Sedimentologic, rock-magnetic, and X-ray fluorescence data from two marine sediment cores in Hudson Strait suggest that a red, hematite-rich clay layer was deposited throughout the strait during the final collapse of the Laurentide Ice Sheet in the vicinity of northern Hudson Bay and western Hudson Strait. This layer, which can be recognized by its reddish-pink color (10YR6/2 to 5YR4/2) and relatively high-hematite proportions (low magnetic susceptibility and magnetite-to-hematite ratio), is dated from 8000 to 7900 14C yr B.P. at both ends of the strait. The Dubawnt Group, a Proterozoic bedrock unit in northern Hudson Bay, is the most likely source of this stratigraphic isochron. In eastern Hudson strait, the recognition of this red unit and other distal glaciomarine sediments from 8400 to 7900 14C yr B.P. indicates that little sediment from the nearby Labrador Dome reached eastern Hudson Strait during this 500-yr interval. This time interval immediately postdates the Noble Inlet advance, a northward flow of Labrador ice across eastern Hudson Strait onto southern Baffin Island from ca. 8900 to 8400 14C yr B.P. One explanation for the lack of Labrador sediments is that the northern margin of the Labrador dome was cold-based for up to 500 yr following the Noble Inlet advance.

Postglacial precipitation δ18O history has been reconstructed for two regions of Canada. Long-term shifts in the oxygen-isotope composition of annual precipitation (δ18Op) in southern Ontario appear to have occurred with a consistent isotope–temperature relation throughout the past 11,500 14C yr. The modern isotope–temperature relation in central Canada near present boreal treeline evidently became established between 5000 and 4000 years ago, although the relation during the last glacial maximum and deglaciation may also have been similar to present. In the early Holocene, however, unusually high δ18Op apparently persisted, in spite of low temperature locally, probably associated with high zonal index. A rudimentary sensitivity analysis suggests that a small reduction in distillation of moisture in Pacific air masses traversing the western Cordillera, perhaps accompanied by a higher summer:winter precipitation ratio, could have been responsible for the observed effect. Equivalent isotope–temperature “anomalies” apparently occurred elsewhere in western North America in response to changing early-Holocene atmospheric circulation patterns, suggesting that a time-slice map of δ18Op for North America during this period might provide a useful target for testing and validation of atmospheric general circulation model simulations using isotopic water tracers.

Using data from glacial geomorphology, tephra-soil stratigraphy and mineralogy, palynology, and radiocarbon dating, a sequence of glacial and bioclimatic stades and interstades has been identified for the past ca. 50,000 yr in the Ruı́z-Tolima massif, Cordillera Central. Six cold stades separated by warmer interstades occurred before 48,000, between 48,000 and 33,000, between 28,000 and 21,000, from ≥16,000 to ca. 14,000, ca. 13,000–12,400, and ca. 11,000–10,000 yr B.P. Although the radiocarbon ages are minimum-limiting ages obtained from tephra layers on top of tills, the tills are not significantly older because most are bracketed by dated tephra sets in measured stratigraphic sections. Two minor moraine stages likely reflect glacier pauses during cold intervals ca. 7400 yr B.P. and slightly earlier. Finally, glaciers readvanced between the 17th and 19th centuries. In contrast to the glacier cover (ca. 34 km2) on volcanoes of the massif during the last glacial maximum (LGM) the ice cover expanded to 1200 km2 and was still 800 km2 during late-glacial time (LGT). Glacier reconstructions based on the moraines suggest depression of the equilibrium line altitude (ELA) by ca. 1100 m during the LGM and 500–600 m during LGT relative to the modern ELA which lies at ca. 5100 m in the Cordillera Central. Glaciers in this region apparently reached their greatest extent when the climate was cold and moist, e.g., during stades corresponding to marine isotope stage 3; glaciers were still expanding during the LGM ca. 28,000–21,000 yr B.P., but they shrank considerably after 21,000 yr B.P. because of greatly reduced precipitation.

An extensive salt karst system has developed in Mount Sedom salt diapir, Israel, during the Holocene. Multilevel vadose caves were 14C dated using wood fragments embedded in alluvial deposits. The oldest date of each cave is used to constrain the age of the salt exposure. The upper portion of the southeastern escarpment was the first to rise above base level ∼7100 yr B.P. Caves in the surrounding area indicate gradual landscape exposure around this initial karstified area between 7000 and 4000 yr B.P. The northern part of the mountain experienced a similar exposure history, lagging some 3000 yr after the southern part. This lag may be attributed to the narrow width of the diapir in the north, which increases viscous drag at the borders of the rising diapir.

A10Be inventory and14C ages of material from a core from northernmost Illinois support previous interpretations that this area was ice free from ca. 155,000 to 25,000 yr ago. During much of this period, from about 155,000 to 55,000 yr ago, 10Be accumulated in the argillic horizon of the Sangamon Geosol. Wisconsinan loess, containing inherited 10Be, was deposited above the Sangamon Geosol from ca. 55,000 to 25,000 yr ago and was subsequently buried by late Wisconsinan till deposited by the Lake Michigan Lobe of the Laurentide Ice Sheet. The Sangamonian interglacial stage has been correlated narrowly to marine oxygen isotope substage 5e; our data indicate instead that the Sangamon Geosol developed during late stage 6, all of stages 5 and 4, and early stage 3.

High-resolution paleomonsoon proxy records from peat and eolian sand–paleosol sequences at the desert–loess transition zone in China denote a rapid oscillation from cold–dry conditions (11,200–10,600 14C yr B.P.) to cool–humid conditions (10,600–10,200 14C yr B.P.), followed by a return to cold–dry climate (10,200–10,000 14C yr B.P.). Variations in precipitation proxies suggest that significant climatic variability occurred in monsoonal eastern Asia during the Younger Dryas interval. Late-glacial climate in the Chinese desert–loess belt that lies downwind from Europe was strongly influenced by cold air from high latitudes and from the North Atlantic via the westerlies. The inferred precipitation variations were likely caused by variations in the strength of the Siberian high, which influenced the pressure gradient between land and ocean and therefore influenced the position of the East Asian monsoon front.

A calcium oxalate (whewellite)-rich crust occurs on exposed limestone surfaces in dry rock and open air shelters in the Lower Pecos region of southwest Texas. The crust, which also contains gypsum and clay, formed over silica-rich limestone during the Holocene. SEM and optical photomicrographs reveal similarities between whewellite microstructures and the lichen Aspicilia calcarea. This desert lichen is known to produce calcium oxalate, and has been found in several sites in the region. The ubiquity of the whewellite-rich crust in the Lower Pecos shelters suggests that the lichen flourished in the past. Since A. calcarea is a desert species, the virulence of the organism likely peaked during xeric climate episodes then waned during mesic periods. Thus, radiocarbon ages of whewellite would correspond to dry climate periods experienced in the region, while periods with few or no 14C data would indicate wet climate episodes. A preliminary paleoclimate reconstruction based on fourteen AMS14 C dates indicates the Lower Pecos experienced dry to wet climate fluctuations during the late Holocene. This reconstruction generally agrees with other models established for Texas.

Stable isotopes, geochemical, lithological, and micropaleontological results from cores from the far northwest (FNW) Pacific and the Okhotsk and Bering seas are used to reconstruct the regional environment for the last glaciation, the deglacial transition, and the Holocene. δ18O records of planktonic foraminifera of the region show two “light” shifts during deglacial time, provoked by the freshening of the surface water and climate warming. These north Pacific terminal events (T1ANP and T1BNP) with ages of 12,500 and 9300 yr B.P., respectively, occur almost simultaneously with two episodes of accelerated glacier melting around the North Atlantic. Along with the isotopic shifts, the CaCO3 content in regional sediments increased abruptly (1A and 1B carbonate peaks), probably due to changes of productivity and pore water chemistry of surface sediments. Organic matter and opal concentration increased during the transition (between T1ANP and T1BNP events) in the sediments of the FNW Pacific and the southern part of the Bering Sea and opal content increased in the Holocene in the Bering and Okhotsk Seas. δ13C records of cores from the Okhotsk and Bering seas and the FNW Pacific do not contradict the hypothesis of increased intermediate water formation in the region during glaciation. During deglaciation, accumulation of the coarse terrigenous component decreased in sediments of the Bering Sea and the FNW Pacific before the T1ANP event, probably as a result of rising sea level and opening of the Bering Strait.

Luminescence ages help provide a time scale for recurrent surface ruptures on a normal fault in the southern Arava, Israel. The dated samples come from alluvium and colluvium along a fault scarp north of Elat. The ages, obtained mainly by infrared stimulation of sand-size potassium feldspar grains, accord with geomorphic age estimates from soils and degraded scarps. Four large earthquakes occurred between 37,000 and 14,000 yr ago. Five smaller earthquakes occurred thereafter continuing until recent times. The mean recurrence interval for earthquakes of magnitudes greater than 6 is about 4000 yr.

A new and significant site of organic silty sand has been found beneath the Valders till at Valders Quarry in northeastern Wisconsin. This is now the earliest known late-glacial site associated with red till ice advances in the western Great Lakes area. Leaves of terrestrial plants washed into a small depression provide a date of 12,965 ± 200 yr B.P. (WIS-2293), which is significantly older than the Two Creeks Forest Bed (ca. 11,800 yr B.P.). Percentage and concentration pollen diagrams suggest that the site was open and distant from a closed Picea forest. No wood or Picea needles have been found. This date is statistically indistinguishable from 12,550 ± 233 yr B.P., the mean of three dates for the end of inorganic varve sedimentation at Devils Lake, 160 km southwest at the terminus of the Green Bay Lobe. Assuming that the Green Bay lobe vacated its outermost moraine in the interval from 13,000 to 12,500 yr B.P., only a short time was available for retreat of the ice margin over 350 km, drainage of red sediment from Lake Superior into the Lake Michigan basin, readvance of over 250 km, retreat of at least 80 km, and advance to this site. The time for these events appears to have been too short to resolve by current radiocarbon technique. This extremely rapid collapse of the Green Bay lobe has a calibrated age of about 15,000 cal yr B.P., about that of the dramatic warming seen in the Greenland ice cores.

Most conceptual models of atmospheric circulation deduced from bottom sediments or isopollen maps off NW Africa assume the occurrence of intensified NNE trade winds about 18,000 yr B.P. in latitudes of 28°–35°N, and oversimplify the glacial atmospheric circulation over Africa. An alternative method for reconstructing paleowinds of the last glacial maximum in these latitudes was recently put forward, and uses sedimentological records from the Canary Islands and coastal regions of Morocco. The continental data do not agree with the previous models and show the prevalence of westerlies. All the data from deep sea cores (reduction of sea surface temperature, increase of biogenic opal accumulation, distribution patterns of pollen or dinoflagellate cysts, and xeric conditions on the adjacent continent) can be explained without increased activity of the trade winds, but with a discharge current of cold meltwater from the European and North American ice sheets. The model is backed up by a comparison with the present-day Humboldt Current off subtropical South America.

Climate variability on the Yucatan Peninsula during the past 3500 yrs is reconstructed from the measurement of δ18O in monospecific ostracods and gastropods in a 6.3-m sediment core from Lake Punta Laguna, Mexico. This late Holocene record is divided into three periods based on changes in mean δ18O values. From ∼3310 to ∼1785 14C yr B.P. (Period I), low mean δ18O values indicate relatively wet conditions (i.e., low evaporation to precipitation ratio, E/P). Mean oxygen isotopic values increased ∼1785 14C yr B.P., and the interval between ∼1785 and ∼930 14C yr B.P. (Period II) was distinctly drier than the periods before or after. The climate during the latter part of Period II was persistently dry, with exceptionally arid events centered at ∼1171, 1019, and 943 14C yr B.P. (equivalent to 862, 986, and 1051 A.D.). This interval of frequent drought was recorded at several other localities in Mexico and Central America, and coincided with the collapse of Classic Maya civilization. Following the last arid event, δ18O values decreased abruptly at ∼930 14C yr B.P. (beginning of Period III), signaling a return to wetter conditions that have generally prevailed to the present, with the exception of a dry episode centered at 559 14C yr B.P. (1391 A.D.). The paleoclimatic record from Punta Laguna provides evidence that multi-decadal and millennial-scale changes in E/Poccurred on the Yucatan Peninsula during the late Holocene. These wet/dry episodes may have influenced cultural evolution in Mesoamerica.

Climatic reconstruction of glacial to interglacial episodes from oxygen isotopes in sediment cores from the Nordic seas is complicated by strong local meltwater contributions to the oxygen isotope changes. Combination of benthic and planktic foraminiferal isotope data with foraminiferal abundances and ice-rafted debris (IRD) allows separation of local and global effects and subdivision of the marine oxygen isotope events 6.2–5.4, which include the last interglaciation, into: (1) a meltwater phase after glacial stage 6, recorded by large amounts of IRD and low foraminiferal abundance, indicating surface water warming; (2) an IRD-free period with high deposition rates of subpolar foraminifera and other CaCO3 pelagic components, recognized here as the “full” interglaciation; and (3) a phase with the recurrence of IRD and the demise of subpolar species. Comparison of ice-core records and marine data implies that the global climate during the last full interglaciation and that during the postdeglacial Holocene were similar. The records show no significantly different variations in the proxy data. In contrast, the oxygen isotopes of planktic foraminifera and ice cores indicate significant differences during each of the deglacial transitions (Terminations I and II) that preceded these two interglaciations. These suggest that during Termination II the climatic evolution in the Nordic seas was less affected by abrupt changes in ocean–atmosphere circulation than during the last glacial to interglacial transition.

Modern ocean–atmosphere interactions in the northeastern Pacific Ocean have a significant effect on the climate of the west coast of North America. We present radiolarian microfossil-based temperature reconstructions for the eastern North Pacific spanning the past 20,000 yr to examine possible correlations and linkages between continental climate change and changes in sea surface temperature (SST) in the northeastern Pacific Ocean on millennial time scales. The reconstructions indicate that the regional pattern of ocean circulation off the west coast of North America was further south 15,000 cal yr B.P. than it is today, and reached its present location 13,000 cal yr B.P. The North Pacific Drift and Transition Zone were further south as a result of a more southerly North Pacific high pressure cell prior to 13,000 cal yr B.P. While two continental paleoclimate records from northwestern North America show regional differences, they also can be correlated to the SST changes. A coastal site at 48°N shows similar patterns in summer temperatures, as observed in offshore marine records of SSTs. However, an inland continental record seems to reflect more-regional-scale changes in sea surface conditions showing a thermal maximum centered at 10,000 cal yr B.P which is observed in the marine transect south of 42°N. We conclude, based on the pattern of oceanographic change as reflected in radiolarian assemblages, that changes in the past latitudinal position of the North Pacific Drift played a significant role in controlling continental climate immediately to its east, as it does in the present environment. We also conclude that during the past 20,000 yr much of the evolution of oceanographic change is related to the migration of the atmospheric pressure cells (the North Pacific high and Aleutian low) of the northeastern Pacific.

The ages of some tsunami deposits can be determined by optical dating, a key requirement being that the deposits are derived from sediment that was reworked and exposed to daylight by tidal currents, waves, wind, or bioturbation during the last years before the tsunami. Measurements have been made using 1.4 eV (infrared) excitation of K-feldspar grains separated from samples of prehistoric tsunami sand sheets and modern analogs of tsunami source sediments at four sites in Washington state and British Columbia. Source sands gave equivalent doses indicative of recent exposure to daylight. Tsunami sand at Cultus Bay, Washington, yielded an optical age of 1285 ± 95 yr (calendric years before A.D. 1995, ±1σ). At 2σ, this age overlaps the range of from 1030 to 1100 yr determined through a combination of high-precision radiocarbon dating and stratigraphic correlation. Tsunami sands at three sites near Tofino and Port Alberni on Vancouver Island, British Columbia, have optical ages of 260 ± 20, 325 ± 25, and 335 ± 45 yr. Historical records and radiocarbon dating show that the sand at each of the three sites is between 150 and 400 yr old. These optical ages support the hypothesis that the Vancouver Island sands were deposited by a tsunami generated by a large earthquake on the Cascadia subduction zone about 300 yr ago.

Mordern alluvial pollen varies with geomorphic setting and depositional facies in sediments of the Yellow, Hutuo, and Luan rivers and in Baiyangdian and Hengshuihu lakes. Most of the arboreal pollen is derived from the mountains, whereas most of the nonarboreal pollen is derived from the plain itself. Alluvium dominated by Pinus pollen and Selaginella spores was deposited during a flood. Hydrodynamic sorting of alluvial pollen exists in the sediments of floodplain, central bar, natural levees, and point bar. In reconstructing the ancient vegetation and past climate based on pollen in alluvium, it is important to consider sedimentary facies and geomorphologic setting.

Deep-sea sediments recovered from six sites visited during the International Arctic Ocean Expedition of 1991 were examined to determine sedimentation rates in the eastern Arctic Ocean basin. The dearth of age-diagnostic biogenic material in these sediments precludes the application of biostratigraphic methods, but ages can be deduced using paleomagnetism, in conjunction with measurements of radiocarbon and carbonate concentration. Although no one of these techniques gives an unambiguous determination of age, the interpretation most consistent with these diverse data implies that sedimentation rates in the eastern Arctic are, in general, a few centimeters per thousand years. Such estimates of sedimentation rate are an order of magnitude greater than those previously determined from many sediment cores taken from the Canada Basin. However, one site examined on the Morris Jesup Rise shows a relatively low rate of sediment accumulation (less than 0.6 cm/103 yr) suggesting that, although higher than in the Canada Basin, sedimentation rates in the eastern Arctic can be highly variable.

The greatest storm in the written history of California struck the region in the winter of 1861–1862. The unusual weather began on Christmas Eve, 1861, and persisted for some 45 days as a series of middle-latitude cyclones made landfall along the California coast. Episodes of very cold and very warm temperatures occurred both during the storm and in the spring of 1862 as meridional flow prevailed. Heavy precipitation swelled the Santa Ana River to more than triple the highest estimated discharge in this century. High water levels in coastal streams between Los Angeles and San Diego persisted into the spring. Lakes were created in the Los Angeles Basin and the Mojave Desert. Arroyos were cut. Sediments from the flood may be preserved in offshore basins.

The degree of soil development associated with geomorphic surfaces in the San Timoteo Badlands area (STB), California allows correlation of the surfaces. Soil development indices, based on field descriptions and laboratory analysis, provide a basis for comparison of these soils to each other and to dated soil chronosequences at Cajon Pass, Merced, and Anza, California. The soils in this study record a complex tectonic and climatic history, include preserved surfaces intermediate in soil development to those at Cajon Pass, and do not preserve a record of the major late Pleistocene to early Holocene aggradational event observed elsewhere in the region. Pedogenesis is similar to that of regional soils in southern California. With time, the profiles develop progressively thicker argillic horizons, and they increase in redness, clay, and secondary iron oxide content, and amount and thickness of clay films. With increasing soil age the ratio of dithionite-extractable Fe to total Fe increases, while the ratios of Fe2+to Fe3+and of oxalate- to dithionite-extractable Fe decrease. These trends are more subdued in the STB than elsewhere in the region, perhaps because the slightly higher temperature and lower precipitation of the STB may slow the rate of transformation of ferrihydrite to hematite. We use weighted means of soil development indices and rubification index values to compare the soils of this study to those at Cajon Pass, Merced, and Anza. Wide-range age estimates using the maximum limits of these values result in 27,500–305,000 yr for surface Q1, 43,000–570,000 yr for surface Q2, and 300,000–700,000 yr for surfaces Q3 and Q4. Weighing most heavily the nearest dated chronosequences, Anza and Cajon Pass, yields best estimates of age. This method constrains the time of formation of the Q1 surface to between 27,500 and 67,000 yr and the Q2 surface to 43,000–67,000 yr, and does not change the age estimates of the older surfaces.

The mid-Holocene eastern hemlock [Tsuga canadensis L. (Carr.)] decline has been recently attributed to the activity of insect defoliators. N. Bihiry and L. Filion, Quaternary Research 45,312–320 (1996). In this study, soil hydromorphic conditions were investigated for the period 6800–3200 yr B.P. using micromorphological data from a peat section from a swale in a paludified dunefield in southern Québec. After a short period of plant colonization in shallow pools between 6800 and 6400 yr B.P., mesic conditions predominated in the interdune before the decline (6400–4900 yr B.P.), as evidenced by strong bioturbation and abundance of excrements from the soil fauna. During the decline, a shift from mesic to wet conditions occurred (4900–4100 yr B.P.), although xeric to mesic conditions persisted on dune ridges until at least 4200 yr B.P. Wetness culminated when beaver occupied the site (4100–3750 yr B.P.). Hemlock needles with chewing damage typical of hemlock looper (Lambdina fiscellaria) feeding were identified at levels dated 4900, 4600, and 4200 yr B.P., respectively, implying that the hemlock decline was associated with at least three defoliation events. The ca. 400-yr interval between these events likely represents the time required for this late-sucessional tree species to recover.