The configuration of Alpine accumulation areas during the last glacial maximum (LGM) has been reconstructed using glacial–geological mapping. The results indicate that the LGM ice surface consisted of at least three major ice domes, all located south of the principal weather divide of the Alps. This implies that the buildup of the main Alpine ice cover during oxygen isotope stage (OIS) 2 was related to precipitation by dominant southerly atmospheric circulation, in contrast to today's prevalent westerly airflow. Such a reorganization of the atmospheric circulation is consistent with a southward displacement of the Oceanic Polar Front in the North Atlantic and of the associated storm track to the south of the Alps. These results, combined with additional paleoclimate records from western and southern Europe, allow an interpretation of the asynchronous evolution of the different European ice caps during the last glaciation. δ18O stages (OIS) 4 and 3 were characterized by location of the Polar Front north of 46°N (Gulf of Biscay). This affected prevailing westerly circulation and thus, ice buildup in western Scandinavia, the Pyrénées, Vosges, and northern Alps. At the LGM, however, the Polar Front lay at ∼44°N, causing dominating southerly circulation and reduced precipitation in central and northern Europe.

The surficial geology of the tectonically stable Bahamian archipelago preserves one of the most complete records of middle to late Quaternary sea-level-highstand cycles in the world. However, with the exception of deposits from marine isotope substage (MIS) 5e, fossil corals for radiometric dating of this rich stratigraphic sequence are rare. This study utilizes the previously published, independent lithostratigraphic framework as a testing ground for amino acid racemization in whole-rock limestone samples. At least six limestone–soil couplets provide a relative age sequence of events that encompass as many interglacial–glacial cycles. D-Alloisoleucine/L-isoleucine data fall into six clusters, or “aminozones.” On the basis of independent dating and the inferred correlation with global MIS, the ages of several aminozones are known, while the ages of others are calculated from calibrated amino acid geochronology. This study demonstrates the utility of the whole-rock aminostratigraphy method for dating and correlating widespread emergent marine deposits, constitutes the first regional geochronological framework for the Bahamas, and highlights major sea-level events over the past half million years.

A succession of stratigraphic codes (1933, 1961, 1983) has guided attempts to refine classifications and naming of stratigraphic units for Quaternary deposits of the Great Lakes region. The most recent classifications for the late Quaternary of the Lake Michigan lobe (1968) and the eastern Great Lakes (1972) have been widely used, but later work has created the need for revision. An attempt has been made to integrate the two previous classifications following the diachronic system of the 1983 Code of Stratigraphic Nomenclature. A new nomenclature for the higher, more broadly recognized units was presented in 1997. We here present the diachronic nomenclature for finer subdivisions recognized in the eastern and northern Great Lakes. Following the interglacial Sangamon Episode, the three parts of the Wisconsin Episode are further subdivided as follows: the Ontario Subepisode (former Early Wisconsinan) comprises the Greenwood, Willowvale, and Guildwood phases; the Elgin Subepisode (former Middle Wisconsinan) comprises the Port Talbot, Brimley, and Farmdale phases; and the Michigan Subepisode (former Late Wisconsinan) consists of Nissouri, Erie, Port Bruce, Mackinaw, Port Huron, Two Creeks, Onaway, Gribben, Marquette, Abitibi, and Driftwood phases. Succeeding interglacial time to the present is the Hudson Episode.

The volume and surface area of glacial Lake Agassiz varied considerably during its 4000-year history. Computer models for seven stages of Lake Agassiz were used to quantify these variations over the lake's early history, between about 11,000 and 9300 14C yr B.P. (ca. 13,000 to 10,300 cal yr B.P.). Just after formation of the Herman strandlines (ca. 11,000 14C yr B.P.), the volume of Lake Agassiz appears to have decreased by >85% as a consequence of the abrupt rerouting of overflow to its eastern outlet from its southward routing into the Mississippi River basin. This drainage released about 9500 km3 of water into the North Atlantic Ocean via the Great Lakes and Gulf of St. Lawrence. Following closure of this eastern routing of overflow, the lake reached its maximum size at about 9400 14C yr B.P. with an area of >260,000 km2 and a volume of >22,700 km3. A second major reduction in volume occurred shortly after that, when its volume decreased >10% following the opening of the Kaiashk outlet to the east into the Great Lakes, and 2500–7000 km3 of water was released into the North Atlantic Ocean. These discharges may have affected ocean circulation and North Atlantic Deep Water production.

Uplift of the Tibetan Plateau is manifest not only in widespread denudation, but also by an increased deposition rate of sediment, near or far from the exhumed regions. Our results indicate that the mass accumulation rate (MAR) of eolian dust increased between ∼1.1 and ∼0.9 myr ago. We associate this increase in MAR and median grain size with uplift of the Tibetan Plateau and its adjacent regions during this period. This Middle Pleistocene uplift can also be evidenced by the age of volcanism in the marginal region, the existence of thick conglomerate deposits surrounding the uplifted plateau, and the increased sedimentation rate of lacustrine deposits in the Qaidam Basin (northeastern Tibetan Plateau) between ∼1.1 and ∼0.9 myr ago. The correlation between the loess and marine records indicates that after ∼0.9 myr ago, these two records correlate well. This good correlation probably suggest that the Middle Pleistocene upheaval event not only brought the plateau into the cryosphere, but also enhanced the coupling of regional-scale Chinese loess transportation and deposition to the global ice volume variations through its effects on glacial grinding, rock denudation, and east Asian monsoonal circulation.

Glacial sediments and landforms preserved beside the Strait of Magellan record repeated advances of an outlet of the Patagonian Icefield during and following the last glacial maximum (LGM; ∼25,000–14,000 14C yr B.P.). Ice-marginal landform assemblages consist of thrust moraine complexes, kame and kettle topography, and lateral meltwater channels, very similar to those found at the margins of modern subpolar glaciers. Taken together with other forms of paleoenvironmental evidence, the landform assemblages show that, during the LGM and late-glacial time, permafrost occurred near sea level in southernmost South America. This finding implies that mean annual temperatures were ∼7–8°C lower than at present, somewhat lower than those reconstructed by current glacier–climatic models. Comparison with precipitation–temperature relationships for modern glaciers suggests, in addition, that precipitation levels were lower than today. Reduced glacial-age precipitation may have resulted from a precipitation shadow induced by the Patagonian Icefield, an equatorward migration of the average position of westerly cyclonic storm tracks in the southern midlatitudes, or both these factors.

Glacial lake sediments and glacial geomorphology in Valle de las Morrenas, a glacial trough on the north face of Cerro Chirripó, Costa Rica, provide evidence on high-altitude Pleistocene conditions in Central America. The most recent glacier in the valley (Chirripó stage I) receded very rapidly near the end of the Younger Dryas chronozone. Radiocarbon dates on basal organic sediments from lakes beneath upper, middle, and lower limits of that glacier fall close together, and two-sigma calibrated ages overlap for the period 9700–9600 cal yr B.P. Earliest datable transition sediments from the central lake date to 12,360–11,230 cal yr B.P. Larger, older moraines, and associated trimlines, allowed reconstruction of three paleoglaciers (Chirripó stages II, III, and IV). Computer analysis of hypsometry using published tropical-glacier vertical mass balance profiles yields ELAs of 3506–3523, 3515–3537, and 3418–3509 m, respectively; Chirripó II ELA-estimate positions applied to Chirripó I yield an ELA of 3538–3546 m. We infer minimal temperature depressions of 7.4–8.0°C for the Chirripó I–IV stages. Modeling the behavior of modern tropical glaciers yields basinwide net accumulation estimates of 440–620, 550–830, and 960–1760 mm yr−1 for the Chirripó II, III, and IV stages.

Geomorphic and chronological evidence from Cornwall Island in the Canadian High Arctic Archipelago provides direct evidence for the age and dynamics of the center and northern flank of the Innuitian Ice Sheet that covered the islands during the Late Wisconsonian glacial maximum. Dispersal of erratics and glacial landforms indicate that ice flowed north across the island and converged with ice flowing northwest from Norwegian Bay. Cornwall Island was initially deglaciated at 9000 14C yr B.P. in near synchrony with widely separated sites in adjacent parts of the archipelago. This regional chronology suggests rapid breakup of a marine-based Innuitian Ice Sheet that was destabilized by rapid eustatic sea-level rise and ice thinning during the early Holocene. This evidence provides strong support for a recently proposed ice divide spanning the central part of the Canadian High Arctic and indicates that most, if not all, of the region was glaciated during the Late Wisconsinan.

In order to examine high-frequency variations of East Asian winter monsoon in Quaternary climatic extremes, two typical loess–paleosol sequences in the Chinese Loess Plateau were investigated. Sandy layers in the loess deposits, the “Upper sand” and “Lower sand” (layers L9 and L15, respectively), which represent a high-resolution record of paleomonsoon changes, have been sampled at intervals of 5–6 cm from sections at Luochuan and Xifeng. The grain size and magnetic susceptibility was measured for all samples. The grain-size results (a proxy of winter monsoon strength) indicate that the winter monsoon strength fluctuated on a millennial timescale during cold climatic extremes, with climatic events of a few hundred to a few thousand years. However, the winter monsoon was relatively stable during warm periods. The magnetic susceptibility signal (a proxy of summer monsoon intensity) is practically constant over the same period. This is tentatively explained by the assumption that the summer monsoon intensity was too low to be recorded in the magnetic susceptibility signal. The intensified winter monsoon events show periodicities in a range of 1000 to 2770 yr, with a dominant cycle of approximately 1450 yr. The detection of this oscillation in older glacial stages strongly suggests that it may be a pervasive cycle of the cold climatic phases of the Quaternary. Millennial-scale variations of the winter monsoon may be caused by instability of the westerly jet, which is determined by temperature differences between the polar and the equatorial regions.

Radiocarbon dated lacustrine sequences in Perú show that the chronology of glaciation during the late glacial in the tropical Andes was significantly out-of-phase with the record of climate change in the North Atlantic region. Fluvial incision of glacial-lake deposits in the Cordillera Blanca, central Perú, has exposed a glacial outwash gravel; radiocarbon dates from peat stratigraphically bounding the gravel imply that a glacier advance culminated between ∼11,280 and 10,990 14C yr B.P.; rapid ice recession followed. Similarly, in southern Perú, ice readvanced between ∼11,500 and 10,900 14C yr B.P. as shown by a basal radiocarbon date of ∼10,870 14C yr B.P. from a lake within 1 km of the Quelccaya Ice Cap. By 10,900 14C yr B.P. the ice front had retreated to nearly within its modern limits. Thus, glaciers in central and southern Perú advanced and retreated in near lockstep with one another. The Younger Dryas in the Peruvian Andes was apparently marked by retreating ice fronts in spite of the cool conditions that are inferred from the ∂18O record of Sajama ice. This retreat was apparently driven by reduced precipitation, which is consistent with interpretations of other paleoclimatic indicators from the region and which may have been a nonlinear response to steadily decreasing summer insolation.

Core OL-97A, retrieved from the depocenter of Owens Lake, represents a depositional history spanning the past 1000 yr. Among the 17 elements analyzed in the acid-leachable fractions of 315 salt-free samples (at ∼3 yr/sample), Mg and Li, which come chiefly from authigenic Mg-hydroxy-silicates, were found to have concentration variations reflecting lake salinity and climatic changes during the past. A total of 231 isotopic measurements on carbonates from the same samples in the upper 181 cm show that δ18O and δ13C values range from −5.66 to 0.12‰ (PDB) and 1.38 to 4.28‰ (PDB), respectively. The rate of change with time in δ18O records the rate of change in lake's volume due to climate fluctuations, whereas variations in δ13C reflect mainly variations in biological productivity, nutrient supply, and dissolved carbonate in the lake. Results indicate an effectively dry climate between A.D. 950 and 1220, corresponding to the Medieval Climatic Anomaly (a warm period in northern Europe), during which Owens Lake approached playa conditions. Wet climates prevailed during A.D. 1220–1480, producing relatively large and deep lakes. Beginning about A.D. 1550, the regional climate turned colder but had frequently oscillating precipitation. Six wet/dry cycles with ∼50-yr duration occurred between A.D. 1480 and 1760, during the later half of which Owens Lake became a playa. Since ∼A.D. 1880, the lake level has steadily dropped from its historic high stand under strong impact of human activity.

Late-Quaternary travertine at two sites near Stuttgart formed entirely during interglacial periods. The travertine contains structures from growth induced by bacteria, and such structures have been dated by 230Th/U mass spectrometry. The resulting ages from both sites imply growth episodes of short duration, with growth rates up to 5 mm yr−1, at 99,800 ± 1300 yr B.P. (2σ n = 8) and 105,900 ± 1300 yr B.P. (2σ n = 7). These episodes were likely part of marine isotope stage (MIS) 5.3. Deposition of silt interrupted travertine growth at one of the sites ∼105,000 yr B.P. Likely correlatives of this silt are the St. Germain I-B stade recorded in the Grand Pile peat bog and a cold episode ∼1000 yr long recorded by δ18O values in the GRIP ice core. Travertine also formed during stage 5.5 (∼115,000 yr) and during the early Holocene. We found no evidence for travertine accumulation in stages 2, 3, 4, and 5.1. At both sites, the Sr/U ratio and the initial 234U/238U activity ratio resemble those of modern spring water. However, the sites differ in the chemical composition of spring water and in stratigraphic sequence of travertine accumulation.

Rb concentrations, analyzed at 20-cm intervals from the Luochuan sequence of loess and paleosols, are sensitive to the loess–paleosol alternation controlled by monsoon climate. Because it is geochemically immobile, Rb can be well preserved in the loess–paleosol sequence after deposition, and its concentration depends mainly on properties of the winter monsoon-blown dust and on intensity of the summer monsoon-induced pedogenesis. A curvilinear relation has been developed between the measured Rb-concentration and the apparent sedimentation rate for the last glacial–interglacial cycle. This relation provides a time scale that corresponds well with the presently accepted ages for paleomagnetic reversals of Brunhes/Matuyama and Jaramillo events. With allowance for reduced Rb concentrations caused by early Pleistocene climate, the Rb-based time scale is also consistent with the boundary ages of other major paleomagnetic reversals of the past 2.58 myr.

In the west-central part of Lago Argentino, the Puerto Bandera moraines are clearly detached from longer, more prominent moraines of the last glaciation and from shorter and smaller Neoglacial moraines. Scientists have long speculated about the age of the Puerto Bandera moraines. Detailed geomorphologic studies in the western area of Lago Argentino, including stratigraphic profiles at Bahı́a del Quemado in the northern branch (Brazo Norte), indicate that the Puerto Bandera moraines were deposited by three pulses of ice. Each of the three pulses is represented by single moraine ridges and belts of tightly arranged ridges. The timing of the three glacier advances was established by radiocarbon dating, including data published by John Mercer. The oldest moraine system, formed during the Puerto Bandera I substade, was deposited ca. 13,000 14C yr B.P. Moraines of the Puerto Bandera II substade were deposited ca. 11,000 14C yr B.P. The youngest moraine system was deposited during a minor readvance, shortly before 10,390 C14 yr B.P., and thus appears to have occurred some time during the European Younger Dryas interval. After this third substade, the ice tongues retreated into the interior branches of Lago Argentino and have remained there since. Evidence found at Bahı́a del Quemado, together with data provided by other authors, attests to a significant climatic change by the middle Holocene, which we believe occurred during the Herminita advance, the first Holocene glacial readvance recognized within the area.

Stratigraphy and dating provide an 18,000 yr record of erosion and deposition by the middle Gila River. Prior to 18,000 cal yr B.P., the Gila River cut into its floodplain and created a deep, wide channel. Sand and gravel accumulated within this channel until 4250–4400 cal yr B.P. By this time, the channel of the Gila River was narrow and deep. Around 5000 cal yr B.P., fine-grained sediments began to accumulate on the floodplain. At 800–950 cal yr B.P., there was a major period of channel widening. After this erosional episode, the channel again narrowed as it filled with sand and gravel until around 200 cal yr B.P. On the floodplain, deposition continued from about 5000 cal yr B.P. to 500 cal yr B.P. A period of stability and soil formation occurred on the floodplain between 500 and 200 cal yr B.P., after which overbank deposition resumed and buried the soil. Channel widening again occurred in the late 19th century. The changes observed in the sedimentologic history of the Gila River show that during the late Pleistocene, the Gila River was a competent stream capable of carrying and depositing coarse sediment loads. Later deposition of sand and gravel during the first half of the Holocene implies an increase in sediment yield from upstream watersheds. Changes during the last 4000 years reflect the response of the river to climatic perturbations, the timing of large floods, internal landscape thresholds, and human impacts.

A variety of stratigraphic analyses (particle grain size, iron concentration, loss on ignition, and macrofossils) from sediments obtained from two marsh sites are used to reconstruct a middle to late Holocene record of stream flow into San Francisco Bay. Browns Island, a freshwater/brackish site, is located at the confluence of the Sacramento and San Joaquin rivers and is dominated by stands of Scirpus americanus. Peyton Hill is a brackish site located near Carquinez Straits and is dominated by stands of Scirpus robustus. Twenty-five AMS 14C dates provide chronostratigraphic control. During the Holocene, discharge from the Sacramento and San Joaquin rivers was broadly comparable to modern flows; however, an extended period of higher flow began 3800 cal yr B.P. and continued for almost two millennia. At this time Browns Island supported Phragmites communis, a freshwater species, and Peyton Hill supported S. americanus. At least two floods, recognized by discrete increases in sand and silt, occurred at 3600 and 530 cal yr B.P.

Analysis of multiple proxies shows that eastern equatorial Atlantic upwelling was subdued during isotope stage 5.5, more intense during stages 4, 5.2, 5.4, and 6, and most intense early in stage 2. These findings are based on proxy measures from a core site about 600 km southwest of Liberia. The proxies include total organic carbon content, the ratio of peridinoid and oceanic organic-walled dinoflagellate cyst species, accumulation rates of calcareous dinoflagellates, estimates of sea surface paleotemperatures, the difference in stable oxygen isotope composition between two species of planktonic foraminifera that live at different water depths, and the abundance of the planktonic foraminifera Neogloboquadrina dutertrei. Most of these parameters consistently vary directly or inversely with one another. Slight discrepancies between the individual parameters show the usefulness of a multiple proxy approach to reconstruct paleoenvironments. Our data confirm that northern summer insolation strongly influences upwelling in the eastern equatorial Atlantic Ocean.

A comparison is made of the Holocene records obtained from water isotope measurements along 11 ice cores from coastal and central sites in east Antarctica (Vostok, Dome B, Plateau Remote, Komsomolskaia, Dome C, Taylor Dome, Dominion Range, D47, KM105, and Law Dome) and west Antarctica (Byrd), with temporal resolution from 20 to 50 yr. The long-term trends possibly reflect local ice sheet elevation fluctuations superimposed on common climatic fluctuations. All the records confirm the widespread Antarctic early Holocene optimum between 11,500 and 9000 yr; in the Ross Sea sector, a secondary optimum is identified between 7000 and 5000 yr, whereas all eastern Antarctic sites show a late optimum between 6000 and 3000 yr. Superimposed on the long time trend, all the records exhibit 9 aperiodic millennial-scale oscillations. Climatic optima show a reduced pacing between warm events (typically 800 yr), whereas cooler periods are associated with less-frequent warm events (pacing >1200 yr).

Four muddy sand sheets occur within a tidal marsh peat at Swantown on the west coast of Whidbey Island, Washington. The two largest sand sheets pinch out about 100 m inland and became thinner and finer-grained landward. All four sand sheets contain marine microfossils and have internal stratification. They record repeated inundation of the marsh over a short time period by distinct pulses of sediment-laden ocean water, consistent with deposition by a tsunami wave train. The layers have been radiocarbon-dated to 1160–1350, 1400–1700, 1810–2060, and 1830–2120 cal yr B.P. The overlap in age between the two youngest layers and inferred great earthquake events at the Cascadia plate boundary, some 250 km to the west, suggests they were emplaced by tsunamis from this source area. The two older layers do not correlate with plate-boundary events. They may be products of tsunamis caused by earthquakes on local faults in the Strait of Juan de Fuca or by submarine landslides in this area.

Echo-sounding data recorded in the Black Sea in 1969 imaged a chain of hills 5- to 150-m high at a depth of 2000–2200 m that resemble hills on the lower continental rise. Like those hills, the features in the western Black Sea may have been created by bottom currents. The easterly flowing currents inferred to have formed the hills may be related to a catastrophic flood of the Black Sea from the Sea of Marmara 7150 yr ago.