Late Weichselian litho-, bio-, and chronostratigraphy (14C and varves) in southeastern Sweden provide a detailed picture of the deglaciation pattern and dynamics, shore displacement, late-glacial sedimentation, and history of the landscape, vegetation, and climate. Two plausible glacial models were tested against lithologic, chronologic, and climatic data. Permafrost at and outside the ice margin and topographic conditions beneath the ice apparently caused inward spread of frozen glacier-bed conditions. This led to a buildup of a large zone of debris-rich basal ice. A climatic amelioration about 12,700 yr B.P. changed the temperature profile in the ice sheet. Deposition of basal melt-out till began at the previously frozen glacier bed, and a rapid recession of the clean ice set in; thin exposed debris-rich basal ice which was separated from the active ice margin about 150 yr later. In this zone of stagnant ice there followed a 200– 300-yr period marked by subglacial and supraglacial melt-out and resedimentation, forming a large hummocky/transverse moraine. The mild climate favored rapid plant immigration, and a park-tundra was established. The gradual closing of the landscape was interrupted by a 100- to 150-yr period of tundra vegetation and a cool, dry climate, with local vegetational differences caused by differences in soil moisture. About 12,000 yr B.P. a second climatic amelioration set in, and during the next 1000 yr a birch (and pine) woodland gradually developed. Soils stabilized and Empetrum heaths became abundant as the climate gradually deteriorated at the end of this period. By 11,000 yr B.P. the area had become a tundra again with scattered birch stands, dominated by herbs such as Artemisia, Chenopodiaceae, grasses, and sedges. Some 500 yr later a birch/pine woodland again succeeded, and within about 500 yr the vegetation changed to a rather closed woodland as the climate ameliorated further. However, the time lag between climatic and vegetation change was considerable.

Detailed records of δ18O, δ13C, percentage and mass accumulation rate of CaCO3, and eolian percentage, mass accumulation rate, and grainsize generated for core RC11-210 from the equatorial Pacific reveal the timing of paleoclimatic events over the past 950,000 yr. The CaCO3 percentage record shows the standard Pacific correlation of high CaCO3 content with glacial periods, but displays a marked change of character about 490,000 yr ago with older stages showing much less variability. The carbonate mass flux record, however, does not show such a noticeable change. Sedimentation rates vary from about 0.5 to 3.0 cm/1000 yr and, during the past 490,000 yr, sections with enhanced sedimentation rates correspond to periods of high CaCO3 percentage. Eolian mass accumulation rates, an indication of the aridity of the source region, are usually higher during glacial times. Eolian grainsize, an indication of the intensity of atmospheric circulation, generally fluctuates at a higher frequency than the 100,000-yr glacial cycle. The mid-Brunhes climatic event centered at 300,000 yr ago appears as a 50,000-yr interval of low intensity and reduced variability of atmospheric circulation. Furthermore, the nature of this entire record changes then, with the younger portion indicating less variation in wind intensity than the older part of the record. The late Matuyama increase in amplitude of paleoclimatic signals begins 875,000 yr ago in the eolian record, 25,000 yr before the δ18O and CaCO3 percentage amplitude increases about 850,000 yr ago.

A marine terrace up to 6 m above present mean sea level is preserved along 15 km of the Kidnappers coast, Hawke's Bay, North Island, New Zealand. Radiocarbon dates for 10 shell samples collected from intertidal marine deposits overlying an uplifted shore platform have a total age range of 590 yr (from 2280 ± 50 to 2750 ± 70 yr B.P.). Sample age decreases landward suggesting a small relative sea-level rise, probably as a result of tectonic subsidence prior to uplift. Rates of contemporary shore platform formation preclude the preservation of a marine terrace subjected to constant, aseismic uplift. Preservation of a subhorizontal uplifted shore platform mantled by intertidal marine deposits with in situ fauna is best explained by abrupt coseismic uplift 2300 yr B.P. and preceded by tectonic subsidence. The presence of a single marine terrace does not permit the calculation of past earthquake magnitude or recurrence, but the NNE tilt of the 2300-yr-old marine terrace indicates continued growth of the north-trending late Pleistocene Kidnappers Anticline.

The rhyolitic Lake Tapps tephra was deposited about 1.0 myr ago, shortly after culmination of the early phase of the Salmon Springs Glaciation in the Puget Lowland. It is contained within sediments that were deposited in ponds or lakes in front of the reteating glacier. An herb-dominated tundra existed in the southern Puget Lowland at that time. Lake Tapps tephra is most likely the product of an eruption that in part was phreatomagmatic. It forms an early Pleistocene stratigraphic marker across the southern sector of the Puget Lowland and provides a link between Puget lobe sediments of the Cordilleran Ice Sheet and sediments deposited by Olympic alpine glaciers.

Stabilized parabolic dunes in the central St. Lawrence Lowland are oriented NE-SW, in the postulated direction of dune-building winds coming from anticyclonic air circulation induced by the retreating Laurentide Ice Sheet about 10,000 yr ago. The eolian chronology reconstructed from several sections in mixed dune-peatland environments indicates that postglacial plant colonization, characterized by a fortuitous assemblage of arctic-subarctic and boreal elements, preceded dune formation during Champlain Sea regression around 10,000 yr B.P. Confined peatlands and small forests were buried by eolian sands between 10,000 and 7500 yr B.P. under dry and temperate conditions. This eolian episode lasted about 2500 14C yr and ended when cyclonic air circulation similar to the present humid climatic regime was established following the breakup and disappearance of the Laurentide Ice Sheet over Hudson Bay and peripheral areas. Dune stabilization, through paludification of well-drained eolian sands about 7500 yr B.P., suggests a major shift in climate toward wetter conditions that have been characteristic during most of the Holocene in eastern North America. Minor eolian erosion induced by wildfire was recorded during late Holocene time (about 1250 yr B.P.). Anthropogenic perturbation (logging and agriculture practice) was also responsible for recent very local eolian activity.

Accelerator mass spectrometry 14C analyses of organic matter extracted from rock varnishes on morainal boulders yield limiting minimum ages for three crests of the Tioga glaciation. At Pine Creek in the eastern Sierra Nevada, varnish started to form on boulders of the outermost Tioga moraine before 19,000 yr B.P., and varnish originated on the innermost Tioga moraine before 13,200 yr B.P. Comparisons with lake-level, paleohydrological, paleoecological, colluvial, and rock varnish micromorphological data indicate that central-eastern California and western Nevada experienced a moisture-effective period during the late Pleistocene but after the Tioga maximum, and perhaps as Tioga glaciers receded from the mouth of Pine Creek canyon. Varnishes on Tahoeage morainal boulders at Pine Creek have cation-ratio ages of about 143,000–156,000 yr B.P., suggesting that the Tahoe glaciation should not be correlated with oxygen-isotope stage 4 in the early Wisconsin, but rather with stage 6. Varnishes on morainal boulders of an older glaciation at Pine Creek are dated by cation ratio at about 182,000–187,000 yr B.P.

Climatic changes in the Qinghai-Xizang Plateau of China were studied by analyzing the composition of peat and layers of sand and gravel distributed along the southern slopes of Nianqing-Tanggula and Gangdise Mountains, cross sections of deposits near a number of interior lakes in Xizang, past glacial variations on the southern slope of Nianqing-Tanggula Mountain, and landform changes south of the Yaluzangbu River. Such geologic evidence suggests a division of five climatic periods since the beginning of the Holocene: (1) The Wumadung interval, 10,000–7500 yr B.P., slightly cold and dry; (2) Qilongduo interval, 7500-3000 yr B.P., warm and moist; (3) the mid-Neoglacial period, 3000-1500 yr B.P., cold, except between 2500 and 200 yr B.P. when it was warmer; (4) the Dawelong interval, 1500-300 yr B.P., mild; and (5) the Little Ice Age, 300-0 yr B.P., cold. These changes progressed in a similar but not identical pattern as those in the northeastern part of China and in the northern region of Europe.

Thermoluminescence (TL) ages were obtained for five loess samples taken from the Zhaitang section near Beijing, China, using the coarse-grain quartz technique. The paleodose values have been determined by the method of total sample bleaching and regeneration of the TL growth curve. The method appears to be suitable for the age determination of loess samples up to about 150,000 yr where the annual dose-rate values are of the order 3–4 mGy/yr. This limit is a function of the total accumulated dose. The ages are in good agreement with those obtained by a fine-grain TL technique and are consistent with geological and geomagnetostratigraphic evidence.

Morphological and chemical properties of soils developed on moraines of granitic composition, and forested with lodgepole pine, in Bear Valley, Idaho, change significantly with age and slope position. Soil development on Pinedale and Bull Lake moraine slopes of similar curvature and steepness was assessed at the summits, shoulders, backslopes, and footslopes of both catenas, and at the toeslope of the Bull Lake catena. Many soil properties show trends in development with both age and slope position. These properties include clay content, clay film development, color, plagioclase weathering, and dithionite-citrate-extractable iron (Fed). The degree of development of all these properties is greater on the Bull Lake catena relative to that on the Pinedale catena. This trend reflects increased soil development with age. On both catenas, development of soil properties is commonly highest at the footslope or toeslope sites and minimum at the backslope site. Downslope changes are attributed to both colluviation and pedogenic processes, including sorting of sediments during downslope transport, greater accumulation of eolian materials transported to downslope sites by surface processes, and greater in situ weathering of mineral grains at the footslope and toeslope sites.

Morphological changes in Melanopsis shells and the sudden disappearance of Viviparus were observed in a continuous core, representing about 700,000 yr, taken from the L-12 corehole in the Hula Basin, northern Jordan Valley, Israel. Two major events in δ18O values are recorded. The first, at about 435,000 yr B.P., is accompanied by a shift in δ13C values and indicates a shallowing phase with slightly higher evaporation. This event is concurrent with the abundant occurrence of a monotypic “distorted” Melanopsis, suggesting an “ecophenotypic event” during deposition of the lake's Pleistocene lower Lacustrine beds. The second major isotopic event is evidenced in the lower part of the section as a gradual depletion in 18O toward the bottom of the corehole. Minor shifts of 1–20‰ in the oxygen isotope content along the core may represent changes in the δ18O values of the water from glacial to interglacial periods representing the major climatic cycles of the Pleistocene. Five of these cycles were observed along the core.

Three to seven oriented paleomagnetic samples were collected from 16 sites in the Nelson Bay and Bridgewater formations at Portland, Victoria, which contains the recently discovered Nelson Bay local fauna (L.F.). The entire section has reversed polarity. These results, along with Globorotalia truncatulinoides within the section, and the presence of underlying middle Pliocene-dated basalts, indicate that the Portland section, and the included Nelson Bay L.F., was deposited within the late Matuyama Chron between 1.66 and 0.73 myr ago. This represents the first well-documented pre-14C Pleistocene mammalian fauna in Australia calibrated in direct stratigraphic context with absolute dating methods. In addition, the reversed polarity for the Bridgewater Formation confirms the previous hypothesis that the depositional history of this beach-sand deposit is time-transgressive across the Brunhes-Matuyama boundary.

Theoretical investigations on models of the developmental processes of erosional features show that the time variations in erosional force have more sensitive influence on the time fluctuation of eroded material, such as sediments in lakes, than that of the quantity representing erosional features, such as volume of mountains. Time sequences of grain-size distributions from the Pleistocene sediments in Lake Biwa, central Japan indicate dominant periodicities of about 40,000 and 20,000 yr, very close to those predicted from the Milankovitch theory. The time variations of the grain-size parameters which may be indicators of erosional forces are very similar to that of the caloric summer insolation in the Northern Hemisphere.

Previous paleoecological research on the Southern High Plains resulted in development of a late Quaternary chronology of “pollen-analytical episodes” and the proposal that boreal forest existed in the region in the late Pleistocene. These interpretations are now considered untenable because (1) a number of the radiocarbon ages are questionable, (2) there are serious problems of differential pollen preservation and reproducibility of the pollen data, (3) there is an absence of supporting geological or paleontological data, and (4) the soils of the area contain none of the distinctive pedological characteristics produced by a boreal forest. Available data suggest that the region was primarily an open grassland or grassland with some nonconiferous trees through most of the Quaternary.

The chronostratigraphy of mostly Holocene sediments in the Northern Negev desert of Israel was studied through amino acid epimerization analysis (alloisoleucine/isoleucine ratio, A/I) of individual land snail shells and 14C analysis of bulk samples of land snail shells contained within the sediments. Analysis of 31 deposits shows a very strong correlation between A/I and 14C age. A/I analyses indicate that age mixtures occur within all colluvial and in some fluvial deposits. Consequently, radiocarbon dates of bulk samples, which give an average age, overestimate the time of deposition. Sedimentation rates were generally rapid in both fluvial and colluvial deposits. Colluviation shows a maximum in the early Holocene and decreases progressively thereafter, in accordance with the change in average rainfall. Accumulation of fluvial sediments shows a distinct mid-Holocene maximum and apparently relates to the interaction between variation in rainfall, extent of bedrock exposure, and vegetation density. Late Holocene fluvial deposition may relate to agricultural activity. The understanding of the chronology of sedimentation is dependent on an integrated approach employing amino acid epimerization analysis to study age variation within a layer and 14C analyses for establishment of an absolute chronology.

Selected radiocarbon data on surficial materials from the Lahontan basin, Nevada and California, provide a chronology of lake-level variation for the past 50,000 yr. A moderate-sized lake connected three western Lahontan subbasins (the Smoke Creek-Black Rock Desert subbasin, the Pyramid Lake subbasin, and the Winnemucca Dry Lake subbasin) from about 45,000 to 16,500 yr B.P. Between 50,000 and 45,000 yr B.P., Walker Lake rose to its sill level in Adrian Valley and spilled to the Carson Desert subbasin. By 20,000 yr B.P., lake level in the western Lahontan subbasins had risen to about 1265 m above sea level, where it remained for 3500 yr. By 16,000 yr B.P., lake level in the western Lahontan subbasins had fallen to 1240 m. This recession appears synchronous with a desiccation of Walker Lake; however, whether the Walker Lake desiccation resulted from climate change or from diversion of the Walker River is not known. From about 15,000 to 13,500 yr B.P., lake level rapidly rose, so that Lake Lahontan was a single body of water by 14,000 yr B.P. The lake appears to have reached a maximum highstand altitude of 1330 m by 13,500 yr B.P., a condition that persisted until about 12,500 yr B.P., at which time lake level fell ≧100 m. No data exist that indicate the level of lakes in the various subbasins between 12,000 and 10,000 yr B.P. During the Holocene, the Lahontan basin was the site of shallow lakes, with many subbasins being the site of one or more periods of desiccation. The shape of the lake-level curve for the three western subbasins indicates that past changes in the hydrologic balance (and hence climate) of the Lahontan basin were large in magnitude and took place in a rapid step-like manner. The rapid changes in lake level are hypothesized to have resulted from changes in the mean position of the jet stream, as it was forced north or south by the changing size and shape of the continental ice sheet.

Surface soil samples from the forested Chuska Mountains to the arid steppe of the Chinle Valley, Northeastern Arizona, show close correlation between modern pollen rain and vegetation. In contrast, modern alluvium is dominated by Pinus pollen throughout the canyon; it reflects neither the surrounding floodplain nor plateau vegetation. Pollen in surface soils is deposited by wind; pollen grains in alluvium are deposited by a stream as sedimentary particles. Clay-size particles correlate significantly with Pinus, Quercus, and Populus pollen. These pollen types settle, as clay does, in slack water. Chenopodiaceae-Amaranthus, Artemisia, other Tubuliflorae, and indeterminate pollen types correlate with sand-size particles, and are deposited by more turbulent water. Fluctuating pollen frequencies in alluvial deposits are related to sedimentology and do not reflect the local or regional vegetation where the sediments were deposited. Alluvial pollen is unreliable for reconstruction of paleoenvironments.

Three radiocarbon-dated cores from small lakes provide pollen records of postglacial vegetation change in the central Mackenzie River basin. These data are combined with previously published records to reconstruct vegetation development from central Alberta to the Mackenzie delta. The paleoecological data demonstrate that a continuous corridor of herb-and-shrub-dominated vegetation extended between Beringia and the Great Plains of North America from deglaciaiton until about 10,000 yr B.P. The initial vegetation was replaced by spruce forests between 10,000 and 8500 yr B.P. Significant vegetation change continued until approximately 500 yr B.P. when the muskeg-dominated landscape typical of the present subarctic was established. The initial herb-and-shrub vegetation was likely a reprise of conditions which existed during earlier Pleistocene deglaciations. These conditions may have provided a favorable environment for animals adapted to nonforested vegetation and promoted migration southward from Beringia and northward from the Great Plains during the last deglaciation and earlier Pleistocene deglaciations. The sequence of postglacial vegetation change in the Mackenzie basin likely reflects a combination of climatic change, edaphic change, and plant migration patterns and interactions.

The discovery, identification, and radiocarbon dating of the large achatinid land snail Limicolaria kambeul chudeaui at six places in northwestern Sudan, north of the 100-mm isohyet, provide convincing evidence that at least 300 mm annual rainfall existed there 6000 yr ago. Since then the rainfall north of ca. 20°N lat in the eastern Sahara has certainly been <300 mm and probably <200 mm. Accelerator mass spectrometry has allowed the organic fraction, presumably conchiolin, to be accurately radiocarbon dated whereas carbonate fractions are ca. 600 yr too old. Comparative studies show the subfossil form, L. k. chudeaui, to be the most primitive in its species complex and to occupy a narrow east-west belt across Africa north of 15°N and west of 35°E. The northern limit of living forms lies within the southern part of the Sahelian zone in forest or forest-savanna. They cannot live in open grassland. Therefore, the distribution of L. k. chudeaui marks the former northern position of this zone during the middle Holocene, indicating a latitudinal shift of at least 5° (500 km).

Study of Holocene ostracodes and diatoms from Elk Lake, in North-Central Minnesota, indicates that the local climate of the mid-Holocene can be subdivided into three intervals. Throughout interval 1 (ca. 7800 to 6700 yr B.P.), climate was colder and much drier than today. During intervals 2 and 3 (ca. 6700 to 4000 yr B.P.) average mean-annual air temperatures approached the modern mean (3.7°C), but warm summers persisted throughout interval 2, whereas during interval 3 warm summers fell into discrete episodes. Furthermore, average mean-annual precipitation was about 85 and 90% of modern during intervals 2 and 3, respectively. Transition times between the principal intervals were less than 50 yr. The expected effects of a retreating Laurentide Ice Sheet that initially maintained a winter-style circulation, followed by transitional climate states, and finally a near-modern circulation pattern may explain these local climatic events.

A 15-m sedimentary core from Lake Salpeten provides the first complete Holocene sequence for the lowlying Peten District, Guatemala. Today, Lake Salpeten is a brackish, calcium sulfate lake near saturation surrounded by tropical semievergreen forest. The basal pollen record depicts sparse juniper scrub surrounding a lake basin that held ephermal pools and halophytic marshes. The lake rapidly deepened to > 27 m in the early Holocene and may have been meromictic, because nearly 2 m of gypsum “mush” was deposited. Mesic forests were quickly established and persisted until the Maya entered the district 3000 yr ago and caused extensive deforestation. Any climatic information contained in the pollen record of the Maya period is thus masked, but a regional pollen sequence linked to the archaeological record is substantiated because environmental disturbance was pervasive. Local intensification of occupation and population growth are seen as an increased deposition of pollen of agricultural weeds and colluviation into the lake, while the Classic Maya collapse is marked by a temporary decline in Compositae pollen. Effects of perturbations induced by the Maya persist in the pollen and limnetic record 400 yr after the Spanish conquest.