Four glacial drifts that are interstratified with lava flows and tephra layers on the upper slopes of Mauna Kea demonstrate that an ice cap formed repeatedly at the summit of the volcano during the middle and late Pleistocene. The oldest drift (Pohakuloa Formation) probably was deposited shortly after eruption of a lava flow having a K/Ar age of 278,500 ± 68,500 yr. Drift of the Waihu Formation, marked by a belt of subdued end moraines, is correlated with hyaloclastite cones and associated lava flows that were erupted beneath an ice cap about 170,000–175,000 yr ago. One of four younger subglacially erupted lavas at the crest of the volcano has a K/Ar age of 41,300 ± 8300 yr. Tephra layers that antedate the last glaciation are about 29,700 to 37,200 14C yr old and underlie dune sand that is believed to correlate with drift of the Makanaka Formation deposited during the last ice advance. The late Makanaka ice cap, which covered an area of about 70 km2 and was as much as 100 m thick, is reconstructed from end moraines and limits of erratic stones that encircle the summit region. The ice cap disappeared from the summit before about 9080 yr ago. Postglacial lavas and tephra overlie the youngest drift on the upper south flank of the mountain and buried a widespread post-Makanaka soil on the lower south rift zone about 4500 14C yr ago. The island of Hawaii is subsiding isostatically due to crustal loading by Quaternary volcanic rocks, with subsidence near the midpoint of Mauna Kea estimated as about 2.5 ± 0.5 mm/yr. A curve depicting an inferred long-term subsidence rate has been used to adjust equilibrium-line altitudes (ELAs) of former ice caps that are calculated on the basis of reconstructed glacier topography and an assumed accumulation-area ratio of 0.6 ± 0.05. The results indicate that ELA depression was greatest during Waihu glaciation, least during Pohakuloa glaciation, and that the ELA during late Makanaka glaciation was somewhat lower than that of the early Makanaka advance. Available radiometric dates show that late Makanaka glaciation correlates with stage 2 of the marine oxygen-isotope record, and suggest that early Makanaka, Waihu, and Pohakuloa glaciations correlate, respectively, with isotope stages 4, 6, and 8. Because ice caps could have formed on Mauna Kea only after the snowline was lowered many hundreds of meters below its inferred present level, episodes of Hawaiian glaciation probably were restricted to times of maximum ice volume on the continents. The asymmetry of the late Makanaka ice cap and the southeast-descending gradient of its equilibrium line are consistent with a southeast (tradewinds) source of precipitation during the last glaciation. Although departures of glacial-age temperature and precipitation from present values are difficult to assess quantitatively, growth of former ice caps on Mauna Kea most likely was due to enhanced winter snowfall and to reduced ablation rates brought about by lower air temperature and increased cloudiness.

Estimates made of lowland precipitation and evaporation in Britain during the last (Devensian) glaciation suggest that during the cold periods of the Devensian precipitation was probably low, between 260 and 370 mm/thermal maximum of the Upton Warren Interstadial lowland precipitation was probably in the range 450 to 650 mm/year. Two summer precipitation regimes are identified during the cold periods, one with high values and the other with low. The high summer precipitation variant leads to moist conditions with July and August precipitation values similar to those at the present day, and global circulation models suggest that the moist regime may have existed at the time of the maximum advance of the ice sheets. On the other hand, the low summer precipitation variant leads to a dry summer with wind action creating aeolian deposits, and this variant probably existed at earlier times in the glacial period. About 6500 yr B.P., in the Atlantic period, forest conditions probably caused increased evaporation which more than compensated for the increased precipitation of the time, causing low runoff conditions. The clearance of British forests by man since 6500 yr B.P. has probably led to an absolute increase in runoff values even though precipitation values have fallen.

Dated macrofossil evidence documents the widespread occurrence of woodland in what are now desert lowlands of southwestern North America from the last pleniglacial (ca. 20,000 yr B.P.) to late glacial/Holocene transition (12,000–8000 yr B.P.). The composition of the Pleistocene woodlands indicates that they had already differentiated geographically in modern form, though immensely more extensive than today. The pinyon-juniper woodland (Pinus monophylla, Juniperus osteosperma) of the Mohave Desert province had not yet penetrated the central Great Basin, but extended from southern Nevada south through the vast lowlands of the Mohave and westernmost Sonoran Deserts to southeastern California and Baja California. The strongly xerophytic Mohavean woodland was characterized by a very well-marked altitudinal and latitudinal zonation with juniper-Joshua tree (Yucca brevifolia) sorting out below pinyon-juniper woodland, and with live oaks restricted to the upper level along the lower Colorado River drainage. Southeastward, the Sonoran Desert province was similarly zoned, but with the more slender-leaved Pinus edulis var. fallax as pinyon and with more live oaks in the upper zone. However, the pleniglacial woodland of the Chihuahuan Desert province was almost unzoned, inasmuch as the less xerophytic species of pinyon and live oaks prevailed over the entire span of available elevation; the pinyon was the very slender-leaved P. cembroides var. remota.

The overall paleozonation indicates a strong northwest-to-southeast gradient of increasing summer rain with decreasing distance from the monsoonal source area over the Gulf of Mexico, as at present, but augmented pluvially along the same gradient. A key piece of evidence is the counterintuitive latitudinal-zonational anomaly between about 30 and 40° N in southwestern North America; the lower limits of modern vegetational zones are depressed with decreasing latitude (e.g., ca. 500 m lower at 34° than at 36° N). The axis of the gradient actually extends from northwest to southeast, paralleling the monsoonal gradient of increasing summer rain, which no doubt causes the apparent anomaly. During the Wisconsinan glacial, the latitudinal anomaly was greatly steepened, a fact requiring a pluvial increase in precipitation over the Southwest. The monsoonalpluvial pattern is supported by the Neotoma record of a northwest-to-southeast gradient of increasing diversity of evergreen oaks requiring summer rain, and by a parallel segregation of pinyon species. Equability of seasons during the last glacial is also suggested by the Neotoma macrofossil data.

Possible physical mechanisms relating orbital-element variations (i.e., the “Milankovitch mechanism” of insolation regime changes) to Quaternary glacial/interglacial transitions are explored quantitatively. These include the seasonal cycle of albedo and the zonal distribution of thermal inertia. These mechanisms can interact with the perturbations to zonal average and seasonal average insolation caused by orbital-element variations to cause a global annual temperature residual, even though such variations can cause only a negligible change in global annual insolation.

Numerical model experiments with a zonal energy balance model show that the relative interactions between insolation regime changes and seasonally and latitudinally varying albedo and latitudinally varying thermal inertia are roughly of comparable magnitude. Encouraging agreements between model experiments and data are evident, but these (and others') simulations are still a long way away from providing a satisfying explanation of the physical processes that could fully explain the apparent connections between orbital-element variations and Quaternary glaciations.

It seems likely that no single physical process can be identified as predominate, and rather, the hypothesized physical connection between insolation regime changes and glacial/interglacial transitions will have to be built on the interactions of a number of processes on both short and long time scales.

Measurements of saturation isothermal remanent magnetization (SIRM) for recent ombrotrophic peats show that the rate of deposition of magnetic particulates from the atmosphere varies strongly in response to the persistent, small-scale, spatial changes in vegetation and microtopography characteristic of many raised bogs. Hummock environments may trap over an order of magnitude more magnetic particulates than do Sphagnum lawns and pools. These results, thought to reflect the interception of subhorizontally moving particles by the hummock forming plants, have potentially important implications for studies of pollen deposition (“influx”) rates in peat.

The eight massive Pleistocene explosive volcanic eruptions which occurred in western North America produced rhyolitic ash layers estimated to have covered from 0.38 to 2.76 × 106 km2 of the western and central portions of the continent. The surface albedo increases in the Northern Hemisphere resulting from these light-colored ash covers varied from around 0.06 to 0.41% assuming ash albedos based on color of around 53 to 65%. These albedo increases resulted in hemispheric temperature decreases of from around 0.07° to 0.41°C with greater cooling in and adjacent to the ash-covered regions. Such albedo-induced temperature declines lasted for at least several decades and reenforced the substantial posteruption cooling caused by volcanic aerosols and by a feedback decrease in atmospheric precipitable water. The magnitude and critical location of these temperature declines may have contributed to summer snow survival in the sub-Arctic plateaus and to a consequent triggering of major Pleistocene glaciations.

This report considers the scientific opportunities and problems for paleoclimatic research for three time intervals, 0-1000 yr, 0-30,000 yr, and 0-1 my. Each of these intervals is appropriate to understanding different parts of the climate problem. The most recent interval is one for which there exists the possibility of determining year-by-year changes in certain parts of the climatic system over a time span which overlaps the instrumental record. The interval 0 to 30,000 yr is one for which the response of many parts of the climate system during a major change in climate can be examined, and for which laC dating provides good chronological control. The interval 0 to 1.000,000 yr is one for which there is reasonably good chronological control. a well-developed global record of changes in the ocean, and the opportunity of investigating glacial-interglacial changes in global climate which appear to have a cyclic or quasi-cyclic component.

A pollen analytical investigation of the sediments of Berry Pond, Berkshire County, Massachusetts, has demonstrated a sequence of pollen assemblage zones similar to those detected elsewhere in New England. From about 13,000 to 12,000 yr B.P. the vegetation of the region was treeless, probably tundra. By 11,500 yr tundra had been replaced by open boreal forest. Closed boreal forest became dominant by 10,500 yr. Boreal forests were replaced by mixed coniferous and deciduous forests with much white pine about 9600 yr ago. A “northern hardwoods” complex with much hemlock, beech, and sugar maple succeeded the mixed forests 8600 yr ago. Hemlock declined very rapidly approximately 4800 yr ago and was replaced by birch, oak, beech, ash, and red maple. This decline may have been biologically rather than climatically induced. There is a slight maximum of pine (much of it pitch pine) from 4100 to 2600 yr ago, perhaps indicative of warmer and/or drier conditions. There were slight changes in the forests about 1600 yr ago as chestnut immigrated and spruce and larch increased slightly. European land clearance and subsequent land abandonment are detectable in the uppermost levels.

Haploxylon pine(s) and Artemisia dominated the initial vegetation in front of the receding Okanogan Lobe until ca. 10,000 yr B.P., as revealed by two pollen records in north-central Washington. After 10,000 yr B.P. the macroclimate became warmer throughout the Okanogan drainage as diploxylon pines and Artemisia increased. The Mount Mazama eruption at ca. 6700 yr B.P. is recorded as two stratigraphically separate and petrographically distinct tephra units at Bonaparte Meadows. While there are apparent short-term changes in the vegetation coincident with the ashfall(s), Artemisia continues to dominate the Okanogan Valley until ca. 5000 yr B.P. By 4700 yr B.P. the modern vegetation, dominated by Pseudotsuga menziesii, had become established around Bonaparte Meadows.

Spatial anomalies of tree-ring chronologies can provide information on high-frequency spatial anomalies in paleoclimate representing droughts, colder-than-normal intervals, and other synoptic-scale features. Examples are presented in which 65 tree-ring chronologies are calibrated with spatial anomalies in North American meteorological records of seasonal temperature and precipitation, and with sea-level pressure over the North American and North Pacific sectors. Multivariate transfer functions are obtained that scale and convert the past spatial variations in the tree-ring record into estimates of past variations in the meteorological record. Objective verifications of the reconstructions are obtained using independent meteorological observations for time periods other than those used in the calibration. Historical information or other proxy data from the 19th century are also used for verifying the decadal (or longer) and regional reconstructions and for identifying strengths and weaknesses of the various sources of information. The reconstructed winter and summer temperatures for the United States and southwestern Canada and winter precipitation for the Columbia Basin and California during the 17th through 19th centuries were found to differ from the 20th century means with large-scale variations evident. Extreme winters similar to 1976–77 are also identified and found to be more frequent in the past, especially in the 17th century. The climatic reconstructions in this time domain are dominated by high-frequency, synoptic-scale fluctuations that can be interpreted as cyclonic-scale changes in atmospheric circulation. Such reconstructions may be useful for testing various climatic models and estimates developed primarily from 20th-century meteorological data against the longer estimated record for the 17th through 19th centuries.

A 21-kg sample of plant detritus from the Two Creeks forest bed at the junction of Kewaunee and Manitowoc Counties, Wisconsin, has produced a small, but well-preserved fauna of Coleoptera, Hymenoptera, and Diptera. Among the Coleoptera the majority of Carabidae (ground beetles) indicate moderately moist to fairly dry, open ground conditions with a sparsely vegetated substrate. Another significant beetle group is represented by the Scolytidae (bark beetles) nearly all of which inhabit spruce. The restricted number of Coleoptera species (21) is believed to represent an in situ assemblage and is part of about 50 taxa. Most of the Coleoptera identified to the species level have a transcontinental distribution, with the exception of two carabids which are extreme western forms, and nearly all reside within the boreal forest. The beetles indicate that the mean July temperature at Two Creeks approximately 11,850 yr B.P. was 14° to 16°C.

The development of reliable paleoclimatic maps at a global scale requires data at the following three levels of analysis: (1) well-recorded observations of evenly positioned, well-dated geological evidence (Level I), (2) paleoclimatic estimates derived from this evidence by well-defined quantitative repeatable methods (Level II), and (3) maps synthesizing the estimates from several independent sources of geological evidence (Level III). Our paper describes much of the currently available paleoclimatic data from unglaciated terrestrial areas at ca. 18,000 yr B.P. and illustrates the quantity and quality of the data at both the Level I and the Level II stages of analysis. Although the scarcity of well-dated evidence for this time period precluded any major Level III syntheses of the information, comparisons were drawn where possible between the geological evidence and the climatic conditions simulated by general-circulation model experiments of Gates (1976a, b) and Manabe and Hahn (1977). Of the more than 320 sites with data from 18,000 yr B.P., only 65 are well-dated with bracketing dates within the interval of 23,000 to 13,000 yr B.P., whereas about 100 are undated or poorly dated. We concentrated our survey on palynological and paleobotanical evidence and also thoroughly reviewed the evidence for water levels in lakes at 18,000 yr B.P. In areas with few of these sources of evidence, data on former snowlines, periglacial features, and eolian deposits were included, but the survey of these data is far from complete. Maps of the assembled data reveal the consistency of the paleoclimatic estimates in “data-rich” areas and also show which areas required additional information. The maps show that conditions were colder than present at 18,000 yr B.P. for all sites with temperature estimates. Estimated temperature depressions varied from ca. 1° to 12°C or more, depending on the location of the sample, the type of geological evidence, and the method of temperature estimation. Interpreted hydrological conditions were more variable spatially than the temperature estimates. The southwestern U.S. was moister than present, whereas the southeast may have been drier. Europe and the northern Mediterranean across to Afghanistan were drier than present, but northwest Africa was wetter. Australia was mainly drier than present, but several sites there as well as in Africa show significant climatic changes between 21,000 year period and 16,000 yr B.P. This latter evidence suggests that considerable variability may have occurred during the several thousand-centered on 18,000 yr B.P. Accurate time control is therefore required for the geological data used to study the climate dynamics of 18,000 yr ago. Large portions of South America and Asia as well as significant portions of the other continents lack the data base, or at least the well-dated base, required to define the 18,000 yr B.P. climate. In the few areas where comparisons were made with the Ice Age climates simulated by general-circulation models, general agreement existed between the geological evidence and the model simulations. Many critical comparisons were thwarted, however, by the lack of model simulations for all seasons at 18,000 yr B.P. Difficulty in validating precipitation anomalies in the tropics also arose because surface-albedo values, which are a vital input to the general circulation models, are estimated from the same evidence that is used to validate the results of the models.

Pollen and plant macrofossils preserved in lake sediment from Lake West Okoboji, Dickinson County, Iowa, indicate how the vegetation of that area changed during the late glacial and postglacial. A closed coniferous forest, dominated by spruce and larch trees, produced the Picea-Larix pollen assemblage zone. Fir trees were a minor constituent of this forest; pine trees were probably absent. Black ash trees increased in abundance at Lake West Okoboji and by 13,500 yr ago were an important constituent of the forest. The sediment accumulation rate and the pollen influx were low throughout this time. Birch and alder pollen peaked in abundance approximately 11,800 yr ago. Pollen influx increased rapidly as birch and alder replaced coniferous trees on the uplands. A deciduous forest, containing abundant oak and elm trees, replaced the birch-alder-coniferous forest. This forest inhabited northwestern Iowa from approximately 11,000 to 9000 yr B.P. Nonarboreal species became prevalent between approximately 9000 and 7700 yr B.P. as prairie began to replace deciduous forest on the uplands. Charred remains of Amorpha canescens and other upland species attest to the presence of prairie fires as an aid in establishing prairie and destroying the forest. The pollen influx declined. The warmest, driest part of the postglacial occurred in northwestern Iowa from approximately 7700 to 3200 yr ago. Lake level fell 9 to 10 m, and prairie extended to the edge of the lake. Wet-ground weeds inhabited areas near lake level which were alternately flooded, then dry. Pollen influx was approximately 100 grains/cm2/yr during the driest time in this dry interval.

Deciduous trees, particularly oaks, returned after approximately 3200 yr B.P. Prairie continued to occupy the uplands but trees were more common in the lowlying wet areas. Settlement by Europeans in northwestern Iowa about 1865 is marked by an increase in weed pollen. Macrofossil deposition changed in 1910 in response to the stabilization of lake level.

This paper presents selected world maps of lake-level fluctuations since 30,000 yr B.P. These are based on a literature survey of 141 lake basins with radiocarbon-dated chronologies. The resulting patterns are subcontinental in scale and show orderly variations in space and time. They reflect substantial changes in continental precipitation, evaporation, and runoff, which are due to glacial/interglacial fluctuations in the atmospheric and oceanic circulations. In the tropics, high lake levels are essentially an interglacial or interstadial phenomenon, although there are important exceptions. Since extensive lakes during the Holocene corresponded with relatively high sea-surface temperatures, and therefore presumably with high evaporation rates on land, they are interpreted as the result of higher precipitation. Tropical aridity culminated in most areas at, or just after, the glacial maximum, although the present day is also characterized by a below-average abundance of surface water. In extratropical regions the mapped patterns are more complex. They vary markedly with latitude and proximity to major ice sheets. In these areas, evidence is at present insufficient to evaluate the relative contributions of precipitation and temperature to the observed lake-level record.

The Y-5 ash is the most widespread layer in deep-sea sediments from the eastern Mediterranean. This ash layer was previously correlated with the Citara-Serrara tuff on Ischia Island and dated at approximately 25,000 yr B.P. New data on the glass chemistry of the Y-5 ash and pyroclastic deposits from the Neopolitan volcanic province suggest that the layer is correlative with the large-volume Campanian ignimbrite and not with the deposit from Ischia Island. The volume of the Y-5 ash is approximately 65 km3 which is comparable in magnitude to the volume of the Campanian ignimbrite. An interpolated age of approximately 38,000 yr B.P. is estimated based on sedimentation rates derived from δ18O stratigraphy. There is a discrepancy between this estimate and previously reported radiocarbon ages which range from 24,000 to 35,000 yr B.P. We propose that the “Campanian tuff ash layer” should be adopted as the full stratigraphic name for the Y-5 ash. The deep-sea ash layer is divisible into two units in proximal localities, probably correlating with two major phases of the eruption: plinian and ignimbrite.

A quantitative analysis of radiolarian species in 57 deep-sea surface sediment samples from the South Atlantic Ocean produced four geographically distinct assemblages (tropical, polar, gyre margin, and subtropical). The distributions of these assemblages or factors coincide with present-day patterns of sea-surface temperatures or water masses.

These four assemblages were used to construct a transfer function relating radiolarian distribution in the surface sediments to present-day winter and summer temperatures using standard regression techniques. As a test of the quality of this transfer function, temperatures were estimated on surface sediment samples from the eastern South Pacific. The temperatures produced by the transfer function compared favorably with the observed (present-day) winter and summer sea-surface temperatures at these sites.

Temperature estimates produced by a radiolarian-based transfer function, factor distributions of radiolarian assemblages, and variations in calcium carbonate were used to reconstruct the oceanographic conditions in the South Atlantic during the last glacial maximum (18,000 yr B.P.). This study suggests that while the position of the Subtropical Convergence at 18,000 yr B.P. was very similar to its present position, the Antarctic Polar Front shifted northward 1° to 3° of latitude in the eastern South Atlantic and 3° to 5° of latitude in the western South Atlantic. The largest temperature changes occurred in the subantarctic region and along the eastern portion of the Subtropical Gyre.

Calculations based on temperature-corrected oxygen-isotope ratios from deep-sea cores yield a glacioeustatic sea-level fall in excess of 50 m during the first 10,000 yr of the last glaciation, and generally support the local regression of about 70 m inferred from tectonically rising New Guinea beaches. We propose that this rapid glacial buildup depended on high-latitude cooling, and large increases of high-latitude regional winter precipitation in the Laurentide and the Fennoscandian-Barents Sea areas, and that these factors were caused by a critical alteration of North Atlantic Drift currents and their associated subpolar atmospheric circulation. In support of this, faunal data from northeast North Atlantic deep-sea cores show that the glacial buildup was accompanied by a sudden loss of most of the North Atlantic Drift from the Greenland-Norwegian Sea, a factor favoring reduced heat input into the higher latitudes. Subpolar mollusk and foraminifera fauna from elevated marine deposits on the Baffin Island coast, and northwest North Atlantic core data suggest a continuation or an associated restoration of subpolar water west of Greenland as far north as Baffin Bay, a factor favoring precipitation in the northeast Canadian region. Heat transport and atmospheric circulation considerations suggest that the loss of the northeast North Atlantic Drift was itself a major instrument of high-latitude climate change, and probably marked the initiation of major new ice-sheet growth.