PP33B-0919 1340h
Middle Pleistocene glaciation of Britain; new evidence and links with the ocean record
Extensive field mapping and analytical research over the last c. 20 years has lead us to proposed a new, five-glaciation model for the Middle Pleistocene in Britain, rather than the one- or two-glaciation models that have previously existed. The key discoveries are predominantly lithological, morphological and lithostratigraphical, with a limited number of new geochronologic determinations. The previously used pollen biostratigraphy has been found to be flawed and is rejected. The results show that British-sourced glaciation is dominant and evidence for Scandinavian ice cover in Britain can only be found in one occasion. Timing of the glacial episodes is based on existing age determinations and correlation with climatically-forced river aggradations and the model for global ice-volume. On the basis of this evidence glacier ice extended to lowland Britain from the highlands of Wales in MIS 18, from the highlands of Scotland in MIS 16, 12, and 10, and from the mountains of Scandinavia and Scotland in MIS 6. Britain was first separated from Europe during MIS 6 because of the importance of Scandinavian ice blocking the North Sea and creating a large proglacial lake that drained southward to the English Channel and eroded a gap to create the Straits of Dover.
PP33B-0920 1340h
A Glacial Chronology for the Fish Creek Drainage of Boulder Mountain, Utah
The Colorado Plateau physiographic province has several mountain ranges and high plateaus that supported valley glaciers and small ice caps during the late Quaternary. Although the glacial deposits in these ranges have been studied to some degree, there are only three quantitative age determinations for glacial deposits from all of the ranges on the Colorado Plateau. The timing of late Pleistocene glacial advance for this large and unique physiographic province remains largely unknown. Boulder Mountain, located on the western edge of the Plateau in south central Utah, has a well-preserved record of late Pleistocene glaciation. The mountaintop is a relatively flat volcanic upland $\sim$180 km$^{2}$ in area at an elevation of $\sim$3300 m. Glacial evidence on the summit surface in the form of freshly polished and striated bedrock outcrops, sculpted whaleback ridges and massive rouches moutonnees indicate that a thick ice cap occupied the mountaintop sometime in the recent past. Several reentrants around the mountaintop have deeply carved glacial valleys and steep sided latero-terminal moraine sets that were deposited by outlet glaciers spilling off the main ice cap. Using $^{3}$He exposure age dating and a correction for non-cosmogenic $^{3}$He, we determined corrected $^{3}$He exposure ages for pyroxenes from basaltic-andesite boulders on the most well preserved moraines in the Fish Creek drainage of Boulder Mountain. $^{3}$He exposure ages indicate a last glacial maximum (LGM) advance at 21,800 to 20,600 $^{3}$He years and a later and smaller advance at 16,500 to 15,400 $^{3}$He years. Our chronology is similar to other cosmogenic records from mountain ranges in the western U.S. (e.g. Wind River Range, Wyoming; Wallowa Mountains, Oregon) and is consistent with previous suggestions of a regional synchroneity of mountain glacier fluctuations during the LGM and associated deglaciation for ranges significantly south of the Laurentide margin.
PP33B-0921 1340h
The glacial record of New Zealand's Southern Alps
We present detailed mapping and surface exposure dating using in-situ Be-10 and C-14 of the moraine set of Lake Pukaki, New Zealand's Southern Alps, spanning from the penultimate glaciation, over several Last Glacial Maximum (LGM) moraines, the late glacial event to Holocene glacial advances. New Zealand, a mountain ridge in the middle of the Southern Ocean, has one of the best preserved moraine records world-wide, offering the opportunity to reconstruct amplitude and timing of climate changes from Southern mid-latitudes, an area where paleoclimate data is scarce. The extensive mapping effort by G. Denton and colleagues (http://wyvern.gns.cri.nz/website/csigg/) provides a unique background for sample selection for Surface Exposure Dating. Our extensive data set (>40 samples analyzed so far) indicate that (i) the LGM in New Zealand terminated clearly prior to the Boelling/Alleroed warming, (ii) the late glacial advance is within uncertainties consistent with the timing of the Younger Dryas cold reversal; (iii) there occurred an early Holocene glacial event of the same amplitude than the Little Ice Age. This latter event is the first Holocene glacial event from the Southern Hemisphere dated by in-situ Be-10 and C-14.
PP33B-0922 1340h
Exposure Dating Sampling Strategies in Regions Previously Covered by Cold-Based, low-Erosive Glaciers
Cosmogenic nuclide exposure dating has proven to be useful and reliable in many geomorphological studies. However, difficulties are to be expected when using a single nuclide approach on surfaces in blockfields and/or other areas previously covered by cold-based glaciers. Because of minimal subglacial erosion, bedrock surfaces may have an inherited component of cosmogenic nuclides and the total concentration will represent a composite signal. This can be avoided by using two cosmogenic nuclides with significant different half-lives (e.g. 10Be and 26Al), where their ratio in theory should detect periods of partial erosion and/or burial. However, this requires both long exposure periods prior to glacial overriding and long duration of subsequent burial, in order to detect exposure discontinuities. To circumvent this, we combine single nuclide exposure dating (10Be) and optically stimulated luminescence (OSL) dating in sample settings that include contemporaneous erosion and deposition. By using this multiple dating method approach, we expect to obtain reliable exposure age estimates and also to be able to identify burial by cold-based ice. The field area in east-central southern Norway has numerous OSL dated sedimentary deposits ranging in age from the present to 160 ka, often with evidence of glacial overriding (e.g. erratic boulders, thin tills etc.). Such deposits are commonly connected to lateral meltwater channels, glaciofluvially washed valley slopes, overflow gaps and canyons, all which have the potential to have been sufficiently eroded to remove previously accumulated cosmogenic isotopes. Three scenarios can be envisaged when comparing exposure and OSL ages from our time-synchronous sampling settings: 1) If bedrock exposure ages are older than the age of the corresponding glaciofluvial/glaciolacustrine deposit, insufficient glaciofluvial erosion can be assumed. 2) If the ages agree, the glaciofluvial erosion can be interpreted as having been large and the exposure ages can be taken as reliable estimates. 3) If the exposure ages are younger than the corresponding OSL-dated deposits, three interpretations are plausible: a) glaciofluvially eroded surface has later been covered by cold-based, low-erosive glaciers (should leave field evidence both on bedrock and sedimentary deposits), b) bedrock can be highly weathered and hence lost cosmogenic nuclides (possible to identify in the field), and c) sediments have not been zeroed during transport/deposition (should be revealed by inverted and/or inconsistent OSL dates on the deposit). Glacial erratics deposited on glaciofluvial landforms strongly argue that non-destructive glacial burial has taken place, however, the degree of erosion and removal of pre-last exposure signal can be questionable. The first 20 10Be exposure dates from this area are expected before the end of 2004.
PP33B-0923 1340h
Weichselian Glaciation History in the 'dry Valleys' of East-Central Southern Norway
East-central southern Norway is situated within the zone of inferred cold-based (low-erosive) ice-sheets during the Late Weichselian maximum in central Scandinavia. The region has high bedrock coverage of Quaternary deposits, while morphological features are mainly glaciofluvial deltas, kame terraces, shorelines, marginal moraines and various indicators of ice-marginal meltwater drainage. In the studied area, former meltwater flow patterns at high altitudes indicate drainage from east, south and west towards the 'dry valleys' of Doralen, Haverdalen, Grimsdalen and upper Folldal north of the Rondane mountains, and then further northwards across the main watershed to Trondelag. At present, the region is among the driest in Scandinavia with most of the precipitation related to south/south-easterly winds. With a suggested ice divide south of Rondane during the build-up of continental ice sheets in central Scandinavia, the `rainshadow' may have been even more pronounced during glaciations. Hence, the area north of Rondane is suggested to be a key area for studying glacier inception and wastage of Weichselian ice sheets in central Scandinavia. Based on detailed field investigations, both stratigraphical settings and landforms have been dated by use of Optically Stimulated Luminescence (OSL) dating. The results indicate that no period during the Weichselian obtained a large enough lowering of the regional glacier equilibrium-line altitude (ELA) to produce glacier ice at the terrain surface in the valleys north of Rondane. All glaciers that entered this area were produced elsewhere. Three main glacier events have been recorded: the largest occurred about 110-90 kyr ago during the early Weichselian, the next largest took place during early parts of the Middle Weichselian about 70-60 kyr ago, and the third largest occurred during the Late Weichselian maximum c. 20 kyr ago. Based on OSL-dated sections, long periods in between the major glacier advances were dominated by ice-dammed lakes in the upper valley of Folldal and its tributaries. Periglacial features like fossil rock glaciers and ice-wedge casts are related to several of the periods with glacier advances and/or periods with ice-dammed lakes.
PP33B-0924 1340h
Severnaya Zemlya, Arctic Russia: a Middle to Late Quaternary Kara Sea Ice Sheet(s) Nucleation Area
With a position on the north-eastern flank of any Kara Sea-based ice sheet, stratigraphic data from the Severnaya Zemlya islands (ZS) is crucial for any ice-sheet reconstruction. This areas was, however, excluded during the recently terminated QUEEN project, covering the glacial history of the last glacial cycle from the Kola Peninsula in the west to the Taymyr Peninsula in the east. In order to resolve this we launched a research programme with two field seasons (2002 and 2003) on SZ, with key locations on October Revolution Island. Our main effort was put into the stratigraphy along the Ozernaya River. Here we mapped four marine sequences, divided by glacial tills. The lowermost marine sequence is shallow-marine deltaic sand (M-1), containing numerous in situ fossil mollusc shells, as well as driftwood and whalebones. It is underlain by the lowermost till unit (T-1) and tectonized by a glacier which advanced southwards from inland SZ towards the Kara Sea, depositing the second till (T-2). Clayey, deeper-water marine sediments (M-2) in turn overlie this till, containing numerous in situ fossil mollusc shells. Clast fabrics from the above-lying till (T-3) show that also this glacier advance was southwards into the Kara Sea. The next marine unit (M-3) shows a transition from marine clay to sandy sublittoral deposit and beach gravels, containing numerous in situ fossil mollusc shells. The beach gravels, in turn, are covered by a thin glacial till (T-4), showing incorporation of the underlying beach gravels and a clast fabric indicating glacier advance southwards into the Kara Sea. Our youngest marine sequence (M-4), not present in our key locality, reveal a transition from marine clay into well-developed beach ridges, which could be mapped over long distances on altitudes not higher than c. 40 m a.s.l. The data indicate at least four Late Quaternary major glaciations on October Revolution Island, spanning - according to our extensive but not yet completed dating programme - Marine Isotope Stages (MIS) 10 to MIS 4. During stadial conditions local ice caps grew and expanded and probably coalesced with ice from Taymyr and Novaya Zemlya to form Kara Sea-based ice sheets. The significant isostatic depression, causing subsequent deposition of marine sediments and formation of raised beaches at altitudes up to 100-130 m a.s.l. suggests that these glaciations were on regional scales, not only expansions of local glaciers, and that SZ was an ice-sheet nucleation area.
PP33B-0925 1340h
Holocene Climate Change Over Central Asia and the Himalaya: Latitudinal Dependence On the Monsoon and Midlatitude Eddy Activity
Climate over Asia's mid-high latitudes is dominated by cyclone activity whereas over Asia's subtropical southeastern flank it is dominated by the monsoon. Asian climate is therefore highly sensitive to changes in the planet's orbital configuration, which can alter both the midlatitude eddy activity as well as the strength of the south Asian monsoon. The latitudinal extent of the annual mean Hadley (0-30N), Ferrel (30N-60N), and polar (60N-90N) cells provide natural divisions within which to distinguish those regions influenced by changes in the monsoon from those dominated by midlatitude eddy processes. Although the dynamics within these zonal bands are normally coupled quite tightly, the Himalaya in the central longitudes of Asia largely isolate the dynamics of the tropics-subtropics from those of the mid-high latitudes. Through a suite of numerical AGCM simulations that span the Holocene at 500-year intervals, we investigate the latitudinal dependence of Asian climate change over the past 10,000 years. In the monsoon-dominated tropics, the core of the westerly summer monsoon jet is further north than today during the early Holocene and gradually shifts southward as its strength decreases to today's value, with concomitant changes in precipitation and snowfall over northeast India and the Himalaya. In midlatitudes, increased synoptic eddy activity produces colder and wetter (snowier) conditions during the early Holocene as compared to today. There is a distinct shift towards warmer and drier climate during the mid-Holocene when atmospheric carbon dioxide levels are increasing most rapidly, even though synoptic eddy activity is still relatively high. This mid-Holocene shift towards warmer and drier climate has been seen in a number of proxy records from central Russia.
PP33B-0926 1340h
Ambiguities in Interpreting the Beryllium-10 Record From Marine Sediments
Beryllium-10 from Blake Outer Ridge (BOR) late Pleistocene marine sediments at Leg 72, site 1061 B.P. was measured in order to obtain a record of the cosmic-ray flux incident upon the Earth. The variations in the concentration of $^{10}$Be per gram of sediment in BOR sediments and that of the oxygen isotope record of the North Atlantic are strikingly similar highlighting the effects of climate on the deposition of $^{10}$Be in marine sediments. In particular, an abrupt change in the oxygen isotope record from Marine Isotope Stage (MIS) 6 to MIS 5 also is observed in $^{10}$Be record (as normalized to gram of sediment) as a result of an abrupt change in the sedimentation rate. In addition, the long-term trends of $^{9}$Be and $^{10}$Be concentrations, both source dependent, are inversely correlated and thus the $^{10}$Be/$^{9}$Be normalization is ambiguous at best. The effects of climate on $^{10}$Be deposition was best removed by normalization of $^{10}$Be to the mass of the authigenic fraction of the sediment from which it was contained. The $^{10}$Be content of the sediments at site 1061 was correlated to another site on the Blake Outer Ridge, CH88-10P, and to a site in the Gulf of California Leg 64, site 480. $^{10}$Be normalized to the authigenic mass at these sites show that the $^{10}$Be concentrations within the sea are bounded within a specific range consistent with the expected secular variations of the Earth's dipole field.
PP33B-0927 1340h
Evidence of Early Holocene Glacial Advances in Southern South America from Cosmogenic Surface Exposure Dating
$^{10}$Be and $^{36}$Cl cosmogenic nuclide surface exposure dating of erratic boulders reveal two glacier advances in southern South America (46°S) during the Early Holocene. Seven of ten boulders from the outer moraine yield a weighted mean of 8.5$\pm$0.7 ka and five of six boulders from the inner moraine yield a weighted mean age of 6.2$\pm$0.8 ka (2$\sigma$ uncertainties). The four outliers are anomalously old (interpreted to contain inherited cosmogenic isotopes from prior exposure) and are identified on the basis of Chi-Squared statistics and bi-modal probability distribution curves. These glacial advances are likely the result of a northward migration of the southern westerlies causing an increase in precipitation and/or a decrease in temperature at this latitude. Reconstructions of equilibrium line altitudes (ELA) at the times of moraine deposition based on Accumulation Area Rations (AARs) are about 300 m lower than modern. This ELA depression is not particularly sensitive to the AAR used, and corresponds to conditions 2.4$\deg$C cooler (if no change in precipitation), or 1000 mm/a wetter (if no change in temperature) than the modern climate. The older advance precedes the currently accepted initiation of Holocene glacial activity in southern South America by about 3000 years, and appears to be temporally synchronous with the "8.2 ka event" recorded in Greenland and many other parts of the world. The younger advance is slightly older than, but indistinguishable from, documented neoglacial advances and climate changes in southern South America and four other continents. If there are causal links between these temporally synchronous, globally distributed events, then millennial scale climate changes appear to involve reorganization of global weather systems (such as migration of the southern westerlies), or may be externally forced (e.g. solar variability).
PP33B-0928 1340h
Deglacial History of the Ecuadorian Andes and Implication for Climate Variations: Preliminary Results
Paleoclimate reconstructions are essential for evaluating the future evolution of natural climate variability and for determining climate sensitivity to external forcing. Reconstructing climate conditions from the Last Glacial Maximum (LGM) to the Holocene represents a unique opportunity to understand climate variability from full glacial conditions to modern warm conditions. The primary goal of our project, is to verify if the changes in temperature and precipitation driving the glacier event in the tropics during the well-documented Little Ice Age (LIA), may also account for the glaciations related to the LGM and the late glacial period. This inter-disciplinary project brings together specialists in glacial geology, surface exposure dating, and climate modeling. Our first trip to Ecuador took us to the Papallacta Valley at the rim of the Potrerillos Plateau. We developed detailed maps of the snowline lowering in the valley and took samples in well-exposed sections for radiocarbon dating. We used our maps and the age constraints on the deglacial history of the Papallacta Valley to estimate the possible combinations of changes in climate parameters related to reconstructed snowline variations. This local study represents the first step in a broader project that will cover most of the Ecuadorian Andes. We will also provide direct dating ($^{3}$He, $^{10}$Be, and $^{36}$Cl) of the moraine sequences deposited during the retreat of the glaciers during the late Pleistocene. By the time of the project completion we want to evaluate the nature of the driving forces underlying the LGM and the late glacial event in view of the relatively well understood mechanisms behind the termination of the LIA, and we want to compare the produced data to mid- and high- latitude areas in order to evaluate the regional footprint of dimension and timing of glacier response to climate change.
PP33B-0929 1340h
Age Models for a Continous 250-kyr Quaternary Lacustrine Record from Bear Lake, Utah-Idaho
The Quaternary sediments sampled by continuous 120-m-long drill cores from Bear Lake (Utah-Idaho) comprise one of the longest lacustrine sequences recovered from an extant lake. The cores serve as a good case study for construction of age models for sequences that extend beyond the range of radiocarbon dating. From a wide variety of potential age indicators, we selected a combination of radiocarbon ages, magnetic excursions (correlated to a standard sequence), and a single Uranium-series age to develop a "basic" age model. Data omitted from the basic age model include amino acid age estimates, which have a large amount of scatter, and tephrochronology correlations, which have large uncertainties. The basic age model suggests that the cored Bear Lake record is continuous back to about 250 ka. We also compared ages derived from the basic age model for inferred major glacial-interglacial climate events with events from other paleoclimate records (e.g., SPECMAP and Devils Hole). These comparisons support the accuracy of the basic age model and provide a degree of justification for using climate correlations as control points in age modeling. Doing so yields a "tuned" age model that represents our current best estimate of the chronology of Quaternary sediment deposition in Bear Lake.
PP33B-0930 1340h
A 13,000-yr high-resolution record of Rocky Mountain Holocene climate change
A high-resolution study of 13,000 years of climatic changes recovered from Foy Lake, Montana, sampled continuously at 5 - 10 yr intervals throughout the record, demonstrates dramatic changes in fossil diatom assemblages throughout the lake's history. The power of examining lake records at this resolution is detailed by many sharp transitions in fossil diatom assemblages and multiple scales of climate variability imprinted on the sediment record. At 11,550 ybp, near the termination of the Younger Dryas, low-nutrient species give way to higher-nutrient, planktonic diatom species. At roughly 10,000 ybp, alkalinity increases dramatically, resulting in a major assemblage shift and $\sim$1,200 years where the diatom record is repeatedly dissolved, likely because of enhanced alkalinity during extremely dry intervals. From ca 8,800 - 7,100 ybp, the fossil diatom assemblage shows cycles between benthic and tychoplankton species, indicative of cycles of lake-level change in a relatively shallow lake. At $\sim$8,200 ybp, a 200-yr period of dominance by planktonic diatoms suggests a pronounced increase in lake level in the midst of several thousand years of otherwise very dry conditions. Decadal sampling throughout the mid-Holocene indicates that composition of the diatom assemblage remained constant, but high-frequency changes and sharp transitions in dominant species were common. For about 1,000 years, ending at approximately 6,100 ybp, a sudden increase in the percentage of planktonic diatoms suggests a moderate increase in effective moisture, followed by a $\sim$2000 yr period of tychoplankton and benthic species indicating increased alkalinity and reduced moisture. Planktonic diatoms again dominate the record from ca 4,400 - 2,800 ybp. These patterns of change indicate that mid-Holocene climate was neither exceedingly stable nor consistently dry in the northeastern Rocky Mountain region. From 2,800 - 1,400 ybp, the diatom record, dominated by benthic flora, is characterized by long periods of heavy dissolution suggesting higher alkalinity during an intensely dry period. A dramatic transition occurs at 1,400 ybp, when the diatom record becomes overwhelmed by a single planktonic diatom species, which dominates throughout much of the remainder of the late Holocene and from which we infer a modern maximum in effective moisture.
PP33B-0931 1340h
Record of Holocene Lake Levels in Bear Lake Utah/Idaho
Sediment grain-size data and other observations indicate that Holocene Bear Lake experienced large lake-level changes, and that on average the lake was about 20 m below the modern full-lake level. Bear Lake, in northeastern Utah and southern Idaho, lies in an active half graben at an elevation of about 1800 m, covers an area of 280 km$^{2}$, and is 63 m deep. During most of the Holocene the Bear River bypassed Bear Lake, but river water was diverted into the lake circa 1912. In 1909 lake elevation was about 1805.5 m, and modern full-lake level is close to this datum. Holocene sediments are comprised of about 70% endogenic carbonate, 25% detrital siliciclastic material (largely quartz), and small amounts of other material. Except for one calcite interval, the endogenic carbonate is aragonite. After removal of carbonate, biogenic silica, and organic material, grain-size data were acquired using a laser particle-size analyzer. Analyses of 30 samples of the uppermost 1.5 cm of sediment, taken along 4 depth profiles, demonstrate progressive decrease in grain size with water depth. Median grain sizes range from $>$200 $\mu$m in 1 m of water to $<$6 $\mu$m in depths greater than 30 m. Differences among the depth/grain-size relations defined by these profiles are probably related mostly to differences in lake-bottom morphology and in position with respect to fetch and prevailing winds. Paleo-depths were calculated for Holocene sediments from two $^{14}$C dated cores using the depth/grain-size relation defined by the profile closest to the coring sites. These modeled depths indicate that the lake was commonly about 20 m below modern levels and that there were three periods when lake levels were significantly lower. Two of these low-water periods occurred in the early Holocene (both occurring between 10.2 and 9.2 cal. ka) and one in the late Holocene (ending at about 2.8 cal. ka). A bench and scarp on the west side of the lake, observed in reflection seismic profiles, are consistent with a prolonged period with water 20 m to 25 m below modern levels. Low-water levels are further corroborated by shell gravels in one of the cores that coincide with the coarsest grained siliciclastic material and by a horizon containing roots dated at about 9.9 to 10.1 cal. ka from a third core taken from a depth of about 40 m. At face value, the grain-size data suggest that only once during the Holocene did water depth reach or exceed the modern level, although thin intervals of deep-water deposition may have been missed. Shortly after 9.2 cal. ka, water depth increased to a maximum greater than the modern full level. The period of deep water lasted about 1000 years and coincides with an interval in which aragonite is replaced by calcite, indicating that the water was not only deeper but also fresher. Isotopic data (e.g., $^{87}$Sr/$^{86}$Sr) indicate that the Bear River was flowing into the lake during this period. The high-water level coincides with formation of an elevated shoreline 8 m above modern lake level (dated by Laabs and Kaufman at 9.2 cal. ka, GSA Bull., 2003). During much of the Holocene Bear Lake was topographically closed and water levels probably reflected climatic conditions. Because the lake had attained its modern level prior to diversion of Bear River water, effective precipitation in the late 19$^{th}$ and early 20$^{th}$ centuries must have been greater than for the average Holocene climate. The Holocene high stand may also be climatically controlled, but the possible roles of tectonics and surficial processes in diverting the river into the lake cannot be discounted.
PP33B-0932 1340h
Got Varves?: Reconstructing Holocene Climate Change in Seneca Lake, NY
The sedimentary deposits of Seneca Lake, one of eleven Finger Lakes in New York State, contain a valuable record of post-glacial climate and environmental change. Paleoenvironmental interpretations depend on knowing what transport and depositional processes controlled the formation of the laminae. In this study, we examine Holocene rhythmites in two profundal cores collected from the northern half of Seneca Lake to determine possible mechanisms of formation of the alternating olive gray-black layers. Magnetic susceptibility was measured at a 2 cm interval prior to splitting the $\sim$5 m cores. Split cores were described, photographed, and sampled for loss-on-ignition and grain size analysis at a 10 cm interval. When possible, core one was sampled on a lamination-by-lamination basis whereas core five was sampled at a one cm interval. Sequential LOI was used to estimate the total organic and carbonate content of sediments. Grain size analyses were performed using a Coulter LS 230 laser diffractometer after removal of calcium carbonate. Temporal control of paleoenvironmental changes in cores will be established by two accelerator mass spectrometer radiocarbon dates. The cores contain proglacial pink clay overlain by Holocene mm- to cm-scale alternating olive gray and black laminations of fine-grained sand and mud. Magnetic susceptibility changes are distinct, ranging from 1-22 x 10$^{-6}$ SI units in core one and 4-57 x 10$^{-6}$ SI units in core five. Sediment is dominantly composed of siliciclastic material (50-99 weight $%$) with varying amounts of calcite (0.1-75 weight $%$) plus lesser amounts of organic matter (0.5-4.1 weight $%$). A distinct variation in sediment color, organic matter, and carbonate content, and median grain size ($\phi$ $_{50}$) occurs in the laminated sediment. The olive gray layers are characterized by well-sorted very fine silt and clay with a $\phi$ $_{50}$ $<$6 $\mu$m, high carbonate content ($>$25%), and low organic matter content ($<$1%). The black laminae are generally higher in organic matter ($>$1%), lower in carbonate content ($<$25%), and have a $\phi$ $_{50}$ $>$6 $\mu$m. Samples are well-sorted and positively skewed. Cores were correlated based on their carbonate content, magnetic susceptibility, and median grain size. We interpret these laminae to reflect changes in sediment sources. The olive gray layers may reflect deposition during the late spring through the summer. Fine-grained calcite precipitates mainly in the summer during maximum phytoplankton productivity in temperate, mid-latitude lakes. Carbonate content and thickness variations of the gray laminae may provide clues to the length of the productivity season, and thus perhaps the duration of the stratification season. A decrease in carbonate content may reflect cooler climate conditions. The coarser-grained black, siliciclastic-dominated layers may reflect deposition by underflows and/or turbidity flows during spring melt or episodic storms. It is unlikely the rhythmites in Seneca Lake are annual because the number of laminae preserved is insufficient to span the entire Holocene if resolved annually. Bioturbation or current re-working may blur the annual signal. Fabric analysis of thin sections will be used to further constrain the environmental conditions during deposition.
PP33B-0933 1340h
Diatom Records of Holocene Environmental and Climatic Change in Southeastern British Columbia
The sediments of three lakes, spanning a latitudinal gradient and situated within the Interior Cedar-Hemlock biogeoclimatic zone of southeast British Columbia, have been analyzed for independent records of Holocene climate and vegetation history in order to evaluate the role of climate as a control on range expansion of western hemlock and western redcedar. Holocene climate reconstructions are based on multiple proxies, including the diatom and biogenic silica stratigraphies presented here. To our knowledge, these sites provide the first continuous Holocene paleoecological and paleoclimatic records from the region between the Interior Plateau and the Rocky Mountains. Stratigraphic variability in the diatom sedimentary records reflects lake response to changes in water balance, catchment development, and atmospheric inputs of silica (i.e. volcanic ash). At Eleanor Lake, the diatom record shows significant assemblage shifts at $\sim$7500, 6300, and 3700 cal yr BP. Increased abundance of planktonic and tychoplanktonic diatoms and an increase in the abundance of chrysophyte cysts relative to diatoms at 6300 cal yr BP is consistent with an increase in effective moisture at this time. At Mirror Lake, in the south, an assemblage dominated by small benthic species is replaced by a predominantly planktonic assemblage at 6900 cal yr BP, suggesting an earlier increase in lake level. After 6000 cal yr BP, planktonic diatoms further increase in dominance. These dates are consistent with reconstructions of Holocene climate in south-central British Columbia, which suggest a shift to moister conditions ca. 7000 to 6000 cal yr BP. The pollen record indicates the expansion of western hemlock and western redcedar near Mirror Lake at 3550 cal yr BP and 3250 cal yr BP respectively. At Eleanor Lake, hemlock expansion is dated at 4350 cal yr BP. The data indicate a difference of 1950 to $>$ 3000 years between the lacustrine record of increased effective moisture and the vegetation response and suggest that non-climatic factors played a significant role in the Holocene range expansion of these trees.
PP33B-0934 1340h
Simulated and Inferred LAI, NPP, and Biomes in North America Since the Last Glacial Maximum.
Vegetation structure and productivity are sensitive to climate change and are an important source of feedbacks to the climate system. Here we employ multiple lines of evidence to reconstruct variations in leaf area index (LAI), net primary productivity (NPP), and biomes. LAI determines the total canopy surface area available for light interception, gas exchange, and water loss, and NPP, the increase in plant carbon per unit area, measures the flux of carbon into the terrestrial biosphere. BIOME4, an equilibrium biogeography and biogeochemistry vegetation model, is used to simulate LAI, NPP, and biome distributions in North America for the past 21,000 years at 1,000-year time-steps. BIOME4 was coupled asynchronously to the Hadley Center Unified Model with a mixed-layer ocean model forced by variations in orbital boundary conditions, physiography, and atmospheric CO2 concentration (Kaplan et al. 2002). BIOME4 models LAI as a trade-off between maximizing light interception and minimizing water loss and assigns the LAI that maximizes NPP. Past LAI's and biomes, independently estimated from fossil pollen assemblages using the modern analogue technique, are compared to model results. In unglaciated eastern North America, canopy closure of the full-glacial conifer forests and woodlands in response to ameliorating climatic conditions resulted in a 80% increase in LAI's between 21 ka and 11 ka. After 8 ka, large areas of tundra and forest-tundra developed in deglaciated regions. The BIOME4 simulations show good agreement with the LAI's and biome distribution inferred from fossil pollen records. Sensitivity analyses with BIOME4 indicate that both climate and CO2 played important roles in regulating vegetation structure and productivity.
PP33B-0935 1340h
Tree Ring Stable Isotopic Evidence for Climate Change in Coastal Wetlands of the Northeastern United States
Recent studies of the stable isotopic ($\delta$$^{13}$C, $\delta$$^{18}$O, $\delta$D) composition of tree rings have provided several robust paleoenvironmental records that span several millennia. These records are exceptionally important to human populations because they resolve climatic shifts that occur on the scale of years. Reconstructing Holocene climatic conditions for the northeastern United States is crucial for understanding natural variations in regional climate and for defining the limits of what can be expected during episodes of abrupt climatic change. Annually-homogenized bulk wood $\delta$$^{18}$O values from an 800-year old Eastern White Pine (Pinus strobus) log recovered from Hundred Acre Cove, Rhode Island, range from 20.0 to 23.8 % (V-SMOW) over the 55 year period of growth. The record of $\delta$$^{18}$O contains two well-defined, multi-annual negative excursions that are spaced approximately 40 years apart. Isotopic results from a ~5000-year old Eastern White Pine log and a modern cedar will also be interpreted.
PP33B-0936 1340h
Tree Rings as Climate Proxies in Susquehanna River Basin Streamflow
Tree rings have been used throughout the world to reconstruct past climate records. However, no attempt has been made to reconstruct streamflow records for the Susquehanna River Basin. Having previously found a statistical relationship between tree rings and streamflow records in the basin, the researchers have refined this relationship and used it to reconstruct streamflow in the basin back to 1700. Tree ring chronologies were obtained from NOAA's Paleoclimatic International Tree Ring Data Bank for the period 1700 to 1981. Streamflow records were obtained from the USGS website for Harrisburg, PA for the period 1891 - 1981. Six tree ring sites (Figure 1) were used to determine the significance using time series and regression analyses. The highest significance was found among the r-values for tree ring located in Salt Springs Park (.587) and from the combination of all six sites (.584). The tree ring sites were lagged so that the streamflow of year 1 produced the tree rings of year 2. Figure 3 shows the correlation between Salt Springs and Harrisburg streamflow using z-scores as the time series. Figure 4 gives the calibration period of the streamflow data. Figure 5 gives the actual reconstruction of the streamflow records. On the graph there are major peaks and valleys indicating major historical events such as floods and droughts. There is also some decadal variability indicated. The reconstructed data for the 1800s indicates that significant trends were occurring throughout the century.
PP33B-0937 1340h
A High-Resolution Speleothem Climate Record for the Past 5.2ky From Western Costa Rica
The El Niño Southern Oscillation (ENSO) is the most important source of interannual climate variability in the tropics today. While some data suggest that the frequency of strong ENSO events has apparently increased through the Holocene (e.g. Rodbell et al., 1999), these studies need corroboration and a thorough understanding of ENSO variations over the Holocene remains a goal of paleoclimate studies. In Central America, meteorological data show that El Nino events are associated with decreased rainfall relative to normal or La Nina years, and that rainfall amount is strongly anti-correlated to rainfall \delta$^{18}$O via an amount effect. Isotopic studies of stalagmites in other regions have shown their utility in studying changes in precipitation and here we apply these techniques to study Central American climate variation and by inference changes in mean ENSO state on sub-decadal timescales. Stalagmite CN-1 was recovered from Nicoa Cave in Barra Honda National Park located on the Nicoya Peninsula in Costa Rica, on the western side of Central America. An age model for CN-1 was constructed using 7 U/Th measurements and indicates continuous deposition from 5200 y B.P. to present. A preliminary oxygen and carbon isotope data set covers the past 1500 y at 6 year resolution. Ultimately we will cover the past 5.2 ky at similar resolution. The isotopic values range greatly, from -6.5 to -10.2 \permil for \delta$^{18}$O, and from -4 to -10.5 \permil for \delta$^{13}$C. More depleted values are interpreted to be indicative of wetter periods and periods of less-frequent strong ENSO events. A preliminary comparison of the CN-1 isotope record with the Lake Pallcacocha ENSO record shows a good match between more enriched isotope values and periods of more frequent strong El Nino events, particularly for \delta$^{13}$C. The data also show a strong multidecadal-scale cyclicity.
PP33B-0938 1340h
A High-Resolution Absolute-Dated Penultimate Glacial Monsoon Record From Hulu and Dongge Caves, China and Global Correlations of Events Surrounding Termination II
We have obtained the oxygen isotope record of three stalagmites from Hulu Cave (eastern China, $32\deg$30'N, $119\deg$10'E), offering a high-resolution absolute-dated record of Asian monsoon climate over much of the penultimate glacial period. Previously, stalagmites were analyzed from Dongge Cave (southern China, $25\deg$17'N, $108\deg$5'E), located approximately 1200 km to the southwest of Hulu Cave, providing a detailed account of $\delta$$^{18}$O variations over about the same time period (Yuan et al., 2004; Kelly et al., in review). The $\delta$$^{18}$O patterns from the Hulu and Dongge stalagmites appear to be broadly similar across the overlapping portions of the records, indicating that the stalagmites are recording a regional climate signal. Asian Monsoon climate during the penultimate glacial period follows orbitally induced insolation changes, though it is punctuated by numerous sub-orbital scale climate events. The heaviest $\delta$$^{18}$O values during the penultimate glacial period were initiated at $\sim$136 ka B.P. in both records, and are largely maintained for the next $\sim$6 kyr until abrupt strengthening of the monsoon at Monsoon Termination II. This interval is generally characterized by less variability than seen during the earlier portions of MIS 6, with $\delta$$^{18}$O values remaining around -5$\permil$. Kelly et al. (in review) have called this extended interval of time the "Weak Monsoon Interval" (WMI). The WMI, however, is punctuated by one distinct climate event, occurring at 134.5 $\pm$ 1.0 ka B.P. in the Hulu record and 133.6 $\pm$ 1.0 ka B.P. in the Dongge record. This event may correlate with warming in the Alps at 135 $\pm$ 1.2 ka B.P., as indicated by an interval of speleothem growth from Spannagel Cave (Sp\"{o}tl et al., 2002). We also demonstrate that the monsoon intensity correlates well with atmospheric CH$_{4}$ concentrations over the last two glacial cycles. In particular, we correlate the abrupt jump in CH$_{4}$ concentrations at $\sim$129 ka B.P. with our Monsoon Termination II. Based upon this correlation, we show that the full Termination II atmospheric CO$_{2}$ rise, the full Termination II Antarctic temperature rise, and the gradual portion of the CH$_{4}$ rise occur almost entirely within the Weak Monsoon Interval, an extended interval of heavy $\delta$$^{18}$O between $\sim$136 and 129 ka B.P. Due to the previously established relationship between Greenland temperatures and monsoon intensity (Wang et al. 2001), this interval likely corresponds to low Greenland temperatures. Therefore, we infer that atmospheric CO$_{2}$, Antarctic temperature, and the initial portion of the CH$_{4}$ rise preceded the increase in monsoon precipitation and warming in Greenland.
PP33B-0939 1340h
Neogene Flowstones of the Interior Low Plateau of Kentucky: Isotopic, Geochemical, and Petrographic Constraints on Near-Surface Hydrology and Climate
Flowstones exposed in quarries and caves within the Interior Low Plateau of Kentucky and neighboring regions preserve a record of near-surface hydrologic conditions that potentially spans the late Cenozoic pre-glacial and glacial periods. Detailed isotopic ($\delta$$^{13}$C, $\delta$$^{18}$O), geochemical (Mg, Sr, Ba, Fe, Mn), and petrographic analysis of flowstone sequences recovered from across the region may provide unique insight into the characteristics of climate shifts and may ultimately provide a temporal framework for regional climate change. Analytical results will be presented from several flowstones. Preliminary results from a flowstone preserved in a quarry at Rocky Knob, Casey County, Kentucky, indicate a dynamic range for both $\delta$$^{18}$O (-7 to -2.5 $\permil$ V-PDB) and $\delta$$^{13}$C (-13 to -7 $\permil$ V-PDB). Much of the 18 cm-thick flowstone records relatively wet conditions. The upper 3 cm of record preserve evaporation-dominated conditions, defined by trace-metal anomalies and significantly higher enrichment in isotopic compositions that are linearly correlated on a $\delta$$^{18}$O vs. $\delta$$^{13}$C plot. Linkages between trace-metal concentrations, crystal growth and morphology, and isotopic compositions will be explored and interpreted.
PP33B-0940 1340h
Multi-Site Evidence for Marine Nitrogen Fixation in Mid-Cretaceous Black Shales
High concentrations of organic carbon in Cretaceous black shales imply levels of sustained export production of organic matter that are unknown in the modern ocean where marine productivity is usually limited by availability of dissolved nitrate. However, if a mid-water anoxic zone expands upward into the photic zone, then nitrogen-fixing cyanobacteria can flourish. These organisms produce organic matter having an isotopic composition close to atmospheric nitrogen (0 per mil). We have compared the carbon and nitrogen isotopic and total organic carbon compositions of Albian to Santonian black shale sequences from the Demerara Rise in the equatorial Atlantic, the Kerguelan Plateau in the southern Indian Ocean, the Hatteras Rise in the western North Atlantic Ocean, the Angola Basin in the eastern South Atlantic Ocean, and the Cape Verde Rise in the eastern North Atlantic Ocean . Nitrogen isotope compositions that become lighter as organic carbon concentrations increase indicate that organic matter production was enhanced by a consortium of primary producers that included nitrogen-fixers. Expansion of an intensified oxygen minimum zone into the photic zone probably permitted coexistence of algae and of cyanobacteria, the latter functioning best under low-oxygen conditions and not being limited by nitrate availability. Improved preservation of the exported organic matter in an intensified near-surface oxygen minimum zone is implied by C/N ratios that increase to 40 as organic carbon concentrations increase. Periods of wetter climate evidently created periods of increased surface stratification of Cretaceous oceans that led to enhanced cyanobacterial primary productivity, magnified organic matter export, and deposition of the organic-carbon-rich black shales. Our multi-site comparison suggests that climate-related gradients in the degree of surface stratification led to associated gradients in export production of organic matter.
PP33B-0941 1340h
Aptian Facies Diversity Around The Tethys Realm: Global Oceanic Factors Vs. Local Physiographic Conditions
Comparison of Aptian stage succession in southeastern France, northeastern Mexico, and central Italy, indicates that deposits of C$_{org}$-rich facies are diachronous and may show sharply contrasting successive conditions of deposition. For instance, the La Pe\~{n}a Formation in northeastern Mexico spans the interval that includes OAE 1a (Selli/Goguel event), and is mostly organic rich marls, but the section at La Huasteca Canyon also shows conspicuous red to purplish-colored carbonate-poor layers intercalated with the dark gray C$_{org}$-rich shales and marly limestones. While the marls average 68.74%wt of CaCO$_{3}$ and TOC values of 0.59%wt, the reddish beds show CaCO$_{3}$ values as low as 17.04%wt and high TOC values up to 2.63%wt. The apparent discrepancy between these layers questions the origin of the reddish layers that have been overlooked. Although initial EDS analysis on these reddish beds confirms a high iron oxide content (8-9%wt), it remains uncertain whether they represent true oxic conditions at that time, or whether they have been modified by subsequent diagenesis. In southeastern France, the Provencal platform shows in the Early Aptian Urgonian facies a light gray, discontinuous and locally diachronous marly interval, which is a weak indication of possible relation with the worldwide anoxic event. Exact chronological correlation of this layer with the Selli-Goguel Event, and relative timing of these conditions with the northeastern Mexican platform is still pending. Previous studies show considerably high TOC values of 2-2.5%wt for the lower Aptian series at other section within the Provencal platform. In the Umbria-Marche Basin of central Italy, the well-known Selli level or OAE1a is characterized by a regional C$_{org}$-rich marker bed indicative of anoxic/dysoxic conditions. However, in the Gubbio sections, intermittent maroon to red-colored intervals that strike out within the predominantly black shale series are compatible with oxic environments. Geochemical analyses report iron content with values up to 9.15%wt of Fe$_{2}$O$_{3}$ in these red shales, values that are strikingly similar to the reddish beds from NE Mexico. The reddish-brown beds present within the black shale facies in the Aptian series clearly indicate periods of local variations to well-oxygenated conditions within the recurrent low-oxygen waters. Thus, facies differences with variations of coeval oxic and dysoxic conditions, imply that in certain areas, local factors controlled sediment type, and overprinted global forcing mechanisms that depleted oxygen level elsewhere.
PP33B-0942 1340h
Geochemistry and sedimentary environments of Permian deposits in Eastern-Central Iran
Permian of Eastern-Central Iran is represents of Jamal and Khan Formations. These formations are studied in three areas named Kalmard, Shotori and Shirgesht. Khan Formation composed of predominantly delataic to eolian coarse to very coarse silisiclastics that interbedded with thin shallow limestone. Jamal Formation includes predominantly shallow, but sometimes deeper water carbonate sediments. The purpose of this study was to determine the depositional environment and diagenetic history of each of the facies that were recognized within the Permian succession. Sedimentological and microfacies study and advanced geochemical study have been performed in this project. Macroscopic and microscopic evaluations of hand samples taken at approximately 4-5 meter intervals allow determination of composition and extent of cementation and dolomitization. 900 samples for microfacies studies and 100 samples for geochemical analysis have been collected. The principal microfacies have been identified in thin sections. O$^{18}$ and C$^{13}$ isotopes were utilized for better understanding of diagenetic processes. Ca, Mg, Mn, Na, Sr and Fe amounts and ratios were used to compliment the isotopic studies and recognition of original mineralogy. Geochemical data regarding elemental compositions of Khan and Jamal formations represent lower Sr ( 840 ppm ) and Mg ( 4600 ppm ) values and higher Fe ( 1500 ppm ) and Mn ( 250 ppm ) contents compared with tropical carbonates. Therefore the existence of a cool- water conditions during sedimentation of Khan and Jamal formations and probable calcitic precursor for them can be considered. Variations in $\delta$$^{18}$ O (-4 to -10 PDB) and $\delta$$^{13}$C (1 to 4 PDB) values and their relationship with trace elements of carbonates suggesting influence of burial and meteoric diagenesis that are compatible with petrographic studies. Also plot of Sr/Ca ratios versus Mn values indicative of influence of diagenetic processes in a roughly mid-closed environment. Lower $\delta$$^{13}$C contents and higher values of Mn and Fe of Khan Formation bulk carbonates than that of Jamal formation can be explained by presence of siliciclastic components. These data confirm sedimentation of Khan and Jamal formations in two different basins in Permian time. Khan Formation was deposited near to shore while sedimentary environment of Jamal formation is a carbonate shelf.
PP33B-0943 1340h
Of Island Arcs and Icebergs? Strontium and Carbon Isotope Stratigraphy of Ethiopian Snowball Earth Sequences
Neoproterozoic low-grade metasedimentary exposures uniquely preserved in synclinoria of the Tigrai region of northern Ethiopia offer excellent opportunities to assess the sedimentary response to extreme climatic changes encompassed by the Snowball Earth Hypothesis within the framework of the tectonic evolution of the Arabian-Nubian shield. Basement exposures consist of two major sequences; an older predominantly island arc metavolcanic sequence named the Tsaliet Group, and a younger metasedimentary succession named the Tambien Group. An intervening phyllite unit displays a transitional lithology variously interpreted by previous workers as metasedimentary and metavolcanic. We report Sr and C isotope results for Tambien Group carbonates from three prospective Snowball Earth successions flanking the Mekelle outlier: Negash, Mai Kenetal, and Samre. These localities display excellent exposures of the transition from Tsaliet metavolcaniclastics into distinctive carbonate exposures of the Tambien Group. Distinctive volcanic polymict conglomerates occur close to the basal Tambien contact in the upper Tsaliet Group in two localities. Presence of evaporite pseudomorphs and persistence of microbially laminated lithologies including stromatolites suggests that Tambien carbonate deposition occurred in a shallow marine setting that was initially hypersaline. A distinct black limestone unit, with abundant organic matter and indications of variable energy conditions in a slope setting, occurs higher in the section. Accompanying this transition in gross carbonate depositional style is a pronounced enrichment of carbon isotopes from values between -4 and -1 per mil in the lower interval to values between +4 and +7 per mil in the overlying black limestone. A similar excursion to more radiogenic $^{87}$Sr/$^{87}$Sr compositions is observed from values between 0.7060 to 0.7064 in the lower interval to values "plateauing" between 0.7067 and 0.7069 in the black limestone. Thus far, only the Negash syncline preserves a diamictite of probable glacial origin (i.e., dropstones, faceted clasts, cobbles of diverse composition) where a pronounced negative carbon excursion from values of +7 to -2 per mil accompanies the transition from upper black carbonate to diamictite deposition. The diamictite is the youngest preserved unit within the investigated structures so it is unknown if a superjacent cap carbonate was ever deposited. Zircon evaporation dates from granites intruding the base of the sequence but post-dating early deformation of the entire sequence suggest that the Tambien Group is older than 613Ma, and therefore diamictite genesis is incompatible with younger Varanger, or `Ediacaran' glacial intervals. Ages of ~800 Ma for Tsaliet equivalents in Eritrea provide an upper age constraint. The range of measured Sr isotopic compositions best overlaps Sturtian portions of available reference curves (ca. 720-750 Ma or a little older) and strongly supports the prospect that the Negash diamictite is the product of a Sturtian glacial episode.
PP33B-0944 1340h
Modeling of Seawater Geochemistry in Snowball Earth-Hothouse Earth Environments
Abundant evidence exists that during the Neoproterozoic era, the Earth was covered by glaciers and sea-ice into the tropics. This cold phase of Earth's geologic history has come to be known as "Snowball Earth," which interestingly alternated with unusually warm intervals, now called "Hothouse Earth." The cause(s) of these alternating phases of Earth's geologic history are controversial. For example, were these alternating climates caused by obliquity changes, continental drift, weathering, changes in the carbon cycle, or some combination of these factors? Remnants of these geologic times are clearly imprinted in geologic formations with banded iron formations and carbonate deposits overlying glacial tillites. Banded iron formations strongly suggest that seawater was anoxic in order for ferrous iron to have accumulated to concentrations high enough to precipitate distinct iron layers. Carbonate deposits suggest that the oceans were supersaturated with carbonate minerals. One of the mechanisms proposed for bringing the Earth out of these deep freezes is the release of greenhouse gases from gas hydrates that could have contributed to atmospheric warming. To better understand and constrain the possible interpretations of what occurred during these glacial/interglacial periods, we used the FREZCHEM model to simulate processes potentially controlling seawater geochemistry. The FREZCHEM model is structured to predict the equilibrium aqueous geochemistry of chloride, sulfate, carbonate, and nitrate salts, strong acids, and gas hydrates over the temperature range from $<$-70 to 25°C and the pressure range from 1 to 1000 bars. During the Snowball Earth phase, defining seawater solids are likely to have been siderite(FeCO3), methane hydrate, and ice. Warming of a Snowball Earth to a Hothouse Earth and reacting with oxygen causes the conversion of siderite to ferrihydrite [Fe(OH)3] and the precipitation of dolomite. Only shallow gas hydrate deposits ($<$ 1 km) are likely to have influenced global warming during these alternating climate shifts.
PP33B-0945 1340h
Radiocarbon ages and biomarker compositions of organic matter in size-fractionated sediments from the Washington Margin
In order to examine relationships between sources of organic matter, molecular compositions, diagenetic alteration, and particle size in modern marine depositional regimes, seven surface sediments from depths of 70 to 2700 m were collected along an E-W transect from the mouth of the Columbia River, across the Washington continental margin, to the Cascadia Basin in the N.E. Pacific. The sediments were separated by grain size ($>$250, 100-250, 63-100, 38-63 and $<$38 um) and, using SPLITT (split flow thin cell) fractionation for the finest particles, by sinking rate ($>$1 m/d and $<$ 1m/d). This method was recently tested and validated using a variety of natural sediments [1]. Results indicated that many biogeochemical processes governed by particle dynamics could be examined using this approach.Consistent with earlier reports [2], sediments from the Washington shelf are dominated by large particles ($>$250 um) whereas fine particles ($<$38 um) are predominant in slope and Cascadia Basin sediments. With the exception of inner-shelf stations, most TOC resides in the fine fraction and ranges around 1.5-3%. Radiocarbon ages of TOC ranges from 1190 yrs BP to 4320, with older ages distant from the coast. Organic carbon in the finest grain-size fractions is systematically youngest. The more rapidly sinking split ($>$ 1m/d) is younger than the < 1m/d fraction. Stable carbon isotopic measurements on TOC indicate an increase in delta 13C values from the shelf to the Cascadia Basin (from -25 to -22 permil, suggesting a transition from terrestrial to autochthonous marine sources. The results of biomarker composition in grain-size fractions and bulk-phase organic matter will also be presented with those from radiocarbon and in delta13C values fractions. References [1] Coppola, L., Gustafsson, Andersson, P., Larsson, J. and Axelsson, P. (submitted) Limnol. Oceanogr. Methods. [2] Keil, R.G., Tsamakis, E., Fuh, C.B., Giddings, J.C. and Hedges, J.I. (1994) GCA 58, 879-893
PP33B-0946 1340h
High Resolution Calcareous Nannofossil Fluctuations Across the Oligocene/Miocene Boundary, ODP Hole 1168A
Ocean Drilling Program (ODP) Leg 189 drilled five sites around Tasmania to investigate the timing and nature of the development of the Tasmanian Seaway. The Oligocene/Miocene interval at Site 1168, located on the western continental slope of Tasmania, was relatively expanded with an average linear sedimentation rate of 5 cm/k.y. A high resolution study of calcareous nannofossils was conducted at a 5 k.y. sample interval. Fluctuations in nannofossil assemblages, diversity, absolute abundance and flux were calculated across the Oligocene/Miocene interval at Site 1168. These data were combined with physical property data to further elucidate changes across this interval. Multivariate statistical analysis was conducted on several time slices through the 2 m.y. interval of interest in an effort to identify paleoceanographic fluctuations.
PP33B-0947 1340h
The Spectral Evolution of Plio-Pleistocene $\delta^{18}$O
Using a new stack of 57 benthic $\delta^{18}$O records, the evolution of $\delta^{18}$O over the last 5.3 Myr is described in terms of mean, variance, spectral density, and the shape of glacial cycles. The variance of $\delta^{18}$O exhibits an exponential increase with time and a strong response to long-term modulations in orbital forcing. From 5.3--1.7 Ma, the detrended variance of $\delta^{18}$O in the 41-kyr band has a correlation of 0.94 with obliquity modulation. The precession components of $\delta^{18}$O are also sensitive to orbital modulations, with the effects of modulation in the 23-kyr band increasing after 1.5 Ma. However, variance in the 100-kyr band of $\delta^{18}$O is anticorrelated with modulation of the 100-kyr eccentricity cycle and is not unique to the late Pleistocene. The dominance of the 100-kyr cycle in the late Pleistocene might represent the extension of trends which began before the onset of northern hemisphere glaciation (NHG), namely a steady increase in the 100-kyr variance and its sensitivity to 100-kyr power in eccentricity. Neither the onset of NHG nor the mid-Pleistocene revolution is associated with unique or abrupt transitions in the spectral response of $\delta^{18}$O relative to insolative forcing. However, a climate transition at 1.6 Ma results in a sudden change in the shape of glacial cycles, a trend toward warmer interglacials, changes in the climate's sensitivity to modulations of obliquity and precession, and a greater lag in the precession response of benthic $\delta^{18}$O. The initiation or strengthening of a climate feedback sensitive to precession may have triggered this transition.
PP33B-0948 1340h
Intensified North Atlantic Deep Water During Middle Pliocene Warm Period?
We examined the hypothesis of intensified thermohaline circulation/North Atlantic Deep Water (NADW) production during the Middle Pliocene warm period (3 - 3.3 Ma) by reconstructing deep ocean temperature from four sites (3.0 - 4.4 km water depth) in the western equatorial Atlantic using ostracode Mg/Ca ratios. Results indicate an overall steeper thermal gradient compared to modern conditions along the depth transect. Much of the shift is due to warmer water temperatures at the shallower sites, a pattern of warming similar to that observed in the deep North Atlantic during the warm early Holocene. Such a pattern is consistent with an intensification of NADW production relative to modern conditions. In addition to the shift in mean temperatures, the results suggest orbital-scale variability in deep Atlantic temperature during the middle Pliocene, especially at the shallower sites. This in turn suggests further intensification, as well as intermittent weakening, of NADW during the middle Pliocene warm period.
PP33B-0949 1340h
Sedimentary Nitrogen Stable Isotopes and Variations in Nutrient Cycling in the Holocene Black Sea
Interpreting the evolution of organic matter production and preservation in anoxic basins such as the Holocene Black Sea depends on developing an understanding of changes in nutrient cycling within the water column with time. The organic fraction of sediments may preserve evidence of such changes in nutrient utilization. One model proposes changes in phosphorus availability as a driver for changes in algal productivity in the Black Sea. Nitrogen, the other macronutrient commonly considered to limit algal growth, is the focus of this study as we examine the nitrogen content and stable isotope variations of Black Sea sediments to determine what role it may have played in temporal changes in productivity and organic matter accumulation. High-resolution samples from five gravity cores collected by the RV Knorr 1988 expedition were analyzed for $\delta$$^{15}$N-total and $\delta$$^{13}$C-organic as well as their percent composition of organic carbon and total nitrogen. One core, GC71, was subjected to sequential extractions with KCl and hydrogen peroxide to remove exchangeable ammonia and labile organic matter respectively. The KCl extraction did not remove a statistically significant amount of ammonia, having no measurable effect on the percent nitrogen or $\delta$$^{15}$N of the solid samples. The hydrogen peroxide extraction removed ca. 95% of the organic carbon and 85% of the nitrogen, leaving a relatively nitrogen-enriched residual material, probably due to ammonium fixed within the clay lattice. The fixed nitrogen has a minimal effect on the bulk nitrogen isotope values, suggesting the bulk nitrogen values are similar to the organic nitrogen signal. All cores examined were from below the modern Black Sea chemocline and are thought to have remained under anoxic bottom water continuously since soon after the incursion of saline Mediterranean water ca. 7800 years ago. Water depths for these cores range from 411 meters along the south margin of the sea to 2088 meters in the eastern Black Sea basin, and the samples analyzed span the past 10,000 years. This study incorporates new nitrogen isotopic data, higher resolution carbon isotopic data, and C/N ratios with previously published bulk organic carbon, Rock-Eval pyrolysis, regional climate, and molecular organic geochemical data from other sources. The results of this study reveal three intervals in sedimentary units I, IIa, and IIb, preserved in multiple cores, with less enriched $\delta$$^{15}$N values from +0.3 to +1.5 permil, separated by positive excursions with $\delta$$^{15}$N values between +3 and +4.5 permil. These intervals are the result of basin-wide processes as they occurred synchronously across the Black Sea. $\delta$$^{13}$C values suggest a primarily marine source for organic matter throughout units I and II, and fluctuations of $\delta$$^{15}$N do not seem to correlate with regional vegetation or precipitation changes; thus, assuming a relatively constant $\delta$$^{15}$N for inorganic and organic nitrogen riverine influx, changes in $\delta$$^{15}$N are due primarily to marine activity. We believe the lower $\delta$$^{15}$N values in units I and IIa were caused by decreases in water column denitrification due to a deeper chemocline and greater oxygen availability for respiration. The lower $\delta$$^{15}$N values in unit IIb may be due to nitrogen fixation caused by the release of excess phosphorus from the sediments due to the bottom water anoxia.
PP33B-0950 1340h
Age Models for Hemipelagites and Turbidites From the Cap Timiris Channel off Mauritania
This study on sediments from a submarine channel system demonstrates that geochemical data allow reliable age-depth correlation even of highly turbiditic cores and is an attempt to improve our understanding of how turbidity processes are related to sea-level / climatic changes. The recently active Cap Timiris Channel System incises the continental margin off NW-Africa and is characterised by turbidite transport. Gravity cores from three levee sites and one intra-channel location (GeoB 8509) were analysed applying sedimentological and geochemical methods. The hemipelagites at Site GeoB 8509 are interrupted to approximately 45 % by turbidites. Downcore X-ray fluorescence analyses were performed on all cores in order to fit age models by correlating their elemental data to dated reference cores, once turbiditic sequences had been removed from the dataset. These age models, in turn, enabled us to reliably estimate emplacement times of the intercalated turbidites. Core GeoB 8502-2 from a levee position in the lower reaches of the Cap Timiris Channel System dates back to about 300 kyr. Except for one turbidite in the Holocene, turbidites in this core are linked to glacial stages. The highly turbiditic core GeoB 8509-2, on the other side, postdates to about 15 kyr and comprises Holocene and late Pleistocene sediments, including the African Humid Period (14.8-5.5 kyr) and the Younger Dryas cooling event. Turbidites in this core are associated especially with the rapid sea-level rise at the Pleistocene/Holocene transition, whereas they are less frequent subsequently. The three most recent turbidite events occurred approximately 2.5, 1.7, and 1.2 kyr ago, respectively.
PP33B-0951 1340h
The San Francisco Bay Biota Since the Last Sea Level Maximum: Comparing Fossil and Recent Assemblages of Benthic Foraminifera
Comparisons of fossil and Recent foraminiferal assemblages in San Francisco Bay can give us information about how the Bay's ecosystems have changed or remained resilient during the glacial cycles of the Pleistocene. This information, in turn, can help us to predict the potential impact of future environmental change on this system. In this study, we compared the species proportions, overall similarity, and species diversity of the Recent foraminiferal assemblage at one site in South San Francisco Bay with the fossil foraminiferal assemblage from Pleistocene samples taken at several sites in South San Francisco Bay, in order to test the hypothesis that the foraminiferal assemblage in San Francisco Bay has not changed significantly since the last interglacial interval. Comparisons of total sample similarity indicate that the upper layer of the Yerba Buena mud was deposited during conditions similar to today's South San Francisco Bay. The presence of a large number of dead tests of Elphidium gunteri Cole and Elphidiella hannai (Cushman and Grant), together with the lack of living populations of these species in South San Francsico Bay, make it difficult to interpret the presence of these taxa in similar abundances in the Plestocene and today. Possibly due to the introduction of the Japanese foraminifer Trochammina hadai Uchio, the modern assemblage has shifted away from the dominance of Elphidium excavatum (Terquem) found in the Pleistocene. Species diversity has neither decreased nor increased between the Pleistocene and the present day in South San Francisco Bay, suggesting that environmental change and human activity has not affected species diversity of the benthic foraminiferal fauna. These data show that comparisons between fossil and Recent assemblages are powerful tools in interpreting paleoenvironments, and may help us to understand the impact of human activity on estuarine and other marine systems. More research is needed on the living populations of E. gunteri and E. hannai today to be able to interpret the significance of these populations in the Pleistocene bay.
PP33B-0952 1340h
Diatom, Total Organic Carbon, And Sediment Geochemistry Record From Southampton Bay, A Brackish Tidal Marsh In Northern San Francisco Bay, California
Sediments from a 3.8-m-long core collected from the marsh in Southampton Bay in northern San Francisco Bay, California, record both locally and regionally controlled variations in the flow of the Sacramento-San Joaquin river system through the Carquinez Strait. Because this river system drains more than 40 percent of California, variation in its discharge is an indication of changes in the magnitude and timing of precipitation in the region. Sedimentation rates at the site range from a high of 2.5 mm/yr from about 3,800 to 3,300 cal yr B.P. to about 0.8 mm/yr from about 1,800 cal yr B.P. to the present. Relatively high levels (20-30 percent) of total organic carbon (TOC) indicate that the primary interval of marsh accretion was between 3,300 and 300 cal yr B.P. A rapid decrease in TOC about 300 years ago suggests a decrease in marsh accretion, non-accretion, or erosion. Prior to the establishment of the marsh, between 3,800 and 3,300 cal yr B.P., and after about 300 cal yr B.P., overall accretion rates were lower, TOC values fell below 10 percent, and freshwater diatoms comprised less than 20 percent of the assemblage. These lower accretion rates occurred during periods of increased river flow which resulted in seasonal erosion and/or limited deposition of sediment. During the interval of marsh accretion, the abundance of freshwater diatoms parallels the TOC data, with freshwater taxa accounting for between 60 and 70 percent of the assemblage. The diatom and TOC data were supplemented with analysis of the silt and clay fraction using inductively coupled plasma-mass spectrometry and cold-vapor atomic absorption spectrometry. During the interval of marsh accretion high TOC, Na, Ca, and Mg values are higher, whereas during the intervals of lower TOC, the amount of Al, K, Fe, and Ti are higher. It is suggested that marsh accretion results in a greater contribution of sediments from local sources ({\it i.e.} Great Valley sequence and mafic units). During periods of lower rates of marsh accretion a greater part of the sediment budget is derived from weathered granitic sources in the Sacramento and San Joaquin Valleys. Values for Hg and Ag were more than an order of magnitude above background levels between 100 and 150 years ago, indicative of sediments deposited during the period of hydraulic mining in the central Sierra Nevada.