A paleoclimate record for the past 250,000 years from Summer Lake, Oregon, USA. 1. chronology and magnetic proxies for lake level

This study presents the age control and environmental magnetism components of a new, late Pleistocene paleoclimate record for the Great Basin of western North America. Two new cores from the Summer Lake sub-basin of pluvial Lake Chewaucan, Oregon, USA are correlated to basin margin outcrops on the basis of tephrochronology, lithostratigraphy, sediment magnetism and paleomagnetic secular variation. Eleven tephra layers were found in the cores that correlate to tephra identified previously in the outcrop. The Olema ash was also found in one of the cores; its stratigraphic position, relative to 3 dated tephra layers, indicates that its age is 50-55 ka, somewhat younger than has been previously reported. The Summer Lake sediments are divided into deep and shallow lake lithosomes based on sedimentary features. The stratigraphic position of these lithosomes support the tephra-based correlations between the outcrop and the cores. These sediments contain a well resolved record of the Mono Lake Excursion (MLE) and an earlier paleomagnetic excursion as well as a high quality replication of the paleosecular variation immediately above the MLE.Relative sedimentation rates increased dramatically toward the depocenter during intervals of low-lake level. In contrast, during intervals of high-lake level, relative sedimentation rates were comparable along the basin axis from the basin margin to the depocenter. The magnetic mineralogy of the Summer Lake sediments is dominated by pseudo-single domain (titano)magnetite and intervals of high/low magnetite concentration coincide with lithosomes that indicate high/low lake levels. Magnetic grain size also varies in accord with bulk sediment grain size as indicated by the silt/clay ratio. To a first order, variations in magnetic parameters, especially those attributable to the concentration of magnetic minerals, correlate well with global glacial/interglacial oscillations as indicated by marine oxygen isotope stages. This relationship can be explained by increased dissolution of (titano)magnetite minerals as lake level dropped and the lake became more productive biologically. This inference is supported by a correspondence between lower concentrations of magnetite with higher levels of total organic carbon and vice-versa.     

Variation in the composition of Lake Bonneville marl: a potential key to lake-level fluctuations and paleoclimate

Lake Bonneville marl provides a stratigraphic record of lake history preserved in its carbonate minerals and stable isotopes. We have analyzed the marl in shallow cores taken at three localities in the Bonneville basin. Chronology for the cores is provided by dated volcanic ashes, ostracode biostratigraphy, and a distinctive lithologic unit believed to have been deposited during and immediately after the Bonneville Flood.A core taken at Monument Point at the north shore of Great Salt Lake encompasses virtually the entire Bonneville lake cycle, including the 26.5 ka Thiokol basaltic ash at the base and deposits representing the overflowing stage at the Provo shoreline at the top of the core. Two cores from the Old River Bed area near the threshold between the Sevier basin and the Great Salt Lake basin (the main body of Lake Bonneville) represent deposition from the end of the Stansbury oscillation ( 20 ka) to post-Provo time ( 13 ka), and one core from near Sunstone Knoll in the Sevier basin provides a nearly complete record of the period when Lake Bonneville flooded the Sevier basin (20–13 ka).In all cores, percent calcium carbonate, the aragonite to calcite ratio, and percent sand were measured at approximately 2-cm intervals, and ¹⁸O and ¹³C were determined in one core from the Old River Bed area. The transgressive period from about 20 ka to 15 ka is represented in all cores, but the general trends and the details of the records are different, probably as a result of water chemistry and water balance differences between the main body and the Sevier basin because they were fed by different rivers and had different hypsometries. The Old River Bed marl sections are intermediate in position and composition between the Monument Point and Sunstone Knoll sections. Variations in marl composition at the Old River Bed, which are correlated with lake-level changes, were probably caused by changes in the relative proportions of water from the two basins, which were caused by shifts in water balance in the lake.This is the second paper in a series of papers published in this issue on Climatic and Tectonic Rhythms in Lake Deposits.     

The ostracode record from Harris Lake, southwestern Saskatchewan: 9200 years of local environmental change

Holocene paleoenvironments of Harris Lake, southwestern Saskatchewan, are reconstructed from the ostracode stratigraphy of a 10.4 m sediment core. Twenty three taxa, representing nine genera, were identified and counted from 113 samples. At each depth, a theoretical faunal assemblage was derived from the raw counts. The mean and variance of chemical, climatic and physical variables were inferred from modern analogues of the fossil assemblages, using existing autecological data from 6720 sites, mostly in western Canada. These data suggest four paleoenvironments: an early-Holocene (9240–6400 years BP) variable climate supporting aspen parkland vegetation; the warm dry hypsithermal (6400–4500 years BP); a short transitional period of ameliorating climate and expanding subboreal forest (4500–3600 years BP); and the present environment since 3600 years BP. A change in regional climate with the draining of Glacial Lake Agassiz (ca. 8500 years BP) and landsliding in the watershed (ca. 4000 years BP) caused relatively rapid environmental change. The ostracode record generally corroborates the interpretations of other proxy data previously published for Harris Lake. Most of the discrepancy involves the timing and severity of maximum Holocene warmth and aridity. Peak aridity interpreted from the pollen data is earlier than in the other proxy records. Both the diatoms and ostracodes indicate highest paleosalinity between ca. 6500 and 5000 years BP, but maximum salinity in the diatom record occurs between ca. 6000–5700 years BP, whereas the ostracode-inferred salinity is relatively low at this time and peaks later at ca. 5000 years. Neither of these reconstructions suggests the short episodes of hypersalinity interpreted from the mineralogy.     

Stable isotopes and sediments from Pickerel Lake, South Dakota, USA: a 12ky record of environmental changes

Sedimentological parameters and stable O- and C-isotopic composition of marl and ostracode calcite selected from a 17.7-m-long core from the 8-m-deep center of Pickerel Lake, northeastern South Dakota, provide one of the longest (ca. 12ky) paleoenvironmental records from the northern Great Plains. The late Glacial to early Holocene climate in the northern Great Plains was characterized by changes from cold and wet to cold and dry, and back to cold and wet conditions. These climatic changes were controlled by fluctuations in the positions of the Laurentide ice sheet and the extent of glacial Lake Agassiz. We speculate that the cold and dry phase may correspond to the Younger Dryas event. A salinity maximum was reached between 10.3 and 9.5 ka, after which Pickerel Lake shifted from a system controlled by atmospheric changes to a system controlled by groundwater seepage that might have been initiated by the final withdrawal of Glacial Lake Agassiz. A prairie lake was established at approximately 8.7 ka, and lasted until about 2.2 ka. During this mid-Holocene prairie period, drier conditions than today prevailed, interrupted by periods of increased moisture at about 8, 4, and 2.2 ka. Prairie conditions were more likely dry and cool rather than dry and warm. The last 2.2 ka are characterized by higher climatic variability with 400-yr aridity cycles including the Medieval Warm Period and the Little Ice Age.Although the signal of changing atmospheric circulation is overprinted by fluctuations in the positions of the ice sheet and glacial Lake Agassiz during the late Glacial-Holocene transition, a combination of strong zonal circulation and strong monsoons induced by the presence of the ice sheet and high insolation may have provided mechanisms for increased precipitation. Zonal flow introducing dry Pacific air became more important during the prairie period but seems to have been interrupted by short periods of stronger meridional circulation with intrusions of moist air from the Gulf of Mexico. More frequent switching between periods of zonal and meridional circulation seem to be responsible for increased climatic variability during the last 2.2 ka.     

A diatom-based reconstruction of drought intensity, duration, and frequency from Moon Lake, North Dakota: a sub-decadal record of the last 2300 years

Diatom assemblages preserved in sediment cores from closed-basin lakes can provide high-resolution records of past hydrologic and climatic conditions, including long-term patterns in the intensity, duration, and frequency of droughts. At Moon Lake, a closed-basin lake in eastern North Dakota, a comparison of diatom-inferred salinity and the precipitation-based Bhalme-Mooley Drought Index (BMDI) over the last 100 years was highly significant, suggesting that the diatom record contains a sensitive archive of past climatic conditions. A sub-decadal record of inferred salinity for the past 2300 years indicates that extreme droughts of greater intensity than those during the 1930s 'Dust Bowl' were more frequent prior to A.D. 1200. This high frequency of extreme droughts persisted for centuries and was most pronounced from A.D. 200–370, A.D. 700–850 and A.D. 1000–1200. A pronounced shift to generally wetter conditions with less severe droughts of shorter duration occured at A.D. 1200. This abrupt change coincided with the end of the 'Medieval Warm Period' (A.D. 1000–1200) and the onset of the 'Little Ice Age' (A.D. 1300–1850).     

A 1500-year record of climatic and environmental change in ElkLake, Clearwater County, Minnesota II: geochemistry, mineralogy, and stableisotopes

Most of the sediment components that have accumulated in ElkLake, Clearwater County, northwestern Minnesota, over the past 1500 years areauthigenic or biogenic (CaCO₃, biogenic SiO₂, organicmatter, iron and manganese oxyhydroxides, and iron phosphate) and are deliveredto the sediment–water interface on a seasonal schedule where they are preservedas distinct annual laminae (varves). The annual biogeochemical cycles of thesecomponents are causally linked through the carbon pump, and are recapitulatedin longer-term cycles, most prominently with a periodicity of about 400 years.Organic carbon is fixed in the epilimnion by photosynthetic removal ofCO₂, which also increases the pH, triggering the precipitation ofCaCO₃. The respiration and degradation of fixed organic carbon inthe hypolimnion consumes dissolved oxygen, produces CO₂, and lowersthe pH so that the hypolimnion becomes anoxic and undersaturated with respectto CaCO₃ during the summer. Some of the CaCO₃ produced inthe epilimnion is dissolved in the anoxic, lower pH hypolimnion and sediments.The amount of CaCO₃ that is ultimately incorporated into thesediments is a function of how much is produced in the epilimnion and how muchis consumed in the hypolimnion and the sediments. Iron, manganese, andphosphate accumulate in the anoxic hypolimnion throughout the summer.Sediment-trap studies show that at fall overturn, when iron-, manganese-, andphosphate-rich bottom waters mix with carbonate- and oxygen-rich surfacewaters, precipitation of iron and manganese oxyhydroxides, iron phosphate, andmanganese carbonate begins and continues into the winter months.Detrital clastic material in the sediments of Elk Lake depositedover the last 1500 years is a minor component (<10% by weight) that ismostly wind-borne (eolian). Detailed analyses of the last 1500 years of the ElkLake sediment record show distinct cycles in eolian clastic variables (e.g.aluminum, sodium, potassium, titanium, and quartz), with a periodicity of about400 years. The 400-yr cycle in eolian clastic material does not correspond tothe 400-yr cycles in redox-sensitive authigenic components, suggesting that theclastic component is responding to external forcing (wind) whereas theauthigenic components are responding to internal forcing (productivity),although both may ultimately be forced by climate change. Variations in theoxygen and carbon isotopic composition of CaCO₃ are small but appearto reflect small variations in ground water influx that are also driven byexternal forcing.     

Record of Late Pleistocene Glaciation and Deglaciation in the Southern Cascade Range. II. Flux of Glacial Flour in a Sediment Core from Upper Klamath Lake, Oregon

During the late Wisconsin, glacial flour from alpine glaciers along the east side of the Cascade Range in southern Oregon was deposited in Upper Klamath Lake. Quantitative interpretation of magnetic properties and grain-size data of cored sediments from Caledonia Marsh on the west side of the lake provides a continuous record of the flux of glacial flour spanning the last 37 000 calendar years. For modeling purposes, the lake sediments from the 13-m core were divided into three sedimentary components defined from magnetic, geochemical, petrographic, and grain-size data. The components are (1) strongly magnetic, glacial flour made up of extremely fine-grained, fresh volcanic rock particles, (2) less magnetic lithic material made up of coarser, weathered volcanic detritus, and (3) non-magnetic biogenic material (largely biogenic silica). Quantitative interpretation is possible because there has been no significant postdepositional destruction or formation of magnetic minerals, nor alteration affecting grain-size distributions. Major steps involved in the interpretation include: (1) computation of biogenic and lithic components; (2) determination of magnetic properties and grain-size distributions of the non-glacial and glacial flour end-members; (3) computation of the contents of weathered and glacial flour components for each sample; (4) development of an age model based on the mass accumulation of the non-glacial lithic component; and (5) use of the age model and glacial flour contents to compute the flux of glacial flour. Comparison of the glacial flour record from Upper Klamath Lake to mapped glacial features suggests a nearly linear relation between flux of glacial flour and the extent of nearby glaciers. At 22 ka, following an extended period during which glaciers of limited size waxed and waned, late Wisconsin (Waban) glaciers began to grow, reaching their maximum extent at 19 ka. Glaciers remained near their maximum extent for 1000 years. During this period, lake sediments were made up of 80% glacial flour. The content of glacial flour decreased as the glaciers receded, and reached undetectable levels by 14 ka.     

A 2500 year sediment record from Fayetteville Green Lake, New York: evidence for anthropogenic impacts and historic isotope shift

A series (N = 12) of short (< 1 m) sediment cores were collected from meromictic Green Lake in Fayetteville, New York to investigate potential anthropogenic impacts on the watershed during historic time and environmental change over the past ~ 2,500 years. Stratigraphic data document an abrupt basinwide change during the early 1800's A.D. from brown laminated sediments to grey varved sediments separated by a transition zone rich in aquatic moss. Deforestation of the region by European settlers during the early 1800's A.D. resulted in a flux of nutrients and increased biological productivity followed by a 7fold increase in sediment accumulation rates. Elemental geochemical data document the anthropogenic loading of lead to the to the lake basin via atmospheric fallout. Stable oxygen isotope (¹⁸O calcite) data also provide evidence for an abrupt shift in the isotopic composition of lake water ~ 150–200 years ago. This isotopic shift could have been a local phenomenon related to an increased supply of summer enriched precipitation following removal of forest vegetation, or it might have reflected broader scale climatic changes. We hypothesize that the ¹⁸O calcite shift was the result of the polar front jet stream migrating from a more southerly prehistoric position to a contracted, northerly configuration ~ 150–200 years ago. Such a shift could have been natural, associated with the end of the Little Ice Age or it may have been anthropogenically forced.     

A record of vegetation dynamics and lake level changes from Lake Emakat,northern Tanzania,during the last c. 1200 years

Analyses of down-core variations in pollen and charcoal in two short cores of lake sediment and wood samples taken from the in situ remains of Nuxia congesta from Lake Emakat, a hydrologically-closed volcanic crater lake occupying the Empakaai Crater in northern Tanzania, have generated evidence of past vegetation change and lake level fluctuations. Eight AMS radiocarbon (¹⁴C) dates on bulk samples of lake sediment provide a chronological framework for the two cores and indicate that the sediment record analysed incorporates the last c. 1200 years. The in situ remains of a Nuxia congesta tree, now standing in deep water, were dated with three additional AMS ¹⁴C dates, suggesting tree growth within the interval ∼1500–1670 AD. Down-core variations in pollen from terrestrial taxa, particularly the montane forest trees Hagenia abyssinica and Nuxia congesta, indicate a broad period of generally more arid conditions in the catchment to c. 1200 AD and at a prolonged period between c. 1420 and 1680 AD. Variations in pollen from plants in lake margin vegetation indicate low lake levels, presumably as a result of reduced effective precipitation, contemporary with indications of relatively dry conditions mentioned above, but also during the late 18th and the late 19th centuries. The presence of charcoal throughout both cores indicates the frequent occurrence of vegetation fires. An increase in burning, evident in the charcoal data and dated to the early to mid second millennium AD, could relate to an expansion of human population levels and agricultural activity in the region.