Surfacing on ice of frozen-in marine bottom materials


Thomas, Ml.H.

Journal of the Fisheries Research Board of Canada 31(7): 1195-1200

1974


In late winter and early spring, benthic materials have been observed in patches (maximum size 400 m × 10 m) on the surface of thick fast ice. They parallel the shores of estuaries and are commonest in the southern Gulf of St. Lawrence. Materials in the patches are often in strikingly natural postures and include layers of bottom sediment and plants and animals (eelgrass, shellfish [infauna and epifauna] starfish, flounders, and sticklebacks). Apparently these are frozen to the lower surface of the ice in early winter at low tides, when ice comes in direct contact with them. When the tide rises, the ice floats and raises frozen-on materials with it. Thereafter, the floating ice is thickened by freezing of water to its lower surface, and the bottom materials are thus frozen in between two layers of ice. At the same time the ice above the frozen-in materials is thinned by ablation (sublimation and melting) at its upper surface. Eventually the benthic materials are exposed on the surface ofthe ice. Allied phenomena have been observed in the Bay ofFundy region. Antarctica is the only other place where this has been reported to occur regularly. These phenomena and conditions favoring their occurrence in the two areas are compared.

Surfacing
on
Ice
of
Frozen-In
Marine
Bottom
Materials
J.
C.
MEDCOF
Department
of
the
Environment,
Fisheries
and
Marine
Service
Biological
Station,
St.
Andrews,
N.
B.
AND
M.
L.
H.
THOMAS
Department
of
Biology
University
of
New
Brunswick
in
Saint
John
Saint
John,
N.
B.
MEDCOF,
J.
C.,
AND
M.
L.
H.
THOMAS.
1974.
Surfacing
on
ice
of
frozen-in
marine
bottom
materials.
J.
Fish.
Res.
Board
Can.
31:
1195-1200.
In
late
winter
and
early
spring,
benthic
materials
have
been
observed
in
patches
(maximum
size
400
m
x
10
m)
on
the
surface
of
thick
fast
ice.
They
parallel
the
shores
of
estuaries
and
are
commonest
in
the
southern
Gulf
of
St.
Lawrence.
Materials
in
the
patches
are
often
in
strikingly
natural
postures
and
include
layers
of
bottom
sediment
and
plants
and
animals
(eelgrass,
shellfish
[infauna
and
epifauna]
starfish,
flounders,
and
sticklebacks).
Apparently
these
are
frozen
to
the
lower
surface
of
the
ice
in
early
winter
at
low
tides,
when
ice
comes
in
direct
contact
with
them.
When
the
tide
rises,
the
ice
floats
and
raises
frozen-on
materials
with
it.
Thereafter,
the
floating
ice
is
thickened
by
freezing
of
water
to
its
lower
surface,
and
the
bottom
materials
are
thus
frozen
in
between
two
layers
of
ice.
At
the
same
time
the
ice
above
the
frozen-in
materials
is
thinned
by
ablation
(sublimation
and
melting)
at
its
upper
surface.
Eventually
the
benthic
materials
are
exposed
on
the
surface
of
the
ice.
Allied
phenomena
have
been
observed
in
the
Bay
of
Fundy
region.
Antarctica
is
the
only
other
place
where
this
has
been
reported
to
occur
regularly.
These
phenomena
and
conditions
favoring
their
occurrence
in
the
two
areas
are
compared.
MEDCOF,
J.
C.,
AND
M.
L.
H.
THOMAS.
1974.
Surfacing
on
ice
of
frozen-in
marine
bottom
materials.
J.
Fish.
Res.
Board
Can.
31:
1195-1200.
Nous
avons
observe,
a
la
fin
de
l'hiver
et
au
debut
du
printemps,
des
materiaux
ben-
thiques
en
secteurs
(superficie
maximale
400
m
x
10
m)
a
la
surface
d'epaisse
glace
solide.
Cette
glace
est
placee
parallelement
aux
rives
des
estuaires
et
se
rencontre
le
plus
com-
munement
dans
le
sud
du
golfe
Saint-Laurent.
Les
materiaux
ont
souvent
des
postures
d'un
naturel
frappant
et
comprennent
des
couches
de
sediment
du
fond,
ainsi
que
des
plantes
et
des
animaux
(zostere,
coquillages
[endofaune
et
opifaune],
etoiles
de
mer,
plies
et
epinoches).
II
semble
que
ces
materiaux
se
congelent
sur
la
face
inferieure
de
la
glace
au
debut
de
l'hiver,
a
matte
basse,
au
moment
oh
la
glace
vient
en
contact
direct
avec
eux.
Avec
le
flux,
la
glace
fiotte
et
souleve
avec
elle
les
materiaux
congeles
qui
y
adherent.
Par
la
suite,
la
glace
flottante
s'epaissit
par
congelation
de
1'eau
sur
sa
face
inferieure,
de
sorte
que
les
materiaux
benthiques
congeles
sont
enrobes
entre
deux
couches
de
glace.
En
meme
temps,
la
glace
situee
au-dessus
des
materiaux
s'amincit
par
ablation
(sublimation
et
fonte)
sur
sa
face
superieure.
Eventuellement,
les
materiaux
benthiques
apparaissent
a
decouvert
a
la
surface
de
la
glace.
On
a
observe
des
phenomenes
de
meme
nature
dans
la
region
de
la
baie
de
Fundy.
L'Antarctique
est
la
seule
autre
region
oh
on
a
rapporte
le
phenomene
comme
se
produisant
regulierement.
Nous
comparons
ces
phenomenes
et
les
conditions
qui
les
favorisent
dans
les
deux
regions.
Received
February
5,
1973
Regu
le
5
fevrier,
1973
Accepted
March
26,
1974
Accepte
le
26
mars,
1974
OUR
attention
was
first
drawn
to
exposures
of
benthic
materials
on
ice
by
Dr
A.
W.
H.
Needier
during
the
winter
of
1939-40.
These
were
on
Bide-
ford
River,
Prince
Edward
Island,
opposite
the
Printed
in
Canada
(J2839)
Fisheries
Research
Sub-Station
(N
lat.
46°
37';
E
long.
63°
55':
tidal
amplitude
1.3
m).
They
were
a
long-familiar sight
to
residents
of
that
area
and
we,
R.
E.
Drinnan
(personal
communication
1971),
and
S.
E.
Vass
(personal
communication
1972)
have
seen
them
many
times
since.
Patches
were
1195
1196
JOURNAL
FISHERIES
RESEARCH
BOARD
OF
CANADA,
VOL.
31,
NO.
7,
1974
often
small
consisting
of
only
a
few
mussels
but
sometimes
they
were
extensive.
Unusually
large
patches
were
observed
on
Bideford
River
in
Feb-
ruary
1970
and
allied
phenomena
were
observed
in
1948
in
the
St.
Croix
River,
N.B.,
(N
lat.
45°
4';
E
long
67°
4':
tidal
amplitude
9
m).
A
search
of
literature
indicates
that
McMurdo
Sound,
Antarctica,
is
the
only
other
area
in
the
world
where
such
surfacings
are
regularly
reported.
Observations
BIDEFORD
RIVER
1970
Figure
1(a-f)
shows
the
shapes,
sizes,
positions,
and
compositions
of
patches
exposed
following
a
period
of
mild
weather
in
February
1970,
during
which
the
surface
of
the
ice
had
melted
rapidly.
These
patches
were
mostly
ribbon-shaped,
paralleling
shores,
and
commonest
over
oyster
and
mussel
beds.
The
largest
measured
400
x
10
m.
The
inner
borders
of
patches
were
about
the
same
distance
from
shore
as
the
extreme
low-water
mark
and
the
outer
edges
were
over
depths
of
about
20
cm
below
low
water
(levels
based
on
known
levels
of
various
benthic
communities
in
the
area
[Thomas
1969;
Hughes
and
Thomas
1971]).
Some
patches
were
composed
largely
of
eelgrass
and
associated
flora
(Fig.
la).
The
eelgrass
blades
were
matted
over
their
still-attached
rhizomes.
However,
most
patches
consisted
mainly
of
mussel
shells
and
mussels
still
clumped
together
by
byssus
threads
in
life-like
postures
(Fig.
lb).
They
were
present
in
uncounted
thousands
but
all
were
dead
and
the
shells
of
many
gaped
so
wide
as
to
show
the
meats
within.
Oysters
and
oyster
shells,
al-
though
less
numerous,
were
conspicuous
because
of
their
light
color.
As
far
as
could
be
determined,
the
oysters
were
also
dead
and
most
were
gaping.
One
patch
measuring
50
x
10
m
(Fig.
lc)
was
estimated
to
contain
3500
whole
oysters
valued
alive
at
$150-200.
Starfish
were
much
less
common
(Fig.
le).
The
only
vertebrates
observed
were
smooth
and
winter
flounders
(Fig.
lf)
and
four-spined
stickle-
backs.
Table
1
lists
organisms
found
in
the
patches.
Along
shoreward
edges
of
patches
there
were
sometimes
small
areas
of
bottom
sediment,
mostly
muddy
sand
up
to
10
cm
deep.
These
were
in
such
a
remarkably
undisturbed
state
as
to
still
show
the
burrow
openings
of
soft-shell
clams
(Fig.
1d).
These
sediments
included
many
infauna
such
as
soft-shell
clams
up
to
45
mm
long,
small
razor
clams,
and
trumpet
worms,
all
of
which
live
within
the
top
10
cm
of
the
substrate.
Ravens
and
crows
were
observed
feeding
in
the
patches,
mostly
on
fish
and
the
meats
of
dead,
gaping
bivalves.
Borings
with
an
ice
auger
showed
that
the
ice
below
the
patches
was
up
to
1
m
thick.
ST.
CROIX
RIVER
1948
In
February
1948,
ice
cakes
were
stranded
on
the
shores
of
St.
Andrews
Harbor,
New
Brunswick.
These
were
up
to
1.5
m
thick
and
had
evidently
come
from
upper
parts
of
the
river
following
the
spring
breakup
a
few
days
previous.
Many
cakes
showed
alternate
layers
of
ice
and
bottom
sediments.
The
sediment
layers
(up
to
10
cm
deep)
closest
to
the
surface
of
the
cakes
(Fig.
2a)
were
usually
thicker
than
those
below
(Fig.
2a).
The
strong
sun
had
partly
or
wholly
melted
the
surface
ice
layer
from
some
cakes
thus
exposing
the
uppermost
sediment
layers
as
dark
caps
which
were
also
thawing
(Fig.
2a).
The
edges
of
some
thawed
caps
had
already
crumbled
leaving
soft-shell
clams
up
to
35
mm
long
completely
exposed.
Toward
the
centers
of
these
caps,
the
upper
ends
of
shells
of
clams
in
normal
posture
could
be
seen
(Fig.
2b).
Several
of
these
clams
were
placed
on
bottom
in
a
marked,
underwater
area
and
within
half
an
hour
most
of
them
were
actively
burrowing
or
had
completely
buried.
The
only
other
animal
remains
observed
in
the
sediment
caps
were
empty
shells
of
Macoma
(prob-
ably
M.
balthica).
OTHER
OBSERVATIONS
IN
PRINCE
EDWARD
ISLAND
In
March
1959,
R.
E.
Drinnan
(personal
com-
munication
1971)
observed
large
patches
on
Bide-
ford
River
near
the
Biological
Substation.
He
did
not
measure
them
but
his
photographs
show
that
they
were
smaller
than
those
seen
in
1970.
In
April
1962,
S.
E.
Vass
(personal
communication
1972)
also
observed
patches
on
Bideford
River.
Their
lengths
varied
from
15
to
100
m
and
their
widths
from
3
to
15
m.
By
ordinary
standards
these
would
also
be
considered
large
but
again
they
fell
far
short
of
the
great
sizes
observed
in
1970.
A
sound-looking
oyster
that
Vass
found
in
one
of
the
1962
patches
revived
after
a
day
in
an
aqua-
rium
and
was
still
active
several
weeks
later.
Vass
also
reported
that
ice-auger
borings
for
hydrographic
observations
brought
up
sand
and
eelgrass
from
middepths
in
the
ice
in
Bideford
River
in
March
1961;
in
Tuplins
Creek
in
February
1962
and
again
in
Bideford
River
in
March
1962.
These
materials
are
assumed
to
have
been
from
frozen-in
patches.
No
estimates
of
sizes
of
these
subsurface
patches
were
made
and
there
were
no
follow-up
observations
to
find
whether
they
surfaced
before
the
ice
disappeared.
Istil
....
-
"oF
-
.-•,
_
ti
.
.
'
r
a
.w•
4
-
1111k,
410
iG
,
11.
FIG.
1.
Benthic
materials
on
surface
of
Bideford
River
ice
1
m
thick,
February
1970.
Photographs
by
M.
L.
H.
T.
a)
General
view
of
a
patch
(200
x
4
m)
paralleling
the
shore
and
the
Biological
Substation
breakwater;
mostly
eelgrass
and
filamentous
green
algae.
b)
Close-up
of
patch
mostly
of
clumps
of
mussels;
sizes
judged
from
12
cm
wide
rubber
boot.
c)
Close-up
of
patch
mostly
of
eelgrass;
oysters
and
oyster
shells
about
10
cm
long.
d)
Close-up
of
shoreward
edge
of
a
patch
showing
melting
ice
and
a
layer
of
sandy
intertidal
sediment
10
cm
deep;
small
dark
holes
are
openings
of
soft-shell
clam
burrows;
clam,
mussel,
and
oyster
shells,
and
tubes
of
trumpet
worms
also
present.
e)
Starfish
(6
cm
diam);
matted
filamentous
algae;
eelgrass
and
partly
melted
ice
in
part
of
patch
shown
in
a).
f)
Winter
flounder
(length
12
cm)
picked
by
crows
or
ravens;
matted
filamentous
algae;
eelgrass
and
bits
of
unmelted
ice
in
part
of
patch
shown
in
a).
Medcof
and
Thomas
J.
Fish.
Res.
Board
Can.
_
,
TABLE
1.
Organisms
found
in
surfacing
patches
on
Bideford
River
ice,
1970.
Common
name
Scientific
name
Comments
Plants
Eelgrass
Zostera
marina
Abundant
(leaves
and
rhizomes)
Common
periwinkle
Common
Northern
Moon
Shell
Eastern
Mud
Snail
Blue
Mussel
American
Oyster
Bay
Quahaug
Little
Macoma
Common
Razor
Clam
Common
Soft-Shell
Clam
Common
Starfish
Sand
Shrimp
Gould's
Trumpet
Worm
Invertebrates
Littorina
littorea
Lunatia
heros
Nassarius
obsoletus
Mytilus
edulis
Crassostrea
virginica
Mercenaria
mercenaria
Macoma
balthica
Ensis
directus
Mya
arenaria
Asterias
vulgaris
Crangon
septemspinosus
Pectinaria
gouldii
Vertebrates
Few
Few
Common
Abundant
Common
Few
Few
Few
Common
Few
Few
Few
Four
Spine
Stickleback
Apeltes
quadracus
Rare
Smooth
Flounder
Liopsetta
putnami
Rare
Winter
Flounder
Pseudopleuronectes
americanus
Rare
-err
FIG.
2.
Stranded
St.
Croix
River
ice
cakes,
February
1948.
Photographs
by
J.
C.
M.
a)
Side
view
of
upper
part
of
cake
0.7
m
thick;
three
frozen-in
mud
strata
5,
7,
and
1
cm
thick;
uppermost
stratum
exposed
by
melting;
no
benthos
found
in
mud.
b)
Diagonal
view
of
melting
and
crumbling
upper
edge
of
another
cake;
uppermost
stratum
of
mud—gravel
10
cm
thick,
partly
exposed;
two
soft-shell
clams
in
natural
posture
protrude
from
undisturbed
sediment
(upper
right);
six
clams
with
postures
disturbed
by
crumbling
sediment
(middle
of
photograph);
dial-
diameter
of
pocket
watch,
3.7
cm.
Medcof
and
Thomas
J.
Fish.
Res.
Board
Can.
MEDCOF
AND
THOMAS:
BENTHOS
ON
ICE
1197
We
also
have
reports
from
local
residents
of
sur-
facing
patches
in
other
parts
of
Malpeque
Bay
(e.g.
Trout
River
and
Grand
River),
in
Bedeque
Bay,
P.E.I.,
and
in
inlets
of
the
New
Brunswick
coast
near
Cape
Tormentine.
The
phenomenon
may
be
widespread
in
eastern
Canada.
Theories
of
Surfacing
Process
The
following
explanation
of
surfacing
was
proposed
by
A.
W.
H.
Needier
(Department
of
the
Environment,
Fisheries
and
Marine
Service,
St.
Andrews,
N.B.,
personal
communication
1940).
In
early
winter,
salt-water
ice
forms
all
over
Bideford
River.
It
rises
and
falls
with
the
tides
and
at
low
tide
its
shoreward
edges
bend
and
rest
on
the
intertidal
beach
sometimes
for
several
consecutive
hours.
If
there
is
a
sharp
frost
during
one
of
these
periods
there
is
a
freeze-on
of
patches
of
bottom
to
the
underside
of
the
ice.
When
the
tide
rises
the
ice
floats
and
there
is
a
liftoff
of
frozen-on
patches
of
superficial
bottom
materials
with
the
ice.
Subsequently
there
is
a
freeze-in.
Water
freezes
onto
the
lower
surface
of
the
ice
to
produce
a
"sandwich"
of
bottom
materials
between
two
ice
layers.
In
the
meantime
(unless
there
have
been
heavy
snow
falls)
the
upper
layer
of
ice
will
have
grown
thinner
by
sublimation
and
melting.
Eventually,
the
upper
layer
of
the
ice
sandwich
may
be
completely
lost,
thus
leaving
bottom
materials
exposed
on
the
surface
of
thick
ice.
Unknown
to
Needier,
Debenham
(1920)
proposed
a
similar
explanation
for
allied
phenomena
observed
in
McMurdo
Sound,
Antarctica.
R.
E.
Drinnan
and
S.
E.
Vass
(personal
commu-
nication
1971
and
1972)
conducted
an
experiment
in
Bideford
River
that
illustrated
early
phases
of
this
phenomenon.
They
placed
white,
"styrofoam"
markets
under
the
ice
at
2-wk
intervals
from
December
28,
1962
to
March
12,
1963.
The
markers
froze
in
and,
when
raised
on
March
26,
showed
ice
accumulation
both
below
and
above.
This
technique
might
also
reveal
late
phases
of
the
Debenham—
Needier
process,
as
did
the
February
1948
observa-
tions
in
the
St.
Croix
River
(Fig
2a).
By
1965,
surfacings
had
been
observed
in
parts
of
McMurdo
Sound
that
were
too
deep
to
permit
the
ice
sheet
to
contact
bottom.
Debenham
explained
them
with
an
hypothesis
he
had
previously
rejected
in
explaining
surfacings
over
shallower
waters
(Swithinbank
1970).
This
was
that
frazil
(Barnes
1928)
ice
forms
and
grows
on
and
around
benthic
materials
until
its
buoyancy
raises
them
to
the
bottom
side
of
the
permanent
ice
sheet
where
they
are
frozen
on.
This
hypothesis
was
confirmed
later
by
scuba
divers
who
observed
benthic
materials
in
masses
of
frazil
ice
0.5
m
thick,
rising
to
the
lower
surface
of
the
ice
(Dayton
et
al.
1969).
Hind
(1875)
described
freeze-ins
of
bottom
materials
in
Nova
Scotia
rivers
and
Kindle
(1924)
described
and
explained
the
presence
of
a
great
variety
of
bottom
materials
found
on
ice
in
northern
Canada
and
other
parts
of
the
arctic.
However,
these
phenomena
are
quite
different
from
those
seen
in
Bideford
River
and
in
McMurdo
Sound.
Observations
show
that
frazil
ice
(anchor
ice
of
Dayton
1970)
never
occurs
on
bottom
in
Bideford
River
because
after
the
surface
ice
forms
bottom
water
temperatures
are
constantly
above
freezing
(Needier
1941).
Thus
Bideford
River
freeze-ins
must
develop
as
postulated
by
Debenham
in
1920
and
by
Needier
in
1940.
Swithinbank
(1970)
also
favors
this
same
explanation
for
inshore
surfacings
in
McMurdo
Sound.
Conditions
Favoring
Surfacing
in
Eastern
Canada
Surfacing
of
benthic
materials
on
ice
is
the
end
result
of
a
long
chain
of
events.
Some
of
these
(e.g.
freeze-ons
on
lift-offs)
are
difficult
to
study
and
we
have
few
data.
But
those
we
have
and
general
knowledge
of
the
area
indicate
that,
although
there
are
many
influential
factors,
the
following
six
conditions
are
the
most
significant
in
contributing
to
extensive
surfacings.
1.
Thick
ice
—Bottom
materials
do
not
seem
to
freeze
onto
thin
ice
Apparently
it
must
be
thick
enough
to
press
hard
against
bottom
for
freeze-ons
and
to
supply
the
buoyancy
necessary
for
the
lift-
off
of
frozen-on
materials
when
the
tide
rises.
This
conclusion
is
supported
by
observations
in
the
upper
intertidal
zone
in
Bideford
River.
Freeze-
ons
occur
regularly
there
but
water
levels
at
high
tide
are
insufficient
to
produce
lift-offs.
The
ice
seems
to
stay
in
place
there
frozen
to
the
bottom
until
it
rots
in
spring.
There
are
no
reported
sur-
facings
at
this
beach
level.
Our
observations
indicate
that
ice
must
be
30
cm
or
more
thick
to
produce
a
freeze-on
(the
first
step
towards
a
surfacing).
These
thicknesses
are
found
only
well
after
the
beginning
of
winter.
2.
"Open"
winters
Lack
of
an
insulating
layer
of
snow
favors
freeze-ons
if
the
ice
is
resting
on
bottom
and
favors
thickening
and
freeze-ins
if
1198
JOURNAL
FISHERIES
RESEARCH
BOARD
OF
CANADA,
VOL.
31,
NO.
7,
1794
the
ice
is
floating.
Lack
of
snow
cover
also
favors
surfacing
because
it
favors
ablation
of
the
upper
surface
of
ice
by
sublimation
and
melting.
Conversely,
a
heavy
snowfall
following
a
lift-off
of
bottom
materials
could
prevent
thickening
of
ice
from
below
and
freeze-ins
of
bottom
materials.
It
might
even
reverse
the
whole
process
and
en-
courage
melting
and
loss
of
frozen-on
bottom
materials
lifted
off
following
previous
bottom
contacts.
3.
Intense
cold
in
midwinter
The
frequency
and
extent
of
freeze-ons
appears
to
be
higher
in
midwinter
than
in
late
winter
when
the
ice
is
so
thick
that
its
insulating
effects
may
prevent
freeze-
ons
of
great
extent.
This
is
illustrated
by
Fig.
2a
which
shows
that
thick
layers
of
sediment
are
close
to
the
top
of
the
ice
cake
and
a
thin
one
far
from
the
top.
4.
Diurnal
tidal
periods
When
other
condi-
tions
are
favorable,
the
likelihood
of
freeze-ons
and
their
thickness
and
horizontal
extent
should
vary
directly
with
the
length
of
the
period
of
and
extent
of
contact
of
ice
with
bottom.
These
vary
with
regular
neap
and
spring
tidal
cycles
and,
in
some
places,
with
cycles
of
the
moon's
declination
to
the
north
or
south
of
the
earth's
equator.
In
the
Bide-
ford
River
area
and
in
many
other
parts
of
the
Canadian
Atlantic
coast,
moon
declination
has
dominant
effects
(Anderson
1930.
When
declination
approximates
zero,
tides
are
semidiurnal.
When
it
is
maximum
or
near
maximum,
tides
are
diurnal.
When
Bideford
River
tides
are
diurnal
there
is
one
high
tide
of
normal
length
each
day
and
the
rest
of
the
day
is
occupied
by
one
extended
period
of
low
water.
Unpredictable,
meteorological
conditions
in
the
Gulf
also
have
important
effects
on
timing
and
levels
of
tides
at
any
phase
in
their
cycles
(Lauzier
1952).
With
various
combinations
of
influential
factors,
inshore
ice
may
be
in
uninterrupted
contact
with
bottom
for
long
periods
or
for
only
brief
periods.
We
speculate
that
most
freeze-ons
that
lead
to
small-scale
surfacings
are
brought
about
by
com-
binations
of
conditions
1,
2,
and
3
listed
above,
during
normal
spring-tide
phases
of
tidal
cycles.
We
further
speculate
that
massive
surfacings,
like
those
observed
in
March
1959,
April
1962,
and
February
1970,
result
from
freeze-ons
that
occur
when
there
is
coaction
of
all
factors
favoring
long
stands
at
low
tide;
for
example
when
factors
1-3
coincide
with
the
spring
phase
and
the
diurnal
phase
of
tidal
cycles
and
with
unusual
meteorologi-
cal
conditions
favoring
extended
periods
of
low
tide.
But
there
are
two
others
that
are
basically
important.
5.
Abundance
of
epibenthos
Light-weight
ob-
jects
rising
above
the
sediment
surface
are
most
likely
to
be
frozen
onto
ice
during
bottom
contact.
In
other
words,
epibenthos
should
be
commoner
in
patches
than
bottom
sediments
with
their
in-
fauna.
This
deduction
is
supported
by
Bideford
River
observations.
Most
patches
occurred
just
below
low
water
mark;
they
were
composed
of
epibenthos
(eelgrass,
mussels,
and
oysters)
and
epibenthos
are
most
abundant
at
that
same
level
(Thomas
1969).
Sediment
occupied
relatively
small
areas
and
was
found
only
along
the
shoreward
edges
of
patches.
6.
Low
salinity
surface
water
Fresh
water
has
a
higher
freezing
point
than
salt
water.
Thus
a
surface
layer
of
brackish
water
or
fresh
water
under
the
ice
favors
freeze-ons
at
low
tide
(Swithin-
bank
1970).
Similar
conditions
also
favor
rapid
freeze-ins
at
high
tide
by
ice
formation
below
frozen-
on
bottom
materials.
Haloclines
in
the
quiet
water
under
Bideford
River
ice
are
often
sharp
(Needier
1941;
M.
L.
H.
Thomas
unpublished
data)
especially
in
winter.
Thus
hydrography
as
well
as
other
features
of
the
Bideford
River
area
are
peculiarly
favorable
to
the
surfacing
phenomenon.
There
is
no
reason
why
surfacings
should
not
occur
on
inshore
ice
in
bodies
of
fresh
water
if
their
water
levels
fluctuated
appreciably.
However,
we
have
neither
seen
nor
heard
of
such
phenomena.
Comparison
of
Conditions
in
Bideford
River
and
McMurdo
Sound
Conditions
of
surfacing
of
bottom
materials
on
ice
in
the
two
places
in
the
world
where
it
is
known
to
occur
regularly
are
compared
in
Table
2.
In
spite
of
differences,
similarities
are
striking.
As
Debenham
(1961)
suggested,
there
are
prob-
ably
many
places
in
the
northern
hemisphere
where
surfacing
of
bottom
materials
may
be
observed
if
we
look
for
them.
Smith's
observation
on
ARLIS
II,
a
floating
ice
island
in
the
arctic,
may
be
an
ex-
ample
(Gow
et
al.
1965).
Survival
of
Frozen-in
Organisms
Oysters,
soft-shell
clams,
mussels,
and
peri-
winkles
all
survive
freezing
in
air
for
considerable
periods
and
at
surprisingly
low
temperatures
(Fried-
man
1933;
Kanwisher
1955).
In
some
intertidal
species
the
lower
lethal
temperatures
can
be
de-
MEDCOF
AND
THOMAS:
BENTHOS
ON
ICE
1199
pressed
by
acclimation
(Somme
1966).
Evidence
suggests
that
survival
times
are
extended
if
the
animals
are
protected
from
long
air
exposures
and
blue
mussels
and
rough
periwinkles
(Littorina
saxatilis)
seem
to
be
specially
hardy
when
frozen
into
ice.
Kanwisher
(1955)
reported
survivals
of
these
species
after
8-mo
freeze-ins
in
Labrador
and
the
arctic
during
which
ice
temperatures
fell
to
-20
C
and
probably
lower.
In
this
context
our
1948
observations
on
soft-
shell
clams
from
St.
Croix
River
ice
cakes
and
the
revival
of
an
oyster
from
a
surfacing
patch
on
Bideford
River
(S.
E.
Vass
personal
communication
1972)
might
have
been
anticipated
and
observations
by
M.
L.
H.
Thomas
(unpublished
data)
seem
quite
in
order.
He
showed
that
epibenthos
of
the
upper
intertidal
zone,
such
as
saltmarsh
cord
grass
(Spartina
alterniflora),
filamentous
green
algae
(unidentified),
rough
periwinkles,
and
bidentate
ear
shells
(Melampus
lineatus),
all
survive
lengthy
annual
freeze-ins.
The
ice
at
this
level
rarely
lifts
off
after
it
has
frozen
to
the
beach
and
thus
permits
survival
of
stable
populations
of
these
organisms
at
that
beach
level.
Ribbed
mussels
(VolseIla
demissa)
also
survive
here
in
frozen
sediments
but
avoid
freeze-ins
in
ice
because
of
their
burrowing
habits.
From
all
this
we
conclude
that
dead
molluscs
found
in
surfacing
patches
survived
the
freeze-ins
and
died
from
air
exposure
after
surfacing.
Freeze-Ins
of
Fish
These
was
debate
as
to
whether
large
fishes
(some
of
them
decapitated)
found
surfacing
on
the
McMurdo
Sound
ice
shelf
(Swithinbank
et
al.
1961)
were
alive
when
they
were
frozen
in.
They
were
likely
the
lost
prey
of
seals
(Gow
et
al.
1965;
Dayton
1970)
and
dead
when
frozen
in.
TABLE
2.
Observed
and
probable
conditions
attending
surfacings
in
McMurdo
Sound
and
Bideford
River.
Characteristic
McMurdo
Sound
Bideford
River
Ice
cover
Permanent
(nature)
Maximum
ice
30-33
thickness
(m)
Form
and
sites
of
Isolated
patches;
exposures
over
both
shallows
and
depths
up
to
200
m
Materials
exposed
Bottom
invertebrates
Primary
causes
of
freeze-ins
Mechanisms
of
freeze-ons
Period
of
ice-bottom
contact
Mechanism
of
surface
exposure
Period
between
freeze-ins
and
exposures
Exposure
period
of
surfaced
materials
Seasonal
(Dec.-May)
0.75-1.2
Small
patches
and
long
ribbons;
only
near
low
water
mark
Bottom
invertebrates
and
eelgrass
both
common;
fish
up
to
25
cm
long;
bottom
sediments.
Cold
weather
and
low
tides
Ice-bottom
contacts
Hours
do
1-4
mo
Weeks
(mostly
sponges)
predominate;
fish
up
to
142
cm;
kelp;
sodium
sulphate
crystals
bottom
sediments
Cold
water
and
low
tide
Ice-bottom
contacts
or
lift-offs
of
frazil
ice
with
frozen-in
bottom
materials
Hours
Ablation
of
upper
surface
of
ice
50-70
yr
a
Many
years
b
'Ablation
annually
removes
the
upper
0.5
m
of
the
ice
cover.
b
Carbon
dating
for
a
fish
from
a
McMurdo
Sound
exposure
was
1100
B.P.
(Swithinbank
et
al.
1961).
1200
JOURNAL
FISHERIES
RESEARCH
BOARD
OF
CANADA,
VOL.
31,
NO.
7,
1974
The
fishes
observed
on
Bideford
River
ice
were
mostly
flounders
that
in
life
often
lie
half-buried
in
the
bottom.
They
were
probably
alive
when
they
were
trapped
during
ice-bottom
contacts
at
low
tide.
It
is
assumed
that
the
sticklebacks
(Table
1)
were
similarly
trapped.
Acknowledgments
We
thank
Dr
Needier
for
drawing
our
attention
to
the
surfacing
phenomena
in
Prince
Edward
Island,
for
information,
and
for
stimulating
discussions
of
observations.
We
also
thank
Mr
S.
E.
Vass
and
Mr
R.
E.
Drinnan
for
personal
communications,
Dr
C.
W.
Swithinbank
for
constructive
comments
and
liter-
ature
references,
and
Dr
D.
G.
Wilder
for
critical
review
of
the
draft
of
this
paper.
This
work
was
supported,
in
part,
by
a
National
Research
Council
of
Canada
grant
(A6389)
to
M.
L.
H.
T.
ANDERSON,
F.
1936.
Tide
tables
for
the
Atlantic
coast
of
Canada
for
the
year
1937.
Can.
Hydrogr.
Serv.,
Dep.
Marine.
96
p.
BARNES,
H.
T.
1928.
Ice
engineering.
Renouf
Publ.
Co.,
Montreal,
Que.
364
p.
DAYTON,
P.
K.
1970.
Rafting
of
animals
to
the
surface
by
ice,
p.
289-290.
In
M.
W.
Holdgate
[ed.]
Antarctic
ecology.
Vol.
1.
Academic
Press,
London
and
New
York.
604
p.
DAYTON,
P.
K.,
G.
A.
ROBILLIARD,
AND
A.
L.
DE
VRIES.
1969.
Anchor
ice
formation
in
McMurdo
Sound,
Antarctica,
and
its
biological
effects.
Science
163:
273-274.
DEBENHAM,
F.
1920.
A
new
mode
of
transportation
by
ice:
the
raised
marine
muds
of
Victoria
Land
(Antarctica).
Q.
J.
Geol.
Soc.
Lond.
75:
51-76.
1961.
A
fish
story
from
the
Antarctic.
Geogr.
Mag.
34:
360-368.
FRIEDMAN,
M.
H.
1933.
The
freezing
and
storage
of
live
clams
and
oysters.
Annu.
Rep.
Biol.
Board
Can.
1932:
23-24.
Gow,
A.
J.,
W.
F.
WEEKS,
G.
HENDRICKSON,
AND
R.
ROWLAND.
1965.
New
light
on
the
mode
of
uplift
of
the
fish
and
fossiliferous
moraines
of
the
McMurdo
ice
shelf,
Antarctica.
J.
Glaciol.
5:
813-828.
HIND,
H.
V.
1875.
The
ice
phenomena
and
the
tides
of
the
Bay
of
Fundy.
Can.
Mon.
Nat.
Rev.,
8:
189-203.
HUGHES,
R.
N.,
AND
M.
L.
H.
THOMAS.
1971.
Clas-
sification
and
ordination
of
shallow-water
benthic
samples
from
Prince
Edward
Island,
Canada.
J.
Exp.
Mar.
Biol.
Ecol.
7:
1-39.
KANWISHER,
J.
W.
1955.
Freezing
in
intertidal
animals.
Biol. Bull.
109:
56-63.
KINDLE,
E.
M.
1924.
Observations
on
ice-born
sediments
by
the
Canadian
and
other
Arctic
expedi-
tions.
Am.
J.
Sci.
(Ser.
5)
7:
251-286.
LAUZIER,
L.
1952.
Effects
of
storms
on
the
water
conditions
in
the
Magdalen
shallows.
J.
Fish.
Res.
Board
Can.
8:
332-339.
NEEDLER,
A.
W.
H.
1941.
Temperatures
and
salini-
ties
under
the
ice
in
a
shallow
inlet.
J.
Fish.
Res.
Board
Can.
5:
236-243.
SOMME,
L.
1966.
Seasonal
changes
in
the
freezing
tolerance
of
some
intertidal
animals.
Nytt.
Mag.
Zool.
(Oslo)
13:
52-55.
SWITHINBANK,
C.
1970.
Ice
movement
in
the
Mc-
Murdo
Sound
area
of
Antarctica.
Proc.
Int.
Symp.
Antarctic
Glaciological
Exploration,
Hanover,
U.S.
A.,
3-7
September
1968.
ISAGE
Publ.
No.
86:
472-487.
SWITHINBANK,
C.
W.,
D.
G.
DARBY,
AND
D.
E.
WOHL-
SCHLAG.
1961.
Faunal
remains
of
an
Antarctic
ice
shelf.
Science
133:
764-766.
THOMAS,
M.
L.
H.
1969.
Zonation
of
shore
biota
in
Bideford
River,
Prince
Edward
Island.
Fish.
Res.
Board
Can.
Tech.
Rep.
155:
59
p.