-
ELSEVIER Sedimentary Geology 111 (1997) 177-197
Sedimentary Geology
Subglacial deformation associated with fast ice flow, from the
Columbia Glacier, Alaska
Jane K. Ha r t *, B a m a b y S m i th
Department of Geography, University of Southampton, Southampton,
S017 1B J, UK
Received 30 January 1996; accepted 3 June 1996
Abstract
The Columbia Glacier is a large well studied tidewater glacier
which has one of the highest non-surging modem glacier flow rates.
Previous studies from ice drilling have tentatively suggested the
presence of a deforming bed. This study examined the recently
deglaciated (since 1974) foreland and concluded there were features
indicative of deforming bed conditions, i.e., streamlined
subglacial bedforms, squeeze-type push moraines and crevasse
diapirs. It was also argued that the elongation ratio and height of
streamlined subglacial bedforms can be used as a proxy for shear
strain and thickness of the deforming layer, respectively, and when
these two parameters are combined together they produce a complex
but predictable eigenvalue pattern in the tills comprising the
landforms.
Keywords: deforming bed; deforming bed till; flutes; Columbia
Glacier; fast ice flow
1. Introduction
There has been a great deal of recent research concerned with
understanding the causes of fast ice flow which can occur at any
temporal or spatial scale (see Table 1). On a small scale,
short-term velocity increases (over short time periods 1-10 years,
Dowdeswell et al., 1991) in valley glaciers are known as surges.
There have been many the- odes to explain this phenomenon; however
recent research can be sunarnarised into two models, based on
changes in subglacial water flows (Kamb et al., 1985) or the
subglacial deforming bed (Clarke et al., 1984). On a larger scale,
it has been shown that ice streams (which are the faster-moving
parts of ice
* Corresponding author. Fax: +44 1703-593729; E-mail: jhart@
soton.ac.uk
sheets) change their velocities over longer (102 or 103 years)
time periods (e.g., Ice Stream B, Antarc- tica; Shabtaie and
Bentley, 1987). It has also been shown that during the Quaternary
many of the large ice sheets moved over a deforming bed which led
to an increase in glacier velocity (Boulton and Jones, 1979;
Clarke, 1987; Boulton and Hindmarsh, 1987), but the spatial or
temporal scale of this fast ice is not known. In contrast there are
some glaciers and ice streams today which seem to have consistently
flowed at a fast rate, these include Jakobshavn Isbrae, Greenland
(Lingle et al., 1981) and the Columbia Glacier, Alaska. In this
paper we look in detail at the sediments and landforms deposited by
the Columbia Glacier and use these to interpret the nature of the
subglacial processes associated with fast-moving ice.
0037-0738/97/$17.00 ~, 1997 Elsevier Science B.V. All rights
reserved. PI IS0037-0738(97)00014-6
-
178
Table 1 Fast ice velocities
J.K. Hart, B. Smith~Sedimentary Geology 111 (1997) 177-197
Glacier type Glacier Area Velocity Reference (km 2 )
Non-surging valley glacier
Surging valley glacier
Antarctic ice Streams
Floating tongue tidewater glacier
Grounded tidewater glacier
Quaternary ice stream lobes
Fels Glacier, Alaska 18 70-100 rrdyr
Variegated Glacier, Alaska 20 Quiescent phase = 0-73 m/yr phase
Surge phase = max. 23,000 m/yr
Eyjabakkajokull, Iceland 40 Surge phase = 10,950 m/yr
Ice Stream B 15,000 800 m/yr
Rutford Ice Stream 15,000 >400 m/yr
Jakobshavn Isbrae, 9,000 8360 rrdyr Greenland
Columbia Glacier, Alaska 1,100 Average rate = 1277-3285 rrdyr
Seasonal high-velocity waves = 58,000-91,000 m/yr
Fastest advance rate = 7000 m/yr Laurentide Ice Sheet Lake
87,500 Michigan lobe
Raymond et al. (1995)
Sharp (1988a) Kamb et al. (1985) Sharp (1988b)
Whillans et al. (1987, 1993) Doake et al. (1987)
Lingle et ai. (1981)
Kamb et ai. (1994) Krimmel and Vaughan (1987)
Mickelson et al. (1981)
2. The Columbia Glacier
The Columbia Glacier is a large fast-flowing tide- water glacier
(approx. area 1100 km2; Ferguson, 1992; see Table 1), that has
undergone dramatic retreat since the early 1970s (Post, 1975; Leth-
coe, 1987). It flows from the southern side of the Chugach
Mountains in south central Alaska (Fig. 1) into Prince William
Sound. It is one of the best stud- ied tidewater glaciers in Alaska
with a photographic record of the margin available since 1899
(Gilbert, 1910; Field, 1937; Post, 1975; Brown et al., 1982; Meier
et al., 1985; Meier and Post, 1987; Calkin, 1988) and very detailed
surveys of the glaciology (Meier et al., 1994; Kamb et al., 1994).
This includes a recent study by Humphrey et al. (1993) who in-
ferred a 65 cm basal deforming layer at the base of the Columbia
Glacier from the bending of a drill stem used to study the basal
properties.
The Columbia Glacier has been so well studied because icebergs
from its tidewater margin flow over a large, partially submerged,
neoglacial end-moraine into the Valdez shipping lanes. This moraine
traps
most of the icebergs during low tide, but at high tide some of
the icebergs are able to overtop the moraine, and there are fears
that a channel may be cut into the moraine allowing much higher
iceberg release.
Since the recent glacier retreat, the eastern margin and Heather
Island are readily accessible to exam- ine the subglacial
sediments. The area examined in this study was the eastern margin
(Fig. 1) which was deglaciated between 1974 and 1986. The aim of
this study was to try to understand the geomorphology as- sociated
with fast ice flow, and attempt to use the ge- omorphic data to
suggest the nature of the subglacial environment beneath the modern
Columbia Glacier.
3. The eastern Columbia subglacial surface
The area studied (Fig. lb, c) revealed a remark- ably well
preserved subglacial surface, overlying a bedrock knoll. The
surface is dominated by flutes, but there are also moraines and
other transverse fea- tures, which we shall discuss in turn. At
most of the sites, till fabric and shape analysis was carded out.
For the till fabric investigations at each site a
Fig. 1. Columbia Glacier: (a) location in Alaska; (b) insert
showing a detailed view of the Columbia Glacier; (c) detailed ages
of the moraines; (d) schematic map of the area studied with the key
sites.
-
J.K. Hart, B. Smith~Sedimentary Geology 111 (1997) 177-197
179
d)
Columbia Glacier
f P
Slte 6
0 o ° Esker
~ Flutes
0 Drumlin
Icebergs
Iceberg scour marks
,,dP~,,._~ Iceberg push moraines
//J Moraines
l) Large discontk~uous ridges
~ :~ Camp
100 metres l
N
-
180 J.K. Hart, B. Smith~Sedimentary Geology 111 (1997)
177-197
minimum of 30 clasts with an axial ratio greater than 1.5 : 1
were sampled. The initialised eigenval- ues were then calculated.
These eigenvalues (S1, $2 and $3) summarise fabric strength along
the three principal directions of clustering (after Mark (1973) and
Dowdeswell and Sharp (1986)). A fabric with no preferred
orientation (weak fabric) would have equal eigenvalues, whilst a
strong fabric would have a high value in the direction of maximum
clustering (S1) which is usually the direction of tectonic
transport, and a low value in the direction of least clustering
(S3).
Shape analysis was carried out by the measure- ment of 50
clasts, including the measurement of lengths of the three axes and
the overall roundness (very angular, angular, semiangular,
semirounded, round and well rounded). The results are plotted on a
triangular graph using the technique of Sneed and Folk (1958) and
Benn and Ballantyne (1993) and the values for C40 (percentage of
clasts with c :a axes ratio of less than or equal to 0.4) and RA
(percentage of very angular and angular clasts) were
calculated.
3.1. Modern-day iceberg 'push moraines'
The sea between the study site and the Columbia Glacier was
filled with icebergs and bergy bits (Fig. 2). These ice blocks are
moved ashore at high tide, and either left stranded or moved back
down at low tide. The result of this was the production of two
interesting styles of geomorphology super- imposed on the older,
mainly subglacial, surface. The first was iceberg scratches left by
the movement of the bergs (approx. 0.5 m wide and 0.3 m deep) (Fig.
3a, b), similar to those described by Belderson et al. (1973) and
Bennett and Bullard (1991); and the second were a series of iceberg
push moraines (0.3 m high) (Fig. 3c, d), similar to those described
by Nichols (1953) and John and Sugden (1975). These were small
arcuate ridges pushed up by the icebergs, with striations on the
sub-iceberg surface. The shape and style of deformation patterns
associated with the iceberg push moraines were very similar to
'squeeze' push moraines (proglacial) and flutes (subglacial) as-
sociated with a deforming bed and described from many modem active
glaciers (Price, 1970; Sharp, 1984; Boulton, 1987; Hart,
1995a,b).
3.2. Well preserved flutes and marginal features associated with
site 1
The field sites were located on a peninsula which the ice
retreated from in 1981 (Fig. lc, d). It con- sisted of a
till-covered bedrock knoll whose surface is lineated into flutes
which range in length from 0.3 m to over 300 m, most of which have
a stoss- side striated clast (Fig. 4a). Associated with the flutes
are tree trunks that have been overridden by the glacier and are
now aligned parallel with the flutes and the ice flow direction
(Fig. 4b). There are also many striated 'bullet-shaped' stones
which have been described by numerous authors (Boulton, 1978;
Krtiger, 1979; Sharp, 1982) which are also aligned with the ice
flow direction.
Fig. 5 shows flute 1 (90 m long and 0.3 m high), and associated
landforms that were investigated in detail. Fig. 6a and Table 2
show the eigenvalues and roundness results from four sites along
the length of the flute. These all have very high eigenvalues
(average values S1 = 0.742; $3 = 0.0803), simi- lar to those found
at other flute sites (e.g., Rose, 1989; Benn, 1994; Benn, 1995;
Eklund and Hart, 1996; shown in Table 3), with S1 values all
oriented close to the ice flow direction. Fig. 6b shows the
elongation of the flute compared with results from Rose (1987).
Roundness analysis carried out at two of the sites indicates that
the flute was composed of very rounded clasts with low RA values,
typical of subglacial material (Table 2).
The flute is traversed by four small transverse ridges, that we
will discuss in turn. The outermost moraine (M1) was 0.8 m high and
1 m wide with an arcuate plan form, and contains clasts with a very
low RA value which is typical of subglacial material. A further
roundness test was carried out on a nearby end-moraine of similar
dimensions but 0.6 m down beneath the surface (Mlb), which gave
results very similar to those from moraine 1.
Moraine 2 was less continuous than moraine 1, and had similar
dimensions. However, flute 1 was superimposed on the top of moraine
2. The shape indices of the clasts that composed moraine 2, were
very similar to those in moraine 1 (Table 2). Moraine 3 overlies
the flute and is of similar dimensions to the other two moraines.
However, the clasts within this moraine were far more angular.
-
J.K. Hart, B. Smith~Sedimentary Geology 111 (1997) 177-197
181
= ~i~i!i /iiiiiii~i~ =ii I~I~
Fig. 2. Photograph of the ice margin, glacier is in the
background, and in the foreground is the iceberg-choked bay
(photograph taken from the western edge of the peninsula studied -
- Fig. lb, c, d).
Moraine 4 is slightly more complex, and the junc- tion with the
flute is more muted. A fabric and round- ness were taken at this
site (M4a) which showed that the moraine was composed of very
rounded material with slightly weaker fabric orientation than those
on the main body of the flute. However, this transverse feature has
two styles of longitudinal ridges associ- ated with it. Firstly,
features that are located on the down-glacier side of the ridge
which are composed of very rounded clasts (M4b) and, secondly,
ridges composed of very angular clasts located on both sides of the
moraine 4 ridge (M4c).
3.3. Interpretation o f site 1
A number of alternative models for flute forma- tion have been
proposed. Boulton (1976), Amark (1980), Benn (1994, 1995), and Hart
(1995b) have argued that flutes from when saturated subglacial
sediment infills grooves in the ice base. These grooves are formed
by the presence of large ob- stacles in the deforming bed (usually
large clasts) and the saturated subglacial sediment flows into the
low-pressure area in the lee of the obstruction. Al- ternatively,
Schytt (1962) has argued that although saturated till is squeezed
up into cavities in the lee of boulders, the release of pressure
from the surround-
ing ice causes the till in the cavity to refreeze to the basal
ice which will later melt-out to form a flute. Gordon et al. (1992)
have suggested that flutes form from the preferential melting of
the debris-rich basal ice in the lee of a large clast.
Although it has been shown that processes and thus fabrics in
the debris-rich basal ice are very simi- lar to those in the
deforming layer (Hart, 1995b), and thus differentiation of
flute-forming processes may be difficult, we suggest that flutes
are more likely to form associated with deforming bed conditions,
because:
(1) Most marginal subglacial observations of modern-day
flute-forming are associated with a deforming layer (with no
debris-rich basal ice), e.g., Breidamerkurjtkull, Iceland (Boulton,
1976; Benn, 1995); Vestari-Hagafellsjtkull, Iceland (Hart, 1995a),
Turtmann Glacier, Switzerland (Roberts, 1995); Exit Glacier, Alaska
(Hart, 1995b); Root Glacier, Alaska (Hart, 1996).
(2) Flow patterns derived from till fabrics show both flow into
the low-pressure area, and along the direction of the flute,
indicating that the till was moving with the glacier.
(3) Once the debris-rich basal ice melts, it behaves as a
deforming layer (Vivian and Bocquet, 1973; Hart, 1995b), thus it
loses any pattern that it had in
-
182 J.K. Hart, B. Smith/Sedimentary Geology 111 (1997)
177-197
Fig. 3. Effects of the modem-day icebergs: (a) and (b) iceberg
scratches; (c) iceberg push moraines.
-
J.K. Hart, B. Smith~Sedimentary Geology 111 (1997) 177-197
183
Fig. 3 (continued).
the debris-rich ice, and its final sedimentation will reflect
its reorientati
-
184 J.K. Hart, B. Smith~Sedimentary Geology 1ll (1997)
177-197
trunk Fig. 4. Photograph of the subglacial surface: (a) flute 4
in the foreground with a fluted surface in the background; (b) tree
trunks overrun by the glacier and orientated in the ice flow
direction.
there were small fragments of supraglacial sediments deposited
in the dump moraines and crevasse infills.
3.4. Description o f other flutes
The eastern margin of the Columbia Glacier was completely
covered with flutes of all shapes and
sizes (see Fig. ld). The smallest flutes (apart from the
lineations discussed above) tended to be less elongated. Two small
flutes were investigated in detail (flutes 2 and 3). Flute 2 (Fig.
7a) consisted of a large clast with a 1.5 m lee-side tail. Till
fabrics taken in the flute indicate a high fabric strength, and a
flow which is almost parallel with ice direction.
-
J.K. Hart, B. Smith~Sedimentary Geology 111 (1997) 177-197
185
_ f - - - - - ~ e+4%,(~ \ e •
e+4~ x l~ / ' ~ ' . ' ~ e_+4a J~ , -~ . / / ' ~ e_+4a r~'. /h /
/ / ~
\ I~ e+4a \ "~/I k~,,7"e+4a =v=,.~ / . ~ i
L \ - .___ .
angular g~ la j ~ l M4a N materials ~ l~
e e+_2~ ° o/~sed 1~0 M4b flute superimp ~ ~ ; ~ / ~ M4b I on
moraine
" . ~ . ~ . . . ~ . y M 4 e ' ~ / / 0 , , , , 50 metres ~ e
Fig. 5. Detail of flute 1 and associated morainic landforms at
site 1 (stereonets contoured using Kamb's method, eigenvalues are
given in Table 2).
The stoss-side clast is oriented at 60 ° (ice flow direction 110
°) at this point. Flute 3 is a slightly larger flute (Fig. 7b) with
a lower elongation ratio (Fig. 6b). However, the fabric results
within the flutes indicate convergent flow and slightly
lower-strength eigenvalues than flute 2.
Flute 4 is a much larger flute (over 68 m long and 2.5 m high)
(Fig. 3a and Fig. 7c). This flute has medium-strength fabrics which
converge near to the stoss-side clast and then diverge towards the
end of the flute.
3.5. Interpretation of the other flutes
It can be seen that all the flutes have similar high eigenvalues
whatever the size of the flute, but that the elongation ratio
increases as the flutes grows (see Fig. 6b). We would suggest that
these flutes represent different stages in flute evolution. Once a
flute is initiated in the deforming layer due to the presence of a
clast, a small prow is formed on the lee side of the clast (Hart,
1995a) but as time continues, so the flute grows in length.
3.6. Flutes and marginal features with more supraglacial
components associated with the 1986 moraine (site 5)
Site 5 represents a more complex environment, with a greater
supraglacial component (Fig. lc). A detailed map of the area is
shown in Fig. 8a. The site consists of two zones delineated by
small moraines composed of angular clasts, which must represent
dump moraines (possibly annual). Flutes, a very low elongation
ratio linear feature (without a stoss-side clast) and an irregular
linear feature, are found in both areas and underlay the
end-moraines. The east- ern zone consists of a mostly rounded
surface with many oriented and striated clasts (oriented down ice).
The western zone has a far more angular surface, and contains many
small pools of water (Fig. 8b) and linear ridges composed of
angular clasts.
The low elongation ratio linear feature is very interesting
(Fig. 8c) as it contains no stoss-side clast but is oriented down
ice. Unlike the flutes which have very high elongation ratios
(typically over 10), this has a very low ratio (3) (Fig. 6b).
Detailed fabrics were taken of this feature (Fig. 6a, Fig. 8c and
Table 2). The feature itself is oriented 95 ° , and
-
186 J.K. Hart, B. Smith~Sedimentary Geology 111 (1997)
177-197
300-
200 -
1 0 0 -
~ 50 O . m
t - O
o~
1o
5
1-
(b)
0.2-
0.1
0 0.4
(a)
• 6a
• 4d 7b • I) M4a
8a • • 5a • 4e
4b (I • 3a
• 5d lc ~ 5b
i /
8b /
i b•qlm6b / / e~" / 4a l a /
ld • / / /
/
~ 3 b / /
/ /
/ /
/ /
/ /
/
0.5
2b 5e ZO 7a •
/ /
/
0.6 0.7
$1
c1
2a • • 5c • 5f
0.8 0.9 1.0
• Okstindan flutes o OksUndan megaflutes
o Glasgow drumlins (inside Loch Lomond moraine) Glasgow
drumlins
• Glasgow megadrumlins
o Glasgow streamlined hills
f lu te f o r m • • 0
• 0 0 - ~ " "
• 0 . . ~ ' ' ' ~ A & k A~,& 0 • e A 6 ~ A
. - ' " . A~, AA t A~.a,~AA 0 ¢ • " ' " °
. . . . . . . ~t~' k,t~'~" , , , o ° C 2 • . . . - ' " • ° 'P "
. . . . d r u m l i n f o r m C 4 . • ~ oq~° °
C5 0 , , 4 [ e l i i i i i l l i f i ( i f i i i [ i i i i i i i
i ( I I ; I I i l l
1 5 10 100 1000 10000
long [a] axis (metres)
Fig. 6. (a) Graph showing the fabric results. The dashed line
marks the boundary between (Hart, 1994): lodgement tills and
deforming bed tills with a thin deforming layer (below the line);
and deforming bed tills with a thick deforming layer (above the
line). (b) Graph of elongation ratio against length from Rose
(1987) with the Columbia results added (C1 = flute 1; C2 = flute 2;
C3 = flute 3; C4 = flute 4; C5 = drumlin).
-
J.K. Hart, B. Smith~Sedimentary Geology 111 (1997) 177-197
187
Table 2 Till fabric and roundnes,; results
Table 3 Comparative flute and drumlin data
Site & No. S 1 $3 Dir. C40 RA Sites Height Elongation S
1
1-a 0.724 0.101 110 37 0 ratio (a/b) 1-b 0.723 0.103 268
Columbia 1-c 0.793 0.040 88 30 0 flute 1 (C1) 0.3 300 0.742 1-d
0.730 0.077 284 flute 2 (C2) 0.2 1.5 0.712 l-M1 26 8 flute 3 (C3)
0.3 2.8 0.644 1-Mlb 40 12 flute 4 (C4) 1.5 28 0.635 l-M2 26 6 site
5 (drumlin) (C5) 0.4 3 0.712 l-M3 42 26 Rose (1989), A 3 6 0.679
1-M4a 0.603 0.085 233 30 0 Rose (1989), i 0.4 13 0.62 1-M4b 40 12
Benn (1994) 1 6 0.707 1-M4c 78 98 Eklund and Hart (1996) 0.3 17 0.8
2a 0.782 0.028 99 Hart (1997), LP1 1 1.8 0.593 2b 0.643 0.035 108
Hart (1997), LP2 1 2.4 0.733 3a 0.618 0.053 117 Hart (1997), LP3 2
2.6 0.667 3b 0.671 0.046 91 4a 0.711 0.097 260 4b 0.598 0.054
106
4c 0.668 0.049 93 supraglacial environment, with the hollows
repre- 4d 0.615 0.093 76 4e 0.609 0.068 104 senting kettle holes
and l inear angular r idges rep- 5-a 0.571 0.071 76 resenting
crevasse infills. This site has far more 5-b 0.797 0.036 90
supraglacial sediments than the other sites studied 5-c 0.824 0.026
93 because it is closer to the valley walls where the 5-d 0.537
0.032 89 sediment can be seen to be located on the modern 5-e 0.684
0.031 109 5-f 0.753 0.016 98 glacier. It has been suggested by Rose
(1987) that 6a 0.537 0.133 280 there is a continuum of s treamlined
subglacial land- 6b 0.729 0.101 94 forms from small highly
elongated flutes to larger 7a 0.728 0.021 235 less elongated
drumlins, formed by similar processes 7b 0.544 0.082 324 (Fig. 6b).
It has been shown by Hart (1995a) that 8a 0.548 0.068 55
these different forms can exist in a s imilar area, and 8b 0.715
0.110 114 8c 33 20 can be related to the thickness of the deforming
bed. 8d 30 17 This is because although flutes form in ice
groves,
so the fabrics are general ly divergent on the stoss side and
the lee side. There is also a range in fabric strengths from high
to low, with an average S 1 value for the stoss side of 0.731 and
lee side of 0.694.
A further feature at this site is the irregular l inear feature.
This contains both stratified sand and till and has some angular
blocks on top. The width of this feature varied from 2 to 5 m and
the height from 0.5 t o 2 m .
3. 7. Interpretation ,ff site 5
We suggest that the irregular, more angular sur- face of the
western zone probably represents a more
the movement of till into these areas leads to the evacuation of
till elsewhere, part icularly in the stoss side of the flute and
the interflute areas. In this way the ice groove, core clast and
deforming layer are interrelated. Fig. 9 shows that as the
deforming bed thickens so there is more sediment movement and a
heightening of the landform. In this way the sub- glacial landform
represents a theoretical min imum thickness of the deforming
layer.
Which reference to Fig. 6b, it can be seen that the low
elongation ratio l inear feature at site 5 fits into the drumlin
class. Although there have been many theories proposed for drumlin
formation (see Men- zies, 1984), recently most workers have
suggested that they are formed associated with the deform- ing bed
(e.g., Menzies, 1979; Boulton, 1987; Pi-
-
188 J.K. Hart, B. Smith~Sedimentary Geology 111 (1997)
177-197
a) Plan~e r 1.5m
e±2 e ~ trench-fabric taken
~,110 ° ' ~ e~:~,~ 1.25m cr
X-section
Io2m 1.5m
b)
1.
Plan X-section 0.48m
I
0.7m
e
c) e:t2(~ e e±4a ee±2~
e I Plan
e±2o e
e±2a
23.6m 7.8m 12.5m 55m
Fig. 7. Schematic diagrams of the other flutes in the site 1
area: (a) flute 2, (b) flute 3, (c) flute 4.
otrowski and Smalley, 1987; Boyce and Eyles, 1991; Hart, 1995a).
These workers suggest that drumlins form when material flows around
more competent
obstructions within the subglacial deforming layer, which can be
made either of a clast of hard rock, or a more competent mass of
soft sediment, such as
-
J.K. Hart, B. Smith~Sedimentary Geology 111 (1997) 177-197
189
(a) 0
large clast orientated in
palaeo-ice flow direction _
o..~_ 0 0 0 o
~ esker
weak / / ~ \ ~ .,~ute '
~ flat sub/supra %
many rounded glacial % bullet shaped surface
clasts striations
O / Drumlin 0
0 0 20 km I i I
water pool
0 ©
0
o / faint
/
/
Fig. 8. Site 2. (a) Schematic map of the area; (b) photograph of
a hollow in the eastern area; (c) schematic diagram of the drumlin;
(d) photograph of the drumlin (in background with two figures
resting on it) and moraine in foreground.
-
190 J.K. Hart, B. Smith~Sedimentary Geology 111 (1997)
177-197
e e+~c r e e:E2cr e e
/ / X . ht 1.5m
e±6o" ~ e:~2.o" e~4~ e:i:2a
e ¢T (c)
sand and gravel or till. Hart (1995c) has suggested that there
are three types of deforming bed drumlins: deposit ional - - that
form in a manner similar to flutes; deformat ional - - that have a
weak core which is deformed; and erosional - - that have a strong
core. However, it is suggested that all three styles of drumlin
form associated with net subglacial deform- ing bed erosion. The
pattern of fabric flow directions indicates that this drumlin
represents a depositional drumlin (i.e., it has a pattern similar
to that recorded in flutes by Rose, 1989).
We suggest that the irregular linear feature rep- resents a
short subglacial esker, the stratified sand represents deposition
in the conduit associated with flowing water whilst the till
represents the saturated subglacial sediments infilling the conduit
as water pressure dropped. The angular blocks which rest on the
surface probably represent supraglacial debris.
Thus, we suggest that at this site there is more evidence for a
subglacial surface with indications of subglacial deformation in
the form of the flutes, drumlins and also till injections within
the esker. Superimposed on these subglacial landforms is a cover of
supraglacial sediment, reflecting this site's more marginal
position.
3.8. Transverse fea tures
Fig. 8 (continued).
To the north of site 1 there is an area domi- nated by
transverse features (site 6) (Fig. ld and Fig. 10). These consist
of three styles: small inter- secting transverse ridges and flutes,
small continuous ridges, and large discontinuous ridges, which we
will discuss in turn.
(1) Small intersecting transverse ridges and f lutes (feature
6). In the north of the area, in a small valley there are a series
of intersecting flutes and transverse ridges. On inspection it can
be seen (Fig. 10b) that the top surface of the intersection is
composed of a very sandy diamicton, whilst the lower part is the
silt-rich diamicton seen elsewhere in the area. The form of the
flute is both in the silt-rich till and the sandy till and the
flute clearly postdates the for- mation of the transverse feature.
The fabrics in the feature (6a and 6b) (Fig. 6a, Fig. 10a, and
Table 2) indicate a relatively high fabric strength in the silt-
rich till, but a much weaker fabric in the sandy till and both are
oriented in the ice flow direction.
-
J.K. Hart, B. Smith~Sedimentary Geology 111 (1997) 177-197
191
resultant height of landform
: ~ - . - ' - _ , , ~ - . - - _ , , ~ . - . - - 2 , , % - . - -
_ . , % - . - ' - 2 , , ~ . - . - ice
deforming layer
i - - . . i - f - _ . . / _ J - _ . . , _ r - - . . / _ J " - -
- t - J " - -
resultant height ] of landform
ice
deforming layer
Fig. 9. Schemalic diagram to show the relationship between the
size of flutes and the thickness of the deforming layer.
(2) Low continuous transverse ridges (feature 7). The main part
of the ridges are composed of a very sandy diamicton approximately
30 cm thick. Beneath this is the silt-rich diamicton, which is also
in the shape of a small anticline, but of a smaller amplitude than
the shape of the ridge, and with the crest towards the proximal
side of the ridge (Fig. 10b) (7a and 7b in Table 2). The fabrics at
this site show a contrast to the latter site, with the stronger
fabric in the sandy diamicton and the weaker in the silt-rich till.
Also the mean orientation is 40 ° from the ice direction in the
upper sandy unit and 49 ° in the lower till unit.
(3) Large discontinuous ridges (feature 8). To- wards the
southern end on the area, there are three larger (1.5 m) moraines
(Fig. 10c) also located at a small valley. These ridges are made
entirely of till. A fabric was taken on the proximal and distal
side of the ridge (8a and 8b). These indicated a weak fabric,
oriented 40 ° from t]he ice direction on the proximal side, and a
strong fabric, oriented 20 ° from the ice direction on this distal
side. Roundness values taken
both in the ridge (8c) and on the sandy transverse ridge (8d)
show a relatively similar high roundness value.
3.9. Interpretation of the transverse features
Sandy till was not found elsewhere in the study, so we suggest
that this probably represents a fluvial de- posit, that was
subsequently subglacially deformed. The deformation in feature 6
was by fluting, but the compressive deformation in feature 7 is
more equivocal, and could be formed proglacially a bull- dozed-type
push moraines or subglacially (Rogen moraines).
Lundqvist (1989, 1995) and Hattestrand (1995) have argued that
Rogen moraines can contain any glacial or non-glacial sediment, and
like drumlins are a landform produced by many processes. Many
workers, e.g., Lundqvist (1989) and Boulton (1987), have also
demonstrated the link between drumlins and Rogen moraines. Most
workers have argued that Rogen moraines form due to subglacial
com-
-
192 J.K. Hart, B. Smith~Sedimentary Geology 111 (1997)
177-197
• • 0.3m
1.5m
b) ~ e e+~
Ds
1.5m
~ q ~ . / \ ~ -. ~-.~.- ~-+'.~/ \ / oo \ .... ~c~-~..~.
/,,., oo \ / °° \
1.5m
Fig. 10. Schematic diagrams of the transverse ridges: (a)
feature 6, (b) feature 7, (c) feature 8, (d) photograph of feature
8.
pression, caused by subglacial bedrock topography, ice
compression or thermal changes (e.g., Solliel and S¢rbel, 1984,
suggested that they formed at
the transition between warm-based and cold-based ice).
Thus we suggest that the simplest explanation is
-
J.K. Hart, B. Smith~Sedimentary Geology 111 (1997) 177-197
193
Fig. 10 (continued).
that a localised relxeat occurred and outwash sedi- ments were
deposited. This was followed by a small readvance, which in places
fluted the fluvial surface, but in others produced compressive
deformation, which may have occurred either proglacially or sub-
glacially.
4. Discussion
The sediment~¢ and geomorphic evidence from the eastern margin
of the Columbia Glacier includes: flutes, drumlins, lineations, and
eskers oriented par- allel with ice flow; and squeeze-type push
moraines, dump moraines and cavity diapirs orientated per-
pendicular to ice :flow. The vast majority of these features are
indicative of a deforming bed, and to- gether they substantiate the
suggestion of a deform- ing bed proposed fi:om the drilling by
Humphrey et al. (1993).
Rose (1987) has suggested that the streamlined subglacial
bedforms (lineations, flutes and drumlins) represent a continuum
formed by similar processes. Many authors have suggested that the
presence of subglacial bedforms indicates increased ice velocity
(Boulton, 1987; Boyce and Eyles, 1991; Clark, 1994) and thus basal
shear strain and that elongation ratio can be used as a proxy for
this. Hart (1995a) has suggested that the height of bedforms
relates to the
deforming layer thickness, since the landform (be it flute or
drumlin) must be thinner or equal to the deforming bed thickness
(Fig. 9).
Many authors have reported the presence of su- perimposed
subglacial streamlined landforms (Rose and Letzer, 1977; van der
Meer, 1983; Kr/iger and Thomsen, 1984; Rose, 1987, 1989; Hart,
1995a). However, there were only a few occurrences where this was
observed at the Columbia Glacier (e.g., the lineations at site 1,
and the reorientation at site 6). Instead, we suggest that the
surface shows flutes at different stages of evolution. The flutes
are mostly of a similar height, approx. 0.3 m except for the very
large flute at site 5 which was formed in a hollow (and so may
reflect a local thickening of the deforming layer to a minimum
thickness of 1.5 m).
From the discussion above, it was suggested that both elongation
ratio may be a proxy for shear strain, and height may be a proxy
for thickness of the deforming layer. The values from known flutes
and depositional drumlins are shown in Fig. 11 (Table 3). The
interaction of these processes (shear strain and thickness of the
deforming layer) also leads to complex eigenvalue results. It has
been shown by Hart (1994) that as strength of eigenvalues increases
with shear strain in a stable thickness deforming layer, but that
as the deforming layer thickens this leads to a reduction in fabric
strength.
-
194 J.K, Hart, B. Smith~Sedimentary Geology 111 (1997)
177-197
1000
500
c
100
v 50
~ to m 5
C2 E&H C4 Ioi ~
Benn Rose A 0.71, 0.68
C5 0,71
C.3I • LP2 oLd7 0.73 ( 0.64 , LP1
0.59 1
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Height (~ thickness of the deforming layer)
Fig. 11. Graph of elongation ratio against height for the
examples shown in Table 3 (with S1 values also shown). The
subglacial bedforms in this study are as follows: C1 to C5
represent landforms from the Columbia Glacier, Alaska (C1 = flute
1; C2 = flute 2; C3 = flute 3; C4 = flute 4; C5 = drumlin); Rose i
and Rose A represent two flutes from Austre Okstindan Glacier,
Norway (Rose, 1989) (these two flutes were chosen because they
represent the flutes which were studied in the greatest detail); E
and H represent a flute from Isfallsglaci~en, Sweden (Eklund and
Hart, 1996); Benn represents flute A from Slettmarkbreen,
Jotunheimen, Norway (Benn, 1994); LP1, LP2 and LP3 represent
rock-cored (depositional drumlins) from Loch Portain, Isle of Uist,
Scotland (Hart, 1997).
We suggest that this pattern is also reflected in the
development of subglacial streamlined landforms. Within a deforming
layer of constant thickness, the bedform will elongate over time,
but if the deforming layer thickness changes, then the rate of
elongation will change (become faster if the layer thins or slower
if the layer thickens).
We tentatively suggest that these processes (and the
corresponding response of the till fabric) can be described by four
end-members: (a) thin deforming layer~low shear strain (e.g.,
Columbia flute 3 {C3 }); (b) thin deforming layer~high shear strain
(Columbia flute 1 {C1 }); (c) thick deforming layer~low shear
strain (Loch Portain 1, Isle of Uist, UK, Hart (1997) {LP1 }; (d)
thick deforming layer~high shear strain (flute A, Norway, Rose,
1987) {Rose A}. These results show that bedforms formed under a
thick deforming layer tend to have weaker fabric strengths, and
that in both cases shear strain has little effect on eigenvalue
strength (these results are very similar to those shown by Hart,
1994 for tills).
5. Conclusion
Ice drilling experiments by Humphrey et al. (1993), have
suggested that one of the reasons for fast ice flow at the Columbia
Glacier was the pres- ence of a deforming bed. Our investigations
of the
recently deglaciated foreland confirm these results. Evidence
for the deforming bed includes:
(a) subglacial streamlined bedforms at many scales,
(b) squeeze-type push moraines which indicates the movement of
material from beneath the glacier into the foreland, and
(c) crevasse diapirs, which indicate the present of saturated
sediments.
We suggest that the deforming bed in most places was relatively
thin (approx. 0.3 m) (which would be expected because of the
marginal location of the site), and that the cumulative shear
strain was very high, indicated by the presence of flutes over 300
m long. This thickness is consistent with the data collected by
Humphrey et al. (1993) from a loca- tion up-glacier where the
deforming layer would be expected to be thicker (Hart et al.,
1990). However, in local areas, in particular in hollows in the
land- scape, the deforming layer was much thicker (1.5 m) and
taller flutes were formed. This indicates how the deforming bed
changes in response to local condi- tions, and so highlights a
further problem of the spot sampling associated with ice cores.
This irregularity of the deforming bed thickness may also provide
further evidence for the presence of 'sticky spots' beneath the
glacier. These were suggested by Kamb (1991) and Alley (1993) to be
places beneath the
-
J.K. Hart, B. Smith~Sedimentary Geology 111 (1997) 177-197
195
glacier where a discontinuity in the lubricating till of the
deforming layer (usually a bedrock knoll) can support high basal
shear stress.
Thus, we suggest that the landscape on the Columbia foreland was
formed by fast ice flow associated with a deforming bed, and
presumably a similar subglacial surface is present beneath the
Columbia Glacier today.
Acknowledgements
The authors would like to thank Richard Waller and Kirk Martinez
for field assistance and to KM for photography. We would also like
to thank Nancy and Jim Lethcoe, Bob Krimmell and Charles Raymond
for information about the site and the Tatitlek Cor- poration for
permissJion to carry out research in the area. We would also like
to thank Tim Aspden and his colleagues in the Cartographic Unit,
Department of Geography for their excellent figure reproduction.
JKH was funded by NERC grant GR9/991, and BS by the University of
Southampton (Richard Newitt Trust, Roy Queare Bequest and the
Vice-chancellor's fund).
References
Alley, R.B., 1993. In search of ice-stream sticky spots. J.
Glaciol., 39, 447-454.
/Ykmark, M., 1980. Glacial flutes at Isfallsglacitiren, Tarfala,
Swedish Lapland. Geol. Frren. Stockholm Frrh., 102: 251- 259.
Belderson, R.H., Kenyon, N.H. and Wilson, J.B., 1973. Iceberg
plough marks in the northeast Atlantic. Palaeogeogr. Palaeo-
climat. Palaeoecol., 13: 215-224.
Benn, D.I., 1994. Fluted moraine formation and till genesis be-
low a temperate valley glacier: Slettmarkbreen, Jotunheimen,
southern Norway. Sedimentology, 41: 279-292.
Benn, D.I., 1995. Fabric signature of subglacial till
deformation, Breidamerkurjrkull, Iceland. Sedimentology, 42:
735-747.
Benn, D.I. and Ballantyne, C.K., 1993. The description and rep-
resentation of particle shape. Earth Surf. Process. Landforms, 18:
665-672.
Bennett, M.R. and Bullard, J.E., 1991. Iceberg tool marks: an
example from Heinabergsjrkull, Southeast Iceland. J. Glaciol., 37:
181-183.
Boulton, G.S., 1976. The origin of glacially fluted surfaces:
Observations and theory. J. Glaciol., 17: 287-309.
Boulton, G.S., 1978: Boulder shapes and grain-size distributions
as indicators of transport paths through the glacier and till
genesis. Sedimentology, 25: 773-799.
Boulton, G.S., 1987. A theory of drumlin formation by
subglacial
deformation. In: J. Menzies and J. Rose (Editors), Drumlin
Symposium. Balkema, Rotterdam, pp. 25-80.
Boulton, G.S. and Hindmarsh, R.C.A., 1987. Sediment deforma-
tion beneath glaciers: Rheology and geological consequences. J.
Geophys. Res., 92(B9): 9059-9082.
Boulton, G.S. and Jones, A.S., 1979. Stability of temperate ice
caps and ice sheets resting on beds of deformable sediment. J.
Glaciol., 24: 29-44.
Boyce, J.I. and Eyles, N., 1991. Drumlins carved by deforming
till streams below the Laurentide ice sheet. Geology, 19:
787-790.
Brown, C.S, Meier, M.E and Post, A., 1982. Calving speed of
Alaska tidewater glaciers, with application to Columbia Glacier.
U.S. Geol. Surv. Prof. Pap. 1258-C, 13 pp.
Calkin, P.E., 1988. Holocene Glaciations of Alaska (and Adjoin-
ing Yukon Territory, Canada. Quat. Sci. Rev., 7: 159-184.
Clark, C.D., 1994. Large-scale ice-moulding: a discussion of
genesis and glaciological significance. Sediment. Geol., 91:
253-268.
Clarke, G.K.C., 1987. Subglacial Till: A physical framework for
its properties and processes. J. Geophys. Res,, 92(B9):
8942-8984.
Clarke, G.K.C., Collins, S.G. and Thompson, D.E., 1984. Flow,
thermal structure and subglacial conditions of a surge-type
glacier. Can. J. Earth Sci., 21: 232-240.
Doake, C.S.M., Frolich, R.M., Mantripp, D.R., Smith, A.M. and
Vaughan, D.G., 1987. Glaciological studies on Rutford Ice Stream,
Antarctica. J. Geophys. Res., 92(B9): 8951-8960.
Dowdeswell, J.A. and Sharp, M., 1986. Characterisation of peb-
ble fabrics in modern terrestrial glacigenic sediments. Sedi-
mentology, 33: 699-710.
Dowdeswell, J.A., Hamilton, G.S. and Ove Hagen, J., 1991. The
duration of the active phase on surge-type glaciers: contrasts
between Svalbard and other regions. J. Glaciol., 37: 388-400.
Eklund, A. and Hart, J.K., 1996. Glaciotectonic deformation
within a flute from the glacier Isfallsglaci~en, Sweden. J. Quat.
Sci., 11: 299-310.
Ferguson, S.A., 1992. Glaciers of North America, a Field Guide.
Fulcrum Publishing, Golden, Colo., 176 pp.
Field, W.O., 1937. Observations on Alaskan coastal glaciers in
1935. Geogr. Rev., 27: 63-81.
Gilbert, G.K., 1910. Glaciers and glaciation. In: Harriman
Alaska Expedition, Vol IIL Doubleday Page, New York, pp. 71-81.
Gordon, J.E., Whalley, W.B., Gellatly, A.E and Vere, D.M., 1992.
The formation of glacial flutes: assessment of models with evidence
from Lynsdalen, North Norway. Quat. Sci. Rev., 11: 709-732.
Hart, J.K., 1994. Till fabric associated with deformable beds.
Earth Surf. Processes Landforms, 19: 15-32.
Hart, J.K., 1995a. Drumlins, flutes and lineations at Vestari-
Hagafellsjrkull, Iceland. J. Glaciol., 41: 596--606.
Hart, J.K., 1995b. An investigation of the deforming
layer/debris- rich ice continuum, illustrated from three Alaskan
glaciers. J. Glaciol., 41: 619-633.
Hart, J.K., 1995c. Drumlin formation in southern Anglesey and
Arvon, north west Wales. J. Quat. Sci., 10: 3-14.
-
196 J.K. Hart, B. Smith~Sedimentary Geology 111 (1997)
177-197
Hart, J.K. (Editor), 1996. Alaska Expedition 1995. Dep. of
Geography, Univ of Southampton, Discussion Paper No. 40, 161
pp.
Hart, J.K., 1997. The relationship between drumlins and other
forms of subgiacial deformation. Quat, Sci. Rev., 16: 93-108.
Hart, J.K., Hindmarsh, R.C.A. and Boulton, G.S., 1990. Different
styles of subglacial glaciotectonic deformation in the context of
the Anglian ice sheet. Earth Surf. Processes Landforms, 15:
227-241.
H~ittestrand, C., 1995. Rogen moraines in central and northern
Sweden - - distribution pattern and palleoglaciological impli-
cations. Abstracts of the International Quaternary Association
(INQUA) Conference, Berlin, p. 108.
Humphrey, N., Kamb, B., Fahnestock, M. and Engelhardt, H., 1993.
Characteristics of the bed of the lower Columbia Glacier, Alaska.
J. Geophys. Res., 98(B1): 837-846.
John, B.S. and Sugden, D.E., 1975. Coastal Geomorphology of high
latitudes. Prog. Geogr., 7: 53-132.
Kamb, W.B., 1991. Rheological nonlinearity and flow instability
in the deforming bed mechanism of ice-stream motion. J. Geophys.
Res., 96(B10): 16,585-16,595.
Kamb, W.B., Raymond, C.F., Harrison, H.E, Englehardt, H.E,
Ecbelmeyer, K., Humphry, N., Brugman, M. and Pfeffer, T., 1985.
Glacier Surge Mechanism: the 1982-1983 surge of Variegated Glacier,
Alaska. Science, 227: 469-79.
Kamb, W.B., Engelhardt, H., Fahnestock, M.A., Humphrey, N.,
Meier, M. and Stone, D., 1994. Mechanical and hydrological basis
for the rapid motion of a tidewater glacier, 2. Interpreta- tion.
J. Geophys. Res., 99(B8): 15,231-15,244.
Krimmel, R.M. and Vanghan, B.H., 1987. Columbia Glacier, Alaska:
Changes in velocity 1977-1986. J. Geophys. Res., 92(B9):
14,987-14,992.
Kriiger, J., 1979. Structures and textures in till indicating
sub- glacial deposition. Boreas, 8: 323-340.
Krfiger, J. and Thomsen, H.H., 1984. Morphology, stratigraphy,
and genesis of small drumlins in front of the glacier Myrdal-
sjrkull, south Iceland. J. Glaciol., 30: 94-105.
Lethcoe, N.R., 1987. An Observer's Guide to the Glaciers of
Prince William Sound, Alaska. Prince William Sound Books, Alaska,
151 pp.
Lingle, C.S., Hughes, T.J. and Kollmeyer, R.C., 1981. Tidal
flexure of Jakobshavn Glacier, West Greenland. J. Geophys. Res.,
86: 3960-3968.
Lundqvist, J., 1989. Rogen (ribbed) moraine - - identification
and possible origin. Sediment. Geol., 62: 281-292.
Lundqvist, J., 1995. Rogen moraine - - two-step formation of
glacial landforms. Abstracts of the International Quaternary
Association (INQUA) Conference, Berlin, p. 172.
Mark, D.M., 1973. Analysis of axial orientation data, including
till fabrics. Bull. Geol. Soc. Am., 84: 1369-1374.
Meier, M.E and Post, A., 1987. Fast tidewater glaciers. J. Geo-
phys. Res., 92B: 9051-9058.
Meier, M.F., Rasmussen, L.A., Krimmel, R.M., Olsen, R.W. and
Frank, D., 1985. Photogrammetric determination of surface al-
titude, terminus position, and ice velocity of Columbia Glacier,
Alaska. U.S. Geol. Surv. Prof. Pap. 1258-F, 41 pp.
Meier, M.F., Lundstrom, S., Stone, D., Kamb, B., Engelhardt, H.,
Humphrey, N., Dunlap, W.W., Fahnestock, M., Krimmel, R.M. and
Waters, R., 1994. Mechanical and hydrological basis for rapid
motion of a large tidewater glacier, 1. Observations. J. Geophys.
Res., 99(B9): 15,219-15,229.
Menzies, J., 1979. Mechanics of drumlin formation. J. Glaciol.,
27: 372-384.
Menzies, J., 1984. Drumlins - - A Bibliography. Geo Books,
Norwich, 117 pp.
Mickelson, D.M., Acomb, L.J. and Bentley, C.R., 1981. Possible
mechanisms for the rapid advance and retreat of the Lake Michigan
Lobe between 13,000 and 11,000 years B.P. Ann. Glaciol., 2:
185-186.
Nichols, R.L., 1953. Marine and lacustrine ice-pushed ridges. J.
Glaciol., 2: 172-175.
Paul, M.A. and Evans, H., 1974. Observations on the inter- nal
structure and origin of some flutes in glaciofluvial sedi- ments,
Blomstrandbreen, north-west Spitsbergen. J. Glaciol., 13:
393--400.
Piotrowski, J.A. and Smalley, I.J., 1987. The Woodstock drumlin
field, southern Ontario, Canada. In: J. Menzies and J. Rose (Ed-
itors), Drumlin Symposium. Balkema, Rotterdam, pp. 309-321.
Post, A., 1975. Preliminary hydrography and historic terminal
changes of Columbia Glacier. USGS Hydrologic Investigations Atlas
559.
Price, J.R., 1970. Moraines at Fjallsjrkull, Iceland. Arct.
Alpine Res., 2: 27-42.
Raymond, C.E, Benedict, R.J., Harrison, W.D., Echelmeyer, K.A.
and Strum, M., 1995. Hydrological discharges and mo- tion of Fels
and Black Rapids Glaciers, Alaska, USA: impli- cations for the
structure of their drainage systems. J. Glaciol., 41: 290--304.
Roberts, D.H., 1995. Criterion to distinguish between glacioma-
rine and glaciotectonic sedimentation. Unpublished Ph.D. The- sis,
University of Southampton.
Rose, J., 1987. Drumlins as part of a glacier bedform continuum.
In: J. Menzies and J. Rose (Editors), Drumlin Symposium. Balkema,
Rotterdam, pp. 103-116.
Rose, J., 1989. Glacier stress patterns and sediment transfer
as- sociated with the formation of superimposed futes. Sediment.
Geol., 62: 151-176.
Rose, J. and Letzer, J.M., 1977. Superimposed Drumlins. J.
Glaciol., 18: 471-480.
Schytt, V., 1962. Notes on glaciological activities in
Kebnekajse, Sweden, 1962. Geogr. Ann., 44A: 407-412.
Shabtaie, S. and Bentley, C.R., 1987. West Antarctic ice streams
draining into the Ross Ice Shelf: configuration and mass bal- ance.
J. Geophys. Res., 92(B2): 8865-8884.
Sharp, M., 1982. Modification of clasts in lodgement till by
glacial erosion. J. Glaciol., 28: 475-481.
Sharp, M., 1984. Annual moraine ridges at Skalafellsjrkull,
south-east Iceland. J. Glaciol., 30: 82-93.
Sharp, M.J., 1988a. Surging Glaciers: behaviour and mecha-
nisms. Prog. Phys. Geogr., 12: 349-370.
Sharp, M.J., 1988b. Sedimentation and stratigraphy at Eybakka-
jtikull - - an Icelandic surging glacier. Quat. Res., 24: 268-
284.
-
J.K. Hart, B. Smith/Sedimentary Geology 111 (1997) 177-197
197
Sneed, E.D. and Folk, R.L., 1958. Pebbles in the lower Colorado
River, Texas, a study in particle morphogenesis. J. Geol., 66:
114-150.
Solliel, J.L. and S~rbel, L., 1984. Distribution and genesis of
moraines in central Norway. Striae, 20: 63~57.
van der Meet, J.J.M., 1983. A recent drumlin with fluted surface
in the Swiss Alps. ]In: E.B. Evenson, Ch. Schluchter and J. Rabassa
(Editors). Tills and Related Deposits, Balkema, Rotterdam, pp.
105-11)9.
Vivian, R. and Bocquet, G., 1973. Subglacial cavitation phenom-
ena under the Glacier d'Argentiere, Mont Blanc, France. J.
Glaciol., 12: 439-451.
Whillans, I.M., Botzan, J. and Shabtie, S.S., 1987. Velocity of
Ice Streams B and C, Antarctica. J. Geophys. Res., 92:
8895-8902.
Whillans, I.M., Jackson, M. and Tseng, Y., 1993. Velocity
pattern in a transect across Ice Stream B, Antarctica. J. Glaciol.,
39: 562-572.