Late Pleistocene stratigraphy and chronology of lower Seymour Valley,
southwestern British Columbia
OLAVB. LIAN]AND EDWARD
J. HICKIN
Department of Geography and The Institute for Quaternary Research, Simon Fraser University,
Burnaby, B. C . , Canada V5A IS6
Received April 7, 1992
Revision accepted December 18, 1992
A detailed study of the surficial sediments and landforms in the lower Seymour Valley has provided a complete record
of the last glacial cycle. This study shows that the valley was slowly aggrading before 29 ka and rapidly aggrading after 29 ka.
Ice of the Fraser Glaciation reached the mouth of the valley about 22 ka and retreated before 18.5 ka. The valley was subsequently vegetated. Ice once again advanced and occupied Seymour Valley after about 17 ka. Before 11.5 ka ice in Seymour
Valley retreated a final time, and the valley once again became vegetated.
Une Ctude detaillCe des sCdiments de surface et des formes de terrain dans la basse vallCe de Seymour a permis de documenter de manibre complkte le dernier cycle glaciaire. Cette ttude revkle qu'antkrieurement i 29 ka les alluvions s'accumulaient lentement dans la vallCe, tandis que postCrieurement i 29 ka leur accumulation fut rapide. Un glacier de la Glaciation
de Fraser a rejoint l'embouchure de la vallte il y a environ 22 ka, et il s'est retirt de la vallCe avant 18,s ka. La vallee a
kt6 ulterieurement envahie par la vCgCtation. Aprks 17 ka environ, le glacier a de nouveau avancC et occupe la vallCe de
Seymour. Le glacier s'est finalement retirC de la vallBe de Seymour antkrieurement i 11,s ka, et i nouveau la vegetation
a envahi la vallCe.
[Traduit par la redaction]
Can. J. Earth Sci. 30, 841 -850 (1993)
Introduction
In the Fraser Lowland and surrounding mountain valleys,
sediments deposited during the Middle Wisconsinan generally
have been eroded by subsequent glacial and (or) fluvial action,
or are overlain by thick deposits of sediments deposited during
the last (Fraser) glaciation and postglacial time. Exposures of
Middle Wisconsinan sediments in association with Late Wisconsinan drift therefore are rare.
Seymour Valley (Figs.1, 2) contains some of the best
natural exposures of Middle and Late Wisconsinan sediments
in the Fraser Lowland and adjacent mountain valleys. The
exposures in Seymour Valley reveal an almost complete stratigraphic sequence spanning more than 37 000 radiocarbon
years. In this paper we present the first detailed description
and interpretation of these sediments.
The study area
Seymour Valley, located north of Vancouver within the
Pacific Ranges of the Coast Mountains (Fig.l), is one of
several major valleys extending north-south to the Fraser
Lowland. The study area extends about 20 km from Seymour
Falls Dam to Burrard Inlet (Fig. 2).
The Seymour River basin is relatively narrow (-5 km
wide) and deep, rising eastward from near sea level to an elevation of 1455 m (Mount Seymour) and westward to an elevation of 1466 m (Coliseum Mountain, Lynn Ridge). Seymour
Valley probably was formed in the Late Cretaceous by fluvial
incision as the Coast Plutonic Complex began to rise (Armstrong 1990).
Rounded peaks and the characteristic U shape of the valley
provide evidence for ice-sheet and valley glaciation. Additional evidence is supplied by a well-exposed valley fill comprising sediments that represent at least one major glacial
'Present address: Department of Geology, The University of
Western Ontario, London, Ont., Canada N6A 5B7.
Printed in Canada I Imprimt au Canada
cycle and an interglaciation. The top of the valley fill forms
a bench at about 200 m above sea level (asl) between Seymour
Falls Dam and Rice Lake (Fig. 2). North of Rice Lake, glacial
sediments are generally covered by fans and aprons emanating
from tributary valleys. South of Rice Lake, the valley descends,
over a distance of -7 km, to about 15 m as1 across a series
of postglacial wave-cut terraces. Seymour River flows near
the surface of the valley fill at Seymour Falls Dam, but has
incised 100 m at Rice Lake, leaving behind a complex system of terraces. The base level of Seymour River below the
dam presently is controlled by a bedrock canyon between
-2-4 km above the mouth.
Seymour Valley is one of the principal watersheds supplying
drinking water to the Vancouver metropolitan area. For that
reason, access to Seymour Valley has been restricted since
1936. As a result of this closure, no detailed Quaternary
research has been undertaken previously in the valley. In
1987, however, Seymour Valley below Seymour Falls Dam
was opened to the general public as the Seymour Demonstration Forest.
The first published geological report that mentions Seymour
Valley likely is that of Burwash (1918). Burwash's study was
regional, and he did not conduct any detailed studies of the
Quaternary deposits in the valley. He did, however, list Seymour Valley as one of the best "drift sections" in the Vancouver area. Johnston (1923) briefly discussed the raised marine
terraces and bedrock canyon in Seymour Valley, and Armstrong
(1956) produced a surficial geology map (scale 1 : 63 360) that
includes Seymour Valley north to about 49'22'. Wagner (1959)
described fossil shells in the lower valley, whereas Maynard
(1978) interpreted the stratigraphy of a few exposures near the
mouth of the valley.
The surficial geology of Seymour Valley below Rice Lake
has been described generally by Armstrong and Hicock
(1980), and between Rice Lake and Seymour Falls Dam by
Lian (1991). The Quaternary stratigraphy of the Fraser Lowland discussed in this paper is summarized in Fig. 3.
-
CAN. J . EARTH SCI. VOL. 30, 1993
123' 00.
I
FIG. 1. Location of Seymour Valley.
Lithostratigraphy
We have identified 14 lithostratigraphic units in lower Seymour Valley. Of these, 13 (units 1- 13) are of Pleistocene age
and are discussed in this paper; unit 14 represents sediments
deposited during the middle Holocene (see Lian 1991). Locations of measured sections are shown in Fig. 2, and schematic,
composite cross sections of the valley fill are shown in Fig. 4.
Representative stratigraphic sections are presented in Fig. 5;
units 9-11 (sections SVMS-1 to SVMS-3, SVMS-29, and
SVMS-30) are not represented in Fig. 5, as they occur in small
exposures, and contacts with other units are rare. A summary
of the radiocarbon ages is given in Table 1.
Unit 1
Unit 1 is up to -7 m thick and includes three facies:
(i) weakly stratified, compact, subrounded pebble gravel;
(ii) horizontally bedded sand and silt, locally with disseminated
organics; and (iii) highly compressed woody peat beds up to
20 cm thick (Figs. 6, 7).
The unit comprises a series of cycles, each cycle generally
consisting of an upward-fining sequence of pebble gravel,
coarse sand to silt with organic stringers, and peat. In places,
however, beds of disseminated organics are separated by beds
of upward-coarsening sand and silt. At section SVMS-7 seven
peat beds, separated by fluvial sand and gravel, occur in 5 m
of exposure. The contacts between the peat, pebble gravel,
and sand are sharp.
Wood was found in sediments containing disseminated
organics, directly below a woody peat bed ( - 100 m asl) at
section SVMS-6. A 5 cm diameter log (Picea sp.) from this
320 BP (GSCbed yielded a radiocarbon age of 35 700
5069HP). A piece of wood (Abies sp.) from section SVMS-7
-
+
(-99 m asl) gave a radiocarbon age of 37 100 f 340 BP
(GSC-5121 HP); peat slightly higher ( - 101 m asl) in the
300 BP (Beta-46053). The
sequence was dated at 29 440
latter age is supported by a thermoluminescence age of 41
7 ka from mineral sediments within the peat (Lian 1991).
+
+
Unit 2
Unit 2 attains a thickness of -30 m; the contact with unit
1 was not observed. Unit 2 is exposed only at section SVMS-8
and is mostly covered by colluvium and vegetation. Exposures
are limited to the upper 5 m of the unit, and to an 6 m interval near its base. The lower 6 m consists of finely laminated
clayey silt, conformably(?) overlain by horizontally bedded
medium sand. The upper 5 m consists of horizontally bedded,
generally upward-coarsening, fine to coarse sand with some
gravel. No organic material was found in these deposits.
Because of the extent of the covered portion (-20 m) of
section SVMS-8, it is possible that unit 2 represents more that
one unit. However, the upper and lower exposures are similar,
thus we tentatively assign them to one unit.
Although the contact with unit 1 was not observed, it is
thought that unit 2 directly overlies unit 1 at sections SVMS-6,
SVMS-7, SVMS-8, and SVMS-10.
Radiocarbon ages from directly underlying and overlying
sediments indicate that unit 2 was deposited between approximately 29 and 22 ka.
-
Unit 3
18 m at section
Unit 3 has a maximum thickness of
SVMS-9. The contact with unit 2 is gradational. Unit 3 is composed of compact, laminated, blue-gray clayey silt with abundant plutonic dropstones (subrounded to rounded) reaching
50 cm in diameter. No obvious striae or facets were observed
-
843
LIAN AND HICKIN
TIME-
Years BP (X lo3) S T W A E Y H l C
(radiocarbon)
._ _ _ .
5
GEOLOGIC
CUMTIC UNITS
LITHOSTFLATIGFLAPHICUNITS
*
Holocene
Postglacial
Late
Wisconsinan
Fraser
Glaciation
Salish Sediments
and
Fraser River Sediments
.---.10 - - - - .
.---.11 - - - - .
.---.12 - - - - .
.---.13
.,,,.
15
----.
Vashon Drift
*
- - - - 17
. ----.
.---.18 - - - - .
.- - - .20 ----.
.---.26
.-,-. 30 - - - - .
Cowichan Head
Formation
.---.35 - - - -.
cross sections
--- -.41 - -- -.
.---.so
Middle
Wisconsinan
----.
I
FIG. 2. Seymour Valley study area showing the locations of the
measured sections discussed in this paper. Also shown are the locations of the composite schematic cross sections shown in Fig. 4.
on the dropstones. The unit also contains isolated beds of
coarse to medium sand. Stratification is weak near the base of
the unit, but becomes increasingly well defined (laminated)
upward.
wood is found near the base ofthe unit. section SVMS-8,
20 cm piece of wood yielded a radiocarbon age of
a 50
22 320
130 BP (Beta-40686). At section SVMS-9, a slice
of wood from a log (8 cm diameter) yielded a radiocarbon age
of 22 040 t- 130 BP (Beta-38909).
Age unknown
.-
-
-,
,
-,
, -,
,
,
,
,
,
,
,
Cowichan Head
Formation ?
.---.
60 - - - - .
----,62 ----..
I
Olympia
Nonglacial
Interval
Semiahmoo
Glaciation
Semiahmoo Drift
Highbury
Nonglacial
Highbury Sediments
Ea"Y wan"p""a"lnterval
pre-Wismns~nan
WesUynn
Glaciation
Westlynn Drift
Older Sediments
FIG.3. Time-stratigraphic, geoclimatic, and lithostratigraphic units
in the Fraser Lowland (after Armstrong 1984). Lithostratigraphic
units exposed in Seymour Valley are indicated by an asterisk.
+
Unit 4
Unit 4 has a maximum thickness of 15 m. The contact
with unit 3 is sharp. At section SVMS-9, it is composed of a
massive, matrix-supported diamicton. Coarse clasts are about
5 - 10 cm in diameter and generally are subrounded; nearly all
are plutonic. The larger clasts are more common near the base
of the unit. Scattered beds of horizontally and crossbedded
sands are present within the diamicton near the top of the unit.
At section SVMS-11, unit 4 is more complex and only the
lowest 2-3 m resembles the diamicton at section SVMS-9.
The remainder of the unit consists of interbedded massive,
clayey silt containing scattered pebble-sized clasts, massive silt
with no clasts, compact diamicton, and beds of massive, hori-
-
zontally and crossbedded, medium to coarse sand. Some of the
diamicton beds exhibit flow structures; locally the diarnicton
has been injected into surrounding sediments. A thick ( - 3 m)
cobble lag occurs in the lower part of the unit.
Unit 5
Unit 5 is up to -20 m thick and has a sharp contact with
unit 4. Unit 5 consists of laminated silty clay with relatively
few dropstones. The laminations are extremely convoluted at
about 190 m as1 at section SVMS-9. The dropstones are
mainly plutonic, but some are volcanic; the former are subrounded to rounded, whereas the latter generally are subangular.
Near the base of the unit is a single, 10-20 cm thick,
organic-rich bed containing small fragments of wood, char-
CAN. J. EARTH SCI. VOL. 30. 1993
@
CDgl
@
CHF
41 f 7 ka (TL
37 l o o ? 340
Distance (km)
Distance (km)
FIG. 4. Composite schematic cross sections of the Seymour Valley fill showing correlations with lithostratigraphic units (circled numbers)
in the Fraser Lowland; radiocarbon ages and average thicknesses are shown. See Fig. 2 for locations. (a) Cross section A. Vertical exaggeration
7.6x. (b) Cross section B. Vertical exaggeration 6.4 x . CHF, Cowichan Head Formation; QS, Quadra Sand; CDgl, Coquitlam Drift, glaciolacustrine sediments; CDt, Coquitlam Drift, till; VDgl, Vashon Drift, glaciolacustrine sediments; VDt, Vashon Drift, till; PMi, sediments
representing the Port Moody interstade; CS, Capilano Sediments; SS, Salish Sediments; TL, thermoluminescence age; BR, bedrock.
coal, and rare leaf imprints (Fig. 8). Locally, organic material
is concentrated at the surface of this bed. No evidence of soil
development (soil horizons) or in situ vegetation was observed.
The organic bed can be traced along the valley for 3 km at
an elevation of about 175 m asl. Directly overlying this bed
at sections SVMS-9 and SVMS-11 are buried logs, up to
50 cm in diameter, and sticks. All of the buried logs that
were observed were within 2 m of the organic bed, and most
were within 50 cm. A 10 cm diameter log from just above the
organic bed at section SVMS-9 yielded a radiocarbon age of
18 490 90 BP (Beta-38908), and a 50 cm diameter log from
-
-
-
+
+
130 BP (Beta-38907). A
section SVMS-11 dated 17 600
wood fragment ( 20 cm long, 1 cm diameter) found within
the organic bed at section SVMS-13 yielded a radiocarbon age
of 17 910 f 100 BP (Beta-40689).
-
Unit 6
Unit 6 is a compact, 20 m thick, matrix-supported diamicton displaying fissility and uncommon glaciotectonic structures (fractures and faults). Its contact with unit 5 is sharp. The
matrix ranges from clayey silt to fine sand and silt. Clasts
generally are subrounded; none showed obvious striae. Rare
-
LIAN AND HICKIN
H
i>~<o$iB~~.*>t~,
Gcm
SVMS- 1 1
9
-
SVMS 24
5
----
'
3--150-du
-
SVMS 21
Sh
Dmm Sm Fm
Sp Sr, Dmm(r)
GCm
w;rb>>it&!
A
0120=60
W5o+b(l
Dmm
Fld
L~thofaciesCode
Dmm
Dmm(r)
Dms
Dmg
Gcm
Gmm
Gcs
. Sh, Sp
ELEVATlON(m a+l)
MEASURED SECTION NUMBER
,
"
\
Matr~x-supportedmassive diam~cton
Dmm with ev~denceof resedimentation
Matnx-supportedstratified diamicton
Matr~x-supportedgraded diamicton
Clast-supported massive or crudely stratified gravel
Matrix-supportedrnasslve gravel
Clast-supported stratified gravel
SVMS-40
CLAY-SILT (lam~natedw ~ l hdropstones)
SHARP CONTACT
SmIFm
Sr
ShlSp
Massive sandlfines
Sand w~thr~pples
Hor~zontallvstratified sand/ Planar cross-stratified
GRADATIONAL
CONTACT
&SAND (horizontally
-
COVERED (NOT EXPOSED1~
-+GRAVEL
'
-HIGHLY
bedded)
ORGANIC SEDIMENTS (eg, peat)
.- . ..-.
.$SILT-SAND (laminated)
~ C. L- A Y - S I(lammated
L T with convoluted beds)
-=;.:&
::
'
34 320 ? 320 -RADIOCARBON
AGE
41 c7 (TL)-THERMOLUMINESCENCE AGE
FIG. 5. Representative stratigraphic sections (note the different vertical scales). See Fig. 2 for locations.
beds of medium to coarse sand occur within the diamicton.
At section SVMS- 11, two small rounded fragments of wood
(total mass -20 g), which were found in contact with each
other in a sand bed a few centimetres above the contact
with unit 5, yielded an anomalously old radiocarbon age of
>43 500 BP (Beta-38910).
Unit 7
Unit 7 generally is less than 2 m thick. It sharply overlies
unit 6 at sections SVMS-9 and SVMS-11. The unit was also
observed in roadcuts immediately north of Rice Lake and
along Seymour Mainline, although contacts with overlying
and underlying units are not exposed at these localities. Unit
7 consists of beds of stratified sand containing clasts of diamic-
ton, beds of massive pebble gravel with clasts of sand and silt,
and beds of fractured and faulted, laminated silt and clay.
Stratification locally is extremely convoluted, and some beds
are vertical. Plutonic and volcanic boulders (up to 1 m
diameter) are common.
-
Unit 8
Unit 8 is up to 12 m thick. Its contact with unit 7 was not
observed. Locally, the unit is unconformably overlain by fluvial gravel and sand (unit 10; section SVMS-24), or by lacustrine sand and silt (unit 12; section SVMS-21). Unit 8 consists
of laminated clay and silty clay with few or no dropstones and
no visible organics. At exposures where dropstones were
observed, the concentration of stones decreases upward.
846
CAN. J. EARTH SCI. VOL. 30, 1993
FIG.6. Photograph of unit 1 at section SVMS-7. Note the cyclical repetition of gravel and sand-silt. The exposure is about 2 m thick.
TABLE1. Summary of radiocarbon and thermoluminescence ages from Seymour Valley
Lab NO.^
Ageb
Material
Elevation (m asl)
Section
Beta-46052
Beta-40686
Beta-40690
Beta-38911
Beta-38912
Beta-40687
Beta-38907
Beta-40689
Beta-38908
Beta-38909
Beta-40686
Beta-46053
GSC-5069 HP
GSC-5 121 HP
Beta-38910
SVP 1
4 980k60
5 300+70
9 700 170
10120+60
10 350+60
11 420 f 110
17600+130
17 910f100
18 490f 90
22 040f 130
22 320 130
29 440+300
35 700+32OC
37 100+340d
>43 500
41 +7 kae
Stick
Stick
Charcoal fragments
Log
Log
Charcoal fragments
Log
Stick
Log
Log
Log
Peat
Log (Picea sp.)f
Log (Abies sp.)g
Wood fragments (2 pieces)
Mineral sediment in peat
113
113
166
175
175
179
171
172
176
142
141
101
100
99
178
101
SVMS-7
SVMS-7
SVMS-13
SVMS-21
SVMS-21
SVMS-25
SVMS-11
SVMS- 13
SVMS-9
SVMS-9
SVMS-8
SVMS-7
SVMS-6
SVMS-7
SVMS-11
SVMS-7
+
+
NOTE:All samples were collected by O.B. Lian.
aGSC, Geological Survey of Canada Radiocarbon Laboratory (errors f20); HP, high pressure; Beta, Beta
Analytic Inc. (errors fla).
b ~ l ages
l
in radiocarbon years BP, except SVPI, which is a thermoluminescence age.
' 6 " ~ = -24.7%.
d 6 1 3=
~ -23.3%.
eThermoluminescence age; analysis performed at the Thermoluminescence and Optical Dating Laboratory,
De artment of Physics, Simon Fraser University, Burnaby, B.C.; see Lian (1991) for experimental details.
by R.J. Mott (GSC Wood Report 90-41).
gIdentified by R.J.Mott (GSC Wood Report 90-74).
identified
Unit 9
Unit 9 comprises 1-2 m of massive stony clay, locally
interbedded with sand or pebble gravel; the clay is not compact. Stones within the clay are less than 1 cm in diameter. At
section SVMS-2b, unit 9 is overlain by horizontally and cross-
bedded sand and pebble gravel (unit 10); contact with underlying units was not observed. This unit has only been found
in the southern, lower part of the study area (limited exposures
at sections SVMS-2a (18- 19 m asl) and SVMS-2b (44-45 m
asl)).
847
LIAN AND HICKIN
FIG. 7. Close-up of gravel beds shown in Fig. 6. A woody peat bed (-20 cm thick) separates the two gravel beds. A large piece of wood
extracted from this peat bed yielded a radiocarbon age of 37 100 340 BP (GSC-5121 HP). The gravel beds are overlain and underlain by
sand and silt.
+
FIG. 8. Organic-rich bed (B) at section SVMS-11. This bed is overlain and underlain by glaciolacustrine sediments (A) correlative with
Vashon Drift. A large log (C), directly above the organic-rich bed, yielded a radiocarbon age of 17 600
130 BP (Beta-38907). The shovel
is 1 m long.
+
Unit 10
Unit 10 consists of up to 2 m of horizontally bedded and
crossbedded sand and gravel. Its contact with overlying and
underlying units was not observed. These deposits are found
in exposures south of Rice Lake Gate. At sections SVMS-29
and SVMS-30 the sediments consist of well-rounded and wellsorted, pebble gravel interbedded with sand and are associated
with terraces. The highest such terrace occurs at 187 m asl,
-
848
CAN. J. EARTH SCI. VOL. 30, 1993
about 100 m south of Rice Lake Gate.
(Armstrong and Clague 1977; Ryder and Clague 1989).
Unit 11
Unit 11 consists of up to 9 m of foreset-bedded sand and
gravel. Its contacts with underlying and overlying units were
not observed. These sediments are found at the mouth of the
valley and form raised deltas with beds dipping 10- 15" to the
south or southwest. The surfaces of the deltas are at -29 m
as1 at section SVMS-1 and at -49 m as1 at section SVMS-3.
After 29 ka to 22 ka
Unit 2 was deposited during this period. Based on limited
exposure, this unit appears to be composed of fluvial sand and
gravel at the top and lacustrine clayey silt grading upward into
lacustrine, or perhaps fluvial, sand at the base.
A lack of dropstones suggests that the lake in which the
lowermost sediments were deposited was dammed either by
glacial outwash that accumulated at the mouth of the valley or
by distal ice. The upward coarsening of the sediments in the
unit likely indicates deltaic progradation and infilling of the lake.
Bracketing radiocarbon ages indicate that unit 2 is correlative with Quadra Sand (Armstrong and Clague 1977; Clague
1977; Ryder and Clague 1989).
Unit 12
Unit 12 is -3 m thick and is exposed at section SVMS-21.
It consists of horizontally laminated, organic-rich, sand and
silt unconformably overlying unit 8. The two units are separated by a bed of cobble gravel. Unit 12 sediments occupy
what appears to be a channel-shaped feature.
Two logs at the base of unit 12 yielded radiocarbon ages of
10 120 f 60 BP (Beta-38911) and 10 350 ? 60 BP (Beta38912).
Unit 13
Unit 13 is up to -5 m thick and sharply overlies units 12,
7, or 8. It consists of matrix- and clast-supported diamicton
with angular to subangular clasts and horizontally bedded and
crossbedded gravel, locally interbedded with organic-rich pebbly sand. The sand beds, in places, contain lenses of charcoal
fragments. Imbrication was observed in some exposures, suggesting paleoflow roughly perpendicular to the main valley
axis. Unit 13 sediments form aprons along the sides of Seymour Valley and fans emanating from tributary valleys.
At Elsay Creek (section SVMS-25; Fig. 2) charcoal fragments -2 m below the fan surface yielded a radiocarbon age
of 11 420 f 110 BP (Beta-40687) (see Lian 1991 for a
description of this exposure).
Sedimentary history
Based on the litho- and chronostratigraphy presented above,
the Middle to Late Wisconsinan sedimentary history of lower
Seymour Valley can be reconstructed. All of the lithostratigraphic units defined here are correlated with units that
already have been established for the Fraser Lowland.
Before 37 ka to at least 29 ka
Unit 1 sediments were deposited during this period. This
unit consists of fluvial gravel and sand laid down by an aggrading Middle Wisconsinan Seymour River. The high degree of
sorting of the gravel facies indicates that it is unlikely that alluvial fans or debris flows contributed sediment to the immediate
area. The reason for aggradation at this time is unknown.
The nearly massive nature of the gravel beds suggests rapid
deposition. Sharp contacts between these beds and adjacent
sand beds indicate that there were periodic abrupt changes in
energy in the fluvial environment. Each organic bed appears
to represent a recovery of vegetation following an aggradational event.
The total time spanned by this unit suggests a low average
rate of valley aggradation, although each gravel bed probably
was deposited rapidly. After each gravel bed was deposited,
Seymour River may have shifted its course or incised its floodplain, allowing vegetation to become established on formerly
active channels.
Radiocarbon ages indicate that this unit was deposited during the Middle Wisconsinan and therefore correlates with the
Cowichan Head Formation of south-coastal British Columbia
-
About 22 ka to before 18.5 ka
Units 3 and 4 were deposited during this period. Unit 3 is
interpreted as glaciolacustrine and records a proglacial lake
that formed in Seymour Valley behind ice in the Fraser
Lowland during the Fraser Glaciation. This unit is correlative
with Coquitlam Drift (Hicock and Armstrong 1981; Ryder and
Clague 1989).
The diamicton of unit 4 is thought to be glacigenic. The massive diamicton at section SVMS-9 is interpreted to be basal
till, whereas the diamicton beds at section SVMS-11 are likely
flow till deposited in a water body. The massive clayey silt
beds may be mudflow deposits. The sand beds are interpreted
to be ice-contact and proglacial meltwater sediments; the
upward increase in their number at both exposures suggests,
perhaps, an increase in meltwater activity with time.
The lithology of the clasts in the diamicton (virtually all plutonic) implies that they were not transported far. Rather, it
appears that they may have been derived from local fluvial
deposits within Seymour Valley. A nonlocal ice source might
be expected to deposit drift of a more varied lithology. Unit 4
represents the first arrival of Fraser Glaciation ice in the lower
Seymour Valley, although it is not known if this unit was
deposited from Fraser Lowland ice or Seymour Valley ice.
Unit 4 also correlates with Coquitlam Drift (Hicock and
Armstrong 1981; Ryder and Clague 1989).
Before 18.5 ka to at least 17.5 ka
The presence of an organic bed directly below numerous
buried logs, near the base of unit 5, suggests that there was a
period when lower Seymour Valley, to at least 3 km north of
Rice Lake, became revegetated after the initial advance and
apparent retreat of Fraser Glaciation ice. Radiocarbon ages
indicate that a substantial forest existed about 18.5 ka.
The lack of in situ vegetation, such as roots, suggests that
this organic bed is detrital and was reworked from in situ
organic material (soil?). This reworked organic material was
deposited in a lake (represented by unit 5) that formed behind
ice readvancing in the Fraser Lowland.
This period of deglaciation and revegetation has been
termed the Port Moody interstade (Hicock and Armstrong
1985).
After 17.5 ka to before 9.5 ka
This period is represented by units 5-13. Unit 5 accumulated in a glacial lake. The presence of subangular volcanic
dropstones indicates that the ice that transported them originated outside the Seymour River basin and may have flowed
into the valley from the Fraser Lowland.
Unit 6 is thought to be till. It is bracketed by radiocarbon
849
HICKIN
+
ages of 17 600 f 130 BP (Beta-38907) and 11 420
110 BP
(Beta-40687); the latter age was obtained from charcoal fragments in a postglacial alluvial fan at section SVMS-25 (Lian
1991). The two wood fragments in unit 6 at section SVMS-11
which yielded an age of >43 500 BP likely are reworked
material.
Unit 7 is interpreted as ablation till (melt-out and flow till).
There are no radiocarbon ages from this unit, but at section
SVMS-13 charcoal fragments extracted from overlying
organic-rich mudflow sediments (unit 13) yielded an age of
9700 170 BP (Beta-40690), which is a minimum for unit 7.
Units 5 -7 are correlated with Vashon Drift, the climactic
deposit of the Fraser Glaciation (Hicock and Armstrong 1985;
Ryder and Clague 1989).
Unit 8 is interpreted as glaciolacustrine or possibly glaciomarine. An upward decrease in dropstones may suggest that the
sediments were deposited in association with a retreating ice
margin. Radiocarbon ages of 10 120 60 BP (Beta-3891 1) and
10 350 60 BP (Beta-38912) from wood in overlying organicrich, channel-fill sediments (unit 12) at section SVMS-21 indicate that unit 8 was deposited before 10.3 ka.
Unit 9 could be either glaciolacustrine or glaciomarine. However, its massive character, low density, stratigraphic position
(there are no overlying glacigenic units), and geographic position (at the mouth of the valley, below the late-glacial marine
limit in the Fraser Lowland) favour a glaciomarine origin.
Unit 9 may correlate with unit 8, but the two differ in character and geographic position, thus their exact relationship is not
known. No organic material has been found in unit 9 and,
therefore, no radiocarbon ages are available.
Unit 10 is either deltaic or beach gravel and sand (the interpretation is uncertain because of limited exposure). If these
sediments are deltaic, they may correlate with unit 11 (below).
No organic material for radiocarbon dating was found in these
sediments.
Unit 11 was deposited in marine deltas at a time of higher
sea level. It too lacks organic material for radiocarbon dating.
Units 8 - 11 are correlative with Capilano Sediments (Armstrong 1981, 1984; Ryder and Clague 1989).
Unit 12 overlies a cobble lag and is contained within a
channel-shaped depression. This, the presence of horizontal
bedding, and a lack of fluvial structures suggest that the sediments are lacustrine, or perhaps overbank, in origin. Unit 12
was deposited during postglacial time ( - 10.3 ka) and correlates with Salish Sediments (Armstrong 1981, 1984; Ryder
and Clague 1989).
Unit 13 is postglacial alluvial-fan, alluvial-apron, debrisflow and mudflow sediments and is also correlative with Salish
Sediments.
+
+
+
-
Discussion
Exposures of Quadra Sand, Vashon Drift, Capilano Sediments, and Salish Sediments are numerous in the Fraser
Lowland, and these units are well documented and understood. The nature and timing of these units in Seymour Valley
support what is known already and will not be discussed here.
In contrast, exposures of the Cowichan Head Formation and
Coquitlam Drift are much less common, and sedimentary and
organic evidence of the Port Moody interstade (Hicock and
Armstrong 1985), the nonglacial period separating the Coquitlam and Vashon stades of the Fraser Glaciation, has previously been found at only two sites. The relationship between
these units and correlative sediments in Seymour Valley is discussed below.
The Cowichan Head Formation comprises two members
(Armstrong and Clague 1977). The lower member consists of
stratified marine silt and clay overlain by sand deposited into
relatively shallow water during a fall in sea level, presumably
in response to isostatic rebound following the penultimate
(Semiahmoo) glaciation. The upper member consists of fluvial
and organic sediments. In Seymour Valley the lower marine
member is not exposed and may not exist; only the upper terrestrial member is present.
The nearest exposures of correlative sediments are in Lynn
Canyon in neighbouring Lynn Valley (Fig. 2), where they
consist of about 1 m of compressed peat containing a 2 cm
thick sand bed. Also exposed is about 25 cm of an underlying
till deposited during the penultimate glaciation (Hebda et al.
1983; Armstrong et al. 1985; Armstrong 1990). Four radiocarbon ages from the Lynn Valley peat range from 47 800 f
1100 BP (GSC-3290) to 33 100 f 620 BP (GSC-2797) (Armstrong et al. 1985; Armstrong 1990). It appears, therefore,
that in Lynn Valley the Cowichan Head Formation represents
more or less continuous accumulation of peat over a period of
about 15 ka, interrupted by only one small fluvial incursion.
This contrasts with the situation in Seymour Valley where
-2 m of peat and fluvial sediment were deposited over a
period of more than 8 ka. The different sedimentary character and overall thickness of the two exposures may be attributable to their respective proximity to ancestral Seymour River
and Lynn Creek. An alternative explanation is that significant
aggradation during the Middle Wisconsinan was localized to
Seymour Valley, although no evidence to support this has been
found.
Until this study was undertaken, Coquitlam Drift (Hicock
1976; Hicock and Armstrong 1981) had only been identified
at three sites in the Coquitlam - Port Moody area, and in
Chilliwack Valley, where proboscidean remains of Coquitlam
age have been found in glaciofluvial outwash (Hicock et al.
19826).
Coquitlam Drift at the Coquitlam Valley holostratotype
(Hicock and Armstrong 1981) is similar to unit 4 in Seymour
Valley. The former consists of "at least three lodgement tills
separated by glaciofluvial, ice-contact, and glaciomarine
deposits" (Hicock and Armstrong 1981, p. 1444). Hicock and
Armstrong (1981) suggested that by -21.5 ka glacial ice
blocked lower Coquitlam Valley, creating a lake in which
Quadra Sand was deposited.
The sections in both Coquitlam Valley and Seymour Valley
show that ice margins advanced and retreated into water during the Coquitlam Stade. In Coquitlam Valley, marine indicators suggest that the sea invaded the valley at least once during
this period. Since no microfossil or macrofossil studies were
performed on the Seymour Valley sediments, it is not known
whether unit 4 was deposited in a freshwater lake or the sea.
The character of the sediments at section SVMS-9 suggests,
however, that the mouth of the valley was blocked by ice during this time, which in turn suggests that the sediments at section SVMS-11 were deposited in fresh water.
In Coquitlam Valley both pre- and post-Coquitlam Quadra
Sand (Hicock and Armstrong 1981) is present. In Seymour
Valley, however, only pre-Coquitlam Quadra Sand has been
identified.
The Port Moody interstade is represented in Seymour
Valley by a single organic-rich bed underlying buried wood
-
CAN. J.
850
EARTH SCI.
(Fig. 6). Although these sediments and wood appear to b e
reworked, the organic-rich bed is found within 2 m of the
underlying Coquitlam Drift and therefore closely postdates it.
Similar sediments of the same age occur at Port Moody and
Mary Hill (Hicock e t al. 1982a).
Concluding remarks
Seymour Valley contains a n almost complete sedimentary
record spanning the period from the Middle Wisconsinan
through the last glaciation. All of the lithostratigraphic units
established for this part of the Fraser Lowland a r e present in
Seymour Valley.
Of special interest is the presence of the Cowichan Head
Formation and Coquitlam Drift and evidence of the Port
Moody interstade, which until now have been identified at two
sites. This study provides additional information that supports
what is already known about Late Wisconsinan glacial fluctuations in southwestern British Columbia and introduces an important site for future Quaternary paleoenvironmental research in
this region.
Acknowledgments
Assistance in the field was provided by M. Hafer. W e thank
J.J. Clague (Geological Survey of Canada) for facilitating
radiocarbon analysis of two wood samples at the Geological
Survey of Canada laboratory in Ottawa. D.J. Huntley (Department of Physics, Simon Fraser University) provided access to
his laboratory and useful advice for the thermoluminescence
experiment. W e also thank T. McComb (Greater Vancouver
Regional District) for allowing access to the Seymour Demonstration Forest. This paper has benefited from reviews by R.J.
Fulton and J.M. Ryder. J.J. Clague provided many suggestions which greatly improved this paper. This work forms part
of a n ongoing study of the geomorphic history of Coast Mountains river valleys, which is being conducted at Simon Fraser
University and is funded by the Natural Sciences and Engineering Research Council of Canada.
Armstrong, J.E. 1956. Surficial geology of Vancouver area, British
Columbia. Geological Survey of Canada, Paper 55-40.
Armstrong, J.E. 1981. Post-Vashon Wisconsin Glaciation, Fraser
Lowland, British Columbia. Geological Survey of Canada, Bulletin 322.
Armstrong, J.E. 1984. Environmental and engineering applications
of the surficial geology of the Fraser Lowland, British Columbia.
Geological Survey of Canada, Paper 83-22.
Armstrong, J.E. 1990. Vancouver geology. Geological Association
of Canada, Vancouver.
VOL. 30,
1993
Armstrong, J.E., and Clague, J.J. 1977. Two major Wisconsin
lithostratigraphic units in southwestern British Columbia. Canadian Journal of Earth Sciences, 14: 1471 - 1480.
Armstrong, J.E., and Hicock, S.R. 1980. Surficial geology, ~ a n c o u ver British Columbia. Geological Survey of Canada, Map 1486A.
Armstrong, J.E., Clague, J.J., and Hebda, R.J. 1985. QuaternaryFraser Lowland. In Field guides to geology and mineral deposits
in the southern Canadian Cordillera. Edited by D.J. TempelmanKluit. Geological Society of America, Cordilleran Section, Vancouver, pp. 15-9- 15-10.
Burwash, E.M.J. 1918. The geology of Vancouver and vicinity.
University of Chicago Press, Chicago.
Clague, J.J. 1977. Quadra Sand: a study of Pleistocene geology and
geomorphic history of coastal southwestern British Columbia.
Geological Survey of Canada, Paper 80- 13.
Hebda, R.J., Hicock, S.R., Miller, R.F., and Armstrong, J .E. 1983.
Paleoecology of mid-Wisconsin sediments from Lynn Canyon,
Fraser Lowland, British Columbia. Geological Association of
Canada - Mineralogical Association of Canada, Program with
Abstracts, 8: A31.
Hicock, S.R. 1976. Quaternary geology: Coquitlam - Port Moody
area, British Columbia. M.Sc. thesis, University of British Columbia, Vancouver.
Hicock, S.R., and Armstrong, J.E. 1981. Coquitlam Drift: a preVashon Fraser glacial formation in the Fraser Lowland, British
Columbia. Canadian Journal of Earth Sciences, 18: 1443 - 145 1 .
Hicock, S.R., and Armstrong, J.E. 1985. Vashon Drift: definition of
the formation in the Georgia Depression, southwest British Columbia. Canadian Journal of Earth Sciences, 22: 748-757.
Hicock, S.R., Hebda, R.J., and Armstrong, J.E. 1982a. Lag of the
Fraser glacial maximum in the Pacific Northwest: pollen and macrofossil evidence from western Fraser Lowland, British Columbia.
Canadian Journal of Earth Sciences, 19: 2288-2296.
Hicock, S.R., Hobson, K., and Armstrong, J.E. 19826. Late Pleistocence proboscideans and early Fraser glacial sedimentation in
eastern Fraser Lowland, British Columbia. Canadian Journal of
Earth Sciences, 19: 899-906.
Johnston, W.A. 1923. Geology of Fraser River delta map area. Geological Survey of Canada, Memoir 135.
Lian, O.B. 1991. The late Quaternary surficial geology and geomorphology of the lower Seymour Valley, North Vancouver, British
Columbia. M.Sc. thesis, Simon Fraser University, Burnaby.
Maynard, D.E. 1978. Geomorphic constraints to urban residential
development in the Seymour Area, District of North Vancouver,
B.C. M.Sc. thesis, University of British Columbia, Vancouver.
Ryder, J.M., and Clague, J.J. 1989. British Columbia (Quaternary
stratigraphy and history, Cordilleran Ice Sheet). In Quaternary
geology of Canada and Greenland. Edited by R.J. Fulton. Geological Survey of Canada, Geology of Canada, no. 1 , pp. 48-58.
(Also Geological Society of America, The Geology of North
America, vol. K-1.)
Wagner, F.J.E. 1959. Paleoecology of the marine Pleistocene faunas
of southwestern British Columbia. Geological Survey of Canada,
Bulletin 52.