Geological Society, London, Special Publications Fluvial response to Holocene glacier fluctuations in the Nostetuko River valley, southern Coast Mountains, British Columbia Kenna Wilkie and John J. Clague Geological Society, London, Special Publications 2009; v. 320; p. 199-218 doi:10.1144/SP320.13 Email alerting click here to receive free email alerts when new articles cite this service article Permission click here to seek permission to re-use all or part of this article request Subscribe click here to subscribe to Geological Society, London, Special Publications or the Lyell Collection Notes Downloaded by University of Alberta on 28 August 2009 © 2009 Geological Society of London Fluvial response to Holocene glacier fluctuations in the Nostetuko River valley, southern Coast Mountains, British Columbia KENNA WILKIE & JOHN J. CLAGUE* Department of Earth Sciences, Simon Fraser University, 8888 University Drive, Burnaby, British Columbia, Canada V5A 1S6 *Corresponding author (e-mail: [email protected]) Abstract: Mountain rivers, like alpine glaciers, are sensitive indicators of climate change. Some rivers may provide a more complete record of Holocene climate change than the glaciers in their headwaters. We illustrate these points by examining the record preserved in the upper part of the alluvial fill in the Nostetuko River valley in the southern Coast Mountains, British Columbia (Canada). Glacier advances in the upper part of the watershed triggered valley-wide aggradation and complex changes in river planform. Periods when glaciers were restricted in extent coincide with periods of incision of the valley fill and floodplain stability. As many as 10 overbank aggradation units are separated by peat layers containing tree roots and stems in growth position. Twenty-five radiocarbon ages on roots, tree stems and woody plant detritus in several of the peat layers closely delimit periods of aggradation. The oldest phase of aggradation occurred about 6500 years BP and coincides with the Garibaldi Advance documented elsewhere in the southern Coast Mountains. A second phase of aggradation, recorded by several units of clastic sediment, dates to about 2500 years BP, near the peak of the middle Neoglacial Tiedemann Advance. The third phase occurred shortly after 1400 years BP during or shortly after the First Millennium Advance, which has been recently documented in coastal British Columbia and Alaska. The most recent phase of aggradation began about 800 years BP and continued until recently. It coincides with the Little Ice Age, when glaciers in the Nostetuko River basin and elsewhere in the southern Coast Mountains attained their greatest Holocene size. Several periods of peat deposition during the Little Ice Age indicate periods of floodplain stability separated by brief inter- vals of floodplain aggradation that coincide with Little Ice Age glacier advances in western Canada. The results imply that the west fork of Nostetuko River is sensitive to upvalley glacier fluctuations and, indirectly, to relatively minor changes in climate. The proglacial fluvial archive is a largely unexp- subsequently radiocarbon dated. We show that loited source of information on upvalley glacier sediment supply is intimately linked to fluctuations fluctuations. Streams may respond to fluctuations of glaciers at the head of the valley. Radiocarbon of glaciers in their headwaters by aggrading up or ages on stumps at the tops of the peat layers incising their floodplains. The resulting changes closely constrain times of glacier advances within in local base level can be preserved in the valley- the watershed. These times agree with those deter- fill stratigraphy. Although potentially difficult to mined independently by other researchers working decipher, valley-fill stratigraphies may be more elsewhere in western North America. complete than the record of glacier fluctuations derived from landforms and sediments the forefields Study area themselves. At the very least, they complement and strengthen the glacier forefield evidence. The study area is the valley of the west fork of This paper documents the response of the west Nostetuko River in the southern Coast Mountains fork of the Nostetuko River valley, located in the of British Columbia, 220 km north of Vancouver southern Coast Mountains of British Columbia (Fig. 1). The west fork flows 11 km north and east (Canada), to changes in sediment supply during from its source to the main stem of Nostetuko Neoglaciation – the last half of the Holocene. We River (Fig. 2). It is fed mainly by meltwater from have two objectives: first, to add to the knowledge valley glaciers at the edge of Homathko Icefield. of Holocene glacier fluctuations in British A major tributary of the west fork flows from Columbia; and second, and more generally, to Queen Bess Lake, a moraine-dammed lake that demonstrate the potential of fluvial archives for partially drained during an outburst flood in deciphering past alpine glacier activity. Field August 1997 (Kershaw 2002; Kershaw et al. 2004). inspection of the upper part of the sediment fill Queen Bess Lake is impounded by a large com- revealed a series of clastic sediment units interstra- posite moraine produced by at least two advances of tified with peats containing rooted stumps that were Diadem Glacier (Kershaw 2002). The lake formed From:KNIGHT,J.&HARRISON, S. (eds) Periglacial and Paraglacial Processes and Environments. The Geological Society, London, Special Publications, 320, 199–218. DOI: 10.1144/SP320.13 0305-8719/09/$15.00 # The Geological Society Publishing House 2009. 200 K. WILKIE & J. J. CLAGUE Fig. 1. Location of the study area in the southern Coast Mountains of British Columbia (modified from BMGS data; reproduced with permission of the Province of British Columbia). behind the composite moraine during glacier retreat of sections. Fieldwork was conducted during the in the late 1800s and early 1900s. In 1997, a large ice summer of 2004. Detailed topographic maps avalanche fell from the toe of Diadem Glacier into (1:5000 scale), constructed from aerial photographs Queen Bess Lake, generating displacement waves flown in 1998 – one year after the outburst flood – that overtopped and incised the moraine. The result- were used to map deposits and landforms. Locations ing flood eroded sediments in the valley below the of sections, terraces, trimmed colluvial fans and tree dam, causing aggradation upstream and down- stumps exhumed by river incision were located stream of channel constrictions (Fig. 2). It also (+10 m) using a hand-held GPS unit and cross- created exposures of the upper part of the valley referenced with the topographic maps. These data fill, which enabled this study. were subsequently entered into a Geographic Information System (GIS). Methods Detailed sedimentological and stratigraphic logs were made of exposed valley-fill sediments at seven The aggradation history of the west fork of Noste- sites (Fig. 2), and additional observations of sedi- tuko River was determined through stratigraphic, ments and landforms were made at many other sedimentological and geochronological analyses locations. Sections were logged using a metric FLUVIAL RESPONSE TO GLACIER FLUCTUATION 201 sand paper. Annual tree-ring widths were measured to the nearest 0.001 mm along up to four radii for each tree sample using a Velmex-type measuring stage, a Leitz stereomicroscope and the Measure J2X measuring program. Samples were cross-dated to establish floating chronologies by visually com- paring marker rings and by employing the statistical correlation and verification procedures within the ITRDBL (International Tree-Ring Data Bank Library) tree-ring dating software program COFECHA (Holmes 1999; Grissino-Mayer 2001). Segments that were not significantly correlated were re-measured and corrected to account for radial growth anomalies and missing or false rings. The age of the oldest living tree on a surface provides a minimum age for that surface, after cor- rections for local ecesis and sampling height have been applied (McCarthy et al. 1991; Wiles et al. 1999). Ecesis, defined as the time between surface stabilization and germination of the first seedling, has been shown to range from 1 to 100 years in the Pacific Northwest (Sigafoos & Hendricks 1969; Desloges & Ryder 1990; McCarthy et al. 1991; Smith et al. 1995; Wiles et al. 1999; Luckman 2000; Lewis & Smith 2004). Ecesis inter- vals of 1–4 years have been documented in the Coast Mountains at Tiedemann Glacier (Larocque & Smith 2003), and on Vancouver Island at Colonel Foster and Septimus glaciers (Lewis & Smith 2004). Seedlings growing on the floodplain scoured by the 1997 Queen Bess outburst flood were no more than 5 years old when we conducted fieldwork in 2004, suggesting that ecesis in the Fig. 2. Aerial photomosaic of the valley of the west fork west fork valley is 1–2 years. Two years were there- of Nostetuko River, showing locations of studied fore added to the outer ring ages of in situ stumps sections. The aerial photographs were flown in 1998, to correct for ecesis. 1 year after the outburst flood. Sampling height errors occur when annual growth rings are lost due to sampling above the root crown (McCarthy et al. 1991). Larocque & tape and a barometric altimeter (elevation accuracy Smith (2003) proposed a regional correction factor of +5 m). Recorded data included grain size of 1.35 cm year21 for subalpine fir seedlings on (field estimates), sorting, sedimentary structures, valley floors in the Mount Waddington area. This Munsell colour, unit thickness, the nature of unit correction was applied to all in situ stumps. contacts and fossil plant remains. Sampling height corrections cannot be applied to Samples of in situ tree stems, fossil roots and detrital logs. detrital plant fossils were collected from peat layers and rooting horizons, and submitted to Beta Analytic for conventional (radiometric) 14C analy- Results sis. Radiocarbon ages were calibrated using the Geomorphology software OxCal v. 4.0 (Bronk Ramsey 1995, 2001), which is based on the decadal data of Stuiver et al. The west fork of Nostetuko River flows through (1998).