Paleoceanography and Paleoclimatology [PP]

PP41A  MS:Exh Hall B   Thursday
The Past Two Millennia in the Arctic: A Long-Term Context for Present-Day Changes I Posters
Presiding: Y Axford, University at Buffalo; D Kaufman, Northern Arizona University

PP41A-0177 

Late Holocene environmental change at three glacier-fed lakes, southern Alaska

* Kaufman, D S (Darrell.Kaufman@nau.edu), Northern Arizona Univ, Dept of Geology, Flagstaff, AZ 86011, United States * Kaufman, D S (Darrell.Kaufman@nau.edu), Northern Arizona Univ, Environmental Sciences, Flagstaff, AZ 86011, United States Anderson, R S (scott.anderson@nau.edu), Northern Arizona Univ, Environmental Sciences, Flagstaff, AZ 86011, United States Daigle, T A (tad58@NAU.EDU), Northern Arizona Univ, Dept of Geology, Flagstaff, AZ 86011, United States Kathan, K M (kmk238@nau.edu), Northern Arizona Univ, Dept of Geology, Flagstaff, AZ 86011, United States McKay, N P (npm4@nau.edu), Northern Arizona Univ, Dept of Geology, Flagstaff, AZ 86011, United States Michelutti, N N (michelut@biology.queensu.ca), Queen's Univ, Dept of Biology, Kingston, ONT K7L 3N6, Canada Werner, A (awerner@mtholyoke.edu), Dept of Earth and Environment, Mount Holyoke College, South Hadley, MA 01075, United States

Lake-sediment cores and glacial geomorphology were used to infer late Holocene paleoenvironmental changes at three glacier-fed lakes across southern Alaska. The lakes form a 730-km-long transect around 60N lat, and they span the transition zone between two centers of opposite surface air-temperature responses attributed to fluctuations in the strength of the Aleutian Low, the primary indicator of winter climate in the North Pacific. Sediment cores from Hallet Lake in the NE Chugach Range display varying concentrations of biogenic silica (BSi), a measure of overall lake production. A transfer function was developed to infer summer temperature from downcore BSi content. The reconstruction shows clear evidence of first millennium AD cooling, warmth from 1300-1500 AD, Little Ice Age (LIA) cooling between 1750 and 1900 AD, and recent warming beginning ca. 1900 AD. During the last 30 yr, summer temperatures were nearly 2C warmer than the reconstructed mean of the past 2 millennia. Goat Lake is near treeline in the Kenai Mountains, and about 1 km from an outlet glacier of the Harding Icefield. Pollen assemblages show increasing abundances of mountain hemlock from 700-1200 AD, which we interpret as an expansion of treeline. The expansion was terminated around 1230 AD when 10 cm of tephra was deposited in the lake. Treeline above the modern and prior to the LIA is further indicated by a 14C age of 1470 ± 85 AD on logs exposed below till at the present glacier terminus. By 1660 AD the outlet glacier thickened by 150 m where it overtopped its drainage divide and spilled meltwater into Goat Lake, which continued until around 1890 AD. Since then, hemlock pollen has increased to levels comparable to the 1200 AD peak, and the outlet glacier has retreated 1.4 km to the location of the 1470 AD logs. At Cascade Lake, sediment traps installed for 2 yr collected 77% less BSi when spring and summer temperatures were lower, suggesting that BSi flux in the lake is related to growing-season conditions. BSi was at its minimum early during the first millennium AD. It peaked around 700 AD, then decreased during the next 400 yr. BSi flux was relatively constant until the 19th century when it decreased to near-minima values, then attained its highest values of the last 2000 yr late during the 20th century. BSi and hemlock pollen are probably related more strongly to summer conditions than to winter, whereas glaciers respond to a combination of winter and summer climate variability. Late Holocene moraines in the forefields of cirque glaciers around all study lakes were mapped and dated roughly with lichenometry. The moraines delimit maximum glacier positions attained late in the 19th century, when glacier snouts generally descended less than 100 m in elevation relative to their 1950-1970 positions. This limited LIA expansion, together with tree-ring and other independent evidence for decades-long LIA summer cooling of at least 0.8çC in south-central Alaska, indicates a reduction in accumulation-season precipitation during the LIA. A simultaneous reduction in winter precipitation across southern Alaska is difficult to ascribe to a shift in the Aleutian Low pressure system because instrumental data show dipolar responses across this region. This implies a longer- term, more general climate forcing that supersedes inter-decadal variability in the Aleutian Low.

PP41A-0178 

Reconstruction of Centennial and Millennial-scale Climate and Environmental Variability during the Holocene in the Central Canadian Arctic

* Rolland, N (rolland.4@osu.edu), The Ohio State University Nicolas Rolland, David Porinchu, Department of Geography 1036 Derby Hall 154 North Oval Mall, Columbus, OH 43210, United States Porinchu, D (porinchu.1@osu.edu), The Ohio State University Nicolas Rolland, David Porinchu, Department of Geography 1036 Derby Hall 154 North Oval Mall, Columbus, OH 43210, United States MacDonald, G (macdonal@geog.ucla.edu), UCLA Glen M. MacDonald, Department of Geography 1125 Bunche Hall, Los Angeles, CA 90095, United States Moser, K (kmoser@uwo.ca), University of Western Ontario Katrina A. Moser, Department of Geography Social Science Building 1151 Richmond Street North, London, ON N6A 5C2, Canada

The Arctic and sub-Arctic regions are experiencing dramatic changes in surface temperature, sea-ice extent, glacial melt, river discharge, soil carbon storage and snow cover. According to the IPCC high latitude regions are expected to warm between 4oC and 7oC over the next 100 years. The magnitude of warming and the rate at which it occurs will dwarf any previous warming episodes experienced by latitude regions over the last 11,000 years. It is critical that we improve our understanding of how the Arctic and sub-Arctic regions responded to past periods of warming, especially in light of the changes these regions will be experiencing over the next 100 years. One of the lines of evidence increasingly utilized in multi-proxy paleolimnological research is the Chironomidae (Insecta: Diptera). Also known as non-biting midge flies, chironomids are ubiquitous, frequently the most abundant insects found in freshwater ecosystems and very sensitive to environmental conditions. This research uses Chironomidae to quantitatively characterize climate and environmental conditions of the continental interior of Arctic Canada during the Holocene. Spanning four major vegetation zones (boreal forest, forest-tundra, birch tundra and herb tundra), the surface samples of 80 lakes recovered from the central Canadian Arctic were used to assess the relationship of 22 environmental variables with the chironomid distribution. Redundancy analysis (RDA) identified four variables, total Kjeldahl nitrogen (TKN), pH, summer surface water temperature (SSWT) and depth, which best explain the variance in the distribution of chironomids within these ecoregions. In order to provide new quantitative estimates of SSWT, a 1-component weighted average partial least square (WA-PLS) model was developed (r2jack = 0.76, RMSEP = 1.42oC) and applied downcore in two low arctic continental Nunavut lakes located approximately 50 km and 200 km north of modern treeline. This robust midge-inferred temperature reconstruction of the Holocene thermal conditions will then be compared with previous research describing vegetation development in this region. This study provides new and important data which helps to further resolve millennial and centennial-scale climate variability in the central Canadian Arctic during the Holocene.

PP41A-0179 

Deterministic hydrological model to reconstruct the last 400 years of runoff and climate variations using the properties of the varved lake sediments in the Canadian Arctic

* Ciobanas, N (nicoleta.ciobanas@ete.inrs.ca), Institut national de la recherche scientifique, Eau, Terre et Environnement, 490, rue de la couronne, Québec, Qc G1K 9A9, Canada * Ciobanas, N (nicoleta.ciobanas@ete.inrs.ca), GEOTOP-UQAM-McGill, C.P. 8888, Succ. Centre-Ville, Montreal, QC H3C 3P8, Canada Francus, P (pfrancus@ete.inrs.ca), Institut national de la recherche scientifique, Eau, Terre et Environnement, 490, rue de la couronne, Québec, Qc G1K 9A9, Canada Francus, P (pfrancus@ete.inrs.ca), GEOTOP-UQAM-McGill, C.P. 8888, Succ. Centre-Ville, Montreal, QC H3C 3P8, Canada

The aim of this study is to develop a deterministic hydrological model in order to reconstruct the last 400 years of runoff using the properties of varved sediments from Sawtooth Lake-Ellesmere Island. Varved sediments can provide well-constrained records of interannual-to centennial-scale variation of climatic and hydrologic cycles, such as the North Atlantic Oscillation and Arctic Oscillation. Using image analysis technique of microscopic views, we can determine some properties of varved sediments, including the grain size of each year of sedimentation. Grain size and sediment input are mainly controlled by complex hydroclimatic processes within the watershed and such a direct correlation is not obvious. In order to better understand the relation between meteorological data and the waterflow to the lake, we developed a hydrological model. This deterministic model is based on a simple water balance equation at the scale of the watershed and uses the following parameters: the runoff, the evapotranspiration, the snowmelting and the soil's humidity storage. To estimate the evapotranspiration, Penman-Monteith's model was used. The snow melting was determined using the concept of "degree day". The model forced by climatic data successfully reproduced the flow measurements obtained during 1998. Moreover, the predicted runoff correlates well with the varves properties for the last 50 years. Hence, we used our model to reconstruct the climate of the last 400 years using the varved sediment. Our reconstruction compares well with the decadal and multi-decadal variations of the North Atlantic Oscillation and Arctic Oscillation.

PP41A-0180 

4500 Years of Annual Sediment Accumulation Recorded in Lower Murray Lake in the Canadian High Arctic

* Cook, T L (tcook@geo.umass.edu), University of Massachusetts, Department of Geosciences, 611 North Pleasant Street, Amherst, MA 01003, United States Bradley, R S (rbradley@geo.umass.edu), University of Massachusetts, Department of Geosciences, 611 North Pleasant Street, Amherst, MA 01003, United States Stoner, J S (jstoner@coas.oregonstate.edu), College of Ocean and Atmospheric Sciences, Oregon State University, 104 COAS Administration Building, Corvallis, OR 97331, United States

A varved sediment record spanning the last 4500 years was recovered from Lower Murray Lake, northern Ellesmere Island, Nunavut, Canada (81°20'N, 69°30'W). The annually laminated sequence covers the upper 2.3m of a 12m core; similar patterns of sedimentation are not observed elsewhere in the record. The lower section of the record is characterized by irregular laminations and greater variability in grain size. Additionally, the transition at 2.3m is associated with a distinct shift in sediment geochemistry as observed in ITRAX scanning XRF results. Flatbead scanned images of sediment thin sections from the upper, laminated portion of the record were used to establish the varve chronology and measure lamination thicknesses for the past 4500 years. The annual nature of sedimentation in Lower Murray Lake is supported by the occurrence of additional laminations after a second set of cores was collected following a 5 year interval; comparison of the varve chronology with 137Cs and 210Pb profiles and comparison of the paleomagnetic characteristics of the Murray Lake sediments with those from other records that have been independently dated. The pattern of sedimentation in Lower Murray Lake is characterized by centennial scale variability in lamination thickness superimposed on a millennial scale trend of increasing lamination thickness from 4500 years BP to present. Three distinct periods, corresponding to the Little Ice Age (~150 to 300 years BP), ~2550 to 2900 years BP, and ~3750 to 4450 years BP, are characterized by below normal varve thicknesses and fewer extreme sedimentation events (i.e. anomalously thick beds of coarse grained sediment). Preliminary results also suggest that higher mean sediment accumulation observed over the last 30 years is unequaled during any thirty year period in the preceding 3500 years, and is only surpassed during two intervals (centered around ~3550 and 3700 yrs BP) in the entire 4500 year record.

PP41A-0181 

Late Holocene Climate Change Inferred From Varved Proglacial Lake Sediments on Northeastern Baffin Island, Arctic Canada

* Thomas, E K (ekthomas@buffalo.edu), Department of Geology, University at Buffalo 876 NSC, Buffalo, NY 14260, United States Briner, J P (jbriner@buffalo.edu), Department of Geology, University at Buffalo 876 NSC, Buffalo, NY 14260, United States Axford, Y (yaxford@buffalo.edu), Department of Geology, University at Buffalo 876 NSC, Buffalo, NY 14260, United States Axford, Y (yaxford@buffalo.edu), Institute of Earth Sciences and Department of Geosciences, University of Iceland Askja, Reykjavik, NY 101, Iceland

The Arctic has a disproportionately large response to changes in radiative forcing of climate, and glaciers and arctic lacustrine ecosystems respond sensitively to these changes. Lacustrine ecosystems throughout the Arctic are undergoing rapid regime shifts, including dramatically increased primary productivity and changing aquatic floral and faunal assemblages. Our work on organic lake sediments from northeast Baffin Island shows a large increase in primary productivity, changes in insect (Chironomidae) assemblages including the disappearance of cold stenotherms, and a rise in chironomid-inferred summer water temperatures of at least 1.5°C over the past 50 years, reaching temperatures that were unprecedented in the past 5000 years. Here, we pursue the use of varve thickness, an abiotic temperature proxy, to expand our understanding of late Holocene temperature changes on northeast Baffin Island. We obtained a 14C- and 239+240Pu-dated surface core/percussion core pair from a proglacial lake. Together these cores span > 8000 years and the sediments are varved, as verified by the 239+240Pu analysis, for at least the past 700 years. Magnetic susceptibility was high during the early Holocene, decreased to near-zero values during the mid-Holocene and increased during the past 2500 years to reach the highest values seen in the record around 1000 years ago. Loss-on- ignition had an opposite trend, with the highest values in the mid-Holocene. Sedimentation rate was constant during most of the Holocene (0.03 cm yr -1) and increased during the past 1000 years to 0.05 cm yr -1. These parameters indicate that following the absence of an active glacier during the middle Holocene, glacier activity initiated ~2500 years ago and reached peak activity over the last 1000 years. Our ongoing work to obtain a varve-thickness record for at least the last 700 years, and its calibration to a nearby weather station, will be presented.

PP41A-0182 

Climate in North Iceland over the Past 2000 Years: Inferences from Midges and Other Lake- Sediment Proxies

* Axford, Y (yaxford@buffalo.edu), Institute of Earth Sciences and Dept. of Geosciences, University of Iceland, Askja, Reykjavik, 101, Iceland * Axford, Y (yaxford@buffalo.edu), Dept. of Geology, University at Buffalo, 876 NSC, Buffalo, NY 14260, United States Geirsdottir, A (age@hi.is), Institute of Earth Sciences and Dept. of Geosciences, University of Iceland, Askja, Reykjavik, 101, Iceland Miller, G H (gmiller@colorado.edu), INSTAAR and Dept. of Geological Sciences, University of Colorado, UCB 450, Boulder, CO 80309, United States Langdon, P G (P.G.Langdon@soton.ac.uk), Dept. of Geography, University of Southampton, Highfield, Southampton, SO17 1BJ, United Kingdom

Sediment cores from two lakes in north Iceland record environmental changes over the past 2000 years. Shifts in chironomid (non-biting midge) assemblages over the last two centuries at both lakes are coherent with instrumental temperatures recorded at Stykkisholmur in west Iceland. At Torfadalsvatn, which is small and shallow (Zmax=5 m), quantitative temperature inferences based upon an existing transfer function for Iceland compare well with Stykkisholmur temperatures. In contrast, at Stora Vidarvatn (Zmax=48 m) sample scores from detrended correspondence analysis are much better correlated with Stykkisholmur temperatures than are transfer-function based temperature inferences. Paleotemperature inferences from Stora Vidarvatn and Torfadalsvatn sediments bolster prior inferences from Icelandic historical records and other proxy-based climate reconstructions (e.g., Geirsdottir et al., this volume). Reconstructed temperatures during parts of the 10th and 11th centuries AD, the period of rapid Norse expansion in the region, were comparable to temperatures in warm decades of the 20th century. Declining biogenic silica concentrations, increasing C:N, and shifts in midge assemblages between the 13th and 19th centuries suggest declining summer temperatures, decreasing aquatic production, and increasing soil erosion. The most severe climatic conditions occurred in the 19th century, coincident with documented sea ice expansion around Iceland. Iceland sits at the present-day boundary (Polar Front) between Arctic and Atlantic surface currents in the North Atlantic, and Iceland's climate is tightly coupled to sea-surface conditions. The coherency and amplitude of cooling in the 13th through 19th centuries, along with the lag between the peak cold anomaly in north Iceland and presumed extremes in solar and volcanic forcings, may imply changes in surface ocean circulation that would have affected the larger northern North Atlantic region.

PP41A-0183 

Ecological Change in Arctic Lakes: the Limitation of the Palaeolimnological Record

* Anderson, N J (gynja@lboro.ac.uk), Loughborough University, Department of Geography, Loughborough, LE11 3TU, United Kingdom

The evidence for ecological change in arctic lakes is now unambiguous and is considered to have started about 1850 and to be driven by increases in air temperature. In the context of longer-term, regional changes, however, this summary overlooks a three important factors: dating control of the recent sediments can be problematical, spatial heterogeneity of lake responses at the landscape scale is often ignored and finally, other possible drivers exhibit similar trends to that of air temperature, most notably atmospheric pollutants. 210-Pb together with 14-C are widely used to date lake sediment records but are often problematical because of the low sedimentation rates. In particular, in the arctic 210-Pb provides a chronology covering less than 100 years, often with large uncertainties. Importantly, biological change often predates reliable 210-Pb data. Arctic lake districts often contain thousands of lakes, possibly with considerable among lake variability in terms of their response to external drivers. The density of palaeolimnological sampling is often too low to support regional upscaling. Finally, environmental change in the arctic is driven by more than just increases in air temperature. Arctic environments are far from pristine and have been subjected to long-range atmospheric pollution (e.g. Pb, Hg and N) for long- periods: the effects of these other drivers are often ignored, possibly because they are difficult to separate from temperature trends. Deriving independent palaeo-proxies for the drivers of environmental change is problematical because of poor understanding of the controlling biogeochemical and ecological processes. The implications of these factors for understanding ecological change over the last 2000 years are discussed using examples from lakes in SW Greenland, an area with over 20,000 lakes and a range of water chemistries.

PP41A-0184 

A 400-year ground surface temperature history inferred from three temperature depth profiles in Northern Quebec

Richard, F (Richard.Fortier@ggl.ulaval.ca), Dept of Geology, Laval University, Pavillon Adrien-Pouliot, Quebec, QC G1K7P4, Canada * Chouinard, C (cchouin@olympus.geotop.uqam.ca), GEOTOP-UQAM-McGill, University of Quebec at Montreal, POB 8888, sta. "downtown", Montreal, QC H3C3P8, Canada Mareschal, J (mareschal.jean-claude@uqam.ca), GEOTOP-UQAM-McGill, University of Quebec at Montreal, POB 8888, sta. "downtown", Montreal, QC H3C3P8, Canada

The inversion of three temperature profiles measured in permafrost near the Raglan mine in Northern Quebec, Canada, was carried out to infer the ground surface temperature history (GSTH). The site is located in the Katinniq Plateau, a barren rock desert, at the northern tip of the Ungava Peninsula. The 400-500 m exploration boreholes were logged in September 2005 and 2006, more than three years after drilling was completed. Repeat measurements were made in one the boreholes to verify that the boreholes had returned to thermal equilibrium after the drilling perturbation. Thermal conductivity measurements were made on core samples. Radiogenic heat production is small and can be neglected. Marked deviations from steady state affect the temperature profiles in the uppermost 200 m with inverted gradients above 80 m. These deviations are assumed to be caused by recent (< 300 years) variations in ground surface temperature. Three methods were used to infer the GSTH: 1) forward modeling where an assumed GSTH model developed from proxy data is used to calculate the temperature anomalies at depth which are then compared with the measured ones, 2) a standard inversion using an algorithm based on singular value decomposition, and 3) a Monte Carlo inversion. The GSTH models show a marked warming (approx. 1.4K) between the mid-1700s and the early 1900s associated with the end of the Little Ice Age followed by a cooling episode (approx. 0.4K) which lasted 40-50 years. The models suggest also a recent ground surface temperature warming of about 1.5 K over the past 15 years. The ground surface at Raglan has experienced a 1.7 K warming over the past century and a 2.7 K warming since the Little Ice Age minimum (late 1700s).