
observed
for
Martian
impact
ejecta
(Barlow
et
al.,
2007;
Osinski
et
al.,
2011b),
rather
than
as
a
ballistically
deposited
blanket.
Its
absence
at
Enard
Bay
West
might
be
attributed
to
subsequent
erosion
but
the
lack
of
any
erosional
modification
of
the
draping
dust
pellet
layer,
Division
C
of
Branney
and
Brown
(2011),
by
the
encroaching
waters
of
the
lake
that
subsequently
deposited
the
Poll
a
`Mhuilt
Member
(Fig.
8)
indicates
that
the
Stac
Fada
Member
was
already
well
indurated,
with
early
authigenic
feldspar
and
silica
cements,
soon
after
deposition.
The
small
melt
shards
within
the
limestone
at
Enard
Bay
West
might
represent
late-stage
fallout
from
the
impact
plume,
but
if
a
more
substantial
representation
of
the
ejecta
layer
had
once
existed
there
we
would
expect
to
see
greater
evidence
of
it.
Branney
and
Brown
(2011)
compared
the
Stac
Fada
Member
with
volcanic
pyroclastic
deposits
and
considered
that
most
of
it
was
deposited
from
a
catastrophic
density
current
in
which
deposition
accompanied,
rather
than
followed,
transport.
They
interpreted
the
uppermost
unit
(their
Division
C),
which
drapes
topographic
irregularities
without
any
change
in
thickness
or
grain
size
(Fig.
8),
as
a
typical
fallout
deposit
from
a
residual
dust
cloud.
From
the
general
absence
of
bedforms,
such
as
ripples
and
cross
stratification,
through
much
of
the
Stac
Fada
Member
they
inferred
that
there
was
a
high
concentration
of
particles
but
that
the
lower
part
of
the
flow
behaved
as
a
granular
fluid
in
which
the
particles
were
suspended
by
escaping
interparticle
fluids
and
gas.
The
dense
granular
fluid
passed
up
into
a
hot
buoyant
plume
in
which
dust
particles
agglomerated
and,
when
they
became
too
large,
dropped
into
the
turbulent
upper
part
of
the
density
current
beneath
where
they
accreted
concentric
dust
laminae
before
finally
settling
to
the
base
of
the
granular
fluid
zone.
They
attributed
the
absence
of
accretionary
lapilli
in
the
lower
part
of
the
Stac
Fada
Member
at
Enard
Bay
(their
Division
A)
to
the
delay
between
the
arrival
of
the
density
current
and
the
growth
of
accretionary
lapilli
from
dust
particles
settling
out
of
the
developing
plume
above.
As
such
the
marked
reduction
in
thickness
of
the
lapilli
beds
(>3
m
vs.
<1
m)
and
size
of
lapilli
(15
mm
vs.
5
mm)
between
Enard
Bay
and
Stac
Fada,
and
the
absence
of
lapilli
from
any
of
the
sites
further
south,
suggests
a
proximal-distal
transition
both
northwards
and
southwards
of
Enard
Bay.
The
planar
cross
beds
in
the
upper
part
of
the
lapilli
beds
at
Enard
Bay,
and
the
troughs
which
cut
through
the
lapilli
beds
and
are
now
draped
by
the
fallout
layer,
indicate
episodes
of
both
aggradation
and
scouring
in
the
later
stages
of
deposition
of
the
Stac
Fada
Member
there.
As
the
granular
fluid
progressively
degassed
its
viscosity
correspondingly
increased
and
its
fluid
behaviour
would
have
changed.
Several
features
of
the
more
southerly
exposures
of
the
Stac
Fada
Member
suggest
movement
of
a
far
more
viscous
medium
in
the
latter
stages
of
emplacement,
implying
that
these
sites
were
located
further
from
the
impact
site
than
were
Enard
Bay
and
Stoer.
The
arcuate
fractures,
convex
to
the
west,
on
the
upper
surface
of
the
Stac
Fada
Member
at
Stattic
Point
(Fig.
12)
may
represent
ogives,
which
are
curved
pressure
ridges
that
develop
on
the
surface
of
slowly
moving
viscous
materials,
such
as
mudflows,
and
are
convex
in
the
direction
of
flow.
Similarly,
the
curved
or
undulating
planes
within
the
Stac
Fada
Member
at
Stattic
Point
(Fig.
11)
and
other
sites
in
the
southern
part
of
the
outcrop
may
represent
synsedimentary
shear
surfaces
or
thrust
planes
that
developed
within
a
slowly
moving
viscous
medium
in
the
later
stages
of
emplacement.
Curved
shear
planes,
concave
upwards
and
rising
towards
the
surface,
develop
near
glacier
snouts
and
in
viscous
mudflows
as
higher
parts
of
the
flow
override
lower
more
slowly
moving
layers
(Benn
and
Evans,
2010).
The
truncation
or
displacement
of
the
elutriation
pipes
across
these
structures
is
consistent
with
this
model.
Movement
may
have
been
greater
in
the
upper
part
of
the
Stac
Fada
Member
where
viscosity
was
somewhat
lower
than
towards
the
base,
producing
more
closely
spaced
shear
planes
and
the
lobed
and
undulating
upper
surface.
The
random
orientation
of
the
melt
clasts
and
absence
of
shear
planes
at
Stac
Fada
and
Enard
Bay
suggest
that
viscosity
increased
very
rapidly,
effectively
‘freezing’
the
ejecta
deposit
at
these
more
northern
sites
before
the
melt
clasts
could
become
preferentially
aligned
(Amor
et
al.,
2008).
Branney
and
Brown
(2011)
considered
that
the
elutriation
pipes
formed
through
channelling
of
dusty
gas
upwards
through
the
progressively
aggrading
and
degassing
deposit.
The
local
align-
ment
of
melt
clasts
parallel
to
these
pipes
is
consistent
with
degassing
through
a
relatively
low
viscosity
medium
but
the
pervasive
deposition
of
authigenic
feldspar
in
cavities
and
pore
spaces
throughout
the
Stac
Fada
Member
indicates
that
degassing
continued
for
some
time
after
deposition
but
while
the
ejecta
was
still
very
hot.
Hence
the
temperature
reported
for
the
Stac
Fada
Member
by
Parnell
et
al.
(2011)
may
represent
an
early
post-
depositional
temperature
rather
than
the
initial
emplacement
temperature
of
the
ejecta
layer.
7.
Locating
the
impact
site
There
is
no
clear
consensus
among
previous
authors
as
to
the
location
or
proximity
of
the
ultimate
source
of
the
Stac
Fada
Member,
whether
this
was
a
volcano
or
an
impact.
There
have
been
suggestions
for
sources
to
the
north
and/or
west
(Lawson,
1972;
Young,
2002;
Amor
et
al.,
2008,
2011)
or
to
the
east
(Stewart,
2002),
with
distances
of
perhaps
as
little
as
a
few
kilometres
from
the
present
outcrop
(Lawson,
1972;
Young,
2002).
Recognition
that
the
Stac
Fada
Member
was
generated
by
an
impact
necessarily
implies
a
single
source
while
the
relatively
thick,
and
near
continuous,
nature
of
the
Stac
Fada
Member
led
Amor
et
al.
(2008)
to
infer
that
it
was
fairly
proximal
to
the
impact,
although
without
specifying
the
distance.
Nonetheless,
its
stratigraphic
position
above
a
thick
and
undisturbed
sedimentary
succession
(Fig.
3)
suggests
that
the
impact
crater
was
still
a
significant
distance
away,
perhaps
tens
of
kilometres,
while
the
extent
of
the
preserved
outcrop
(Fig.
1)
implies
that
the
crater
from
which
it
was
ejected
must
itself
have
been
tens
of
kilometres
across.
Since
the
present
outcrop
is
effectively
linear
at
least
some
evidence
for
proximal-
distal
changes
along
the
outcrop
might
be
anticipated,
as
has
been
described
in
the
previous
section.
The
abundance
of
accretionary
lapilli
in
the
Stac
Fada
Member
at
Enard
Bay,
and
their
absence
from
sites
further
south,
has
led
previous
authors
to
suggest
that
this
site
was
closest
to
the
source
(Lawson,
1972;
Young,
2002;
Amor
et
al.,
2008;
Branney
and
Brown,
2011),
with
the
smaller
and
less
abundant
accretionary
lapilli
to
the
north,
at
Stac
Fada
and
Cnoc
Breac,
indicating
that
these
might
also
be
relatively
proximal
locations.
The
greater
thickness
(10–15
m)
of
the
Stac
Fada
Member
at
the
two
northern
sites,
Stoer
and
Enard
Bay,
compared
with
sites
further
south
(4–
6
m
thick),
and
the
facies
differences
between
them
(Fig.
3),
also
provides
tentative
evidence
that
the
southern
sites
may
be
more
distal
to
the
impact.
The
presence
of
a
thin
(2
m)
laminated
facies
at
Achiltibuie,
little
more
than
6
km
south
of
Enard
Bay,
might
seem
to
conflict
with
this
interpretation,
but
here
it
rests
directly
on
Lewisian
basement
and
hence
may
represent
just
the
upper
part
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
755

of
the
Stac
Fada
Member,
although
the
apparent
absence
of
accretionary
lapilli
does
cast
some
uncertainty
over
such
an
interpretation.
Despite
the
extent
of
the
Stac
Fada
Member
outcrop
relatively
few
sedimentary
features
have
been
identified
in
association
with
it
that
might
be
used
to
demonstrate
the
direction
of
emplacement.
This
may
be
ascribed
to
the
largely
non-erosive
emplacement
and
relatively
massive
nature
of
the
Stac
Fada
Member,
in
which
a
metre-scale
planar
fabric
and
feldspar-lined
elutriation
pipes
are
the
only
widespread
sedimentary
structures.
Stac
Fada
has
been
the
focus
of
most
analyses
concerning
the
emplacement
direction
of
the
melt-rich
breccia,
with
the
most
robust
evidence
coming
from
the
intrusions
of
melt-rich
breccia
injected
along
bedding
planes
in
the
underlying
strata
(Fig.
5).
Lawson
(1972)
and
Amor
et
al.
(2008)
maintained
that
the
material
was
moving
from
west
to
east,
from
which
the
latter
authors
inferred
an
impact
in
what
is
now
the
Minch
Basin
(Fig.
4).
Subsequent
AMS
(Anisotropy
of
Magnetic
Susceptibility)
analysis
of
the
Stac
Fada
Member
from
several
sites
led
Amor
et
al.
(2011)
to
reaffirm
a
western
location
for
the
impact,
between
Stoer
and
Stornaway.
However,
all
four
of
the
intrusions
exposed
on
the
shore
at
Stac
Fada
thicken
eastwards
in
a
manner
that
is
consistent
with
material
being
forcibly
injected
westwards
along
bedding
planes,
implying
that
the
source
lay
to
the
east
(Figs.
5
and
6).
Young
(2002)
dismissed
any
suggestions
of
these
structures
as
intrusions
and
instead
used
cross-cutting
relationships,
small-
scale
folding
and
flame
structures
associated
with
sandstone
sheets
within
the
Stac
Fada
Member
to
identify
two
distinct
units,
one
supposedly
emplaced
from
the
south-west
and
the
other
from
the
north-east.
Rotation
of
the
rafted
sandstones
during
transport,
as
indicated
by
their
random
palaeomagnetic
orientations
(Irving
and
Runcorn,
1957;
Stewart,
2002),
may
explain
this
apparent
discordance
with
the
source
direction
as
inferred
from
the
intrusions.
In
contrast,
the
oversteepened
sandstone
beds
that
overlie
the
intrusive
wedges
are
firmly
anchored
into
the
pre-
impact
stratigraphy
and
hence
preserve
a
more
robust
record
of
the
emplacement
direction
of
the
ejecta.
At
Enard
Bay
the
planar
cross-beds
in
the
upper
part
of
the
lapilli-bearing
units
indicate
that
material
was
moving
broadly
from
east
to
west
during
deposition,
which
is
consistent
both
with
the
orientation
of
lapilli
long
axes
in
the
upper
part
of
the
Stac
Fada
Member
and
with
the
gently
plunging
troughs
incised
subse-
quently
into
the
lapilli
beds
(Figs.
6
and
8).
At
sites
further
south
the
evidence
for
emplacement
direction
is
more
equivocal.
If
the
curved
fractures
on
the
upper
surface
of
the
Stac
Fada
Member
at
Stattic
Point
(Fig.
12)
are
indeed
ogives,
then
their
convex-
westwards
configuration
indicates
movement
from
east
to
west
(Fig.
6).
Similarly,
if
the
concave-up
surfaces
within
the
Stac
Fada
Member
there
represent
thrust
planes
within
a
viscous
flow,
then
their
ascent
westwards
from
base
to
top
also
is
broadly
consistent
with
emplacement
from
the
east.
Direct
evidence
for
the
impact
location
is
very
sparse
within
the
Stac
Fada
Member
itself
but
additional
evidence
can
be
deduced
from
the
post-impact
succession.
At
all
but
the
southernmost
sites,
Poolewe
and
Loch
Thurnaig,
the
Stac
Fada
Member
is
succeeded
by
up
to
100
m
of
planar-bedded
lake
sediments
of
the
Poll
a
`Mhuilt
Member,
representing
the
most
significant
lacustrine
interruption
within
the
predominantly
fluvial
environment
of
the
Stoer
Group
(Stewart,
2002;
Stewart
and
Parker,
1979).
Amor
et
al.
(2008)
attributed
this
facies
change
to
a
radical
reconfiguration
of
the
regional
drainage
pattern
arising
from
damming
of
the
pre-impact
rivers
by
debris
from
the
impact.
Eastward-draining
rivers
in
the
north
were
impounded
by
ejecta
dams
located
to
the
east
of
the
present
outcrop,
but
they
continued
to
drain
east
into
rapidly
growing
lakes
as
evidenced
by
the
cross
bedding
in
the
lower
part
of
the
Poll
a
`Mhuilt
Member
at
Stoer
(Fig.
6).
In
contrast,
the
prevailing
drainage
in
the
southern
part
of
the
outcrop
was
from
east
to
west
and
so
the
impact
may
have
effectively
beheaded
these
rivers.
In
marked
contrast
to
the
variable
directions
of
the
pre-impact
drainage,
palaeocurrent
azimuths
throughout
the
Meall
Dearg
Formation,
which
succeeded
the
Poll
a
`Mhuilt
Member,
are
broadly
westwards
along
the
entire
outcrop;
to
the
west-north-west
at
Stac
Fada,
almost
due
west
at
Enard
Bay,
and
to
the
south-west
at
Poolewe
(Fig.
2).
Stewart
(2002)
attributed
this
drainage
shift
to
the
effects
of
tectonic
uplift
on
the
eastern
flank
of
the
basin
but
the
abruptness,
scale
and
pattern
of
drainage
reconfiguration
implies
that
it
was
caused
by
a
major
catastrophic
event,
unlike
any
other
encountered
in
the
Torridon
Supergroup,
that
caused
a
regional
doming
centred
several
tens
of
kilometres
to
the
east
of
the
present
outcrop.
For
a
crater
of
final
diameter
40–50
km,
a
figure
perhaps
broadly
consistent
with
the
extent
and
uniformity
of
the
ejecta
outcrop
preserved
today,
hundreds
of
cubic
kilometres
of
material
might
be
ejected
during
excavation
of
a
transient
crater
about
25
km
across
and
8
km
deep
(Collins
et
al.,
2005).
Isostatic
rebound
inevitably
would
follow
the
impact
(Melosh,
1989)
and,
by
analogy
with
post-glacial
isostasy,
significant
uplift
(at
least
many
tens
of
metres)
would
have
occurred
within
just
a
few
thousand
years
(Eronen
et
al.,
2001)
and
been
centred
on
the
crater.
This
phase
of
uplift
may
have
caused
the
draining
of
the
Poll
a
`Mhuilt
Member
lakes
and
perhaps,
in
its
initial
stages,
triggered
the
slumping
and
large-scale
deformation
of
some
of
the
lake
deposits
(Fig.
12).
In
particular
it
can
explain
the
subsequent
establishment
of
the
radial
drainage
system
of
the
Meall
Dearg
Formation
which
is
more
consistent
with
this
model
of
crater-induced
regional
doming
than
the
essentially
linear
basin
margin
uplift
that
was
envisaged
by
Stewart
(2002).
Significantly,
the
broader
spread
of
the
palaeo-
current
data
from
the
Meall
Dearg
Formation
at
Poolewe
(Figs.
2
and
6)
is
consistent
with
its
inferred
more
distal
location
where
the
effects
of
the
regional
doming
would
have
been
correspondingly
reduced.
Reworked
melt
clasts
from
the
Stac
Fada
Member
are
present
in
the
lower
part
of
the
Poll
a
`Mhuilt
Member
but
their
apparent
absence
within
the
Meall
Dearg
Formation,
noted
by
Stewart
(2002),
might
seem
to
conflict
with
the
suggestion
of
an
impact
crater,
and
associated
thick
ejecta
deposits,
to
the
east.
However,
volcanic
glass,
and
by
analogy
impact
glass,
is
thermodynamically
unstable
and
decomposes
more
readily
than
almost
all
associated
mineral
phases.
This
is
particularly
the
case
with
mafic
and
ultramafic
glasses,
which
weather
to
smectite,
allophane
and
other
clay
minerals
(Fisher
and
Schmincke,
1984).
In
deposits
rich
in
melt
glass
the
weathering
front
may
move
downwards,
forming
a
clay-rich
soil,
at
a
rate
of
0.5
m/ka
(Hay,
1960).
The
absence
of
reworked
melt
clasts
in
the
Meall
Dearg
Formation
may,
therefore,
be
due
to
the
elimination
of
this
material
from
surface
exposures
by
rapid
weathering,
producing
fine
clays
that
are
poorly
represented
in
the
well-sorted
sands
of
the
Meall
Dearg
Formation.
8.
The
Lairg
Gravity
Low:
geophysical
evidence
for
a
buried
impact
structure
The
evidence
from
both
the
ejecta
deposit
(Stac
Fada
Member)
and
the
post-impact
succession
(Poll
a
`Mhuilt
Member
and
Meall
Dearg
Formation)
indicates
that
the
impact
crater,
probably
tens
of
kilometres
across,
was
located
several
tens
of
kilometres
to
the
east
of
the
Stoer
Group
outcrop.
An
impact
to
the
west,
as
suggested
by
Amor
et
al.
(2008,
2011),
can
be
discounted
and
the
offshore
gravity
low
in
The
Minch
(Fig.
4)
can
instead
be
attributed
to
a
thick
post-Palaeozoic
sediment
fill
in
the
fault-bounded
Minch
Basin
(Binns
et
al.,
1975).
Today
the
Stoer
Group
outcrop
is
little
more
than
20
km
beyond
the
western
edge
of
the
Moine
Thrust
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
756

Fig. 14. Regional geology of northern Scotland showing the outcrop of the Stoer Group and its relationship to the residual gravity field (contoured at 2 mGal intervals) and gravity lineaments for the Lairg Gravity Low (from Rollin,
2009). Arrows indicate directional azimuths within the Stac Fada Member; for details see Fig. 6. Rose diagrams show palaeocurrent data for the Meall Dearg Formation (partly from Stewart, 2002, Fig. 17) and long axis orientations of
accretionary lapilli. The fine dotted lines are radii projected from the centre of the Lairg Gravity Low through four key exposures of the Stoer Group.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
757

Belt
(Fig.
1),
to
the
east
of
which
any
sediments
or
structures
contemporary
in
age
with
the
Stoer
Group
will
lie
buried
beneath
a
thick
cover,
mostly
of
metasedimentary
rocks
of
the
Moine
Supergroup
(Neoproterozoic).
The
Moine
Thrust
Belt
is
a
major
tectonic
structure
formed
by
the
collision
of
Laurentia
with
Baltica
about
430
million
years
ago
and
representing
the
north-west
limit
of
Caledonian
deformation
in
Scotland.
Moinian
rocks
were
thrust
westwards
across
an
undeformed
foreland
of
Archean
to
Lower
Palaeozoic
rocks
but
thrust
slabs
of
other
stratigraphic
units,
including
Lewisian
basement,
were
also
displaced
and,
in
some
instances,
moved
westwards
along
the
thrust
zone
for
tens
of
kilometres
(Mendum
et
al.,
2009).
Assuming
that
the
putative
Mesoproterozoic
impact
structure
was
not
subsequently
destroyed
by
erosion,
or
by
the
effects
of
Caledonian
tectonics,
it
must
lie
buried
beneath
the
thrust
sheets
to
the
east
and
hence
will
have
no
physical
manifestation
at
the
surface.
The
Lairg
Gravity
Low
lies
to
the
east
of
the
Moine
Thrust
Belt
in
a
location
that
shows
a
remarkable
correspondence
with
that
inferred
for
the
impact
crater
from
the
directional
data
recorded
from
the
Stac
Fada
Member
(Fig.
14).
It
is
one
of
the
most
prominent
features
of
the
onshore
gravity
field
in
northern
Scotland
(Rollin,
2009)
yet
there
is
no
clear
consensus
either
on
its
present
cause
or
its
ultimate
origin.
Rollin
(2009)
summarized
previous
interpretations;
a
Caledonian
granite
pluton,
Laxfordian
(=Archean)
granite
in
the
basement,
a
faulted
block
of
Moine,
or
a
buried
wedge
of
Torridonian
strata.
Butler
and
Coward
(1984)
attributed
the
Lairg
Gravity
Low
to
a
thickening
of
the
Moine
sequence
linked
to
westward
transfer
of
Lewisian
thrust
sheets
to
the
Assynt
area.
Leslie
et
al.
(2010)
similarly
invoked
a
thickening
of
Moine
rocks,
with
a
modeled
5–6
km
thickness
at
the
centre
of
the
Lairg
Gravity
Low.
They
attributed
its
western
flank
to
an
eastward
thickening
of
the
Moine
sequence
above
the
Moine
and
Ben
Hope
thrusts,
its
steep
eastern
flank
to
a
thick
wedge
of
Lewisian
basement
above
the
Achness
Thrust,
and
its
southern
flank
to
dense
Moinian
pelites
above
shallow
basement.
However,
they
acknowledged
that
it
was
difficult
to
explain
all
of
the
observed
gravity
variation
simply
in
terms
of
lithological
contrast
within
the
Moine
succession.
One
of
the
suggestions
made
by
Rollin
(2009),
that
the
gravity
low
might
be
due
to
a
buried
wedge
of
Torridonian
strata
is
similar
to
what
might
be
predicted
for
an
impact
crater
several
tens
of
kilometres
across
that
contains
a
post-impact
sediment
fill
many
hundreds
of
metres
thick
above
a
deeply
fractured
basement.
The
Lairg
Gravity
Low,
particularly
as
seen
in
the
residual
gravity
field
(Figs.
14
and
15a),
appears
as
a
large,
roughly
circular
structure
with
a
concentric
pattern
of
isogals
reaching
a
maximum
depth
of
23
mGal.
This
is
consistent
with
values
for
other
impact
craters
(Pilkington
and
Grieve,
1992)
and
suggests
an
impact
crater
40
km
across.
Shallower
parts
of
the
crater
may
have
been
removed
by
pre-Caledonian
erosion
and/or
by
Caledonian
thrust
faulting,
in
which
case
the
present
extent
of
the
gravity
low
may
not
fully
reflect
the
original
size
of
the
crater.
Several
gravity
lineaments
cross
or
impinge
on
the
Lairg
Gravity
Low
(Fig.
14).
Gravity
lineaments
represent
local
maxima
of
the
gravity
gradient
magnitude
and
commonly
correspond
to
major
faults
or
other
significant
tectonic
structures
(Aydogan,
2011).
A
straight
lineament
along
the
SW
flank
of
the
Lairg
Gravity
Low
may
represent
the
boundary
between
the
Gruinard
and
Assynt
terranes
of
the
Archean
basement
(Kinny
et
al.,
2005;
Park,
2005),
a
pre-
impact
structure
with
which
it
is
approximately
parallel.
Another
Fig.
15.
Comparison
of
gravity
anomalies
over
the
Lairg
Gravity
Low
and
the
Ries
impact
structure,
to
same
scale.
(a)
Lairg
Gravity
Low.
Bouguer
Residual
Gravity
(mGal)
after
removal
of
regional
field
(from
Rollin,
2009).
(b)
Ries
impact
structure.
Bouguer
Residual
Gravity
(mGal)
after
removal
of
regional
field.
(c)
Ries
impact
structure.
Bouguer
Residual
Gravity,
horizontal
gradient
(mGal/km).
The
steeper
gradients
are
associated
with
the
outer
rim
of
the
crater
and
the
inner
ring.
Ries
gravity
data
reproduced
courtesy
of
Prof.
Kord
Ernstson
(impact-structures.com).
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
758

lineament,
along
part
of
the
relatively
straight
western
flank
of
the
gravity
low,
probably
is
associated
with
the
Moine
Thrust
and
hence
is
a
post-impact
structure
that
may
effectively
define
the
crater’s
present
western
edge.
Intriguingly,
the
most
strongly
curved
gravity
lineament
seen
anywhere
in
Scotland
lies
on
the
north-east
side
of
the
Lairg
Gravity
Low
(Rollin,
2009).
It
invites
comparison
with
a
marked
steepening
of
gravity
gradients
in
a
broad
arc
on
the
east
side
of
the
Ries
impact
structure
(Fig.
15c)
that
corresponds
to
the
inner
ring
of
that
crater
(Pohl
et
al.,
1977).
Might
this
curved
lineament
in
Scotland
perhaps
reflect
a
ring
graben
within
a
complex
crater
that
is
represented
by
the
Lairg
Gravity
Low
as
a
whole?
The
Earth
Impact
Effects
programme
(Collins
et
al.,
2005)
calculates
that
a
2.5
km
diameter
stony
impactor
traveling
at
19
km
s
1
and
impacting
at
an
angle
of
658
will
generate
a
complex
crater
close
to
40
km
in
diameter
and
form
an
ejecta
layer
12
m
thick
at
a
distance
of
65
km
from
the
impact
site.
These
figures
show
a
remarkable
correspondence
with
the
thickness
of
the
ejecta
deposit
at
Stoer
and
its
distance
from
the
centre
of
the
Lairg
Gravity
Low.
The
location
of
the
Lairg
Gravity
Low
is
consistent
with
the
scale
of
the
predicted
impact
structure,
with
its
distance
from
the
present
Stac
Fada
Member
outcrop,
with
the
various
directional
azimuths
associated
with
the
Stac
Fada
Member
and
with
the
inferred
proximal-distal
facies
changes
seen
along
the
outcrop.
Furthermore,
the
post-impact
regional
doming
indicated
by
palaeocurrents
in
the
Meall
Dearg
Formation
similarly
appears
to
be
centred
on
the
Lairg
Gravity
Low.
Of
course,
any
suggestion
that
the
Lairg
Gravity
Low
is
a
buried
impact
crater
must
be
considered
within
the
context
of
the
Moine
Thrust
Belt
and
its
potential
effects
on
the
proposed
target
region.
The
presence
of
thrust
slices
of
Lewisian
gneiss
within
the
Moine
Thrust
Belt
is
clear
evidence
that
the
thrust
planes
have
cut
into
the
basement
at
least
locally
and
hence
may
have
impinged
on
the
putative
impact
crater.
Soper
and
Barber
(1982)
proposed
a
steep
and
deep-rooting
Moine
Thrust
descending
to
more
than
20
km
beneath
the
proposed
target
area,
in
which
case
an
impact
crater
now
represented
by
the
Lairg
Gravity
Low
would
originally
have
been
located
several
tens
of
kilometres
further
east
and
would
be
unlikely
to
represent
the
source
of
the
Stac
Fada
Member
unless
a
very
much
larger
impact
is
invoked.
However,
current
consensus
is
that
the
Moine
Thrust
Belt
is
a
classic
example
of
‘thin-skinned’
tectonics,
dominated
by
detachment
of
slabs
of
upper
crustal
rocks
from
their
deeper
crustal
roots,
in
which
the
Moine
Thrust
remains
in
the
upper
part
of
the
crust
beneath
much
of
the
north-west
Highlands
and
does
not
descend
to
deeper
levels
until
at
least
50
km
east
of
the
Moine
Thrust’s
current
outcrop
trace
(Butler,
2010).
Significantly,
this
model
implies
that
the
thrust
planes
are
largely
or
entirely
above
the
level
at
which
the
impact
crater
is
inferred
to
lie
in
the
Lewisian
basement
beneath
Lairg.
9.
Conclusions
The
discovery
of
shocked
mineral
grains
and
reidite
within
the
Stac
Fada
Member
establishes
unequivocally
that
it
is
of
impact
rather
than
volcanic
origin
(Amor
et
al.,
2008,
2009;
Osinski
et
al.,
2011a;
Reddy
et
al.,
2015),
a
finding
that
gains
additional
support
from
its
anomalous
levels
of
iridium,
nickel
and
chromium-53.
Inevitably
this
raises
the
question
of
the
crater’s
location
since
no
trace
of
an
impact
structure
is
evident
at
the
surface
today,
but
it
does
help
to
explain
certain
enigmatic
aspects
of
the
Stoer
Group.
The
large,
angular,
randomly
orientated
gneiss
clasts
immediately
beneath
the
Stac
Fada
Member
at
several
sites
can
now
be
interpreted
as
spallation
ejecta
launched
ballistically
in
the
first
moments
of
the
impact
rather
than
the
product
of
more
conventional
emplacement
mechanisms.
Similarly,
the
abrupt
reconfiguration
of
the
regional
drainage
following
deposition
of
the
Stac
Fada
Member
can
also
be
seen
as
a
direct
consequence
of
the
effects
of
the
impact.
The
Stac
Fada
Member
differs
in
several
respects
from
previously
described
terrestrial
ejecta
deposits.
Melt
clasts
form
a
significant
component
throughout
the
Stac
Fada
Member
at
all
sites
yet
are
markedly
less
abundant
than
in
other
melt-rich
breccias
such
as
the
Ries
suevite,
and
there
appears
to
be
no
representative
of
the
lithic
breccia
seen
at
Ries.
In
this
respect
the
Stac
Fada
Member
is
more
reminiscent
of
a
Single
Layer
Ejecta
deposit,
perhaps
analogous
to
most
ejecta
deposits
on
Mars,
rather
than
a
Double
Layer
Ejecta
deposit
(Barlow
et
al.,
2007;
Sturm
et
al.,
2013).
As
a
working
hypothesis
the
SLE
character
of
the
Stac
Fada
Member,
comprising
a
relatively
dilute
and
cool
melt-rich
breccia,
can
perhaps
be
attributed
to
impact
into
a
thick
(hundreds
of
metres)
sequence
of
unlithified
water-saturated
sandy
sediment
above
Archean
basement.
This
may
have
generated
a
vast
cloud
of
fine
particles
that
remained
suspended
long
enough
to
mix
with,
and
dilute,
the
melt
that
was
ejected
subsequently
from
deeper
levels
in
the
impact
crater.
Investigation
of
the
processes
involved
in
generating
this
unique
ejecta
deposit
is
the
subject
of
ongoing
research.
The
thickness
and
lateral
extent
of
the
Stac
Fada
Member
impact
ejecta
deposit,
and
absence
of
direct
evidence
of
the
crater
nearby,
attest
to
both
the
scale
of
the
impact
crater
and
its
distance
from
the
present
outcrop
(tens
of
km).
Various
sedimentary
structures
associated
with
the
Stac
Fada
Member,
from
melt-breccia
intrusions
along
bedding
planes
beneath
it,
to
cross-bedding
and
erosional
troughs
in
the
lapilli-bearing
unit
at
its
top,
consistently
indicate
emplacement
of
the
ejecta
deposit
from
the
east.
The
location
of
the
impact
crater
as
inferred
from
these
directional
data
corresponds
remarkably
closely
to
the
Lairg
Gravity
Low,
the
centre
of
which
lies
about
50
km
east
of
the
Stoer
Group
outcrop
at
its
closest
point
of
Enard
Bay
(Fig.
14).
Palaeocurrent
data
from
the
Meall
Dearg
Formation
similarly
indicate
that
there
was
a
post-impact
regional
doming,
a
consequence
of
isostatic
rebound,
that
was
also
centred
on
the
Lairg
Gravity
Low.
The
matrix
and
larger
lithic
fragments
of
the
Stac
Fada
Member
indicate
that
shallower
levels
of
the
target
were
composed
of
unconsolidated
sands
surrounding
inliers
of
predominantly
felsic
gneiss
that
appears
indistinguishable
from
along-strike
outcrops
of
the
western
Lewisian.
The
mafic
composition
of
the
melt
clasts
is
not
so
readily
explained
but,
having
been
excavated
from
deeper
parts
of
the
crater,
it
may
indicate
a
higher
crustal
density
at
several
kilometres
depth
below
the
original
target
surface
which
itself
now
lies
buried
beneath
several
kilometres
of
Moinian
metasediments.
Assuming
a
thin-skinned
model
for
the
Moine
Thrust
(Butler,
2010),
the
crater
in
the
underlying
basement
may
have
remained
largely
unaffected
by
Caledonian
tectonics
and
hence
may
preserve
a
thick
Mesoproterozoic
sedimentary
fill.
Although
several
thin
thrust
sheets
of
Lewisian
gneiss
west
of
the
Lairg
Gravity
Low
are
believed
to
have
been
displaced
westwards
by
several
tens
of
kilometres
(Coward
et
al.,
1980),
the
apparent
absence
of
impact
shatter
cones
at
the
two
sites
so
far
examined,
Ben
More
Assynt
(G.
Steel,
pers.
comm.)
and
Loch
Bealach,
suggests
that
these
thrust
sheets
originate
outside
of
the
crater
and
hence
this
may
help
to
constrain
the
magnitude
of
their
lateral
displacement.
The
structural
models
for
the
Lairg
Gravity
Low
proposed
by
Rollin
(2009)
and
Leslie
et
al.
(2010)
draw
on
observations
of
surface
geology
coupled
with
interpretations
of
geophysical
data.
The
suggestion
made
here,
that
the
Lairg
Gravity
Low
is
a
buried
impact
crater
in
the
basement
beneath
the
Moine
Thrust,
is
based
on
the
same
geophysical
evidence
but
would
be
quite
unwarranted
were
it
not
for
the
existence
of
a
thick
and
extensive
impact
ejecta
layer,
the
Stac
Fada
Member,
just
a
few
tens
of
kilometres
to
the
west.
The
evidence
within
the
Stac
Fada
Member
of
emplacement
from
the
east,
the
inferred
proximal-distal
changes
along
its
outcrop,
the
radial
drainage
pattern
evident
in
the
post-impact
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
759

fluvial
succession
of
the
Meall
Dearg
Formation,
and
similarities
between
the
Lairg
Gravity
Low
and
the
gravity
signature
of
the
well-documented
Ries
impact
crater
in
Germany
(Fig.
15)
add
considerable
weight
to
this
conclusion.
Taking
all
of
the
evidence
together,
the
impact
hypothesis
can
explain
both
the
cause
and
origin
of
the
gravity
low,
and
indeed
its
specific
location,
as
the
consequences
of
a
single
event
for
which
the
Stac
Fada
Member
provides
substantial
surface
evidence.
Acknowledgements
I
am
particularly
indebted
to
Geoff
Steel
for
organizing
a
geological
holiday
in
Scotland
on
which
I
first
encountered
the
Stac
Fada
Member
in
2011.
I
had
not
intended
to
spend
more
than
a
day
or
two
looking
at
the
Stoer
Group,
but
those
anomalous
angular
gneiss
clasts
at
Second
Coast
demanded
an
explanation.
.
.
Geoff’s
assistance
in
the
field,
extensive
knowledge
of
the
region’s
geology,
and
perceptive
comments
have
proved
invaluable
during
the
ensuing
research.
Balz
Kamber,
Ian
Sanders,
John
Graham,
and
Hugh
Rollinson
are
thanked
for
helpful
discussions
during
the
course
of
the
project.
Kord
Ernstson
kindly
gave
permission
to
use
the
gravity
maps
of
the
Ries
impact
structure
in
Fig.
15,
for
which
I
am
grateful.
The
manuscript
has
benefitted
considerably
from
the
incisive
and
constructive
reviews
of
Kathryn
Goodenough
and
Maarten
Krabbendam.
References
Alvarez,
L.W.,
Alvarez,
W.,
Asaro,
F.,
Michel,
H.V.,
1980.
Extraterrestrial
cause
for
the
Cretaceous-Tertiary
extinction.
Science
208,
1095–1108.
Amor,
K.,
Hesselbo,
S.P.,
Porcelli,
D.,
Thackrey,
S.,
Parnell,
J.,
2008.
A
Precambrian
proximal
ejecta
blanket
from
Scotland.
Geology
36,
303–306.
Amor,
K.,
Hesselbo,
S.P.,
Porcelli,
D.,
Thackrey,
S.,
Parnell,
J.,
2009.
Stac
Fada
Member
(Torridonian
Supergroup,
Scotland);
a
Mesoproterozoic
proximal
impact
ejecta
blanket.
Meteoritics
and
Planetary
Science
44
(Supplement)
,
A22,
72nd
Annual
Meteoritical
Society
Meeting.
Amor,
K.,
Taylor,
J.,
Hesselbo,
S.P.,
Macniocaill,
C.,
2011.
An
anisotropy
of
magnetic
susceptibility
study
of
the
Stac
Fada
Member
suevite:
constraints
on
the
impact
crater
location.
Meteoritics
and
Planetary
Science
Supplement
46,
A10.
Artemieva,
N.A.,
Wu
¨nnemann,
K.,
Krien,
F.,
Reimold,
W.U.,
Sto
¨ffler,
D.,
2013.
Ries
crater
and
suevite
revisited
–
observations
and
modeling
Part
II:
Modeling.
Meteoritics
and
Planetary
Science
48,
560–627.
Aydogan,
D.,
2011.
Extraction
of
lineaments
from
gravity
anomaly
maps
using
the
gradient
calculation:
application
to
Central
Anatolia.
Earth,
Planets
and
Space
63,
630080903,
http://dx.doi.org/10.5047/eps.2011.04.003.
Barlow,
N.G.,
Sharpton,
V.,
Kuzmin,
R.O.,
2007.
Impact
structures
on
Earth
and
Mars.
In:
Chapman,
M.
(Ed.),
The
geology
of
Mars:
evidence
from
Earth-based
analogs.
Cambridge
University
Press,
pp.
47–70.
Batchelor,
R.A.,
Prave,
A.R.,
2010.
Crystal
tuff
in
the
Stoer
Group,
Torridonian
Supergroup,
NW
Scotland.
Scottish
Journal
of
Geology
46,
1–6.
Beacom,
L.E.,
Anderson,
T.B.,
Holdsworth,
R.E.,
1999.
Using
basement-hosted
clastic
dykes
as
syn-rifting
palaeostress
indicators:
an
example
from
the
basal
Stoer
Group,
northwest
Scotland.
Geological
Magazine
136,
301–310.
Benn,
D.I.,
Evans,
D.J.A.,
2010.
Glaciers
and
Glaciation.
Arnold,
London,
pp.
802.
Binns,
P.E.,
McQuillin,
R.,
Fannin,
N.T.G.,
Kenolty,
N.,
Ardus,
D.A.,
1975.
Structure
and
stratigraphy
of
the
sedimentary
basins
in
the
Sea
of
the
Hebrides
and
the
Minches.
In:
Woodland,
A.W.
(Ed.),
Petroleum
and
the
Continental
Shelf
of
North-West
Europe.
Applied
Science
Publishers,
Barking,
pp.
93–102.
Branney,
M.J.,
Brown,
R.J.,
2011.
Impactoclastic
density
current
emplacement
of
terrestrial
meteorite-impact
ejecta
and
the
formation
of
dust
pellets
and
accretionary
lapilli:
evidence
from
Stac
Fada,
Scotland.
Journal
of
Geology
119,
275–292.
Buchner,
E.,
Gra
¨sslin,
M.,
Maurer,
H.,
Ringwald,
H.,
Scho
¨ttle,
U.,
Seyfried,
H.,
2007.
Simulation
of
trajectories
and
maximum
reach
of
distal
impact
ejecta
under
terrestrial
conditions:
consequences
for
the
Ries
crater,
southern
Germany.
Icarus
191,
360–370.
Butler,
R.W.H.,
2010.
The
role
of
thrust
tectonic
models
in
understanding
structural
evolution
in
NW
Scotland.
Geological
Society,
London,
Special
Publications
335,
293–320.
Butler,
R.W.H.,
Coward,
M.P.,
1984.
Geological
constraints,
structural
evolution
and
the
deep
geology
of
the
NW
Scottish
Caledonides.
Tectonics
3,
347–365.
Collins,
G.S.,
Melosh,
H.J.,
Marcus,
R.A.,
2005.
Earth
Impact
Effects
Program:
a
web-
based
computer
program
for
calculating
the
regional
environmental
conse-
quences
of
a
meteoroid
impact
on
Earth.
Meteoritics
and
Planetary
Science
40,
817–840.
Coward,
M.P.,
Kim,
J.H.,
Parke,
J.,
1980.
A
correlation
of
Lewisian
structures
and
their
displacement
across
the
lower
thrusts
of
the
Moine
thrust
zone,
northwest
Scotland.
Proceedings
of
the
Geologists’
Association
91,
327–337.
Darabi,
M.H.,
Piper,
J.D.A.,
2004.
Palaeomagnetism
of
the
(Late
Mesoproterozoic)
Stoer
Group,
northwest
Scotland:
implications
for
diagenesis,
age
and
relation-
ship
to
the
Grenville
Orogeny.
Geological
Magazine
141,
15–39.
Davison,
S.,
Hambrey,
M.J.,
1996.
Indications
of
glaciation
at
the
base
of
the
Proterozoic
Stoer
Group
(Torridonian),
NW
Scotland.
Journal
of
the
Geological
Society
153,
139–149.
Dulin,
S.,
Elmore,
R.D.,
Engel,
M.H.,
Parnell,
J.,
Kelly,
J.,
2005.
Palaeomagnetic
dating
of
clastic
dykes
in
Proterozoic
basement,
NW
Scotland:
evidence
for
syndeposi-
tional
faulting
during
deposition
of
the
Torridonian.
Scottish
Journal
of
Geology
41,
149–157.
Eronen,
M.,
Glu
¨ckert,
G.,
Hatakka,
L.,
van
de
Plassche,
O.,
van
der
Plicht,
J.,
Rantala,
P.,
2001.
Rates
of
Holocene
isostatic
uplift
and
relative
sea-level
lowering
of
the
Baltic
in
SW
Finland
based
on
studies
of
isolation
contacts.
Boreas
30,
17–30.
Ferrie
´re,
L.,
Osinski,
G.R.,
2013.
Shock
metamorphism.
In:
Osinski,
G.,
Pierazzo,
E.
(Eds.),
Impact
Cratering:
Processes
and
Products.
Blackwell,
Oxford,
UK,
pp.
106–124.
Fisher,
R.V.,
Schmincke,
H.-U.,
1984.
Pyroclastic
Rocks.
Springer-Verlag,
Berlin,
New
York,
pp.
472.
Frei,
R.,
Gaucher,
C.,
Poulton,
S.W.,
Canfield,
D.E.,
2009.
Fluctuations
in
Precam-
brian
atmospheric
oxygenation
recorded
by
chromium
isotopes.
Nature
461,
250–253.
French,
B.M.,
1998.
Traces
of
Catastrophe:
A
Handbook
of
Shock-metamorphic
Effects
in
Terrestrial
Meteorite
Impact
Structures.
Lunar
and
Planetary
Institute
Contribution
No.
954,
,
pp.
120.
Freundt,
A.,
2003.
Entrance
of
hot
pyroclastic
flows
into
the
sea:
experimental
observations.
Bulletin
of
Volcanology
65,
144–164.
Goderis,
S.,
Paquay,
F.,
Claeys,
P.,
2013.
Projectile
identification
in
terrestrial
impact
structures
and
ejecta
material.
In:
Osinski,
G.,
Pierazzo,
E.
(Eds.),
Impact
Cratering:
Processes
and
Products.
Blackwell,
Oxford,
UK,
pp.
223–239.
Goodenough,
K.M.,
Krabbendam,
M.,
2011.
A
Geological
Excursion
Guide
to
the
North-West
Highlands
of
Scotland.
NMSE
Publishing
Ltd.,
,
pp.
224.
Gracie,
A.J.,
Stewart,
A.D.,
1967.
Torridonian
sediments
at
Enard
Bay,
Rosshire.
Scottish
Journal
of
Geology
3,
181–194.
Grieve,
R.A.F.,
Therriault,
A.M.,
2013.
Impactites:
their
characteristics
and
spatial
distribution.
In:
Osinski,
G.,
Pierazzo,
E.
(Eds.),
Impact
Cratering:
Processes
and
Products.
Blackwell,
Oxford,
UK,
pp.
90–105.
Hay,
R.L.,
1960.
Rate
of
clay
formation
and
mineral
alteration
in
a
4000-year-old
volcanic
ash
soil
on
Saint
Vincent,
B.W.I.
American
Journal
of
Science
258,
354–368.
Hergarten,
S.,
Kenkmann,
T.,
2015.
The
number
of
impact
craters
on
Earth:
any
room
for
further
discoveries?
Earth
and
Planetary
Science
Letters
425,
187–192.
Ho
¨rz,
F.,
Ostertag,
R.,
Rainey,
D.A.,
1983.
Bunte
Breccia
of
the
Ries:
continuous
deposits
of
large
impact
craters.
Reviews
of
Geophysics
21,
1667–1725.
Irving,
E.,
Runcorn,
S.K.,
1957.
Analysis
of
the
palaeomagnetism
of
the
Torridonian
sandstone
series
of
north-west
Scotland.
I.
Philosophical
Transactions
of
the
Royal
Society
of
London
A,
Mathematical,
Physical
and
Engineering
Sciences
250,
83–99.
Kenkmann,
T.,
Collins,
G.S.,
Wu
¨nnemann,
K.,
2013.
The
modification
stage
of
crater
formation.
In:
Osinski,
G.,
Pierazzo,
E.
(Eds.),
Impact
Cratering:
Processes
and
Products.
Blackwell,
Oxford,
UK,
pp.
60–75.
Kinnaird,
T.C.,
Prave,
A.R.,
Kirkland,
C.L.,
Horstwood,
M.,
Parrish,
R.,
Batchelor,
R.A.,
2007.
The
late
Mesoproterozoic–early
Neoproterozoic
tectonostratigraphic
evolution
of
NW
Scotland:
the
Torridonian
revisited.
Journal
of
the
Geological
Society
164,
541–551.
Kinny,
P.D.,
Friend,
C.R.L.,
Love,
G.J.,
2005.
Proposal
for
a
terrane-based
nomencla-
ture
for
the
Lewisian
Gneiss
Complex
of
NW
Scotland.
Journal
of
the
Geological
Society
162,
175–186.
Kirkham,
A.,
2003.
Glauconite
spherules
from
the
Triassic
of
the
Bristol
area,
SW
England:
probable
microtektite
pseudomorphs.
Proceedings
of
the
Geologists’
Association
114,
11–21.
Koeberl,
C.,
Claeys,
P.,
Hecht,
L.,
McDonald,
I.,
2012.
Geochemistry
of
impactites.
Elements
8,
37–42.
Lawson,
D.E.,
1972.
Torridonian
volcanic
sediments.
Scottish
Journal
of
Geology
8,
345–362.
Leslie,
A.G.,
Krabbendam,
M.,
Kimbell,
G.S.,
Strachan,
R.A.,
2010.
Regional-scale
lateral
variation
and
linkage
in
ductile
thrust
architecture:
the
Oykel
Transverse
Zone,
and
mullions
in
the
Moine
Nappe,
NW
Scotland.
In:
Law,
R.D.,
Butler,
R.W.H.,
Holdsworth,
R.E.,
Krabbendam,
M.,
Strachan,
R.A.
(Eds.),
Continental
Tectonics
and
Mountain
Building:
The
Legacy
of
Peach
and
Horne.
Geological
Society,
London,
Special
Publication,
335,
pp.
359–381.
McCall,
G.J.H.,
2001.
Tektites
in
the
Geological
Record:
Showers
of
Glass
from
the
Sky.
Geological
Society
of
London,
,
pp.
256.
McEwen,
A.S.,
Preblich,
B.S.,
Turtle,
E.P.,
Artemieva,
N.A.,
Golombek,
M.P.,
Hurst,
M.,
Kirk,
R.L.,
Burr,
D.M.,
Christensen,
P.R.,
2005.
The
rayed
crater
Zunil
and
inter-
pretations
of
small
impact
craters
on
Mars.
Icarus
176,
351–381.
Melosh,
H.J.,
1989.
Impact
Cratering.
A
Geologic
Process.
Oxford
Monographs
on
Geology
and
Geophysics
Series
no.
11.
Clarendon
Press,
Oxford
ix
+
245
pp.
Melosh,
H.J.,
2013.
The
contact
and
compression
stage
of
impact
cratering.
In:
Osinski,
G.,
Pierazzo,
E.
(Eds.),
Impact
Cratering:
Processes
and
Products.
Black-
well,
Oxford,
UK,
pp.
32–42.
Mendum,
J.R.,
Barber,
A.J.,
Butler,
R.W.H.,
Flinn,
D.,
Goodenough,
K.M.,
Krabbendam,
M.,
Park,
R.G.,
Stewart,
A.D.,
2009.
Lewisian,
Torridonian
and
Moine
Rocks
of
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
760

Scotland.
Geological
Conservation
Review
Series,
no.
34,
vol.
34,
Joint
Nature
Conservation
Committee,
Peterborough,
United
Kingdom.
Naldrett,
A.J.,
Cabri,
L.J.,
1976.
Ultramafic
and
related
mafic
rocks;
their
classification
and
genesis
with
special
reference
to
the
concentration
of
nickel
sulfides
and
platinum-group
elements.
Economic
Geology
71,
1131–1158.
Osinski,
G.,
2013.
Processes
and
products
of
impact
cratering;
glossary
and
defini-
tions.
In:
Osinski,
G.,
Pierazzo,
E.
(Eds.),
Impact
Cratering:
Processes
and
Products.
Blackwell,
Oxford,
UK,
pp.
306–309.
Osinski,
G.,
Grieve,
R.A.F.,
Tornabene,
L.L.,
2013.
Excavation
and
impact
ejecta
emplacement.
In:
Osinski,
G.,
Pierazzo,
E.
(Eds.),
Impact
Cratering:
Processes
and
Products.
Blackwell,
Oxford,
UK,
pp.
43–59.
Osinski,
G.,
Pierazzo,
E.
(Eds.),
2013.
Impact
Cratering:
Processes
and
Products.
Blackwell,
Oxford,
UK,
p.
316.
Osinski,
G.R.,
Preston,
L.,
Ferriere,
L.,
Prave,
T.,
Parnell,
J.,
Singleton,
A.,
Pickersgill,
A.E.,
2011a.
The
Stac
Fada
‘‘Impact
Ejecta’’
layer:
not
what
it
seems.
Meteoritics
and
Planetary
Sciences
46,
A181.
Osinski,
G.,
Tornabene,
L.L.,
Grieve,
R.A.F.,
2011b.
Impact
ejecta
emplacement
on
terrestrial
planets.
Earth
and
Planetary
Science
Letters
310,
167–181.
Park,
R.G.,
2005.
The
Lewisian
terrane
model:
a
review.
Scottish
Journal
of
Geology
41,
105–118.
Parnell,
J.,
Mark,
D.,
Fallick,
A.E.,
Boyce,
A.,
Thackrey,
S.,
2011.
The
age
of
the
Mesoproterozoic
Stoer
Group
sedimentary
and
impact
deposits,
NW
Scotland.
Journal
of
the
Geological
Society,
London
168,
349–358.
Pilkington,
M.,
Abdoh,
A.,
Cowan,
D.R.,
1995.
Pre-Mesozoic
structure
of
the
Inner
Moray
Firth
Basin:
constraints
from
gravity
and
magnetic
data.
First
Break
13,
291–300.
Pilkington,
M.,
Grieve,
R.A.F.,
1992.
The
geophysical
signature
of
terrestrial
impact
craters.
Reviews
of
Geophysics
30,
161–181.
Plescia,
J.B.,
Daniels,
D.L.,
Shah,
A.K.,
2009.
Gravity
investigations
of
the
Chesa-
peake
Bay
impact
structure.
Geological
Society
of
America
Special
Papers
458,
181–193.
Pohl,
J.,
Sto
¨ffler,
D.,
Gall,
H.,
Ernston,
K.,
1977.
The
Ries
impact
crater.
In:
Roddy,
D.J.,
Pepin,
R.O.,
Merrill,
R.B.
(Eds.),
Impact
and
Explosion
Cratering.
Pergamon
Press,
New
York,
pp.
343–404.
Pohl,
J.,
Poschlod,
K.,
Reimold,
W.U.,
Meyer,
C.,
Jacob,
J.,
2010.
Ries
Crater,
Germany:
the
Enkingen
magnetic
anomaly
and
associated
drill
core
SUBO
18.
Geological
Society
of
America
Special
Papers
465,
141–163.
Reddy,
S.M.,
Johnson,F
T.E.,
Fischer,
S.,
Rickard,
W.D.A.,
Taylor,
R.J.M.,
2015.
Pre-
cambrian
reidite
discovered
in
shocked
zircon
from
the
Stac
Fada
impactite,
Scotland.
Geology
43,
899–902.
Reimold,
W.U.,
Jourdan,
F.,
2012.
Impact!
Bolides,
craters
and
catastrophes
Ele-
ments
8,
19–24.
Roberts,
A.M.,
Badley,
M.E.,
Price,
J.D.,
Huck,
D.I.W.,
2009.
The
structural
history
of
a
transtensional
basin:
inner
Moray
Firth,
NE
Scotland.
Journal
of
the
Geological
Society
London
147,
87–103.
Rollin,
K.,
2009.
Regional
Geophysics
of
Northern
Scotland
British
Geological
Survey,
CD-ROM.
Rollinson,
H.R.,
Fowler,
M.B.,
1987.
The
magmatic
evolution
of
the
Scourian
complex
at
Gruinard
Bay.
In:
Park,
R.G.,
Tarney,
J.
(Eds.),
Evolution
of
the
Lewisian
and
Comparable
Precambrian
High
Grade
Terrains.
Geological
Society,
London,
Memoir,
27,
pp.
57–71.
Sanders,
I.S.,
Johnston,
J.D.,
1989.
The
Torridonian
Stac
Fada
Member:
an
extrusion
of
fluidised
peperite?
Transactions
of
the
Royal
Society
of
Edinburgh.
Earth
Sciences
80,
1–4.
Sanders,
I.S.,
Johnston,
J.D.,
1990.
Reply
to
‘The
Torridonian
Stac
Fada
Member:
a
discussion’.
Transactions
of
the
Royal
Society
of
Edinburgh.
Earth
Sciences
81,
249–250.
Soper,
N.J.,
Barber,
A.J.,
1982.
A
model
for
the
deep
structure
of
the
Moine
thrust
zone.
Journal
of
the
Geological
Society
139,
127–138.
Stewart,
A.D.,
1997.
Discussion
on
indications
of
glaciation
at
the
base
of
the
Proterozoic
Stoer
Group
(Torridonian),
NW
Scotland.
Journal
of
the
Geological
Society
154,
1087–1088.
Stewart,
A.D.,
2002.
The
Later
Proterozoic
Torridonian
Rocks
of
Scotland:
Their
Sedimentology,
Geochemistry
and
Origin.
Geological
Society,
London,
Memoir,
pp.
130
24.
Stewart,
A.D.,
Parker,
A.,
1979.
Palaeosalinity
and
environmental
interpretation
of
red
beds
from
the
late
Precambrian
(‘Torridonian’)
of
Scotland.
Sedimentary
Geology
22,
229–241.
Sto
¨ffler,
D.,
1977.
Research
drilling
No
¨rdlingen
1973:
polymict
breccias,
crater
basement,
and
cratering
model
of
the
Ries
impact
structure.
Geologica
Bavarica
75,
443–458.
Sto
¨ffler,
D.,
Ryder,
G.,
Artemieva,
N.A.,
Cintala,
M.J.,
Grieve,
R.A.F.,
Ivanov,
B.A.,
2006.
Cratering
history
and
lunar
chronology.
Reviews
in
Mineralogy
and
Geochem-
istry
60,
519–596.
Sto
¨ffler,
D.,
Artemieva,
N.A.,
Wu
¨nnemann,
K.,
Reimold,
W.U.,
Jacob,
J.,
Hansen,
B.K.,
Summerson,
I.A.T.,
2013.
Ries
crater
and
suevite
revisited
–
observations
and
modeling
Part
I:
Observations.
Meteoritics
and
Planetary
Science
48,
515–559.
Sturm,
S.,
Wulf,
G.,
Jung,
D.,
Kenkmann,
T.,
2013.
The
Ries
impact,
a
double-layer
rampart
crater
on
Earth.
Geology
41,
531–534.
Terry,
D.O.,
Chamberlain,
J.A.,
Stoffer,
P.W.,
Messina,
P.,
Jannett,
P.A.,
2001.
Marine
Cretaceous-Tertiary
boundary
section
in
southwestern
South
Dakota.
Geology
29,
1055–1058.
Thackrey,
S.,
Walkden,
G.,
Indares,
A.,
Horstwood,
M.,
Kelley,
S.,
Parrish,
R.,
2009.
The
use
of
heavy
mineral
correlation
for
determining
the
source
of
impact
ejecta:
a
Manicouagan
distal
ejecta
case
study.
Earth
and
Planetary
Science
Letters
285,
163–172.
Tornabene,
L.L.,
Moersch,
J.E.,
McSween
Jr.,
H.Y.,
McEwen,
A.S.,
Piatek,
J.L.,
Milam,
K.A.,
Christensen,
P.R.,
2006.
Identification
of
large
(2–10
km)
rayed
craters
on
Mars
in
THEMIS
thermal
infrared
images:
implications
for
possible
Martian
meteorite
source
regions.
Journal
of
Geophysical
Research
111
,
http://
dx.doi.org/10.1029/2005JE002600.
Turnbull,
M.J.M.,
Whitehouse,
M.J.,
Moorbath,
S.,
1996.
New
isotopic
age
determi-
nations
for
the
Torridonian,
NW
Scotland.
Journal
of
the
Geological
Society,
London
153,
955–964.
Upton,
B.G.J.,
Emeleus,
C.H.,
Heamanc,
L.M.,
Goodenough,
K.M.,
Finch,
A.A.,
2003.
Magmatism
of
the
mid-Proterozoic
Gardar
Province,
South
Greenland:
chro-
nology,
petrogenesis
and
geological
setting.
Lithos
68,
43–65.
Walkden,
G.M.,
Parker,
J.,
Kelley,
S.,
2002.
A
late
Triassic
impact
ejecta
layer
in
southwestern
Britain.
Science
298,
2185–2188.
Young,
G.M.,
2002.
Stratigraphy
and
geochemistry
of
volcanic
mass
flows
in
the
Stac
Fada
Member
of
the
Stoer
Group,
Torridonian,
NW
Scotland.
Transactions
of
the
Royal
Society
of
Edinburgh,
Earth
Sciences
93,
1–16.
M.J.
Simms
/
Proceedings
of
the
Geologists’
Association
126
(2015)
742–761
761