Evidence of late and postglacial seismic activity in the TCmiscouata- Madawaska Valley,
Quebec - New Brunswick, Canada1
W. W. SHILTS
Geological Survey of Canada, 601 Booth Street, Ottawa, Ont., Canada KIA OEB
M. RAPPOL
Prinsengracht 21012, 101 6 HD Amsterdam, The Netherlands
AND
A. BLAIS
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
Department of Earth Sciences, Carleton University, Ottawa, Ont., Canada KIS 5B6
Received September 19, 1991
Revision accepted December 16, 1991
Three types of geological phenomena independently suggest that the TCmiscouata -Madawash Valley was affected by one
or more seismic events following its deglaciation:
(1) Subbottom acoustic profiling of Lac TCmiscouata and Grand lac Squatec revealed disturbance of bottom sediments by
mass transport processes in both lakes. Erosional truncation of preexisting, acoustically laminated sediments and accumulation of hummocky, chaotic deposits over older hummocky surfaces or laminated sequences both result from mass transport
processes. Unidirectional mass flows from several points in these symmetrical basins, in situ disruption of laminated sediment
beneath flat bottoms, and the large area of the lake floors affected suggest strong similarities in sedimentation style with lakes
that have been disturbed during strong earthquakes.
(2) Southeast of Lac TCmiscouata, in Saint-Jacques, New Brunswick, two separate mass flow deposits, made up largely
of coarse ( > 0 . 5 m), angular boulders of local bedrock, occur on opposite sides of the Madawaska River valley. These
deposits have different source areas and transport directions, but occupy more or less the same stratigraphic position within
sediments deposited in glacial Lake Madawaska.
(3) At one site in Saint-Jacques, a near-vertical fault displaces a glacially striated bedrock surface at least 7 cm, suggesting
a response to postglacial compressive stress similar to that observed on outcrops in the nearby epicentral region of the 1982
Miramichi earthquake.
Although the TCmiscouata -Madawaska Valley lacks historical evidence of seismic activity, and many of the phenomena
observed could, individually, have been generated by aseismic processes, we conclude that the close proximity of diverse
features related to mass transport and faulting suggest that the valley has been the locus of seismic activity from the time
of its deglaciation to the recent, but prehistorical, past.
Trois types de phCnomknes gCologiques suggkrent, indkpendamment, que la vallCe de TCmiscouata-Madawaska a subi
un ou plusieurs sCismes depuis sa dkglaciation :
(1) Les sondages acoustiques du trCfonds du lac TCmiscouata et du Grand lac Squatec revklent que les stdiments de fond
furent deranges par des processus de transport de masse dans les deux lacs. Les sondages acoustiques indiquent que ce transport de masse a tronqut les skdiments laminaires prk-existants, et qu'il en a r6sultC une accumulation de dCp6ts chaotiques,
en creux et bosses, au-dessus de surfaces mamelonnkes plus anciennes ou qui recouvrent des skquences de laminites. La direction unique des Ccoulements en masse telle qu'indiqute h plusieurs endroits dans ces bassins symCtriques, la rupture in situ
de skdiments laminaires en-dessous des couches de fond horizontales, et les fonds des lacs affect& sur une grande superficie
suggkrent un style de sdimentation fortement similaire ?i celui des lacs dont les sMiments ont CtC dCrangts par de gros tremblements de terre.
(2) Au sud-est du lac TCmiscouata, dans Saint-Jacques, au Nouveau-Brunswick, deux dCp6ts de flots en masse distincts,
form& principalement de blocs angulaires (>0,5 m) arrachCs au substratum, sont exposCs de chaque c6tC de la vallte de
la rivikre Madawaska. Les matkriaux de ces dCp6ts dkrivent d'aires nourricikres diffkrentes, et les directions de transport
sont distinctes, cependant ils occupent sensiblement la m6me position stratigraphique dans la colonne de sCdiments du lac
glaciailre Madawaska.
(3) A un endroit dans Saint-Jacques, une faille quasi verticale a dCplact une surface strite du substratum d'au moins 7 cm,
ce qui tCmoigne d'une rtaction ?i des forces de compression postglaciaires analogues h celles observCes sur des affleurements
dans la rCgion avoisinante de 1'Cpicentre du tremblement de terre de Miramichi de 1982.
Bien que l'histoire de l'activitt sismique de la vallCe de TCmiscouata-Madawaska ne soit pas bien documentee, et que plusieurs des phCnomknes observCs pourraient, individuellement, avoir CtC causBs par des processus asismiques, nous concluons
que les relations Ctroites entre divers phCnomknes rCsultant du transport en masse et le dCveloppement de failles suggbent
que la vallCe fut le lieu d'activitk sismique depuis le temps de sa dkglaciation jusqu'au pass6 rCcent, mais prkhistorique.
[Traduit par la rCdaction]
Can. J. Earth Sci. 29, 1043-1069 (1992)
Introduction
The Saint John River valley between Edmunston and Grand
Falls, the Madawaska Valley, and the southern segment of the
T
Cvalley together
~
form~a remarkable
~ linear feature
~
'Geological Survey of Canada Contribution 12491.
Printed in Canada 1 ImprimC au Canada
that trends transverse to the regional northeast-southwestthe
striking bedrock stmCture
features extend farther northwest
(Fig.
and
~ southeast.
~ The linear
~ character~of this trans-Appalachian
~
trench suggests a relationship to a structural discontinuity, but
available bedrock maps do not support such an inference (e.g.,
LespCrance and Grenier 1969; St.Peter 1977).
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
CAN. J. EARTH SCI. VOL. 29, 1992
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
SHlLTS ET A t .
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
1046
CAN. J. EARTH SCI , VOL. 29, 1992
Pleistocene ice movements generally were parallel to the axis
of the trench. Around Edmunston, oldest striations indicate ice
movement toward the southeast, and abundant Precambrian
erratics point to a Laurentide source for the glacier ice that
formed them. Later ice movement was toward the northwest
and appears to be related to drawdown caused by ice movement down the St. Lawrence Valley, following the development of a calving bay in the lower St. Lawrence Estuary
(Shilts 1976; Thomas 1977; Chauvin et al. 1985; Lortie and
Martineau 1987; Blais 1989; Blais and Shilts 1989; Rappol
and Russell 1989; Lowell et al. 1990).
Deltaic deposits at several levels in the TtmiscouataMadawaska Valley indicate that a large lake, Lake Madawaska,
occupied the axis of the valley and contiguous low areas during
and after deglaciation. During its earliest and highest phases,
the lake was dammed along an ice front that was retreating
southward toward an ice divide over New Brunswick and Maine.
After ice had disappeared entirely from the valley, the lake was
initially prevented from draining southward by thick drift
deposits that blocked the preglacial valley of the Saint John
River at Grand Falls, New Brunswick. Overflow from this
phase was northward through isostatically depressed cols to
the St. Lawrence River, via either Rivibre Rimouski or
Rivibre des Trois Pistoles (Fig. 1). Southward tilting of the
TCmiscouata-Madawaska depression, due to differential isostatic rebound, and subsequent erosion of the drift dam at
Grand Falls, drained Lake Madawaska, creating the southflowing Madawaska River. The Madawaska became a northern
tributary of the Saint John River, which resumed its preglacial
southward course (see Kiewiet de Jonge 1951; Lougee 1954;
Kite 1979; Rampton et al. 1984; Kite and Stuckenrath 1986).
Methods
Aggregate pits at Saint-Jacques were examined as they were
being actively worked in 1987-1988 as part of a regional
reconnaissance of ice flow directions and glacial history.
Lac TCmiscouata was surveyed in 1987 and 1988 using a
Raytheon RTT-1000A subbottom acoustic profiling (SAP) system deployed from a 5 m long, inflatable launch. Profiles were
generated by 7 kHz acoustic signals fired and received at the
rate of about three per second. Navigation was by line of sight
from headland to headland or to other recognizable features on
shore, and is probably not particularly accurate for traverses
exceeding 2 krn. For the purposes of this survey, however, the
accuracy of navigation was considered to be adequate. Grand
lac Squatec and Lac MatapCdia were surveyed in 1988 using the
methods described above. Bathymetric maps of Lac TCmiscouata
and Grand lac Squatec were compiled by the authors from the
acoustic profiles, assuming water depths approximating the
calibrations on chart paper provided by the manufacturer of
the RTT-1000A system. Sediment thicknesses were determined
assuming that sound velocities do not vary significantly from
water to saturated, fine-grained sediment, an assumption commonly employed in near-surface acoustic surveys (Berryhill
1987), and one confirmed by the first author by deep drilling
through lake bottom sequences to bedrock in the Ottawa area
(unpublished data).
Results
Boulder deposit at Saint-Jacques
Clast-supported boulder gravels form tabular masses or
channel fills over Lake Madawaska sands and silts in pits
recently excavated on either side of the Madawaska Valley at
FIG.2. Location of pits exposing bouldery mass transport deposits
and topographic detail at Saint-Jacques, New Brunswick. Contours in
feet above sea level (1 ft = 0.3048 m).
Saint-Jacques, New Brunswick (Fig. 2). The boulder beds are
both overlain and underlain by shallow-water lacustrine and
deltaic sediments deposited in a lake that had a surface at 170 m
asl, the altitude of the deltaic surfaces found all along the
TCmiscouata-Madawaska Valley.
The open-work and clast-supported boulder gravels consist
mostly of angular block- to pebble-sized clasts with a variable,
but generally minor, matrix. Individual blocks are either fractured irregularly or bounded by joint surfaces (Fig. 3). Fracture
or joint surfaces show virtually no evidence of abrasional
rounding, but some of the blocks have one polished and striated surface (Fig. 3), indicating that they were torn from the
original glacially moulded bedrock surface.
Clasts are almost exclusively of local provenance, but rare,
rounded erratics from the Precambrian Shield and from other
Appalachian sources indicate that some of the glacially derived
surface cover also has been incorporated into the boulder beds.
Those boulder deposits exposed on the northeast side of the
Madawaska Valley (sites 2 and 3) contain, along with the
dominant local slate, diabase boulders presumably derived
from dikes known to cut the local slate north of Saint-Jacques
(St. Peter 1977). Diabase boulders were not observed in boulder deposits west of the river, nor are dikes reported in the
vicinity of site 1.
In pit 3, clast imbrication and the dip of two superimposed,
> 4 m high foreset beds with avalanche slopes indicate a
southward or southeastward transport direction for the boulder
deposits northeast of the river. Southwest of the river, in pit 1,
the inclinations of the floors of the gullies at the base of the
boulder deposit are toward the north, suggesting northward
transport directly down the valley side.
The two infilled gullies observed in pit 1 are cut down into
the upper part of silty, sandy lake sediments deposited in Lake
Madawaska and into the underlying till in places (Fig. 4). The
fact that the boulder deposits postdate all or most of the phase
of deposition in Lake Madawaska argues against a glacial
origin for them.
The contact of the boulder beds with underlying glaciolacustrine sediments and till is erosional, truncating bedding in the
silty sand. The contact is either sharp or shows a well-defined
zone with bedding or shear planes parallel to the gully slope
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
SHILTS ET AL.
1047
FIG.3. (a) General aspect of mass transport deposit lying on Lake Madawaska sands, pit 3 (GSC-205470E,D). (b) Contact of diamicton
with underlying lacustrine sands (GSC-205470-1). (c)Slate boulder with one glacially moulded and striated surface (verticalface) (GSC-205470-H).
(Fig. 4). The latter type of contact suggests deposition of the
bouldery sediment by gravity flow.
The boulder beds are interpreted as rock avalanche or debris
flow deposits, representing two separate flows from different
sites on either side of the river. Similar deposits were not
observed anywhere else in the region, but since the pits were
only recently opened, because of local aggregate demands, it
is entirely possible that other, similar deposits remain unexposed in the valley.
Because the boulder de~ositsare found on both sides of the
river in similar stratigraphic positions, but have different sources
and opposing transport directions, it is possible that failure
leading to their formation was triggered by a single catastrophic
event. A strong seismic shock is a plausible explanation that
could account for such coeval dislodgement of significant
masses of metasedimentary bedrock in a valley with relatively
gentle slopes.
It was further observed in pit 1 that the finer grained laminated lake sediments commonly exhibit extensive intrastratal
contortions (Fig. 4). Where observed, these convolutions are
typical load casts with no unidirectional properties and may be
due to fluidization and liquefaction processes associated with
normal rapid sedimentation processes in the glaciolacustrine
environment. However, similar convolutions in some cases
have been observed to have been caused by strong earthquake
shocks (Sims 1975; Allen 1984, pp. 348-384). At SaintJacques, preferential liquefaction of the finer grained beds also
could have been caused by sudden, rapid loading of saturated,
lake-bottom sediment by the mass of boulders. This would have
occurred in a subaqueous environment because the boulder
beds are overlain by glaciolacustrine sediment in places. The
first author has observed recent loading of lacustrine sediments
by rock avalanches on Vancouver Island, and the bouldery
debris seems to have caused significant deformation in underlying lacustrine and glaciolacustrine sediments (see subbottom
acoustic profiling records in Clague et al. 1989 and Shilts and
Clague 1992).
In the Madawaska Valley pits, many of the postglacial Lake
Madawaska beds that exhibit intrastratal convolutions are cut
by erosional gullies filled with mass transport debris (Fig. 4).
This suggests that the convolutions had been at least partly
formed before the mass transport debris passed over them.
Though loading associated with avalanching may have played
a role in forming some of the intrastratal convolutions, it is
likely that normal lacustrine processes or the postulated seismic shock itself played more important roles.
Faulted bedrock
In pit 1 at Saint-Jacques (Fig. 2), a striated, glacially abraded
slaty bedrock boss was uncovered. It was generally overlain
by shallow-water glaciolacustrine deposits of sandy gravel and
silt, but pockets of till were preserved in depressions on the
outcrop surface.
The flat top of the bedrock boss is cut by a small fault that
has offset the striated surface as much as 7 cm vertically
(Fig. 5). The fault step faces up-glacier with respect to the main
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
1048
CAN. J. EARTH SCI. VOL. 29, 1992
Fw. 4. ( a ) Angular boulders and gravel with minor matrix in central part of gully cut by mass transport debris in pit 1 (GSC-205470-J).
(b) Sharp contact of diamicton with side of gully cut in Lake Madawaska sand and silt; note intrastratal convolutions (GSC-205470-K). (c) Shear
zone between diamicton and lacustrine sand and silt at gully side (GSC-205470-M). (d) Intrastratal convolutions in lacustrine sediment immediately below diamicton (GSC-205470-P).
ice flow direction recorded by striations, but clearly has not
been affected by glacial abrasion. The open fissure along the
fault plane has not been filled with fine sediment, indicating that
the offset postdates both glacial erosion and till
The fault strikes about 150 - 330°, which is slightly oblique
to bedding in the rock and to the orientation of the long axis
of the boss (130 - 3 10"). Several smaller extensional fractures
are associated with the'main fracture. They trend mostly in a
direction normal to the main fracture but show very little vertical or lateral displacement.
In the region around the TCmiscouata-Madawaska depression, Rampton et al. (1984) and Dionne et al. (1988) have
reported similar vertical displacements of glacially striated
bedrock surfaces near Edmunston, New Brunswick and Bic,
Quebec, respectively. Postglacially offset striations are commonly reported in the Appalachians (Oliver et al. 1970; Grant
1989) and have been attributed by some researchers to nearsurface stresses created by rapid glacioisostatic adjustments to
unloading of glacier ice during deglaciation (Adams 1981).
2Because the faulted outcrop lies in the midst of terrain disturbed
by intense construction activity, we asked the owner of the pit if there
had been any dynamiting associated with construction and excavation, on the remote chance that a man-made explosion could have
fractured the outcrop. He informed us that to his knowledge no explosions had taken place in or near the pit.
FIG.5. Postglacial fault in bedrock boss, pit 1, Saint-Jacques. Main
ice movement was from right to left (GSC-205470-F).
According to Quinlan (1984), rebound stress rarely should
be able to dictate focal mechanisms of earthquakes, but it may
act as a trigger for stress accumulated by other processes.
Hasegawa (1988) cites a case where rebound stress was sufficient to generate a magnitude 7 earthquake along a preweakened
zone in the bedrock. Data presented recently at a North Atlantic Treaty Organization conference on earthquakes at and near
1
I
1I
I
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
I
SHILTS ET AL.
1049
contour interval.5 m
FIG. 6 . Bathymetric maps of Lac TBmiscouata and Grand lac Squatec. F7, F8, etc. refer to figure numbers of subbottom acoustic profiles (SAP).
North Atlantic passive margins (Gregersen and Basham 1989)
indicate that considerable strong seismic activity (magnitude
>7) may have occurred during the early, rapid uplift phases
of glacioisostatic rebound (Basham and Gregersen 1989; Wood
1989; Johnston 1989; Hasegawa and Basham 1989).
For almost a century, however, other geologists and geophysicists have suggested that the offset striations observed in
the Appalachians reflect ongoing neotectonic movements (e.g.,
Chalmers 1897; Oliver et al. 1970). They cite the regional
consistency of fault orientations and displacements as evidence
of significant neotectonic displacement. In the case of the fault
at Saint-Jacques, we cannot suggest whether it was generated
by isostatic adjustment, tectonic forces, or some other unrecognized process. Since it is associated with bedrock slopes that
apparently failed catastrophically just after deglaciation, the
causal link with glacioisostatic stresses is strong. Nevertheless,
the Holocene paleoseismic record of lakes in the Ttmiscouata Madawaska Valley, discussed below, suggests that the valley
also may be susceptible to earthquakes generated by neotectonic
processes, well after any important glacioisostatic adjustments
took place.
Small displacements of bedrock caused by surface stresses
generated above the epicentre of the January, 1982, Mirarnichi
earthquake, located about 100 km southeast of Saint-Jacques,
are similar in scale to that observed in the Saint-Jacques pit
(Basham and Adams 1984). Extensive geological observations
and drilling around the Miramichi epicentral area further revealed the near-surface bedrock to be under significant horizontal compressional stress, leading to rock bursts and pop-ups
when the confining overburden was removed (Mackay et al.
1985; Martin 1988). If an earthquake generated the SaintJacques mass flow deposits, fracturing of the bedrock necessary to mobilize it on the relatively low local slopes may have
been caused by similar near-surface compressional stresses.
Subbottom acoustic profiling (SAP)
Lac TCmiscouata occupies an elongated, glacially overdeepened basin, with steeply sloping subaqueous sides and a relatively wide, fairly flat bottom (Fig. 6). Present lake level is
about 150 m as1 and water depth exceeds 65 m in places. The
lake basin comprises three distinct subbasins that can be distinguished on the basis of their bathymetry and predominant type
of sediment fill: (1) Basin "A" occupies approximately the
northern three-eighths of the lake and is generally shallower
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
CAN. J. EARTH SCI. VOL. 29, 1992
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
SHILTS ET AL.
than the southern basins ( <40 m). It is underlain by largely
undisturbed, acoustically parallel-laminated sediment (APLS)
with minor areas of small-scale hummocky topography (Fig. 7).
Its side slopes are gentler than those in the rest of the lake, and
it lacks the wide, flat bottom of the other basins. (2) The
southeastward-trendingcentral subbasin constitutes about onehalf of the lake and extends between the two prominent bends in
the lake's axis. It is deep (>65 m), flat bottomed, and steep
sided (Fig. 8). The abyssal plain of much of this subbasin is
characterized by a hummocky or mamillated surface characteristic of surfaces of sediment flows deposited by mass-transport
processes (Shilts 1984; Hardin 1987). (3) The southeasternmost
eighth of the lake is a separate, deep subbasin with mass transport deposits exclusively on the slopes along its southwest side.
Its sedimentary fdl consists of thin, disturbed or undisturbed
sediment with weak acoustic laminations lying on a smooth,
acoustically opaque surface (Fig. 9). The buried surface is
thought to be an extension of the sand plain that abuts the
south end of the lake. The sand pinches out against an anticlinal ridge of APLS that forms the lake bottom southwest
from the southeastern shore of Grande Baie, and divides the
distinctive sedimentary sequence in subbasin "C" from that in
subbasin "B. "
The SAP survey further revealed that the extensive shallow
platform at the south end of the lake is underlain by sediment
layers dipping gently in a northerly direction at an angle
exceeding that of the bottom slope. These deposits are also
correlated with the postglacial Madawaska sand plain and
were probably formed when Lake Madawaska was essentially
a wide reach of the Saint John River, draining northwestward
through the Madawaska Valley into the south end of Lac
TCmiscouata basin while Lac Ttmiscouata, in turn, was still
draining northward into the St. Lawrence estuary. Northwesterly
paleocurrent directions in the Madawaska sand plain reflect
the temporarily reversed drainage and have been reported by
Kiewiet de Jonge (1951), Lee (1955), and Gauthier and Thibault
(1980). The weakly laminated sediment overlying the sand in
Lac Ttmiscouata represents deposition since it assumed its
modern level and southward drainage.
Northwestward drainage of Saint John River, debouching
into Lac TCmiscouata immediately after deglaciation of the
valley, might also explain the occurrence in the lake of the
crustacean Limnocalanus macrurus, which cannot swim against
a current, but can only move downstream. Dadswell (1974)
discussed this possibility, but dismissed it on the basis of
the absence of the species in lakes of northern Maine and
New Brunswick, which were supposedly connected with Lac
TCmiscouata during deglaciation. However, elevations of these
lakes are well above the level of the Saint John River at the
time of formation of the Madawaska sand plain. If the upper
Saint John drainage basin was deglaciated at a time when
south-facing Laurentide ice was still blocking northward
drainage in the Chaudikre Valley to the west (Blais 1989; Blais
and Shilts 1989), overflow toward the east from the Chaudibre
Basin, via the Daquaam-Famine Valley, could have brought
Limnocalanus into Lac TCmiscouata.
Acoustic sedimentary record
As discussed by Klassen and Shilts (1982) and by Shilts
and colleagues in subsequent papers (Shilts and Farrell 1982;
Larocque 1984; Shilts 1984; Shilts et al. 1989; Clague et al.
1989), the RTT-1000 system is capable of deep penetration
and high resolution of sedimentary structures in silty or clayey
1051
sediments, but normally lacks the power to penetrate sorted
sandy, gravelly sediments, bouldery diamictic sediments, or
sediments charged with gas. Thus, where acoustic records of the
subbottom can be obtained, the limited number of sediment types
that can be penetrated allow the sediment facies to be inferred
with a fair degree of confidence. In addition, experience has
shown that glaciolacustrine or glaciomarine sequences that have
accumulated in the deepest parts of lake basins are generally
acoustically laminated, although exceptions to this generalization
have been noted by the first author in Lac DeschCnes, Quebec,
near Ottawa, and by Mullins et al. (1990) in Kalamalka Lake,
British Columbia; in each of these cases thick accumulations
of glaciomarine (Deschenes) and glaciolacustrine (Kalamalka)
sediments have an acoustically transparent appearance, lacking reflectors. Postglacial or modern sediments, on the other
hand, commonly have few and weak reflectors, and are as a
rule acoustically transparent, whether they consist primarily of
the authigenic organic deposits (gyttja) common in smaller
lakes, or the organic-poor clayey or silty deposits common in
larger lakes.
Interpretations of acoustic profiles from Lac TCmiscouata
and Grand lac Squatec are based on the above generalizations
about RTT-1000 acoustic records, which are supported by
several deep boreholes to bedrock in lakes in Ontario and Quebec and by numerous grab samples and short cores collected
from boats and by divers in conjunction with SAP surveys.
Our level of confidence in our ability to infer sedimentary
characteristics from these subbottom acoustic records is high.
Our confidence in the interpretations of lake-bottom morphology, particularly the sediment deformations and hummocky
surfaces encountered in Lac TCmiscouata and Grand lac Squatec,
is equally high and is based on examination of similar features
in lakes observed to have been damaged during historic earthquakes (Shilts 1984; Clague et al. 1989; Shilts and Clague
1992), as well as on published examples of high-resolution
seismic records from surveys of sediment slumping in areas of
unstable marine or lacustrine sedimentation (e.g., Kelts 1978;
Hardin 1987).
The SAP survey of Lac TCmiscouata reveals that a major
deposit of thick ( > 15 m), acoustically clear sediment with
small-scale hummocky surface covers large parts of the lake's
bottom (Fig. 8). Adjacent to many areas with this sedimentary
characteristic are zones of erosionally dissected, parallellaminated sediments, which form larger scale hummocky
topography. Surveys of nearby Grand lac Squatec show that
it has similar deposits and areas of dissection, but that they are
not nearly so extensive as in Lac TCmiscouata.
Such hummocky and dissected bottom surfaces are extremely
uncommon in the more than 150 lakes we have surveyed in
Quebec, Ontario, Labrador, District of Keewatin, and British
Columbia (Shilts et al. 1989; Shilts and Clague 1992). They
have been noted in only 10 other lakes, of which four were
located within a few kilometres of strong historic earthquakes
(Shilts 1984; Clague et al. 1989).
For example, Lac MatapCdia, located about 100 krn northeast of Lac TCmiscouata in a similar geologic and topographic
setting, shows no evidence of sediment slumping, despite the
similarity of its bathymetry and sedimentary fill to Lac Ttmiscouata and Grand lac Squatec (Shilts et al. 1989) (Figs. 10,
11). Furthermore, Lac MatapCdia is immediately adjacent to
the lower St. Lawrence seismic zone (Adams and Basham
1989), where numerous, but relatively low intensity (magnitude <6) earthquakes have been recorded.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
CAN. 1. EARTH SCI. VOL. 29, 1992
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
S H U B ET AL.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
0
1
I
CAN. J. EARTH SCI. VOL. 29. 1992
1
2
I
km
I
FIG.9. (a)Profile showing typical slump deposits at base of southwest side of southern subbasin C of Lac TCmiscouata. Note weakly acoustically laminated modem sediment overlying smooth, acoustically opaque, sandy substrate. V.E. = 21. (b) Interpretation. See Fig. 8b for key
to description of sedimentation units.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
SHILTS ET AL.
.9
Q ) .
0
C)
za
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
1056
CAN. J. EARTH SCI. VOL. 29, 1992
FIG. 1 1 . (a) Apparently failed, acoustically parallel-laminated sediments in Lac TCmiscouata; preslump sedimentological and morphological
elements are similar to those in Lac MatapCdia. V.E. = 19. (b) Interpretation. See Fig. 8b for key to description of sedimentation.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
SHILTS ET AL.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
1058
CAN. J. EARTH SCI.
The hummocky deposits in Lac TCmiscouata have morphologic and acoustic characteristics similar to those reported for
sediments deposited by mass transport processes in the marine
or lacustrine environment (e.g., Kelts 1978; Prior et al. 1982;
Hardin 1987). The latter are generally attributed to mass transport associated with slumping of unstable sedimentary masses
(i.e., delta foreslopes in fiords or at the edge of continental
shelves, sediments disturbed by diapiric activity, subaerial
landslides, or sediments released by seismic activity).
There are four types of evidence of mass transport sedimentation in Lac TCmiscouata:
(1) Hummocky deposits appear to represent the toes of mass
flows originating in postglacial sediment on the basin sides,
below water level (Fig. 12). These deposits may consist of
individual or coalescing mass flows and are similar to seismically generated sediment flows observed in Lac Tee and Lac
Kipawa, Quebec (Shilts 1984; Shilts and Clague 1992). This
type of deposit is most common in subbasin C in Lac TCmiscouata and in all examples of mass transport in Grand lac
Squatec. In both lakes they can be seen to have deformed
underlying, late or postglacial APLS and the modern, acoustically nonlaminated sediment (ANLS) by drag or by loading.
In the case of the ANLS, deformation can take the form of
anomalous bulges or steps in the normally smooth bottom,
indicating a capacity for this sediment to transmit a compressional shock, generated by rapid loading, laterally (Fig. 12).
(2) A second type of mass transport deposit consists of
hummocky, acoustically chaotic sediment (ACS) with rare,
disoriented internal reflectors and sharp lateral contacts with
channels or kettles in parallel-laminated glaciolacustrine sediment (Figs. 13, 14). Except as noted above, this type of sediment
is acoustically without parallel lamination, but it commonly
inhibits effective acoustic penetration into the underlying APLS.
ACS occurs preferentially in subbasin B of Lac Ttmiscouata,
where several older deposits with the same characteristic hummocky surfaces and massive character can be seen to be buried
by successive mass transport deposits. The close association of
these deposits with channels and precipitous faces that cut across
acoustically laminated glaciolacustrine sediment suggests that
repeated failures of ice-contact faces in APLS (Shilts 1984;
Larocque 1984) and sides of channels cut by earlier mass flows
into APLS or ACS contributed to successive mass flow deposits.
(3) In some cases disruption of acoustic lamination and
development of an exposed or buried hummocky surface within
a particular bed suggests that the sediment was fluidized and
disrupted in place, with little or no lateral transfer of sediment
(Fig. 15). This type of liquefaction, where observed on a small
scale in outcrops of young lacustrine sediment, has been
widely attributed to the effects of strong seismic shocks (Sims
1975; Scott and Price 1988; and others), although aseismic
sedimentation phenomena related to rapid loading or dewatering are in many cases equally plausiblemechanisms for formation of such small-scale features. The large-scale examples of
in situ sediment disruption in Lac Ttmiscouata are probably
not related to glaciotec~onicdisruptions sometimes observed in
proglacial lakes because of their limited area of occurrence,
abrupt boundaries on all sides, and, most importantly, their
associated disru~tionsof overlying Holocene sediments.
The beds dishpted in situ, ciepLted in Fig. 15a, comprise
over 13 m of chaotically deformed sediment, which passes
laterally with an apparent faulted contact on one side into
undeformed, acoustically laminated beds. The configuration
VOL. 29,
1992
of the disruption is such that lateral transport of beds is no
more than a few metres. It is difficult to conceive of a mechanism, other than seismic shaking, that could have created such
massive postdepositional deformation of an acoustically laminated unit. In some profiles, disrupted parallel-laminated beds
are seen to be buried by a thick cover of undisturbed postglacia1 sediment (Figs. 7, 15). Whether these sediments represent
preferential in situ liquefaction of the disrupted bed or an early
postglacial mass transport deposit buried by postglacial sediment is often difficult to determine.
(4) Many areas of Lac TCmiscouata and Grand lac Squatec
are marked by an irregular, hummocky surface that appears to
represent channels formed by erosion by turbidity currents of
both modern and parallel-laminated glaciolacustrine sediment
(Figs. 8, 16). The features are generally cut on slopes and
seem to indicate movement of the eroding medium directly
down the slope. These channels are thought to represent erosion by mass flows composed of sediment dislodged from
basin sides. Thus, the hummocky deposits at the foot of these
eroded areas must be formed not only of sediment dislodged
from high on the basin side, but also of sediment eroded and
incorporated into the flow as it descended the slope (Fig. 17).
Discussion
At three places within or adjacent to the TCmiscouataMadawaska, trans-Appalachian depression, we have found what
we consider to be unequivocal evidence of deposits formed by
mass transport processes that have operated periodically over
a time spanning the Holocene. The question remains, however,
of what process or processes initiated the downslope transfer
of large amounts of sediment and rock. Because of the important geological and sociological implications of our most
favoured explanation, it is important to evaluate the evidence
conservatively, while recognizing that it could have serious
implications for anthropogenic activities in the region.
Saint-Jacques
The rockfalls at Saint-Jacques apparently occurred while
Lake Madawaska, which was a greatly expanded and northwarddraining, immediately postglacial version of Lac TCmiscouata,
was still in existence, that is to say, prior to about 9000 BP.
The angular clasts of local bedrock, some with one striated
face, the paucity of clasts with exotic lithologies, the lack of
matrix, and the smooth channelized contact with underlying
Lake Madawaska sand and silt suggest that these deposits are
debris slides that originated above or near the lake shore. The
deposits are covered in places with thin lacustrine sand and
apparently entered the valley down opposite slopes. In one pit,
they lie near and above striated bedrock that has been faulted
with a vertical displacement of at least 7 cm.
Taken together, these observations support a seismic shock
as the trigger for the debris slides. Judging from the relatively
low slopes of the valley, and the competent nature of the
bedrock, a substantial shock must have been required to dislodge such a high volume of rock. The timing of the shock
would place it in early postglacial time. This suggests that the
hypothetical seismic event(s) may have been related to isostatic rebound.
Lac T&rniscouata
The central subbasin of Lac TCmiscouata has been modified
at least three times by major mass transport sedimentation events
(Fig. 13, 14). Whether sediment disruption and resedimenta-
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
SHILTS ET AL.
FIG. 13. (a) Multiple mass transport deposits in Lac TCmiscouata, one buried surface. Note older, buried hummocky surfaces. V.E. = 20.
(b) Interpretation. See Fig. 8b for key to description of sedimentation.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
CAN. J . EARTH SCI. VOL. 29, 1992
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
SHILTS ET AL.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
CAN. J. EARTH SCI. VOL. 29, 1992
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
SHILTS ET AL.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
CAN. 1. EARTH SCI. VOL. 29, 1992
o oC oU o mo cc
(u)H l d M
SHILTS ET AL.
1065
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
~ G l a c i o l a c u s t r i n ea n d
FIG. 17. Diagram of probable geometry of mass flows on a lake bottom. A, B, and C represent toe of mass flow, hummocky, chaotic mass
flow deposits, and erosional gullies carved by mass flows, respectively.
tion occurred throughout the basin three times or whether the
net result of numerous mass transport events that affected various parts of the basin gives the impression of only two or three
events at any one place is not known. Slumping in the north
and south subbasins is much more restricted than in basin B
and usually indicates one, fairly recent mass transport event.
In the southern basin (C) a single event is marked by deposits
preferentially thrown from the south-southwest side of the
lake, unlike the central and northern basins, where slumping
has occurred on both sides.
Although mass transport deposits can be found buried well
back within the early postglacial sedimentary record, they can
be ascribed a relatively recent age in some parts of the basin,
because acoustically clear, postglacial sediment is displaced or
deformed, and the hummocky depositional and erosional surfaces show no sign of burial by modern sediment. This is
particularly true in the southernmost basin, where the hummocky deposits are as fresh looking as similar ones that were
formed in Lac Tee, Quebec (Shilts 1984; Shilts and Clague
1992) by sediment dislodged during the 1935 Ttmiskaming
earthquake (Hodgson 1936).
We assume that most or all of the mass transport deposits
in Lac TCmiscouata were generated by seismic shocks that
have occurred periodically in postglacial time. Because this
area is historically aseismic, the source of the seismic energy
necessary to dislodge sediment previously was speculated (Shilts
et al. 1989) to have been strong earthquakes in the nearby
(100 krn) Charlevoix area, the most seismically active region
in eastern North America (Adarns and Basham 1989). Reexarn-
ination of the acoustic records and assumptions that lead to the
Charlevoix hypothesis has caused us to reconsider the principal arguments that supported that hypothesis. In doing this,
a "hierarchy" of increasingly less conservative scenarios for
the production of mass transport deposits in Lac TCmiscouata
can be constructed:
(1) The most conservative interpretation of the available
data, historical seismic and SAP records, would suggest that
the slumping observed in Lac TCmiscouata has an aseismic cause
such as those proposed for mass transport deposits in another
trans-Appalachian Lake, Lac Mtgantic, which has similar late
or postglacial sediment fill and bathymetry (Larocque 1984;
Shilts 1984; Shilts et al. 1989). In Lac MCgantic, however, the
slump deposits are limited in extent and related to specific
shoreline features that could have been predicted to be sites
with a high probability of failure. In Lac Ttmiscouata, the
failure of slopes in the central basin is general and apparently
largely unrelated to shoreline phenomena (Fig. 18). In fact,
except for the delta of Rivibre Touladi, present and former
deltas, notoriously unstable environments in many lakes, show
little or no evidence of failure in Lac TCmiscouata. Furthermore, some of the disruption of beds seems to have taken place
in situ, beneath the profundal plane of the lake, and involved
virtually no lateral transport of the 13 m of sediment affected
(Figs. 13, 14). The authors can conceive of no postglacial
aseismic process that could cause such massive postdepositional disruption. The extent to which the bottom and sides of
Lac TCrniscouata are affected by mass transport processes is
unprecedented in most of the more than 150 lakes surveyed
CAN. J.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
1066
EARTH SCI. VOL. 29, 1992
contour interval. 5 m
FIG.18. Map showing areas (shaded) of bottoms of Lac TCmiscouata and Grand lac Squatec that have been affected by mass flow processes.
with SAP in this program, including lakes that were observed
to be severely damaged by strong recent earthquakes on Vancouver Island and near Lac TCmiskaming.
(2) The second most conservative explanation for the sedimentation disruptions in Lac TCmiscouata and nearby Grand
lac Squatec is that they were caused by strong seismic shocks
in the Charlevoix region, as proposed by hilts et al. (1989).
The extent of the damage to Lac TCmiscouata and the unidirectional nature of somevof the slumping are characteristic of
seismically generated subaqueous slumping observed during
historical earthquakes (Shilts 1984; Clague et al. 1989; Shilts
et al. 1989; Shilts and Clague 1992). This, the repetitive nature
of the slumping events at specific sites, the occurrence of
in situ disruptions, and the similarity of morphology and acoustic structure between these sediments and those known to be
seismically disturbed, lead us to believe that these deposits
were seismically generated, but there are arguments against
linking them with seismicity in the Charlevoix area.
(3) The least conservative of the hypotheses that we have
entertained to explain the sediment patterns in Lac Ttmiscouata
is that they were seismically produced by strong earthquakes
and (or) neotectonic displacements in or near the TCmiscouataMadawaska Trench itself. In spite of the lack of historical evi-
dence of seismicity, we think that epicentre locations for a series
of postulated seismic shocks are very near the TCmiscouata
basin. Nearby Grand lac Squatec is damaged, but much less so
than Ttmiscouata, an observation incompatible with the hypothesis that they were both disturbed by shocks at Charlevoix, more
or less equidistant from each of them. Lac MatapCdia, though
located virtually beside an active seismic zone (albeit one with
shocks less severe than the strongest at Charlevoix), is completely undisturbed. Even the north and south subbasins of Lac
TCmiscouata itself, though susceptible to seismically generated mass transport sedimentation, as indicated by the mass
transport deposits in them, are markedly less damaged than the
central basin. While it could be argued that basin A lacks the
bathymetric and sedimentological configuration necessary to
promote sediment failure, the same argument cannot be made
for steep-sided, southern basin C. Thus, we surmise that there
is a strong possibility that the Lac TCmiscouata basin itself has
been subjected to local, repeated, strong seismic shocks and
that it may be at risk for further disturbance. In spite of the
lack of geological evidence for a major structure associated
with the TCmiscouata-Madawaska Trench, there may be
some coincidence of geology, topography, and seismicity that
is so subtle that it has not yet been recognized.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
ET AL.
Conclusions
The principal conclusion we have drawn from our observations of sediments in the Saint-Jacques, New Brunswick,
aggregate pits and of subbottom profiles of Lac TCmiscouata and
Grand lac Squatec is that both sediment sequences are dominated
by mass transport deposits. A second and far more significant
conclusion is that the mass transport deposits were generated
by seismic shocks, probably strong seismic shocks, that originated within or near the historically aseismic TCmiscouataMadawaska depression, or, less likely, at Charlevoix or some
other external, historically seismically active site.
In the Saint-Jacques pits, we base the inference that the
coarse diamictons exposed there are nonglacial mass transport
deposits into Lake Madawaska on (1) their occurrence near the
top of but within the sedimentary sequence in postglacial Lake
Madawaska, (2) the angularity of the clasts and their local
provenance, (3) the presence of clasts with one glacially striated surface, (4) paleocurrent indicators showing flow from
sources on either side of the valley toward its axis, and (5) the
nature of their erosional contacts with and load structures in
underlying, fine-grained sediments. We feel the diamictons
were seismically generated because (1) the relatively gentle
bedrock slopes would require a strong seismic shock to fragment and dislodge their underlying metasedimentary bedrock,
(2) two separate diamictic masses were generated, if not
simultaneously, in close temporal proximity, from both sides
of the valley, flowing toward its axis, and (3) striated bedrock
in one of the pits is faulted with displacement of at least 7 cm,
suggesting either seismically induced fracturing or neotectonic
movement, as has been inferred elsewhere in the Appalachians
for similar features (for example, Oliver et al. 1970; Grant
1989).
For Lac TCmiscouata and Grand lac Sauatec. we conclude
that much of the sediment on their bottoms comprises subaqueously generated mass transport debris for the following reasons:
(1) the small-scale hummocky surfaces typical of subaqueous
slump deposits (Prior et al. 1982; Shilts 1984; Hardin 1987;
Clague et al. 1989) characterize significant areas of both
lakes; (2) the deposits beneath these hummocky surfaces are
acoustically clear or occluded by gas, and contain chaotic
and irregular reflectors; (3) underlying, acoustically parallellaminated sediments are deformed in many places by compressional or tractive stresses generated by the superincumbent
flows; (4) many parts of the slopes above hummocky sediment
on the abyssal plain are scoured and gullied at such great water
depths that only gravity-driven slurries or currents could have
eroded them.
The inference that these mass flows were generated by seismic activity was derived from the following observations: (1)
the deposits occur throughout the lakes, regardless of shoreline types or processes; (2) the deposits often seem to result
from a series of coalescing flow slides, generated simultaneously from several points along the basin side; (3) one identifiable mass transport event in the south end of Lac TCmiscouata
is decidedly asymmetric, all flows being generated from the
southwest side, a characteristic of earthquake-damaged lakes
elsewhere (Shilts 1984; Shilts and Clague 1992); (3) hummocky
mass flow de~ositsare uncommon in lakes with similar sediment fill and dathymetry in this region and elsewhere throughout
Canada; (4) slumping is predominantly from normally stable
areas of quiet water deposition, generally not from notoriously
unstable sedimentary environments such as delta fronts or spits;
1067
and (5) thick, acoustically laminated sedimentary sequences
underlying the profundal plane of the lake have been severely
disrupted in situ.
Furthermore, we feel that the seismic shock that caused the
slumping was probably local and not from Charlevoix, 100 km
away, because (1) sediment in nearby Grand lac Squatec, though
disturbed in a fashion similar to that in Lac Ttrniscouata, has
far smaller areas of disturbance, which should not be the case
if both lakes were shocked by a seismic event centred 100 km
distant; (2) likewise, the north and south ends of Lac TCmiscouata are not nearly so damaged as the central portion, suggesting that the shock was local; (3) lakes located much nearer
to known active seismic zones than Lac TCmiscouata, such as
Lac MatapCdia, show no evidence of earthquake damage.
Finally, we can draw no temporal link between the mass
transport events at Saint-Jacques and Lac TCmiscouata. The
Saint-Jacques event(s) took place in early postglacial time,
before drainage reversed to allow the Saint John River to flow
southward through Grand Falls instead of northward through
Lac TCmiscouata. Thus, vertical stress relief associated with
ice sheet removal, accompanied by isostatic adjustments, may
have played a role in generating late glacial seismic activity,
as suggested by Johnston (1989) and Wood (1989). The disturbances in Lac TCmiscouata, on the other hand, seem to have
been repeated several times, at least two buried slumped surfaces being identified beneath a hummocky surface deposit on
some profiles. The lack of detectable sediment cover over the
uppermost hummocky surfaces suggests that the last slumping
event was fairly recent, probably within the last thousand
years; buried slump deposits within the glaciolacustrine
sequences suggest that seismicity also was present early in
postglacial time, possibly linking the TCmiscouata and SaintJacques events in a loose temporal framework.
We conclude that the Ttrniscouata-Madawaska valley is and
has been a source of strong seismic events, and suggest that further research to core and date the sediments of Lac TCmiscouata
and nearby lakes would yield some evidence of recurrence intervals and possibly establish whether these deposits can be linked
to historical seismicity in surrounding regions.
Acknowledgments
The authors were assisted in the field by Hazen Russell,
Louise Gagnon, and Ineke Prins. They appreciate discussions
of the ideas expressed herein with various colleagues, especially
John Adams and Thomas Lowell. Daren Stebner organized
and carried out preliminary interpretations of the profiles and
drew and conceived the diagram used as a base for Fig. 15.
Wendy Rebidoux provided invaluable assistance in the final
preparation and editing of the figures. We are indebted to
Peter Basham and Kerry Kelts for their careful, critical review
of the manuscript. Although we incorporated most of the
changes they suggested and clarified the text in response to
their comments, the authors assume full responsibility for the
conclusions drawn in the paper as published.
Adams, J. 1981. Postglacial faulting: a literature survey of occurrences in eastern Canada and comparable glaciated areas. Atomic
Energy of Canada Limited, Technical Report 142.
Adams, J., and Basham, P. 1989. The seismicity and seismotectonics
of North America. Geoscience Canada, 16: 3- 16.
Allen, J. R. L. 1984. Sedimentary structures. Their character and
physical basis. Vol. 2. Elsevier, Amsterdam.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
1068
CAN. .I.
EARTH SCI. VOL. 29, 1992
Basham, P. W., and Adams, J. 1984. The Mirarnichi New Brunswick
earthquakes: New surface thrust faulting in the northern Appalachians. Geoscience Canada, 11: 115 - 120.
Basham, H. S., and Gregersen, S. 1989. Earthquakes at and near
North Atlantic passive margins: neotectonics and post-glacial
rebound-a workshop report. Terra Nova, l(2): 118 - 122.
Berryhill, H. L., Jr. 1987. Use of high-resolution reflection profiling
techniques for identifying major late Quaternary continental
shelflslope facies. In Late Quaternary facies and structure, northern Gulf of Mexico: Interpretations from seismic data. Edited by
H. L. Berryhill, Jr. American Association of Petroleum Geologists, Studies in Geology, no. 23, pp. 1 - 11.
Blais, A. 1989. Lennoxville Glaciation of the Middle Chaudiitre and
Etchemin valleys, Beauce region, QuCbec. M.Sc. thesis, Carleton
University, Ottawa, Ont.
Blais, A., and Shilts, W. W. 1989. Surficial geology of Saint-Josephde-Beauce map area, ChaudiBre River valley, QuCbec. In Current
research, part B. Geological Survey of Canada, Paper 89-lB,
pp. 137-142.
Chalmers, R. 1897. Report on the surface geology and auriferous
deposits of south-eastern Quebec. Geological Survey of Canada,
Annual Report 10, pp. 5 - 12.
Chauvin, L., Martineau, G., and LaSalle, P. 1985. Deglaciation of
the Lower St. Lawrence region, Qutbec. In Late Pleistocene history of northeastern New England and adjacent Quebec. Edited by
H. W. Borns, Jr., P. LaSalle, and W. B. Thompson. Geological
Society of America, Special Paper 197, pp. 111 - 123.
Clague, J. J., Shilts, W. W., and Linden, R. H. 1989. Application
of subbottom profiling to assessing seismic risk on Vancouver
Island, British Columbia. In Current research, part E. Geological
Survey of Canada, Paper 89-lE, pp. 237-242.
Dadswell, M. J. 1974. Distribution, ecology, and postglacial dispersal of certain crustaceans and fishes in eastern North America.
National Museum of Natural Sciences (Canada), Publications in
Zoology, no. 11.
Dionne, J.-C., Rousseau, G., Dumais, P., and HCtu, B. 1988. Excursion no 2 : Aspects du Quaternaire dans les regions de SaintFabien et de Trois-Pistoles. In Livret-guide de l'excursion: CBte
sud de l'estuaire du Saint-Laurent. 6e Congrits de llAQQUA,
Rimouski, QuC.
Gauthier, R. C., and Thibault, J. 1980. Deglaciation of the Edmunston
area and reappraisal of Glacial Lake Madawaska intepretation. In
Guidebook to the geology of northeastern Maine and neighbouring
New Brunswick. Edited by D. C. Roy and R. S. Naylor. 720d
Annual Meeting, New England Intercollegiate Geological Conference, Maine, pp. 188-190.
Grant, D.R. 1989. Quaternary geology of the Atlantic Appalachian
region of Canada. In Quaternary geology of Canada and Greenland. Edited by R. J. Fulton. Geological Survey of Canada, Geology of Canada, No. 1, pp. 393-440.
Gregersen, S., and Basham, P. W. 1989. Earthquakes at North
Atlantic passive margins: Neotectonics and postglacial rebound.
Kulwer-Academic Publishers, Dordrecht, Boston, London.
Hardin, N. S. 1987. Mass transport deposits of the upper continental
slope, northwestern Gulf of Mexico. In Late Quaternary facies and
structure, northern Gulf of Mexico: Interpretations from seismic
data. Edited by H. L. Berryhill, Jr. American Association of
Petroleum Geologists, Studies in Geology, no. 23, pp. 241 -284.
Hasegawa, H. S. 1988. Neotectonics and inferred movements in
Canada. Bulletin of the Geological Society of Finland, no. 60,
pp. 3-15.
Hasegawa, H. S., and Basham, P. W. 1989. Spatial correlation between
seismicity and postglacial rebound in eastern Canada. In Earthquakes
at North Atlantic passive margins: Neotectonics and postglacial
rebound. Edited by S. Gregersen and P. W. Basham. KulwerAcademic Publishers, Dordrecht, Boston, London, pp. 483 -500.
Hodgson, E. A. 1936. The Timiskaming earthquake of November 1,
1935-the location of the epicentre and determination of focal
depth. Journal of the Royal Astronomical Society of Canada, 30:
113-123.
Johnston, A. C. 1989. The effect of large ice sheets on earthquake
genesis. In Earthquakes at North Atlantic passive margins: Neotectonics and postglacial rebound. Edited by S. Gregersen and
P. W. Basham. Kulwer-Academic Publishers, Dordrecht, Boston,
London, pp. 581 -599.
Kelts, K. 1978. Geological and sedimentary evolution of Lakes
Zurich and Zug, Switzerland. D.Sc. thesis, Eidgenoessischen
Technischen Hochschule, Zurich, Switzerland.
Kiewiet de Jonge, E. J. C. 1951. Glacial water levels in the Saint John
River Valley. Ph.D. thesis, Clark University, Worchester, Mass.
Kite, J. S. 1979. Postglacial geologic history of the middle Saint John
Valley. M.Sc. thesis, University of Maine, Orono.
Kite, J. S., and Stuckenrath, R. 1986. Postglacial history of the upper
St. John drainage basin. In Contributions to the Quaternary geology of northern Maine and adjacent Canada. Edited by J. S. Kite,
T. V. Lowell, and W. B. Thompson. Maine Geological Survey,
Bulletin 37, pp. 113- 128.
Klassen, R. A., and Shilts, W. W. 1982. Subbottom profiling of
lakes of the Canadian Shield. In Current research, part A. Geological Survey of Canada, Paper 82-lA, pp. 374-384.
Larocque, A. C. L. 1984. Depressions in the bottom of Lac MCgantic, Qutbec-probable stagnant ice features. In Current research,
part B. Geological Survey of Canada, Paper 85-lB, pp. 431 -439.
Lee, H. A. 1955. Surficial geology of Edmundston, Madawaska and
Temiscouata Counties, New Brunswick and Quebec. Geological
Survey of Canada, Paper 55- 15.
Lesptrance, P. J., and Grenier, H. R. 1969. RCgion de SquatecCabano. Ministbre des Richesses naturelles du QuCbec, RG-128.
Lortie, G., and Martineau, G. 1987. Les systbmesde stries glaciaires
dans les Appalaches du QuCbec. Ministitre de 1'Energie et des Ressources, Qutbec, DV 85-10.
Lougee, R. J. 1954. The role of upwarping in the post-glacial history
of Canada, Part 11; The maritime region and St. John Valley. Canadian Geographical Journal, 23: 3 -52.
Lowell, T. V., Kite, J. S., Calkin, P. E., and Halter, E. F. 1990.
Analysis of small-scale erosional data and a sequence of late Pleistocene flow reversal, northern New England. Geological Society of
America Bulletin, 102: 74 -85.
Mackay, D. A., Williams, J. B., Bowlby, J. R., and Grass, J. D.
1985. Miramichi epicentral area-in situ stress pilot project.
Ontario Hydro Research Division, Report 85-185-K.
Martin, G. L. 1988. Neotectonics in the Maritime Provinces. Atomic
Energy Control Board of Canada, Research Report INFO-0265.
Mullins, H. T., Eyles, N., and Hinchey, E. J. 1990. Seismic reflection investigation of Kalamalka Lake; a "fiord lake" on the
Interior Plateau of southern British Columbia. Canadian Journal of
Earth Sciences, 27: 1225- 1235.
Oliver, J., Johnson, T., and Dorman, J. 1970. Postglacial faulting
and seismicity in New York and Quebec. Canadian Journal of
Earth Sciences, 7: 579-590.
Prior, D. B., Bornhold, B. D., Coleman, J. H., and Bryant, W. R.
1982. Morphology of a submarine slide, Kitimat Arm, British
Columbia. Geology, 10: 588 -592.
Quinlan, G. 1984. Postglacial rebound and the focal mechanisms of
eastern Canadian earthquakes. Canadian Journal of Earth
Sciences, 21: 1018 - 1023.
Rampton, R. N., Gauthier, R. C., Thibault, J., and Seaman, A. A.
1984. Quaternary geology of New Brunswick. Geological Survey
of Canada, Memoir 416.
Rappol, M., and Russell, H. 1989. Glacial dispersal of Precambrian
Shield and local Appalachian indicator rocks in Lower St. Lawrence
and western GaspCsie, Qutbec and adjacent New Brunswick. In
Current research, part B. Geological Survey of Canada, Paper
89-lB, pp. 127-136.
Scott, B., and Price, S. 1988. Earthquake-induced structures in
young sediments. Tectonophysics, 147: 165 - 170.
SHILTS ET AL.
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by 74.121.191.133 on 06/07/13
For personal use only.
I
Shilts, W. W. 1976. Glacial events in southern Quebec - northern
New England, a reappraisal. Geological Society of America,
Abstracts with Programs, 8: 267.
Shilts, W. W. 1984. Sonar evidence for postglacial tectonic instability of the Canadian Shield and Appalachians. In Current research,
part A. Geological Survey of Canada, Paper 84-IA, pp. 567-579.
Shilts, W. W., and Clague, J. J. 1992. Documentation of earthquakeinduced disturbance of lake sediments using subbottom acoustic
profiling. Canadian Journal of Earth Sciences, 29: 1018 - 1042.
Shilts, W. W., and Farrell, L. E. 1982. Subbottom profiling of
Canadian Shield lakes-implications for interpreting effects of
acid rain. In Current research, part B. Geological Survey of Canada,
Paper 82-lB, pp. 209-221.
Shilts, W. W., Blais, A., and Adams, J. 1989. Subbottom profiling
of Qutbec Appalachian Lakes and its potential application to
assessing seismic hazard. In Current research, part B. Geological
Survey of Canada, Paper 89-lB, pp. 143 - 154.
1069
Sims, J. D. 1975. Earthquake-induced structures in sediments of Van
Norman Lake, San Fernando, California. Science (Washington,
D.C.), 182: 161 -163.
St. Peter, C. 1977. Geology of parts of Restigouche, Victoria and
Madawaska Counties, northwestern New Brunswick. New Brunswick Department of Natural Resources, Mineral Resources Branch,
Report of Investigation 17.
Thomas, R. H. 1977. Calving bay dynamics and ice sheet retreat up
the St. Lawrence Valley system. Gtographie physique et Quaternaire, 31: 347-356.
Wood, R. M . 1989. Extraordinary deglaciation reverse faulting
in northern Fennoscandia. In Earthquakes at North Atlantic passive margins: Neotectonics and postglacial rebound. Edited by
S. Gregersen and P. W. Basham. Kulwer-Academic Publishers,
Dordrecht, Boston, London, pp. 141 - 173.