A 750-YEAR VARVE RECORD FROM THE MARGIN OF THE DEVON ISLAND ICE CAP, ARCTIC ARCHIPELAGO, CANADA
LAMOUREUX, SCOTT F. Department of Geography Queen's University.
Gilbert, Robert . Department of Geography Queen's University.
Lewis, Ted . Department of Geosciences University of Massachusetts Amherst.
In recent years, a there has been increasing interest in the use of arctic varved lake records (e.g. Lamoureux, 2000, Hughen et al., 2000) to document the long term climate variability from this sensitive region. Together with other high-resolution proxy records (including ice cores and tree rings on the mainland), these records have shown the first detailed indication of recent environmental and climate change from the region (Overpeck et al., 1997). Despite these efforts, relatively few records are available and comparing the details of records from widely separated sites can be problematic, particularly in detail. To address this issue, we initiated research at Bear Lake, Devon Island (75°28’N, 85°10’W) to develop a varve record to compare with the ice core record obtained from the Devon Island Ice Cap (Alt et al., 1985; Koerner and Fisher, 1990). In addition to collecting sediment cores from the lake, a melt season of detailed sediment trapping and watershed process work was carried out in 1999 to identify the processes that control sedimentation in the lake (Lewis et al., 2002 in press).
Bear Lake is a deep (maximum depth 103 m) basin that receives inflow primarily from a river draining c. 20 km of the northwest margin of the Devon Island Ice Cap. Discharge begins in June with snowmelt and reaches a maximum on warm days during July when diurnal fluctuations become pronounced. The river carries substantial fine-grained, carbonate-rich sediment from the ice marginal area. Process work has demonstrated the presence of underflows associated with many high discharge events during July, although there is seemingly not a direct relationship with underflow activity and daily temperatures (Lewis et al., 2002 in press). Lake ice persists until at least mid-August and inhibits thermal stratification in the lake. Discharge decreases in August concomitant with cooler temperatures and reduced ice melt. Lake ice formation is delayed by persistent, strong winds, but probably sets in September, depending on storminess and ambient temperatures.
Sediments from the proximal basin appear as clear couplets of coarser, carbonate-rich silt grading normally into darker clay. The sediments were studied in detail using petrographic thin sections and the rhythmites were counted and measured on three separate occasions using 600 dpi scans. Given the strong seasonality of the sediment delivery to Bear Lake and the consistency of the sedimentary structures, we interpret these couplets to be varves. A profile of 210Pb activity did not provide a consistent means for independently dating the couplets, due to low levels of 210Pb and highly variable sedimentation rates (inferred from couplet thickness). Therefore, although we cannot independently verify the presence of varves, the primary evidence (sedimentology) gives us confidence that the proximal sediments in Bear Lake are varved.
Cores retrieved from two sites in the basin (Cores 18 and 19) show similar sedimentary structures and varve sequences which can be correlated on the basis of structure and thickness changes. Average thickness of the varves is 4.6 mm but is highly variable. Relatively thin varves (<3 mm) are common and contain silt and clay in nearly equal proportions. Thicker varves contain graded units of carbonate-rich silt and in some cases, exceed 50 mm thickness. Compared to the more proximal site (core 18, 100 m depth), core 19 (88 m depth) contains fewer thick varves. Additionally, the distal core contains sections of poorly laminated to massive clay zones varying from a 3-20 millimetres thick. Cross-dating between the two cores reveals that the poorly laminated/massive sections represent years with consistently low accumulation rates in the proximal core and effectively represent short gaps in the distal core, lasting 3-25 years. The gaps can be bridged with the continuous record from the proximal core, allowing cross-dating both cores to 1308 AD. The proximal core continues to 1250 AD and terminates at the core bottom.
Based on observations of underflows during the summer of 1999 (Lewis et al., 2002 in press), the thick carbonate-rich silt units found in the cores are likely produced by the same mechanism and represent sporadic deposition from turbid underflows associated with meltwater production from the ice cap. Although we cannot quantitatively constrain the meteorological conditions that produce these events (to say a daily temperature), they are clearly a product of high meltwater production and suspended sediment entrainment. We can further constrain the intensity of these underflows by their presence or absence in the distal core. Periods when sedimentation was dominated by homopycnal flow produce thin proximal varves and vaguely layered to massive sediments in the distal areas. The presence or absence of underflow deposits in the varves indicates periods in the record without weather conditions suitable for generating underflows, and probably represent cooler years. The most persistent underflow deposition occurred from 1900-1970 and to a lesser extent, during the 1300s and late 1700s. Underflows were consistently infrequent during the 1800s and early 1400s and 1700s. Overall, the 1920-1970s represent the highest sedimentation rates of the entire 750-year record and suggests that deposition has decreased substantially in the late 20th century. Comparison with the melt record and d18O record from the Devon ice core shows many similarities between the two records, particularly during the 20th century.
Comparison of varve thickness with meteorological records from Pond Inlet and Resolute shows a consistent maximum correlation with mean September temperatures (Pond Inlet 1977-88, r2= 0.445, p<0.07, n=12; Resolute 1951-80, r2= 0.237, p<0.10, n=30). Strangely, the maximum correlation between the meteorological record and the Bear Lake varve thickness was during the last two weeks of September (r2=0.530, p< 0.04, n=12). These results suggest that the period of lake ice formation is the most important control over sedimentation rates and we suspect that wave-induced erosion and slumping from the delta front during the open-water period could generate a significant proportion of the interannual variance in recent sedimentation rates. By contrast, the early Pond Inlet weather record (1923-1960 with frequent gaps) shows a modest correlation with July temperatures (r2=0.155, n=30) and essentially no correlation with September temperatures (r2= 0.050, n=30). This latter period covers the highest sedimentation rates of the last 750 years and may indicate an increased importance of melt-generated underflow activity. Regardless of the specific mechanism involved, both indicate that sedimentation is positively correlated with some aspect of summer temperature and hence, the overall correlation to the ice core record.
The results from the Bear Lake varve record underscore the importance of process work for interpreting the hydroclimatic record contained in the sediments, and demonstrate that the linkages between sedimentation and hydroclimatic processes may change with time. While one or several years of this labour-intensive work may be all that is tractable, it is clear that, like the meteorological record, a short-term view of environmental processes can generate misconceptions regarding the linkages between sedimentation processes and catchment controls.
Alt, B.T., Koerner, R.M., Fisher, D.A. and Bourgeois, J.C., 1985. Arctic climate during the Franklin era, as deduced from ice cores. In: P.D. Sutherland (Editor), The Franklin Era in Canadian Arctic History 1845-1859. Archaeological Survey of Canada Paper no. 131, pp. 69-92.
Hughen, K.A., Overpeck, J.T. and Anderson, R.F., 2000. Recent warming in a 500-year palaeotemperature record from varved sediments, Upper Soper Lake, Baffin Island, Canada. The Holocene, 10: 9-19.
Koerner, R.M. and Fisher, D.A., 1990. A record of Holocene summer climate from a Canadian high-Arctic ice core. Nature, 343: 630-631.
Lamoureux, S., 2000. Five centuries of interannual sediment yield and rainfall-induced erosion in the Canadian High Arctic recorded in lacustrine varves. Water Resources Research, 36(1): 309-318.
Lewis, T., R. Gilbert and S.F. Lamoureux, 2002 in press. Spatial and temporal changes in sedimentary processes at proglacial Bear Lake, Devon Island, Nunavut. Arctic, Antarctic and Alpine Research.
Overpeck, J., K. Hughen, D. Hardy, R. Bradley, R. Case, M. Douglas, B. Finney, K. Gajewski, G. Jacoby, A. Jennings, S. Lamoureux, G. MacDonald, J. Moore, M. Retelle, S. Smith, A. Wolfe and G. Zielinski, 1997. Arctic environmental change of the last four centuries, Science, 278: 1253-1256.
Previous | Abstract Index | Next
Copyright © 2001 INSTAAR, Univ. of Colorado